Curbside Consultation in Retina

Curbside
Consultation
in Retina
4 9 Clinica l Que s tions
Editor
S ha ro n Fe k ra t
Associate Editors
Da rius Mo s hfe ghi
De a n Elio t t
Series Editor
Da vid F. Cha ng
SLACK Incorporated
CURBSIDE CONSULTATION IN OPHTHALMOLOGY
SERIES
SERIES EDITOR, DAVID F. CHANG
EDITED BY
SHARON FEKRAT, MD, FACS
ASSOCIATE PROFESSOR, OPHTHALMOLOGY
DUKE UNIVERSITY MEDICAL CENTER
CHIEF, OPHTHALMOLOGY
DURHAM VETERANS AFFAIRS MEDICAL CENTER
DURHAM, NC
ASSOCIATE EDITORS
DARIUS MOSHFEGHI, MD
ASSOCIATE PROFESSOR, OPHTHALMOLOGY
DIRECTOR, VITREORETINAL SURGERY FELLOWSHIP PROGRAM
STANFORD UNIVERSITY SCHOOL OF MEDICINE
STANFORD, CA
DEAN ELIOTT, MD
PROFESSOR, OPHTHALMOLOGY
DIRECTOR, DOHENY RETINA INSTITUTE
KECK SCHOOL OF MEDICINE
UNIVERSITY OF SOUTHERN CALIFORNIA
LOS ANGELES, CA
www.slackbooks.com
ISBN : 978-1-55642-885-2
Cop yright © 2010 by SLACK Incorp orated
Cover p hotographs courtesy of Michael P. Kelly, CPT and Veronica Tom any, Du ke Eye Im aging, Du ke Eye
Center, Du rham , N C, USA.
All rights reserved . N o p art of this book m ay be reprod u ced , stored in a retrieval system or transm itted in any
form or by any m eans, electronic, m echanical, p hotocop ying, record ing or otherw ise, w ithou t w ritten p erm is-
sion from the p u blisher, excep t for brief qu otations em bod ied in critical articles and review s.
The p roced u res and p ractices d escribed in this book shou ld be im p lem ented in a m anner consistent w ith
the p rofessional stand ard s set for the circu m stances that ap p ly in each sp eci c situ ation. Every effort has
been m ad e to con rm the accu racy of the inform ation p resented and to correctly relate generally accep ted
p ractices. The au thors, ed itor, and pu blisher cannot accept responsibility for errors or exclu sions or for the
ou tcom e of the m aterial p resented herein. There is no exp ressed or im p lied w arranty of this book or inform a-
tion im p arted by it. Care has been taken to ensu re that d ru g selection and d osages are in accord ance w ith
cu rrently accep ted / recom m end ed p ractice. Du e to continu ing research, changes in governm ent p olicy and
regu lations, and various effects of d ru g reactions and interactions, it is recom m end ed that the read er carefu lly
review all m aterials and literature provid ed for each d ru g, especially those that are new or not frequ ently
u sed . Any review or m ention of speci c com panies or p rod u cts is not intend ed as an end orsem ent by the
author or p u blisher.
SLACK Incorp orated u ses a review p rocess to evalu ate su bm itted m aterial. Prior to p u blication, ed u cators or
clinicians p rovid e im p ortant feed back on the content that w e p u blish. We w elcom e feed back on this w ork.
Pu blished by: SLACK Incorp orated

6900 Grove Road
Thorofare, N J 08086 USA
Telephone: 856-848-1000
Fax: 856-853-5991
w w w.slackbooks.com
Contact SLACK Incorporated for m ore inform ation abou t other books in this eld or abou t the availability
of ou r books from d istribu tors ou tsid e the United States.
Library of Congress Cataloging-in-Pu blication Data
Cu rbsid e consultation in retina : 49 clinical questions / ed ited by Sharon Fekrat ; associate ed itors, Darius M.
Moshfeghi, Dean Eliott.
p . ; cm . -- (Cu rbsid e consu ltation in op hthalm ology)
Inclu d es bibliograp hical references and ind ex.
ISBN 978-1-55642-885-2 (alk. p ap er)
1. Retina--Diseases--Miscellanea. I. Fekrat, Sharon. II. Moshfeghi, Dariu s M. III. Eliott, Dean. IV. Series:
Cu rbsid e consultation in ophthalm ology series.
[DN LM: 1. Retinal Diseases--d iagnosis. 2. Retinal Diseases--therap y. WW 270 C975 2009]
RE551.C87 2009
617.7’35--d c22
2009027234
For p erm ission to rep rint m aterial in another p u blication, contact SLACK Incorp orated . Au thorization to
p hotocop y item s for internal, personal, or acad em ic u se is granted by SLACK Incorporated provid ed that the
app rop riate fee is p aid d irectly to Copyright Clearance Center. Prior to p hotocop ying item s, p lease contact the
Cop yright Clearance Center at 222 Rosew ood Drive, Danvers, MA 01923 USA; phone: 978-750-8400; w ebsite:
w w w.cop yright.com ; em ail: info@cop yright.com
Dedication
Th is book is ded icated to all ind ividu als w ith eye d isease all over the world. We hop e
that th is endeavor w ill help in the m anagement of the eye cond ition s of m illion s.
viii Contents
Question 13: How Do I Work-Up and Manage a Patient With a

Vitreou s Hemorrhage? ...................................................................................... 51
Pauline T. M errill, M D
Question 14: How Can Optical Coherence Tomography Help Me in My Practice
and What Can It Detect? .................................................................................. 55
Elias Reichel, M D
Question 15: With the Rapid ly Advancing Tech nology of Optical Coherence
Tomography, Is There Still a Role for Fluorescein Angiography? ............. 59
A sheesh Tewari, M D
Question 16: How Imp ortant Is the Prompt Diagnosis of Macu lar Hole? .......................61
William E. Smiddy, M D
Question 17: When Shou ld I Refer a Patient With an Epiretinal Mem brane and What
if There Is Associated Cystoid Macu lar Edem a? ............................................ 63
Gaurav K. Shah, M D
Question 18: How Do I Differentiate a Macu lar Hole From a Lam ellar Hole, an
Epiretinal Mem brane With a Pseudohole, Cystoid Macu lar Edem a,
and Vitreom acu lar Traction, and Why Do I Care? ...................................... 67
SriniVas R. Sadda, M D
Question 19: How Shou ld Patients With a Gas Bu bble Position Their Head
and What Precaution s Shou ld They Follow? ................................................ 71
A bdhish R. Bhavsar, M D
Question 20: What Is Central Serou s Retinopathy and How Do I Treat It? .................... 75
Jennifer I. Lim, M D
Question 21: What Is the Treatm ent Parad igm for Postoperative Cystoid
Macu lar Edem a? ................................................................................................ 79
Rishi P. Singh, M D
Question 22: Wh ich Patients With Un ilateral or Bilateral Dry Age-Related Macu lar
Degeneration Are at H ighest Risk for Developing Wet Age-Related
Macu lar Degeneration? ..................................................................................... 83
Karl G. Csaky, M D, PhD
Question 23: How Do I Cou n sel My Patients With Dry Age-Related Macu lar
Degeneration and What About Variou s Vitam in Supplements? ............... 87
Catherine B. M eyerle, M D
Question 24: What Drugs Can Be Injected Intravitreally and What Ocu lar and
System ic Complications Shou ld I Look For? ................................................... 91
Diana V. Do, M D
Question 25: Why Are There so Many Different Triamcinolone Preparation s and
Wh ich One Shou ld I Use? ................................................................................ 95
Sophie J. Bakri, M D
Question 26: How Do I Mon itor for Steroid-Induced Glaucom a After an
Intravitreal Injection? ........................................................................................ 97
Baruch D. Kuppermann, M D, PhD
Contents ix
Question 27: What System ic Med ications Requ ire Period ic Fu ndu s Evalu ation, What
Am I Looking For, and What Tests Do I Do? ...............................................101
William F. M ieler, M D
Question 28: How Shou ld I Follow Diabetic Macu lar Edem a and When Does It
Requ ire Treatm ent? ..........................................................................................107
Peter K. Kaiser, M D
Question 29: When Shou ld Eyes With Proliferative Diabetic Retinopathy Receive
Panretinal Laser Photocoagu lation? ...............................................................111
Ivan J. Suñer, M D
Question 30: When Shou ld I Refer a Patient With Diabetic Retinopathy for
a Vitrectomy? .....................................................................................................115
Ronald C. Gentile, M D
Question 31: How Do I Follow a Patient Who Has Diabetes and Becomes Pregnant?
What Tests Can I Do?....................................................................................... 123
Judy E. Kim, M D
Question 32: When Do I Refer a Patient With a Branch Retinal Artery Occlu sion
or Central Retinal Artery Occlu sion, What Is the Work-Up, and What
Are the Treatment Options?........................................................................... 127
Todd Klesert, M D, PhD
Question 33: What Do I Do When I See Som eone With a Branch Retinal Vein
Occlu sion and What Are the Treatment Option s? .....................................131
M ichael S. Ip, M D
Question 34: When Do I Refer a Patient With a Central Retinal Vein Occlu sion, What
Is the Work-Up, and What Are the Treatment Option s? ........................... 135
Richard Spaide, M D
Question 35: If I See a Patient With Shaken Baby Synd rome, What Are the
Next Steps?.........................................................................................................141
Christine R. Gonzales, M D
Question 36: What Typ e of Intraocu lar Len s Shou ld Be Con sidered in an Eye
With Vitreoretinal Disease? ...........................................................................145
Kourous A . Rezaei, M D
Question 37: When a Posterior Capsu lar Tear Occu rs Du ring Cataract Su rgery,
What Next?.........................................................................................................149
Timothy W. Olsen, M D
Question 38: When Shou ld a Patient With Dropped Cataract Material or Dislocated
Intraocu lar Lens Be Referred to a Retinal Sp ecialist? ................................153
Steven D. Schwartz, M D
Question 39: How Do I Manage a Suprachoroid al Hemorrhage? ....................................157
Odette M . Houghton, M D
Question 40: When Shou ld I Su spect Endophthalm itis in My Postoperative Cataract
Patient and What Are the Treatm ent Options?............................................161
Bernard H. Doft, M D
x Contents
Question 41: How Do I Follow a Patient With a Presu med Choroidal Nevu s? ...............165
Prithvi M ruthyunjaya, M D
Question 42: How Do I Distingu ish One Pigmented Lesion From Another? ................169
A my C. Schefler, M D
Question 43: How Do I Know if There Is a Retinal Detach m ent or Tu mor When I
Am Doing Ultrasonography?..........................................................................173
Cathy DiBernardo, CDOS, ROUB and James T. Handa, M D
Question 44: After a Closed Globe Inju ry, What Vitreoretinal Find ings Shou ld I
Expect and When Do I Refer?.........................................................................179
Esther M . Bowie, M D
Question 45: What Shou ld I Tell Patients With Retin itis Pigmentosa About Their
Prognosis and Shou ld I Prescribe Vitam in Supplements? ........................183
Raymond Iezzi, M D, M S
Question 46: Which Patients Shou ld I Refer for Electrophysiology Testing and What
Might the Test Show? ........................................................................................187
A mani A . Fawzi, M D
Question 47: How Do I Differentiate All of Those Wh ite Dot Synd romes? ...................191
Sunil K. Srivastava, M D
Question 48: How Do I Work-Up a Patient With Intermed iate Uveitis
or Panuveitis? ....................................................................................................197
Thomas A . A lbini, M D
Question 49: How Do I Differentiate the Infectiou s Posterior Uveitides And What Are
the Cu rrent Treatments? ...................................................................................201
JP Dunn, M D
About the Editors
Sharon Fekrat, MD, FACS joined the Vitreoretinal Su rgical Service at the Du ke Eye
Center in 1998 after completing her ophthalmology and vitreoretinal train ing at the
Wilm er Eye In stitute of the Joh ns Hopkins Med ical Institution s. She is now a tenu red
associate professor of ophthalmology at Du ke and the Ch ief of Ophthalmology at the
Du rham Veterans Affairs Med ical Center. She has co-authored over 60 peer-reviewed
articles and over 30 book chapters, is on the ed itorial board of fou r professional jou rnals,
and has served as site principal investigator for nu m erou s mu lticenter clin ical trials. She
is co-ed itor of Du ke Eye Center’s A ll A bout Your Eyes book for the lay pu blic. Dr. Fekrat is
past president of the North Carolina Society of Eye Physicians and Su rgeons.
Darius Moshfeghi, MD joined Stan ford Un iversity in 2002 after completing fellow-
sh ips in vitreoretinal su rgery at the Cole Eye In stitute of the Cleveland Clin ic and in
ocu lar oncology at the Un iversity of Ten nessee Memph is and St. Jude Ch ild ren’s Research
Hospital. He is an associate professor of ophthalmology where he lead s the ped iatric
vitreoretinal su rgery service and is Co-Director of the ocu lar oncology service. He is the
Fou nder and Director of the Stan ford Un iversity Network for Diagnosis of Retinopathy
of Prem atu rity (SUN DROP) telemed icine outreach in itiative, the largest telemed icine
network for ROP in the Un ited States. He has co-authored over 70 peer-reviewed, indexed
jou rnal articles and 20 book chapters, and serves on the ed itorial board of a professional
jou rnal. He is the Director of the Vitreoretinal Su rgery Fellow sh ip program at Stan ford
University.
D ean Eliott, MD is Director of the Doheny Retina In stitute and Professor of

Ophthalmology at the Doheny Eye Institute and Keck School of Med icine, Un iversity of
Southern Californ ia in Los Angeles, CA. Dr. Eliott completed h is residency at the Wilmer
Eye Institute, Joh n s Hopkin s Hospital, in Baltimore, MD, and he received h is vitreoretinal
fellow sh ip train ing at the Du ke Eye Center, Du ke Un iversity Med ical Center, in Du rham,
NC. He then served as Ch ief Resident and vitreoretinal facu lty mem ber at the Du ke Eye
Center. Before join ing the Doheny Eye Institute, Dr. Eliott was Director of the Retina
Service at the Kresge Eye Institute of Wayne State Un iversity in Detroit, MI. Dr. Eliott has
served as principal investigator for m any clin ical trials, and he has nu merou s pu blication s
in ophthalm ic jou rnals and textbooks.
Contributing Authors
Ron A fshari A delman, M D, M PH (Question Esther M . Bowie, M D (Question 44)
5) Assistant Professor, Ophthalmology
Associate Professor, Ophthalmology Storm Eye Institute
Department of Ophthalmology Med ical Un iversity of South Carolina
Yale Un iversity School of Med icine Charleston, SC
New H aven, CT
Karl G. Csaky, M D, PhD (Question 22)
A nita Agarwal, M D (Question 10) Associate Professor, Ophthalmology
Associate Professor, Ophthalmology Department of Ophthalmology
Retina, Vitreou s and Uveitis Du ke Un iversity
Vanderbilt Eye In stitute Du rham, NC
Nashville, TN
Cathy DiBernardo, CDOS, ROUB (Question
Thomas A . A lbini, M D (Question 48) 43)
Assistant Professor, Clin ical Associate Professor, Ophthalmology
Ophthalmology Wilm er Eye Institute
Bascom Palmer Eye Institute The Joh ns Hopkin s School of Med icine
Un iversity of Miam i, Miller School of Baltimore, MD
Med icine
Miam i, FL Diana V. Do, M D (Question 24)
Assistant Professor, Ophthalmology
Jorge G. A rroyo, M D, M PH (Question 11) Wilm er Eye Institute
Assistant Professor, Ophthalmology The Joh ns Hopkin s School of Med icine
H arvard Med ical School Baltimore, MD
Director, Retina Service
Division of Ophthalmology Bernard H. Doft, M D (Question 40)
Beth Israel Deaconess Med ical Center Retina Vitreou s Con su ltants
Boston, MA Clin ical Professor, Ophthalmology
Un iversity of Pittsbu rgh School of
Steven Bailey, M D (Question 2) Med icine
Assistant Professor Pittsbu rgh, PA
Casey Eye In stitute
Oregon Health and Science Un iversity JP Dunn, M D (Question 49)
Portland, OR Associate Professor, Ophthalmology
Director, Division of Ocu lar Im mu nology
Sophie J. Bakri, M D (Question 25) Wilm er Eye Institute
Associate Professor, Ophthalmology The Joh ns Hopkin s School of Med icine
Vitreoretinal Diseases and Su rgery Baltimore, MD
Mayo Clin ic
Rochester, MN A mani A . Fawzi, M D (Question 46)
A bdhish R. Bhavsar, M D (Question 19) Doheny Eye In stitute
Director, Clin ical Research Keck School of Med icine
Retina Center, PA Un iversity of Southern Californ ia
Attend ing Retina Su rgeon Los Angeles, CA
Director, Posterior Segment Research
Ph illips Eye Institute
Min neapolis, MN
xiv Contributing Authors
Karen M . Gehrs, M D (Question 7) Raymond Iezzi, M D, M S (Question 45)

Associate Professor, Clin ical Associate Professor, Ophthalmology
Ophthalmology College of Med icine, Mayo Clin ic
Un iversity of Iowa Rochester, MN
Iowa City, IA
M ichael S. Ip, M D (Question 33)
Ronald C. Gentile, M D (Question 30) Associate Professor
Associate Adju nct Professor, Un iversity of Wiscon sin
Ophthalmology Fu ndu s Photograph Read ing Center
New York Eye and Ear In firm ary Mad ison, WI
New York, N Y
Peter K. Kaiser, M D (Question 28)
Christine R. Gonzales, M D (Question 35) Professor, Ophthalmology
Retina and Vitreou s Center Cleveland Clin ic Lerner College of
Ash land, OR Med icine
Case Western Reserve Un iversity
Craig M . Greven, M D (Question 1) Vitreoretinal Staff Su rgeon
Richard G. Weaver Professor and Director, Digital OCT Read ing Center
Chairm an Cole Eye Institute
Department of Ophthalmology Cleveland, OH
Wake Forest Un iversity School of Med icine
Win ston-Salem, NC Judy E. Kim, M D (Question 31)
Professor, Ophthalmology
James T. Handa, M D (Question 43) Vitreoretinal Section
Robert Bond Welch Professor Med ical College of Wisconsin
Wilm er Eye Institute Milwau kee, WI
The Joh ns Hopkin s School of Med icine
Baltimore, MD Todd Klesert, M D, PhD (Question 32)
Assistant Professor
Odette M . Houghton, M D (Question 39) Department of Ophthalmology
Assistant Professor, Ophthalmology Un iversity of Min nesota
Department of Ophthalmology Min neapolis, MN
Un iversity of North Carolina
Chap el H ill, NC Baruch D. Kuppermann, M D, PhD (Question
26)
Suber S. Huang, M D, M BA (Question 4) Professor and Ch ief, Retina Service
Ph ilip F. and Elizabeth G. Searle - Su ber S. Vice Chair of Clin ical Research
Hu ang Professor and Vice-Chairm an Gavin Herbert Eye In stitute
Department of Ophthalmology and Visual Department of Ophthalmology
Sciences Un iversity of Californ ia, Irvine
Case Western Reserve Un iversity School of Irvine, CA
Med icine
Director, Center for Retina and Macu lar Linda A . Lam, M D (Question 8)
Disease Assistant Professor, Ophthalmology
Un iversity Hospitals Eye Institute Med ical Director, Doheny Eye Center
Un iversity Hospitals Case Med ical Center Riverside
Cleveland, OH Doheny Eye In stitute
Keck School of Med icine
Un iversity of Southern Californ ia
Los Angeles, CA
Contributing Authors xv
Jennifer I. Lim, M D (Question 20) Franco M . Recchia, M D (Question 3)

Professor, Ophthalmology Ch ief, Division of Vitreoretinal Diseases
Director, Retina Service and Su rgery
Eye and Ear In firm ary Associate Professor, Ophthalmology and
Un iversity of Illinois Visu al Sciences
Ch icago, IL Vanderbilt Eye In stitute
Nashville, TN
Pauline T. M errill, M D (Question 13)
Assistant Professor, Ophthalmology Elias Reichel, M D (Question 14)
Section Director, Uveitis and Ocu lar Associate Professor, Ophthalmology
In flam m ation Vice Chair and Director, Vitreoretinal
Ru sh Un iversity Department of Diseases and Su rgery Service
Ophthalmology Tu fts Un iversity School of Med icine
Illinois Retina Associates New England Eye Center
Ch icago, IL Boston, MA
Catherine B. M eyerle, M D (Question 23) Kourous A . Rezaei, M D (Question 36)

Retina Staff Clin ician Associate Professor, Ophthalmology
National Eye Institute Ru sh Un iversity Med ical Center
National In stitutes of Health Illinois Retina Associates, S.C.
Bethesd a, MD Ch icago, IL
William F. M ieler, M D (Question 27) SriniVas R. Sadda, M D (Question 18)

Professor and Vice-Chairm an Associate Professor, Ophthalmology
Department of Ophthalmology and Visual Keck School of Med icine
Sciences Un iversity of Southern Californ ia
Un iversity of Illinois at Ch icago Director of the Med ical Retina Un it,
Ch icago, IL Ophthalm ic Im aging Un it, and Doheny
Im age Read ing Center
A ndrew A . M oshfeghi, M D (Question 9) Doheny Eye In stitute
Bascom Palmer Eye Institute Los Angeles, CA
Miller School of Med icine
Un iversity of Miam i Steven Sanislo, M D (Question 6)
Miam i, FL Clin ical Professor, Ophthalmology
Stan ford Un iversity School of Med icine
Prithvi M ruthyunjaya, M D (Question 41) Stan ford, CA
Vitreoretinal Su rgery and Ocu lar A my C. Schefler, M D (Question 42)
Oncology Ch ief Resident, Vitreoretinal Su rgery
Director, Ocu lar Oncology Service Fellow
Du ke Eye Center Department of Ophthalmology
Du rham, NC Bascom Palmer Eye Institute
Miam i, FL
Timothy W. Olsen, M D (Question 37)
F. Ph in izy Calhou n Sr. Professor
Chairm an, Emory Eye Center
Emory Un iversity
Atlanta, GA
xvi Contributing Authors
Steven D. Schwartz, M D (Question 38) William E. Smiddy, M D (Question 16)

Ah m an son Professor, Ophthalmology Professor, Ophthalmology
Ch ief, Retina Division Bascom Palmer Eye Institute
Ju les Stein Eye Institute Miller School of Med icine
Associate Professor of Ophthalmology Un iversity of Miam i
David Geffen School of Med icine Miam i, FL
Un iversity of Californ ia
Los Angeles, CA Richard Spaide, M D (Question 34)
Vitreou s, Retina, Macu la Consu ltants of
Gaurav K. Shah, M D (Question 17) New York
Barnes Retina In stitute New York, N Y
Clin ical Associate Professor,
Ophthalmology and Visual Sciences Sunil K. Srivastava, M D (Question 47)
Department of Ophthalmology and Visual Assistant Professor, Ophthalmology
Sciences Emory Un iversity Eye Center
Wash ington Un iversity School of Med icine Atlanta, GA
St. Lou is, MO
Ivan J. Suñer, M D (Question 29)
Christopher N. Singh, M D (Question 12) Retina Associates of Florid a, PA
Retina Associates, PA Tampa, FL
Assistant Clin ical Professor,
Ophthalmology A sheesh Tewari, M D (Question 15)
Un iversity of Kansas Med ical Center Assistant Professor, Ophthalmology
Kansas City, KS Kresge Eye In stitute
Wayne State Un iversity School of Med icine
Rishi P. Singh, M D (Question 21) Detroit, MI
Associate Staff Physician
Cole Eye Institute
Cleveland Clin ic Fou nd ation
Cleveland, OH
Preface
Are you looking for concise, practical an swers to those questions that are often left
u nanswered by trad itional references of the retina? Are you seeking brief, evidence-based
advice for the d aily exam ination of patients? Curbside Consultation in Retina: 49 Clinical
Questions provides qu ick and d irect an swers to question s most com mon ly posed du ring
a “cu rbside consu ltation” between experienced clin ician s. We have designed th is u n ique
reference that offers exp ert advice, preferences, and opin ion s on a variety of clin ical ques-
tions com mon ly associated w ith a variety of vitreoretinal clin ical scenarios. The u n ique
question-and-an swer form at provides qu ick access to cu rrent in form ation related to the
retina w ith the simplicity of a conversation between two colleagues. Im ages and refer-
ences are included to en hance the text and to illu strate clin ical d iagnoses as well as ideas
w ith in the text. Curbside Consultation in Retina: 49 Clinical Questions provides in form ation
basic enough for residents wh ile also incorporating expert pearls that even h igh-volu me
ophthalmologists and specialists w ill appreciate. Optometrists, ophthalmologists, oph-
thalmologists-in-train ing, and even retina specialists w ill benefit from the u ser-friend ly
and casu al form at and the exp ert advice contained w ith in.
Foreword
When we were residents, how great it was to be able to grab a facu lty m em ber by the
sleeve in the hallway and ru n a d ifficu lt case m anagement issue by h im or her. All of u s
know how valu able these types of “cu rbside con su lts” can be, and how they often pu sh
ou r ow n thoughts and actions ju st that extra step fu rther, wh ich resu lts in the best pos-
sible outcome for the patient.
Curbside Consultation in Retina is a type of virtu al re-creation of that collegial environ-
m ent in wh ich we all trained, and that wealth of experience and knowledge that we cou ld
tap into at any tim e. It is a resou rce that allow s question s to be answered by seasoned,
well-resp ected clin icians in a focu sed, practical way. Fekrat and colleagues have assem-
bled a knowledgeable, articu late, and experienced group of contributors and have chosen
clear-cut, specific topics that con front the bu sy clin ician every d ay and p ose sign ificant,
vision-th reaten ing d ilem m as.
One caveat: in The Devil’s Dictionary,1 Am brose Bierce cyn ically defined consult as “to
seek approval for a cou rse of action already decided up on.” The best approach to th is
book, as to clin ical d ilem m as, is an intelligent, alert, and above all, open-m inded one.
Ophthalmologists grapple not on ly w ith tod ay’s signs and symptom s, but w ith their con-
sequences: tomorrow ’s vision and qu ality of life. Ou r approach to th is challenge can be
gu ided by Sir William Osler’s word s, “The best preparation for tomorrow is to do tod ay’s
work sup erbly well.”2 Th is book is an outstand ing add ition to the resou rces that supp ort
u s d ay to d ay in ou r determ ination to live up to th is h igh stand ard.
REFERENCES
1. Bierce A. The devil’s dictionary. London, England: Bloomsbury Publishing; 2004:41.
2. O sler W. The best preparation for tomorrow is to do today’s work superbly well. As quoted in Braude JM.
Lifetim e speaker’s encyclopedia. Englewood Cliffs, NJ: Prentice Hall; 1962:575.
Julia A . Haller, M D
Ophthalmologist-in-Ch ief
Wills Eye In stitute
Professor and Chair of Ophthalmology
Jefferson Med ical College of Thom as Jefferson Un iversity
Ph iladelph ia, PA
Introduction
Th is book was put together ju st for you. Yup, ju st for you. Everyth ing you wanted to
know about 49 com mon ly asked retina questions in an easy form at—ju st as if you called
a retina colleague on the phone and said “Hey, what shou ld I do when I see a patient w ith
one or more cotton wool spots?” or “How do I cou nsel my patients w ith d ry age-related
m acu lar degeneration, and what about variou s vitam in supplements?” or even better,
“How do I d ifferentiate all of those wh ite dot synd romes?” Curbside Consultation in Retina
is exactly what it says it is: cu rbside consu lts for ou r colleagues—those in m ed ical school,
ophthalmology residency train ing, subspecialty fellow sh ip train ing, optometry, ophthal-
mology, and even for retina sp ecialists them selves.
As co-ed itors of Curbside Consultation in Retina, Dariu s, Dean, and I have carefu lly cho-
sen the questions to cover as m any possible likely com mon clin ical scenarios that need
answers on a d ay-to-d ay basis. What is sp ecial about th is book is that of the 49 questions,
there are 50 d ifferent authors who have done their vitreoretinal train ing in d ifferent
institutions and who practice across the Un ited States. Th is book does not represent the
clin ical practice or opin ion of ju st one institution or group. Many authors are well-know n
experts in the specific area that is covered by the particu lar an swer they authored.
Curbside Consultation not on ly provides clin ical opin ion regard ing the an swers to d ifferent
d ay-to-d ay questions at you r fingertips but offers an appropriate depth, and then some,
gained from a “cu rbside con su lt.”
As we all know, Curbside Consultation in Retina is not intended to su bstitute for appropri-
ate vitreoretinal con su ltation. Instead, we hope that it w ill provide su fficient in form ation
that is read ily available to gu ide you r decision m aking in the com mon clin ical scenarios
d iscu ssed herein and prompt an appropriate referral when needed.
Tu rn the pages and let the book d raw you in.
Sharon Fekrat, M D, FACS

1
QUESTION
WHAT ARE SOME TIPS TO

DIFFERENTIATE VISUAL LOSS FROM
RETINAL DISEASE VERSUS CATARACT?
Craig M. Greven, MD
When encou ntering a patient w ith a cataract and visual loss, it m ay be challenging to
determ ine if the lens opacity is the prim ary reason for the visu al symptom s or if retinal
pathology is the etiology of the visual loss. When I am faced w ith th is clin ical d ilem m a,
I u se a stand ard, simple, cost-effective algorith m .
History
The first and most important tool in ou r arm amentariu m is a thorough h istory.
Although the h istory is often u nderemphasized in ou r bu sy clin ics, I find that I can
determ ine from the h istory alone—in the m ajority of cases—whether retinal d isease or
cataract is cau sing the loss of vision. The tim ing of visu al loss can be an important clue.
Acute loss of vision is almost never due to cataract, wh ile ch ron ic visu al loss can be due to
either entity. Fluctu ating visu al loss is more characteristic of retinal d isease than cataract.
The natu re of the visu al complaint m ay also help to d ifferentiate. A gradual blu rring or
d im m ing of the vision is more often associated w ith cataract, wh ile symptom s such as
m etamorphopsia, d istortion, or scotom a are more com mon in eyes w ith retinal d isease.
The circu m stance of blu rred vision m ay also be helpfu l. As an example, d ifficu lty d riving
at n ight or problem s w ith glare from head lights is typically reported by a cataract patient
and less com mon ly in those w ith retinal d isease.
As part of a thorough h istory, past med ical and ocu lar h istories are important and easy
to determ ine. A h istory of long-stand ing d iabetes, renal d isease, card iac d isease, or cere-
bral vascu lar d isease is more likely to lead to posterior segment pathology. Remem ber, a
1
2 Question 1
good h istory can be taken in ju st a few m inutes, costs noth ing extra, and is invalu able in
sorting out th is issue.
Clinical Examination
In tod ay’s bu sy ophthalmology clin ics, visu al acu ity is most often recorded by an oph-
thalm ic tech n ician and not the physician. Th is part of the exam ination often gives impor-
tant in form ation, and the tech n ician shou ld be instructed to report any id iosyncrasies
the patient m ay have when p erform ing Snellen acu ity testing. For instance, some patients
m ay on ly read the first or last few letters of the line, ind icative of a neu ro-ophthalm ic or
retinal d isease. In add ition, patients who move their head from side to side are likely to
have retinal d isease, as they are trying to look arou nd a scotom a. Improvement in vision
w ith pin hole testing and a more myopic refraction are more characteristic of cataract
progression.
Carefu l slit-lamp exam ination w ith d ilated pupils is the best way to exam ine for the
presence and degree of cataract. Th is is important becau se cond itions like oil d roplet
cataract and sm all p osterior subcapsu lar cataracts w ill be m issed w ithout d ilation. In my
experience, even w ith a d ilated pupil, the so-called “oil d roplet” cataract can be m issed.
The best way to see th is type of cataract is w ith the d irect ophthalmoscope (w ith the pupil
w idely d ilated), wh ich w ill reveal the den sely colored nuclear d isc again st the red reflex.
Add itionally, retinoscopy w ill show atypical scissoring. Wh ile cortical cataracts rarely
cau se sign ificant visu al acu ity loss, they m ay cau se symptom s when the cortical sp okes
extend th rough the visual axis.
Fundus Examination
Certain ly one important h int for the clin ician du ring exam ination of the fu ndu s is
whether an adequate view of the posterior p ole is p ossible. If not, then med ia opacity,
either in the crystalline lens or vitreou s, is obstructing the view of the retina. Wh ile it is
relatively easy to determ ine the cau se of vision loss in eyes w ith cataract and a bu llou s
retinal detach ment, a large mu sh room-shaped choroid al melanom a, or an ischem ic cen-
tral retinal vein occlu sion, a nu m ber of su btle retinal cond itions can on ly be detected by
carefu l, m eticu lou s fu ndu s exam ination. Exam ination of the posterior pole w ith binocu-
lar m agn ification w ith either a 78- or 90-D len s, or even a fu ndu s contact len s, is crucial
to ru le out these su btle cond ition s. Microangiopathy of the perifoveal capillary bed in
patients w ith d iabetes, sm all branch retinal vein occlu sion, or less com mon cond itions
such as ju xtafoveal telangiectasia can be seen w ith th is tech n ique. Macu lar d ru sen are a
sign of age-related m acu lar degeneration (AMD) and associated visual loss can occu r due
to either central geograph ic atrophy or occu lt choroid al neovascu larization w ith m in im al
su bretinal flu id. Subtle epiretinal mem branes (ERM) and vitreom acu lar traction (VMT)
m ay be seen on carefu l exam ination. I w ill often ask the patient to describe the slit beam of
light that I am u sing to view the fu ndu s. If he or she answers that it is tilted or has defects
or breaks in it, th is ind icates m acu lopathy as a cau se of the visual loss.
What Are Some Tips to Differentiate Visual Loss From Retinal Disease? 3
Ancillary Testing
Even in the hand s (and eyes) of skilled clin icians, h istory and exam alone can not d iffer-
entiate visual loss from cataract or retinal d isease in all cases. Wh ile nu merou s ancillary
tests are available to facilitate solving the problem, the potential acu ity m eter (PAM) and
optical coherence tomography (OCT) are simple, relatively cheap, non invasive stud ies that
can be of great value.
The PAM projects a lighted visual acu ity chart on the retina th rough any relatively
clear area in the cataract.1 Wh ile u sefu l, the PAM can either over- or u nderestim ate final
visu al outcomes depend ing on the degree of cataract or posterior segment pathology.2
OCT has revolution ized the d iagnosis and treatment of vitreoretinal d iseases, and it is
the best tool to detect su btle m acu lar d iseases like VMT, ERM, early m acu lar hole form a-
tion, and central geograph ic atrophy in AMD. Wh ile often su btle clin ically, these entities
are typically read ily apparent on OCT. Although cataract can attenu ate signal strength,
OCT is still reliable for gross interpretation of retinal morphology.3
What I Tell the Patient

My favorite su m m ation w ith these patients is, “It is impossible to tell whether you r
visu al loss is 90% from cataract and 10% from retina, 50/ 50, or 10/ 90. We can on ly m ake
educated estim ates.” I go on to say that perform ing cataract extraction is the on ly way
to defin itively determ ine the cau se. Th is allow s improved visu alization of the retina to
better assess and potentially treat retinal problem s. In the best interest of you and you r
patients, express you r degree of u ncertainty. It is better to do so at the time of you r evalu-
ation than after 1 or 2 su bsequent su rgeries w ith poor visu al outcomes. A second opin ion
w ith another retinal or cataract su rgeon is always a reasonable option.
References
1. Guyton DL. Instruments for measuring retinal visual acuity behind cataract. O phthalm ology. 1982;89 (8S):98-
103.
2. Uy HS, Munoz VM. Comparison of the potential acuity meter and pinhole tests in predicting postoperative
visual acuity after cataract surgery. J Cataract Refract Surg. 2005;31(3):548-552.
3. van Velthoven ME, van der Linden MH, de Smet MD, Faber DJ, Verbraak FD. Influence of cataract on optical
coherence tomography image quality and retinal thickness. Br J O phthalm ol. 2006;90 (10):1259-1262.
2
QUESTION
WHAT RETINAL FINDINGS WOULD

REQUIRE TREATMENT PRIOR TO CATARACT
OR REFRACTIVE SURGERY?
Steven Bailey, MD
A complete d ilated fu ndu s exam ination shou ld be perform ed prior to cataract or

refractive su rgery. Carefu l in spection of the posterior pole is important to evaluate if
there is any m acu lar d isease cau sing or contributing to visu al dysfu nction. Attention to
the statu s of the posterior hyaloid (detached, attached, or partially detached) is imp ortant.
Peripheral exam ination m ay identify asymptom atic lesions that cou ld becom e visu ally
th reaten ing such as a retinal break or other lesion s of concern, such as a choroid al mela-
nom a.
Retinal Detachment Prevention

Posterior vitreou s detach m ent (PVD) is the principal pred isposing event lead ing to
a rhegm atogenou s retinal detach m ent. Cataract su rgery lead s to vitreou s mod ification
and liquefaction and often a PVD. Th is is the likely mechan ism accou nting for the sig-
n ificantly h igher rates of retinal detach m ent follow ing cataract su rgery. Risk factors for
retinal detach ment follow ing cataract su rgery are vitreou s loss, PVD, presence of lattice
degeneration, posterior capsu lotomy, you ng age, m ale gender, h igh myopia, and retinal
detach m ent in the fellow eye. The retinal detach ment rate follow ing non intraocu lar
refractive su rgery is not h igher than that for the myopic popu lation; however, there are
reports of retinal detach m ents occu rring im med iately follow ing laser in situ keratom i-
leu sis (LASIK).1
There are a nu m ber of peripheral retinal find ings that do not requ ire prophylac-
tic treatment, includ ing sen ile retinosch isis, op ercu lated retinal holes, cystic retinal
tu fts, zonu lar traction tu ft, merid ional fold, enclosed ora bay, and peripheral cystoid
5
6 Question 2
degeneration. Lattice degeneration has been associated w ith 22% to 30% of retinal detach-
ments; however, the risk of lattice degeneration developing into a retinal detach m ent is
on ly 0.3% to 0.5%.2 There is no evidence that prophylactic treatm ent of lattice degenera-
tion in asymptom atic patients prior to cataract su rgery (regard less of vitreou s statu s) is
beneficial.
One m ay con sider prophylactic treatment of lattice degeneration prior to cataract su r-
gery in a patient w ith concu rrent risk factors, such as h istory of retinal detach m ent in the
fellow eye. However, it has been observed that in eyes w ith lattice degeneration and a h is-
tory of a retinal detach ment in the other eye that go on to develop a retinal detach ment,
about 30% of retinal breaks resp onsible for the retinal detach ment develop in otherw ise
healthy retinas.3 Patients w ith mu ltiple risk factors for retinal detach ment shou ld be fol-
lowed closely after cataract su rgery, especially if a PVD develop s.
Consideration shou ld be given to treating asymptom atic retinal breaks, especially if
vitreoretinal traction is evident. If sign s of ch ron icity, such as pigmentary changes, are
present, observation is reasonable. All symptom atic retinal breaks (whether associated
w ith lattice degeneration or not) shou ld be treated and followed for several months prior
to refractive or cataract su rgery.
Diabetic Retinopathy
Those patients w ithout m acu lar edem a and w ith m in im al d iabetic retinopathy are more
likely to ach ieve visu al acu ity of 20/ 40 or better follow ing cataract su rgery compared to
patients w ith more advanced d iabetic retinopathy. The severity of retinopathy at the time
of len s removal is the most pred ictive factor of poor visu al acu ity outcom e. Early pan reti-
nal photocoagu lation for severe nonproliferative and “low-risk” proliferative d iabetic reti-
nopathy prior to lensectomy d id not yield sign ificant resu lts in improved visual acu ities
assessed as part of the Early Treatment Diabetic Retinopathy Study. However, there was
a trend toward better visu al acu ity in the photocoagu lation group, and one m ay consider
panretinal photocoagu lation for severe nonproliferative d iabetic retinopathy prior to cata-
ract su rgery.4 Macu lar edem a shou ld be treated and stabilized prior to cataract su rgery
to prevent worsen ing of m acu lar edem a and associated vision loss. Proliferative d iabetic
retinopathy shou ld be treated and qu iescent prior to cataract su rgery to lower the risk of
severe p ostoperative complication s.
If the cataract prevents treatment of d iabetic m acu lar edem a or proliferative d iabetic
retinopathy, then cataract su rgery followed by prompt treatm ent for d iabetic pathology
is appropriate. In th is setting, the role of adju nctive pharm acotherapy (steroid s or anti-
VEGF agent) near the time of cataract su rgery is cu rrently u nder investigation and can be
considered.
Macular Degeneration
Many patients w ith cataracts m ay also have some degree of m acu lar degeneration.
Any sudden change in vision or new metamorphopsia warrants a d ilated m acu lar exam
w ith a 78- or 90-D lens. Fu rther investigation w ith fluorescein angiography and optical
What Retinal Findings Would Require Treatment Prior to Cataract Surgery? 7
coherence tomography (OCT) can help detect evidence of choroid al neovascu larization
in the presence of sign ificant med ia opacity. Modern treatm ent for neovascu lar m acu lar
degeneration shou ld be in itiated prior to cataract su rgery to ensu re treatment is provided
in a timely fash ion. There is no level I evidence that cataract su rgery is a risk factor for the
development of advanced m acu lar degeneration.
Epiretinal Membrane and Macular Hole

Su btle epiretinal mem branes and m acu lar holes can be d ifficu lt to detect despite care-
fu l fu ndu s exam . Preoperative OCT shou ld be considered if the visu al acu ity is worse
than wou ld be expected from the cataract alone. The p otential benefit of cataract su rgery
in th is setting mu st be determ ined prior to cataract su rgery. Macu lar holes shou ld be
repaired w ith in 6 months of on set if p ossible. If sign ificant m ed ia opacity is evident, cata-
ract su rgery m ay be completed first to improve visualization before vitrectomy, as long as
both su rgeries can be done in a timely m an ner. Phacoemu lsification and lens implanta-
tion com bined w ith vitrectomy for either epiretinal mem brane or m acu lar hole su rgery is
another option.
Acknowledgment of Significant Collaboration

Ch ristina J. Flaxel, MD
Associate Professor, Casey Eye In stitute
Oregon H ealth and Science Un iversity, Portland, OR
References
1. Arevalo JF. Posterior segment complications after laser-assisted in situ keratomileusis. Curr O pin O phthalm ol.
2008;19 (3):177-184.
2. Byer N. Changes in and prognosis of lattice degeneration of the retina. Trans Am Acad O phthalm ol
O tolaryngol. 1974;78 (2):O P114-O P125.
3. Folk JC, Bennett SR, Klugman MR, Arrindell EL, Boldt HC. Prophylactic treatment to the fellow eye of patients
with phakic lattice retinal detachment: analysis of failures and risks of treatment. Retina. 1990 ;10 (3):165-169.
4. Chew EY, Benson WE, Remaley NA, et al. Results after lens extraction in patients with diabetic retinopathy:
early treatment diabetic retinopathy study report number 25. Arch O phthalm ol. 1999;117(12):1600 -1606.
3
QUESTION
WHAT SHOULD I TELL A MYOPE

ABOUT THE RISK OF RETINAL
DETACHMENT AFTER CATARACT
OR REFRACTIVE SURGERY?
Franco M. Recchia, MD
The incidence of rhegm atogenou s retinal detach m ent (RRD) is estim ated to be 1 in
10,000 (0.01%). Increased risk for RRD is associated w ith both ind ividu al and operative
factors.1 The prim ary risk from cataract su rgery is due to the anterior movem ent of the
vitreou s, wh ich accelerates posterior vitreou s detach m ent (PVD) and increases p eriph-
eral traction at the vitreou s base.2 Both m echan ism s m ay lead to the form ation of retinal
breaks and su bsequent RRD. With PVD, up to 15% of eyes m ay have retinal breaks.
Ind ividu al risk factors include m ale gender, you nger age, h igh myopia (axial length
more than 25 m m or myopia >-5 D), and h istory of RRD in the fellow eye. Eyes w ith axial
length s greater than 25 m m have at least a 6-fold increased risk of RRD; the longer the
axial length, the h igher the risk.
Intraoperative risk factors are vitreou s loss and posterior capsu le d isruption. Vitreou s
loss afford s a 5-fold risk of RRD (~5% w ith in 4 years of cataract su rgery). Posterior capsu-
lar d isruption m ay increase the risk of p ostop erative RRD 13-fold. With YAG capsu lotomy
for p ostop erative second ary cataract, there can be a 4-fold increase in RRD.1
Risk also varies by type of cataract su rgery perform ed. The incidence of RRD was
h igher w ith intracapsu lar cataract extraction. After phacoemu lsification, the average inci-
dence is rep orted to be 1.7% but approaches 10% in eyes w ith very h igh myopia (> -15 D).
With clear lens extraction (the removal of the clear crystalline lens and implantation of an
intraocu lar len s to correct refractive error), the risk of RRD ranges from 0% to 8.1%.
Wh ile refractive su rgery is not intraocu lar, procedu res such as laser-assisted in situ
keratom ileu sis (LASIK) can still alter the vitreoretinal interface, thu s increasing the risk
9
10 Question 3
of RRD. One specu lated mechan ism is movem ent of the vitreou s, specifically by com-
pression/ decompression of the globe in the anteroposterior axis when u sing suction. No
study has proven a cau sative association between LASIK and RRD. Such a relationsh ip
wou ld be d ifficu lt to prove since the patients most at risk for RRD (h igh myopes) are the
patients most com mon ly u ndergoing LASIK. However, the largest study in the literatu re
estim ates a risk of 0.06% of RRD after LASIK (approxim ately 3-fold h igher than expected
in a comparable cohort). These eyes d id not have prior retinal pathology requ iring pro-
phylactic retinal treatment. Their pre-LASIK myopia ranged from -1.50 D to -16 D, w ith
an average of -7 D.3
Preop erative retinopexy w ith either laser or cryotherapy to retinal lesions pred ispos-
ing to retinal breaks m ight decrease the risk of RRD, but its benefit has not been proven
conclu sively. It m ay be considered in eyes of patients w ith poor visu al acu ity in the fellow
eye or for patients in whom postoperative follow-up or cooperation is lim ited.
A special clin ical scenario seen w ith grow ing frequency involves the patient born
prem atu rely or who had retinopathy of prem atu rity (ROP). The synd rom e of “adu lt
ROP” includes early-onset cataracts and associated rhegm atogen ic retinal pathology. You
shou ld ask all patients if they were born early (< 32 weeks of gestational age) or if they
were told that they were born w ith a low birthweight. Many patients w ith a h istory of
prem atu re birth or ROP have a myopic refraction. Among these patients, the rep orted risk
of RRD after cataract su rgery approaches 25%. These patients m ay or m ay not have pos-
terior cicatricial changes.4 Visu al acu ity m ay not improve after cataract su rgery becau se
of am blyopia or m acu lar ectopia, and retinal detach m ents are more com mon becau se of
complex vitreoretinal abnorm alities. If cataract su rgery is u ndertaken, the patient shou ld
have close follow-up w ith a retina sp ecialist postoperatively.
In the general popu lation, the overall incidence of RRD in the fellow eye follow ing
RRD in one eye is approxim ately 10%. Th is sign ificant risk means that len s statu s is not
completely protective against retinal detach ment. Moreover, the rate appears to be even
h igher among nonphakic fellow eyes.5 The risk of RRD in fellow eyes can be explained
by the fact that fellow eyes often have sim ilar retinal pathology, such as h igh myopia and
lattice degeneration, as well as probable genetic pred isposition.
All patients shou ld have a thorough d ilated fu ndu scopic exam prior to cataract or
refractive su rgery in order to identify acute retinal breaks requ iring treatm ent or other
pathology that m ay pred ispose to p ostoperative RRD. We tell all patients, not ju st myopes,
that wh ile there is always a risk of RRD, the anatom ic success rate of retinal reattach m ent
is h igh and that visu al acu ity m ay rem ain excellent after prompt repair if the m acu la is
not affected.
Patients shou ld be cou nseled to look for symptom s of increasing floaters, flashes of
light, decreased vision, or defects in their visu al field s. Wh ile these symptom s are most
frequent w ith in 6 to 12 months of cataract extraction, LASIK, or YAG capsu lotomy, they
can occu r at any tim e, and regu lar long-term exam ination is recom m ended. We advise
patients to seek ophthalmologic care as soon as these symptom s occu r in order to opti-
m ize preservation of visu al acu ity.
What Should I Tell a Myope About the Risk of Retinal Detachment? 11

Jan ice C. Law, MD
Assistant Professor, Ophthalmology and Visu al Sciences
Associate Director, Resid ent Education
Vand erbilt Eye In stitute, Nashville, TN
References
1. Lois N, Wong D. Pseudophakic retinal detachment. Surv O phthalm ol. 2003;48 (5):467-487.
2. Jaffe NS, Light DS. Vitreous changes produced by cataract surgery: a study of 1,058 aphakic eyes. Arch
O phthalm ol. 1966;76(4):541-553.
3. Arevalo JF, Ramirez E, Suarez E, et al. Rhegmatogenous retinal detachment in myopic eyes after laser in situ
keratomileusis: frequency, characteristics, and mechanism. J Cataract Refract Surg. 2001;27(5):674-680.
4. Kaiser RS, Fenton GL, Tasman W, Trese MT. Adult retinopathy of prematurity: retinal complications from
cataract surgery. Am J O phthalm ol. 2008;145(4):729-735.
5. Gupta O P, Benson WE. The risk of fellow eyes in patients with rhegmatogenous retinal detachment. Curr O pin
O phthalm ol. 2005;16(3):175-178.
4
QUESTION
HOW DO I MANAGE
LATTICE DEGENERATION?
Suber S. Huang, MD, MBA
Lattice degeneration is a well-know n and important risk factor for the development
of a retinal break and detach m ent. Lattice degeneration u sually occu rs in one or more
circu m ferential row s between the ora serrata and the equ atorial region of the retina. The
“lattice work” appearance is cau sed by the erosion of the internal lim iting mem brane
and by the presence of wh itened blood vessels overlying the retinal pigment epitheliu m
(RPE). Glial cell proliferation is a characteristic that m ay involve the retinal su rface and
extend to the posterior hyaloid of the vitreou s. Increased RPE pigm entation is com mon
and is also seen in associated cond itions of perivascu lar lattice, Stickler synd rome, and
Wagner synd rom e.
Natural History
Longitud inal stud ies demon strate that lattice degeneration can start in early adu lthood
but does not have a well-defined tendency to increase over time. Lattice degeneration is
present in approxim ately 5% to 10% of the general popu lation, and it is bilateral in about
one-th ird of affected ind ividu als. There is no predom inant pattern of genetic in heritance.
However, the prevalence of lattice degeneration in first-degree relatives of patients is esti-
m ated at 23%—nearly 3 times that of the general popu lation.1
The m ajority of lattice lesion s are u su ally stable in appearance, and the development
of new lattice lesion s has not been correlated w ith age or lesion du ration. Although lat-
tice degeneration is present in approxim ately 30% of patients w ith retinal detach ment
(RD), the prevalence of RD for an ind ividu al w ith lattice is low. It is estim ated that the
lifetime risk of developing RD in an em m etropic patient w ith lattice degeneration is 0.3%
to 1.7%.1
13
14 Question 4
Risk Factors for Retinal Detachment

The developm ent of atroph ic holes and the presence of traction tears arising from vit-
reoretinal ad hesion s are associated w ith du ration of d isease and constitute the m ajor risk
factors for RD. Retinal tears associated w ith lattice degeneration occu r at the m argins of
lattice degeneration where there is a weakened chorioretinal ad hesion and an en hanced
vitreoretinal ad hesion. Byer estim ates that 65% of retinal holes app ear before the age of 35
years.1 New atroph ic holes occu r in 12.7% of patients w ith long-term follow-up.1 However,
it has also been noted that m any retinal tears that occu r in lattice degeneration develop in
previou sly norm al-app earing retina.1
In Byer’s series, atroph ic rou nd retinal holes were associated w ith su bclin ical RD in
on ly 6.7% of eyes.1 No patients progressed to clin ical detach m ents. RDs associated w ith
an atroph ic rou nd hole u sually occu r in patients aged 30 years or you nger.1 RD due to
atroph ic retinal holes is seldom associated w ith posterior vitreou s detach ment and is most
frequently characterized by localized “seeping” en largement of areas of su bretinal flu id.
Tractional tears most often occu r at the m argin s of the lattice degeneration. These m ay
lead to true horseshoe tears and are most often associated w ith vitreou s syneresis and
increased vitreoretinal traction. It has been suggested that the greatest risk for progres-
sion of RD occu rs w ith in 6 weeks after a symptom atic tractional break appears.
The risk of RD is sign ificantly increased in patients w ith lattice degeneration when
h igh myopia is present. The lifetim e risk for RD is 0.3% in an em metropic eye w ithout
lattice degeneration, 1.7% in an em metropic eye w ith lattice degeneration, 2.2% in patients
w ith h igh myopia, and 35.9% in eyes w ith both h igh myopia and lattice degeneration.1,2
Prophylactic Treatment of Lattice Degeneration

Natu ral h istory stud ies have been critical in the u nderstand ing of the relative risk of
RD and the need for prophylactic treatment. The low incidence of RD associated w ith
lattice degeneration blu nts the need for prophylactic treatm ent. Prophylactic retinopexy
is not ind icated for patients w ith lattice degeneration w ith asymptom atic atroph ic holes
u nless subclin ical RD is present or if the patient has h igh myopia. If flu id is progressive
or present greater than 1 d isc d iameter su rrou nd ing an atroph ic hole, prophylactic treat-
m ent shou ld be considered. Symptom atic tears have a h igh risk of progression to RD.
Prophylactic treatm ent is recom mended for all acutely symptom atic retinal tears in pha-
kic, pseudophakic, and aphakic eyes.
The fellow eye has had sign ificant study to determ ine the risks and benefits of pro-
phylactic treatment. In a series of u ntreated fellow eyes w ith lattice degeneration and lat-
tice-associated RD in the first eye, the risk of RD is approxim ately 5.1%.3 In patients w ith
myopia and lattice degeneration, the risk is as h igh as 24.6%.3 The overall rate of RD in
the fellow eye is reduced by prophylactic treatment from 5.1% to 1.8% over a 7-year p eriod,
but no beneficial effect was noted in subgroups w ith greater than 6 clock hou rs of lattice
degeneration or 6 d iopters or more of myopia.4 Prophylactic therapy d id not cau se new
retinal breaks or d etach ment, and it was fou nd that 29% of new tears occu rred in areas
not affected by visible lattice degeneration.3,4
How Do I Manage Lattice Degeneration? 15
Summary
Prophylactic therapy is recom mended in symptom atic eyes w ith lattice degeneration,
retinal break associated w ith lattice degeneration, progressive subretinal flu id associated
w ith retinal holes, and in patients w ith h igh myopia and lattice degeneration. Treatment
of the fellow eye in patients w ith h igh myopia and extensive lattice degeneration rem ains
controversial. The risks, benefits, and alternatives of observation versu s treatment shou ld
be d iscu ssed at length w ith each patient.
References
1. Byer NE. Long-term natural history of lattice degeneration of the retina. O phthalm ology. 1989;96 (9):1396-
1401.
2. Byer NE. Cystic retinal tufts and their relationship to retinal detachment. Arch O phthalm ol. 1981;99 (10):1788-
1790.
3. Folk JC, Arrindell EL, Klugman MR. The fellow eye of patients with phakic lattice retinal detachment.
O phthalm ology. 1989;96 (1):72-79.
4. Folk JC, Bennett, SR, Klugman MR, Arrindell EL, Boldt HC. Prophylactic treatment to the fellow eye of patients
with phakic lattice retinal detachment: analysis of failures and risks of treatment. Retina. 1990 ;10 (3):165-169.
5
QUESTION
WHAT RETINAL BREAKS

REQUIRE TREATMENT AND
HOW ARE THEY TREATED?
Ron Afshari Adelman, MD, MPH
Retinal breaks are com mon ly encou ntered in the ophthalmologist’s office. Knowledge
of the d ifferent typ es of breaks and their natu ral cou rse is essential to the clin ical m an-
agement of the patient.
Types of Breaks
A retinal break is any fu ll th ickness defect of the retina and is classified broad ly into
retinal tears and retinal holes. Retinal tears have a tractional component that lead s to a
horseshoe-shap ed or flap tear. The tractional forces m ay increase the size of the break
over tim e or elevate the retina and lead to a retinal detach m ent. Retinal holes lack a trac-
tional component and include atroph ic holes, opercu lated holes, and postnecrotic holes.
Atroph ic holes and opercu lated holes have a relatively low risk for progression to retinal
detach m ent (RD),1 wh ile postnecrotic holes (eg, postherpetic viral in fections) have a
h igher risk of RD.
Which Retinal Breaks Are Risky?

Symptom atic retinal breaks carry a sign ificant risk of RD. The presence or absence of
visu al symptom s, such as floaters, photopsias, and visu al field defects, aid s in risk assess-
m ent.1-3
An acute retinal break carries a greater risk of progression to a retinal detach m ent than
a ch ron ic break. An acute break requ ires treatment. The du ration of symptom can assist
17
18 Question 5
in determ in ing how long a retinal break m ay have been present. The presence of blood
on or arou nd a retinal break ind icates that it is acute and is associated w ith sign ificant
traction and thu s, treatment of such a break wou ld be necessary. On the other hand, an
older break (wh ich is often characterized by pigmentary changes) has a lower risk of RD
and u su ally does not need treatm ent.
The location and size of the break as well as the presence of traction factor into the m an-
agement decision. A retinal break in the superior qu ad rants of the eye has a h igher risk of
progression to RD than a break in the in ferior qu ad rants. A larger break has a h igher risk
than a sm aller break. Up to 50% of u ntreated symptom atic retinal breaks w ith traction
lead to RD. However, im med iate treatment of such breaks decreases the chance of RD to
less than 5%.4 A trau m atic d ialysis or trau m atic tear along the vitreou s base shou ld be
consid ered for treatm ent.
Finally, several pre-existing cond itions are risk factors for retinal break progression to
RD:
• M yopia: About half of RDs occu r in myopic eyes. Moderate and h igh myopes have
a 10-fold increased risk of RD.5
• Lattice degeneration: Lattice degeneration is present in 6% to 8% of the popu lation
and is a risk factor for RD.6,7
• Cataract surgery: Lifetime risk of RD follow ing u ncomplicated cataract su rgery is
about 1%. Vitreou s loss, h igh axial length, and Nd :YAG capsu lotomy increase the
risk by 5-fold.8
• Status of the fellow eye: In patients w ith RD, the fellow eye shou ld be carefu lly exam-
ined for the presence of retinal breaks becau se the fellow eye is at increased risk
for RD.
• Trauma: Trau m atic retinal breaks m ay develop months or years follow ing the
trau m a and m ay lead to RD.9
In general, symptom atic retinal breaks u sually need prompt treatment (Table 5-1). Most
asymptom atic holes can be observed. Asymptom atic horseshoe tears shou ld be treated if
there is a risk factor as described above.
Treatment
The goal of treatm ent is to produce an adequ ate ad hesion between the choroid and the
retina arou nd the retinal break. Usu ally the d irection of traction of flap tears is toward
the ora serrata; therefore, the retina anterior to the tear requ ires thorough treatment. If
necessary, a com bination of slit lamp and binocu lar ind irect laser retinopexy in conju nc-
tion w ith scleral depression shou ld be employed to ensu re 360 degrees of barrier laser
photocoagu lation arou nd the break.10,11 If there is su bretinal flu id anterior to the flap tear
itself, then laser treatment in a “U” shape up to the ora serrata on both sides of the break
wou ld be recom m ended.
TREATMENT OPTIONS
1. Laser retinopexy: Th is method is u sed to treat the m ajority of retinal breaks. If
the break is more posteriorly located, a contact len s (eg, 3-m irror len s) w ith laser
What Retinal Breaks Require Treatment and How Are They Treated? 19
Table 5-1
Treatment Options for Different Types of Breaks

Treatment Option Type of Break
Promptly treated Acute symptomatic horseshoe tears
Tears w ith adjacent hemorrhag e
Usually treated Traumatic retinal breaks
Sometimes treated Asymptomatic horseshoe tears
Asymptomatic dialyses
Fellow eyes w ith atrophic holes, lattice deg eneration, or
asymptomatic horseshoe tears
Rarely treated Operculated breaks
Pig mented chronic breaks
Asymptomatic atrophic round holes
Asymptomatic lattice deg eneration w ithout holes
Asymptomatic lattice deg eneration w ith holes (unless PVD
causes a horseshoe tear)
Adapted and modified from American Academy of Ophthalmolog y Retina Panel. Preferred
Practice Pattern Guidelines: Posterior Vitreous Detachment, Retina Breaks, and Lattice
Deg eneration . San Francisco, CA: American Academy of Ophthalmolog y; 2008. Available at:
http:// w w w.aao.org /ppp. Accessed Aug ust 18, 2009.
delivery by slit lamp can be u sed. For anterior breaks, ind irect laser delivery can be
u sed and sometimes, scleral indentation is necessary. Th ree row s of con fluent laser
arou nd the break is recom mended. Add itional row s anterior to h igh-risk breaks
shou ld be considered. The anterior part of the break is the most risky part and also
the most d ifficu lt part to treat. If there is su bretinal flu id arou nd the break, laser
u sually does not take and treatment in a “U” shap e up to the ora on both sides
shou ld be perform ed. Cryotherapy m ay be u sefu l for more anterior retinal breaks
and those w ith su rrou nd ing subretinal flu id.
2. Cryotherapy: Cryotherapy creates an ad hesion between the retina and choroid.
Cryotherapy is advantageou s in the presence of subretinal flu id su rrou nd ing the
break and also in the presence of med ia opacity that wou ld interfere w ith laser
treatment. It is more pain fu l than laser and thu s requ ires local anesthesia (ie,
su bconju nctival lidocaine). Retrobu lbar lidocaine is not recom mended becau se if
u sed, ocu lar motility is lim ited and wou ld hamp er effective treatment.
3. Vitrectomy: Vitreou s su rgery m ay be necessary in an eye w ith a vitreou s hemor-
rhage that precludes effective laser and cryo treatment of a risky retinal break.
20 Question 5

Gelareh Abed i, MD, MS
Vitreoretin al Fellow
Boston Un iversity School of Med icine, Boston, MA
References
1. Byer NE. What happens to untreated asymptomatic retinal breaks, and are they affected by posterior vitreous
detachment? O phthalm ology. 1998;105(6):1045-1050.
2. Byer NE. Rethinking prophylactic therapy of retinal detachment. In: Stirpe M, ed. Advances in Vitreoretinal
Surgery. New York, NY: O phthalmic Communications Society, 1992:399-411.
3. Neumann E, Hyams S. Conservative management of retinal breaks. A follow-up study of subsequent retinal
detachment. Br J O phthalm ol. 1972;56 (6):482-486.
4. Shea M, Davis MD, Kamel I. Retinal breaks without detachment, treated and untreated. Mod Probl O phthalm ol.
1974;12:97-102.
5. Risk factors for idiopathic rhegmatogenous retinal detachment. The Eye Disease Case-Control Study Group.
Am J Epidem iol. 1993;137(7):749-757.
6. Byer NE. Long-term natural history of lattice degeneration of the retina. O phthalm ology. 1989;96 (9):1396-
1401; discussion 1401-1402.
7. Benson WE, Morse PH. The prognosis of retinal detachment due to lattice degeneration. Ann O phthalm ol.
1978;10 (9):1197-1200.
8. Norregaard JC, Thoning H, Andersen TF, Bernth-Petersen P, Javitt JC, Anderson GF. Risk of retinal detach-
ment following cataract extraction: results from the International Cataract Surgery O utcomes Study. Br J
O phthalm ol. 1996;80 (8):689-693.
9. Cooling RJ. Traumatic retinal detachment: mechanisms and management. Trans O phthalm ol Soc U K.
1986;105(pt 5):575-579.
10. Benson WE, Morse PH, Nantawan P. Late complications following cryotherapy of lattice degeneration. Am J
O phthalm ol. 1977;84 (4):514-516.
11. Delaney WV Jr. Retinal tear extension through the cryosurgical scar. Br J O phthalm ol. 1971;55(3):205-209.
6
QUESTION
WHAT LESIONS IN THE FELLOW EYE OF

A PATIENT WITH A RETINAL BREAK OR
DETACHMENT REQUIRE TREATMENT?
Steven Sanislo, MD
The answer to the question of wh ich lesions to treat in the fellow eye of a patient w ith a
retinal break or detach ment is controversial. If you pose th is question to several d ifferent
retinal specialists, you are likely to get several d ifferent answers. Th is is becau se there is a
paucity of level 1 or level 2 evidence to help m ake these decisions. No prospective, random-
ized clin ical trials regard ing the prevention of retinal detach ment have been published.
In the American Academy of Ophthalmology’s (AAO) Preferred Practice Pattern (PPP)
recom mend ations for prophylactic laser of peripheral retinal lesions, on ly one recom men-
d ation, the treatment of symptom atic flap (horseshoe) tears, is based on level 1 evidence.1
Fellow eyes of patients w ith retinal breaks or detach ment are at an increased risk of
also developing breaks or detach m ent. The goal of prophylactic treatm ent is to reduce th is
risk, if p ossible. To determ ine if treatm ent is likely to have th is effect, I like to look at sev-
eral factors that m ight increase or reduce th is risk. These include the presence or absence
of symptom s, presence of h igh myopia, state of the posterior hyaloid, phakic statu s, typ e
of retinal break or lesion, patient age, and special circu m stances (Table 6-1).
Certain factors probably carry more weight than others, such as the presence of symp -
tom s of photopsias or floaters; breaks w ith elevated flaps, traction, or associated sub-
retinal flu id ; and presence of h igh myopia. If any of these factors are present, I strongly
consider prophylactic laser. Sim ilarly, treatment shou ld be strongly con sidered if there
are sign ificant peripheral retinal lesions and posterior vitreou s detach m ent (PVD) has
not yet occu rred. In a prospective study by Hovland and coworkers, fellow eyes of apha-
kic patients w ith retinal detach m ent were fou nd to be far less likely to develop a retinal
detach m ent if a PVD was present in those eyes.2 I also tend to treat if there is extensive
lattice degeneration in the fellow eye, particu larly if lattice was involved in cau sing the
retinal detach m ent in the first eye.
21
22 Question 6
Table 6-1
Risk Factors
Factor Increased Risk Decreased Risk
Symptoms Symptomatic Asymptomatic
Axial myopia Hig h myopia Low or non -myopic
Posterior hyaloid No PVD PVD
Phakic status Aphakic/ pseudophakic Phakic
Types of lesions Flap (horseshoe) tear Operculated hole
Break w ith subretinal fluid Atrophic hole
Evidence of retinal traction Lattice deg eneration w ith pig ment
demarcation
Lattice deg eneration w ith breaks Breaks w ith surrounding pigment
demarcation
In patients w ithout a h istory of tears or detach ment in either eye, m any lesion s such
as asymptom atic lattice degeneration m ay be observed.3 In those patients where one eye
has had a detach m ent, however, the th reshold for treatm ent is sign ificantly lower. Despite
th is, in certain situ ations, low-risk lesion s can be observed, provided that the patient
u nderstand s the symptom s of tears and detach ment. For in stance, in a pseudophakic
eye w ith a pre-existing PVD, where the fellow eye detached after cataract su rgery, lat-
tice degeneration w ithout breaks or w ith atroph ic breaks m ight be observed. However,
if intraocu lar su rgery is plan ned for that eye, prophylactic laser is probably advised.
Ch ron ic breaks (determ ined by the presence of pigment dem arcation) and op ercu lated
breaks can generally be observed.
In my exp erience, certain sp ecial cond itions w ith h igh risk of retinal detach m ent and/
or increased d ifficu lty of treating retinal detach ment when it occu rs warrant treatm ent
of any sign ificant peripheral lesion s. These include (but are not lim ited to) patients w ith
Marfan synd rome, h igh myopia, and you ng ch ild ren or adu lts w ith developm ental d is-
abilities.
Why not ju st treat all lesion s? It is tempting to treat all lesions in the fellow eye of
patients w ith retinal tears or detach m ent, but th is is probably not a good approach. Even
though peripheral retinal laser treatm ent is generally very well tolerated, there are still
p otential risks. In patients w ithout PVD, laser treatment m ight cau se areas of retinal
th in n ing where breaks can occu r when PVD develop s. Also, som e people believe that
exten sive laser treatment m ay lead to the developm ent of m acu lar epiretinal mem branes.
Add itionally, treating m any patients who are at low risk of developing retinal detach ment
add s to the cost of m ed ical care for patients, in su rance compan ies, and ou r govern m ent.
For an optim al balance of risk and cost versu s benefit, each patient’s ind ividu al situ-
ation mu st be add ressed by evalu ating the variou s factors that increase or reduce that
particu lar patient’s risk of developing retinal detach ment.
What Lesions in the Fellow Eye of a Patient Require Treatment? 23
References
1. Wilkinson CP. Evidence-based analysis of prophylactic treatment of asymptomatic retinal breaks and lattice
degeneration. O phthalm ology. 2000;107(1):12-15.
2. Hovland KR. Vitreous findings in fellow eyes of aphakic retinal detachment. Am J O phthalm ol. 1978;86(3):350-
353.
3. Byer NE. Rethinking prophylactic therapy of retinal detachment. In: Stirpe M, ed. Advances in Vitreoretinal
Surgery. New York, NY: O phthalmic Communications Society; 1992:399-411.
7
QUESTION
HOW SOON SHOULD A PATIENT

WITH FLOATERS BE EXAMINED
AND HOW SHOULD I MANAGE A
PATIENT WITH AN ACUTE POSTERIOR
VITREOUS DETACHMENT?
Karen M. Gehrs, MD
Floaters are a com mon symptom and, fortu nately, most patients w ith floaters do not
have sign ificant vitreoretinal pathology. However, becau se some floaters m ay be associ-
ated w ith vision-th reaten ing cond ition s, it is important to have a protocol for triaging and
m anaging patients w ith floaters in a timely fash ion.
Triage
Most physician s d elegate patient triage to their staff. However, physicians shou ld
provide staff w ith clear gu idelines regard ing the triage of patients w ith floaters. Staff
shou ld be encou raged to d iscu ss w ith a physician any patient who does not fit the triage
gu idelines or any patient who does not follow the gu idelines. In general, when a patient
calls complain ing of the acute onset of floaters, he or she shou ld be advised to com e in the
same d ay. Figu re 7-1 provides add itional suggested triage gu idelines. Once schedu led, a
patient shou ld be given emergency contact in form ation and cou nseled to call back sooner
than schedu led if h is or her symptom s worsen or if new symptom s develop. I requ ire that
the triage encou nter be docu m ented in the patient’s chart.
25
26 Question 7
1. Create a triag e template that prompts staff to ask pertinent questions and record pertinent
positive and neg ative symptoms and history.
2. Checklist for pertinent symptoms:
New onset floaters or a chang e or worsening of chronic floaters
Duration of floaters
Associated symptoms (photopsias, decreased central vision, and peripheral vision loss)
History of prior retinal breaks or detachments in the symptomatic and fellow eye
Family history of retinal detachment
History of trauma in the symptomatic eye
Type and date(s) of prior ocular surg ery (including YAG capsulotomy) in the symptom -
atic eye
Pain, redness, or photophobia, w hich may indicate u veitis, including endophthalmitis,
in a patient w ho has had recent eye surg ery or systemic illness
History of systemic conditions that may predispose to problems that cause floaters (dia -
betes w ith vitreous hemorrhag e, acquired immune deficiency syndrome w ith potential
for cytomeg alovirus retinitis, collag en vascular disease w ith the potential for u veitis)
3. In g eneral, patients w ith new onset floaters of less than 2 w eeks duration should be
instructed to come w ithin 24 hours, preferably the same day. This is especially true if the
patient has associated symptoms of photopsia, central or peripheral vision loss, ocular
pain, redness, photophobia, history of recent eye surg ery or trauma, history of a systemic
condition that could predispose to vitreous hemorrhag e, intraocular infection, or noninfec-
tious intraocular inflammation.
4. Patients w ith subacute floaters (more than 2 weeks but less than 6 weeks) w ithout associ-
ated photopsias; central or peripheral vision loss; and no pain, redness, or photophobia
should be told to come in w ithin 1 week and should call back and come in the same day
if photopsias, vision loss, pain, redness, or photophobia develops before the scheduled
appointment.
5. Patients w ith floaters of more than 6 weeks duration are more difficult to triag e.
In g eneral, if floaters are worsening or if there is associated central or peripheral vision
loss; photopsias; or a systemic condition w ith the potential for infection, hemorrhag e,
or inflammation, the patient should be told to come in the same day.
Evaluation of chronic floaters w ith no associated symptoms or underlying systemic
condition of concern can be scheduled w ithin 1 month w ith instructions to call back
before the scheduled appointment if symptoms worsen or new concerning symptoms
(reviewed w ith the patient in lay terms) develop.
Fig u re 7-1. Sugge ste d guide lin e s fo r tria gin g pa tie n ts w ith flo a te rs.
Evaluation
A complete ocu lar and system ic h istory and ocu lar exam ination w ith d ilation of both
eyes shou ld be p erformed so that cau ses of floaters other than posterior vitreou s detach-
m ent (PVD) are not overlooked. In ou r practice, we have provided gu idelines for the
p ertinent in form ation to be recorded, and we have in structed tech n icians and hou se staff
to d iscu ss pertinent abnorm alities w ith an attend ing physician before d ilation. The fol-
low ing are the key exam ination find ings to be recorded :
• Cu rrent refraction and corrected d istance and near visual acu ity of each eye w ith
pin hole testing for d istance. If pin hole visu al acu ity is not 20/ 20, then I ask that a
d iagnostic refraction be attempted. If th is in form ation is not recorded at the first
How Soon Should a Patient With Floaters Be Examined? 27
visit, it is easy to overlook later. If a patient su bsequently requ ires retinal detach-
ment repair, knowledge of h is or her preop erative visu al acu ity and refraction
allow s me to cou nsel h im or her about visu al prognosis and potential postoperative
refractive changes.
• Visu al field s by con frontation to docu ment visual field loss (or lack thereof).
• Pupil size and reaction to light and testing for an afferent pupillary defect (or docu-
mentation that the patient presented w ith the pupils already d ilated).
• A basic ocu lar motility exam ination. Wh ile form al motility testing w ith prism s
is not u su ally performed in the setting of acute PVD symptom s, any pre-existing
ocu lar m isalign ment in prim ary gaze or lim ited motility shou ld be docu mented.
Such find ings cou ld su bsequently becom e important shou ld the patient requ ire
retinal detach m ent repair.
• Slit lamp exam ination. Both u nd ilated (when p ossible) and d ilated slit lamp exam i-
nation shou ld be p erformed. Record med ia opacities, sign s of cu rrent or past intra-
ocu lar in flam m ation, trau m a or su rgery, iris abnorm alities, len s statu s, posterior
cap su le statu s, and presence or absence of anterior vitreou s cells or pigment.
• Dilated ophthalmoscopic exam ination. I typically exam ine the posterior pole w ith
slit-lamp biom icroscopy and a 90-D lens first. I then inspect the entire fu ndu s w ith
binocu lar ind irect ophthalmoscopy and scleral indentation of the peripheral retina
for 360 degrees u sing a 20-D len s (or 28-D lens for sm all pupils). Th is is su fficient
for most patients. However, if I have not fou nd a break by that point but still su s-
pect a break based on symptom s or other find ings, I perform a contact lens exam i-
nation u sing a Gold m an n 3-m irror fu ndu s lens. I record the presence or absence
of a Weiss ring and the location of all retinal pathology in relation to vascu lar and
other land m arks.
Ancillary Testing
If the entire fu ndu s can not be visu alized becau se of med ia opacity, a sm all pupil, or a
patient’s inability to cooperate, I perform B-scan u ltrasonography of the entire posterior
segm ent. A carefu lly perform ed u ltrasou nd can detect even a sm all peripheral retinal
detach m ent or tear and can detect the presence and extent of a PVD and any associated
vitreoretinal traction.
Optical coherence tomography (OCT) can help identify early posterior vitreou s detach-
m ent and determ ine whether the fovea is detached in the setting of a shallow m acu lar
detach m ent. In one case d iagnosed as a m acu la-off retinal detach m ent becau se of p oor
vision and poor visu alization of the m acu la, OCT demon strated that the central m acu la
was attached and that the decreased vision was due to a focal vitreou s hemorrhage overly-
ing the central m acu la.
28 Question 7
Management
When vitreoretinal pathology is d iscovered, appropriate m anagement is arranged.
Managem ent of the potential pathological cau ses of floaters is beyond the scop e of th is
section. When on ly a PVD is d iscovered, the tim ing of follow-up w ill vary depend ing on
the on set of symptom s and clin ical find ings. In general, the more acute the symptom s,
the sooner I have the patient retu rn. I also see patients back sooner and more frequently if
they have had a prior retinal detach ment or break in either eye, their photop sias continue
to be prom inent, or they have worsen ing symptom s or lattice degeneration. Patients w ith
ch ron ic floaters of more than 6 weeks du ration w ith no retinal breaks can be reassu red
and schedu led for routine follow-up whether a PVD is present or not. I typically recom-
m end that patients w ith the acute on set of floaters of less than 2 weeks du ration and no
retinal breaks retu rn in 7 to 10 d ays. For patients w ith floaters of 2 to 6 weeks du ration
and no evidence of retinal break or lattice degeneration, symptom s are decreasing, and
the patient no longer has photop sias, I typically recom m end routine follow-up. However,
if such a patient still has photopsias or lattice degeneration, I generally exam ine h im or her
at least one more tim e in 2 to 4 weeks before having h im or her resu me routine follow-up.
I ask my staff to teach all patients about PVD symptom s and to give them the American
Academy of Ophthalmology’s educational pamph let on flashes and floaters. I review w ith
patients in lay term s the symptom s that shou ld prompt them to seek u rgent re-evalu ation,
and I provide contact in form ation for after-hou rs evaluation.
Summary
The evalu ation and m anagement of patients w ith floaters or other symptom s of
PVD can u sually be m anaged in a timely fash ion w ith a system atic approach. Wh ile an
occasional patient w ill still develop an u nanticipated retinal detach ment, clear triage
gu idelines coupled w ith a thorough exam ination, appropriately schedu led follow-up,
and patient education shou ld provide a system for detection of sign ificant pathology in a
timely fash ion in the vast m ajority of patients w ith floaters and PVD.
Financial disclosure: Research support from Eli Lilly & Company (principal Investigator); co-investigator
for studies sponsored by A lcon, Regeneron, N ovartis, Ophtherion, A llergan.
Suggested Reading
Byer NE. What happens to untreated symptomatic retinal breaks and are they affected by posterior vitreous detach-
ment? O phthalm ology. 1998;105(6):1045-1049; discussion 1049-1050.
Coffee RE, Westfall AC, Davis GH, Mieler WF, Holz ER. Symptomatic posterior vitreous detachment and the inci-
dence of delayed retinal breaks: case series and a meta-analysis. Am J O phthalm ol. 2007;144 (3):409-413.
Hikichi T. Time course of posterior vitreous detachment in the second eye. Curr O pin O phthalm ol. 2007;18 (3):224-
227.
8
QUESTION
HOW LONG SHOULD I WAIT TO

PERFORM CATARACT SURGERY AFTER
TREATMENT OF DIABETIC MACULAR
EDEMA OR RETINOPATHY?
Linda A. Lam, MD
Controversy has su rrou nded the effect that cataract su rgery has on the developm ent
and progression of d iabetic m acu lar edem a and d iabetic retinopathy. Previou s stud ies
have reported that cataract su rgery in d iabetic eyes was complicated by worsen ing of
m acu lar edem a and retinopathy w ith the development of vitreou s hemorrhage, anterior
segm ent neovascu larization, and worsen ing visual acu ity. Many of these stud ies were
retrosp ective and included older su rgical tech n iques associated w ith intracapsu lar and
extracapsu lar cataract extraction. More recent prospective stud ies in eyes w ith relatively
low-risk d iabetic retinopathy u ndergoing cataract su rgery u sing sm all-incision phaco-
emu lsification tech n iques have show n no sign ificant progression of d iabetic retinopathy
p ostoperatively.
In a patient w ith d iabetes who need s cataract su rgery, there are som e mod ifiable fac-
tors that shou ld be optim ized prior to cataract su rgery in order to lim it the development
and progression of postoperative d iabetic m acu lar edem a and retinopathy:
• Degree of glycem ic control
• Preop erative treatment of severe nonproliferative and proliferative d iabetic reti-
nopathy
• Presence of preoperative d iabetic m acu lar edem a
• Presence of vitreom acu lar traction (VMT)
29
30 Question 8
Norm alizing and stabilizing system ic glucose levels is necessary to reduce the devel-
opment and progression of d iabetic retinopathy after cataract su rgery. Poor preop erative
glycem ic control as well as poor renal fu nction have been correlated w ith h igher rates
of d iabetic retinopathy and d iabetic m acu lar edem a follow ing cataract su rgery. Becau se
transient worsen ing of d iabetic retinopathy m ay occu r du ring the first 6 month s of inten-
sive blood glucose m anagem ent,1 deferring cataract su rgery for at least 6 month s is recom-
m ended in th is scenario.
Patients w ith severe nonproliferative d iabetic retinopathy and proliferative d iabetic
retinopathy shou ld receive pan retinal laser photocoagu lation prior to cataract su rgery.
Cataract su rgery shou ld be delayed for several months after to allow the d iabetic retinopa-
thy to improve and stabilize. Follow ing pan retinal photocoagu lation, m acu lar edem a m ay
su bsequently develop or progress. As a resu lt, waiting several months, and up to 6 months
in some cases, to perform cataract su rgery is reasonable to allow for improvement of the
edem a or retinopathy.
Pre-existing m acu lar edem a is a m ajor contributor to p ostoperative visu al loss in
patients w ith d iabetes. In eyes w ith pre-existing m acu lar edem a, it is necessary to treat
the m acu lar edem a prior to cataract su rgery becau se p ostop erative in flam m ation can
exacerbate the edem a. Effective treatments described in the literatu re for d iabetic m acu-
lar edem a include focal laser, anti-vascu lar endothelial grow th factor (VEGF) agents,
and intraocu lar or su btenon triamcinolone aceton ide. Using fluorescein angiography to
delineate areas of leakage, focal laser m ay be applied as first-line treatm ent. If sign ificant
m acu lar edem a rem ains follow ing focal laser treatm ent, intravitreal (or su btenon) tri-
amcinolone or intravitreal anti-VEGF agent cou ld be u sed to reduce the m acu lar edem a.
Cataract su rgery shou ld then be delayed about 6 months to allow for improvem ent of the
m acu lar edem a.
Eyes w ithout preoperative d iabetic m acu lar edem a that then develop d iabetic m acu lar
edem a p ostoperatively u sually experience resolution of the edem a w ith in the first post-
operative year.2
Optical coherence tomography (OCT) can be u sed to assess the degree of m acu lar
edem a and evalu ate the treatment effect. OCT m ay reveal the presence of coexisting
pathology such as VMT or posterior hyaloid al traction (PH T), wh ich m ay contribute to
p ersistent m acu lar edem a despite treatment. If sign ificant VMT is present, then vitrec-
tomy m ay be needed to elim inate the tractional forces contributing to the edem a. If len s
opacity precludes a clear view of the m acu la for vitreou s su rgery, then con sider either
p erform ing cataract su rgery prior to vitrectomy or perform ing both su rgeries at the same
time. Some retinal su rgeons prefer to have the cataract removed first to allow for any asso-
ciated corneal edem a to subside since corneal edem a m ay m ake the su rgeon’s view of the
m acu la du ring the vitrectomy p ortion of the com bined procedu re suboptim al.
Recent d ata demon strate that the progression of d iabetic retinopathy is not sign ifi-
cantly affected by sm all-incision cataract su rgery.3,4 However, it is imp ortant to treat
any preop erative d iabetic retinopathy and m acu lar edem a and to allow up to 6 month s
for improvement and stabilization prior to elective cataract su rgery. Key factors such as
glycem ic control, presence of u ntreated severe nonproliferative or proliferative d iabetic
retinopathy, m acu lar edem a, and associated vitreom acu lar traction need to be add ressed
prior to proceed ing w ith cataract su rgery.
How Long Should I Wait to Perform Cataract Surgery After Treatment? 31
References
1. Diabetes Control and Complications Trial Research Group. Early worsening of diabetic retinopathy in the
diabetes control and complications trial. Arch O phthalm ol. 1998;116(7):874-886.
2. Murtha T, Cavallerano J. The management of diabetic eye disease in the setting of cataract surgery. Curr O pin
O phthalm ol. 2007;18 (1):13-18.
3. Romero-Aroca P, Fernández-Ballarat J, Almena-Garcia M, Méndez-Marín I, Salvat-Serra M, Buil-Calvo JA.
Nonproliferative diabetic retinopathy and macular edema progression after phacoemulsification: prospective
study. J Cataract Refract Surg. 2006;32(9):1438-1444.
4. Squirrell D, Bhola R, Bush J, Winder S, Talbot JF. A prospective, case-controlled study of the natural history
of diabetic retinopathy and maculopathy after uncomplicated phacoemulsification cataract surgery in patients
with type 2 diabetics. Br J O phthalm ol. 2002;86 (5):565-571.
9
QUESTION
DOES CATARACT SURGERY IN

AN EYE WITH DRY AGE-RELATED
MACULAR DEGENERATION INCREASE
THE RISK OF DEVELOPING CHOROIDAL
NEOVASCULARIZATION?
Andrew A. Moshfeghi, MD
In the early d ays of modern cataract su rgery, the resu lts of large popu lation-based epi-
dem iologic stud ies suggested that there was a correlation between cataract su rgery and
su bsequent development of wet age-related m acu lar degeneration (AMD).1-3 The Blue
Mou ntain s Eye Study (BMES) was an Au stralian popu lation-based cohort study designed
to d eterm ine whether incident late AMD (geograph ic atrophy or neovascu lar AMD) was
present in a h igh prop ortion of nonphakic (pseudophakic or aphakic) compared w ith
phakic eyes over a 10-year follow-up period.1,2 Late AMD was identified in 10 of 132
nonphakic eyes (7.6%), wh ich was sign ificantly more than the incidence of 96 late AMD
observations in 4631 phakic eyes (2.1%).1,2 These find ings suggested an approxim ately
3-fold increased adju sted risk of developing late AMD in nonphakic eyes compared w ith
phakic eyes.
When these d ata from the BMES were pooled w ith those from the Beaver Dam Eye
Study in Wisconsin 3 —wh ich also had previou sly reported an approxim ately 4-fold
increased adju sted risk of neovascu lar AMD developm ent over a 10-year follow-up
p eriod —there was nearly a 6-fold increased adju sted risk of developing late AMD for
nonphakic compared w ith phakic eyes over a 5-year evalu ation p eriod.4
33
34 Question 9
More recently, the Age-Related Eye Disease Study (AREDS), wh ich was a prospec-
tive and controlled cohort study, had concluded that no such association exists.5 AREDS
evalu ated patients w ith early or interm ed iate AMD at baseline.5 Cataract statu s was not
statistically sign ificantly associated w ith the development of neovascu lar AMD.5 It is true
that the original AREDS study was a prosp ective, random ized, and controlled trial and
th is represents a h igher level of scientific evidence than the popu lation-based Beaver Dam
and Blue Mou ntain s Eye stud ies. That said, the present AREDS cohort su b-study was not
specifically designed to answer the question “does cataract su rgery promote the develop -
m ent or progression of AMD?”5 It does, however, have the benefit over the popu lation-
based stud ies in sofar as patients were followed closely w ith serial fu ndu s photographs
th roughout the study follow-up period (and before and after cataract su rgery in the
patients who u nderwent cataract extraction).
Despite th is good evidence show ing no association between cataract su rgery and
AMD progression, th is classic question rem ain s a controversial topic in the modern era
of m icroincisional phacoemu lsification cataract su rgery for anterior segment su rgeons,
vitreoretinal sp ecialists, and patients.
The reason for the p ersistent controversy is ou r clin ical experience. Vitreoretinal spe-
cialists often share cou ntless anecdotes about patients who are referred to them for u nex-
plained vision loss follow ing cataract su rgery who tu rn out to harbor typical choroid al
neovascu larization (CN V) second ary to wet AMD. Often, these patients are not aware of
the fact that their fellow eye also reveals stigm ata of d ry AMD. So, when the patient is told
all of th is by the vitreoretinal sp ecialist ju st follow ing their recent cataract su rgery, it ju st
seem s natu ral to the patient—and to the retina specialist—that the 2 are in fact related.
What is more likely is that the CN V and pre-existing d ry or su btle wet AMD were not
d iagnosed prior to the cataract su rgery. Without carefu l stereoscopic exam ination of the
fu ndu s in an eye w ith reduced med ia clarity due to cataract, it m ay be d ifficu lt for the
anterior segm ent su rgeon to detect pre-existing featu res of AMD: subtle d ru sen, focal
chorioretinal atrophy, geograph ic atrophy, retinal pigment epitheliu m hyperplasia and
m igration into the retina, focal subretinal flu id or hemorrhage, and early CN V or su breti-
nal fibrosis.
Even despite carefu l stereoscopic exam ination by the anterior segment su rgeon, these
AMD featu res m ay be legitim ately overlooked due to subtle or sign ificant med ia opacity.
Ancillary testing in the form of fluorescein angiography and optical coherence tomogra-
phy are necessary to determ ine to what extent, if any, the m acu la deviates from norm al
and if, in fact, CN V is present preop eratively. In most cases, these add itional tests wou ld
u su ally be reserved for patients in whom the sign ificant degree of decreased vision does
not correlate to the relatively m ild appearance of the cataract.
However, it shou ld be kept in m ind that stable occu lt CN V can reduce vision modestly
or not at all. If th is is overlooked preoperatively, the anterior segm ent su rgeon cou ld
erroneou sly attribute the modest level of visual decline to the moderate corticonuclear
cataract. Postoperatively, the anterior segment su rgeon is then left trying to explain 2 con-
trad ictory th ings to the patient: 1) that the CN V/ AMD was not present before the cataract
su rgery (ie, that he or she d id not overlook it) and 2) that the cataract su rgery d id not cau se
the CN V/ AMD to develop. It m ay be especially d ifficu lt to explain away the new find ings
of previou sly u nd iagnosed d ry AMD in the fellow u noperated eye.
Does Cataract Surgery in an Eye With Dry AMD Increase the Risk? 35
What I tell patients w ith early to intermed iate d ry AMD considering cataract su rgery
is that the true answer to th is age-old question is not so important. Futu re stud ies w ill
answer th is question more defin itively. Rather, it is more imp ortant that patients receive
regu lar follow-up retinal exam inations and that they mon itor their ow n vision on a regu-
lar basis.
References
1. Wang JJ, Mitchell PG, Cumming RG, Lim R. Cataract and age-related maculopathy: the Blue Mountains Eye
Study. O phthalm ic Epidem iol. 1999;6(4):317-326.
2. Cugati S, Mitchell P, Rochtchina E, Tan AG, Smith W, Wang JJ. Cataract surgery and the 10 -year incidence of
age-related maculopathy: the Blue Mountains Eye Study. O phthalm ology. 2006;113(11):2020-2025.
3. Klein R, Klein BE, Wong TY, Tomany SC, Cruickshanks KJ. The association of cataract and cataract surgery
with the long-term incidence of age-related maculopathy: the Beaver Dam eye study. Arch O phthalm ol.
2002;120 (11):1551-1558.
4. Wang JJ, Klein R, Smith W, Klein BE, Tomany S, Mitchell P. Cataract surgery and the 5-year incidence of
late-stage age-related maculopathy: pooled findings from the Beaver Dam and Blue Mountains eye studies.
O phthalm ology. 2003;110 (10 ):1960-1967.
5. Clemons TE, Milton RC, Klein R, Seddon JM, Ferris FL 3rd; Age-Related Eye Disease Study Research Group.
Risk factors for the incidence of Advanced Age-Related Macular Degeneration in the Age-Related Eye Disease
Study (AREDS) AREDS report no. 19. O phthalm ology. 2005;112(4):533-539.
10 QUESTION
WHAT SHOULD I DO WHEN I SEE A

PATIENT WITH A HOLLENHORST PLAQUE?
Anita Agarwal, MD
The original description of a retinal arteriolar em bolu s by Hollen horst was the descrip -
tion of a bright orange/ yellow glisten ing intra-arterial plaque, u su ally asymptom atic, that
represented cholesterol particles d islodged from an atherom atou s collection along the
carotid artery, aorta, or card iac valves.1,2 Calcific em boli from d iseased card iac valves and
fibrin platelet th rom bi are believed to be wh ite. However, attempts to categorize em boli
based on their color and appearance have not been very accu rate.3 The calcific em boli
are often larger and m ay lodge at or beh ind the lam ina cribrosa (Figu re 10-1), whereas
cholesterol em boli can fragment into several sm all particles, move dow nstream, and be
asymptom atic.
When I see an em bolu s, I first categorize it as symptom atic or asymptom atic. If the
patient presents w ith a symptom atic central (CRAO) or branch retinal artery occlu sion
(BRAO) w ith visible em bolu s, the first th ing I wou ld do is m anage the arterial obstruc-
tion. If the patient arrived w ith in a short time (less than 24 hou rs) of onset of symptom s,
I perform an anterior cham ber paracentesis and ocu lar m assage and start an ocu lar hypo-
tensive agent becau se som etim es the occlu sion m ay be incomplete or have sup erimposed
spasm of the retinal arteries that m ay be overcome by the sudden lowering of intraocu lar
pressu re. Occasionally, the fibrin platelet coat of the em bolu s m ay spontaneou sly frag-
m ent and resu lt in d issolution of the em bolu s.
Once the acute event is m anaged or if the patient is asymptom atic, I w ill attempt to
determ ine the origin of the em bolu s th rough a thorough m ed ical h istory, includ ing age,
smoking habits, hypertension, d iabetes, hypercholesterolem ia, previou s card iac valvu lar
d isease, card iomyopathy, collagen vascu lar d isorders, fever, anem ia, coagu lation d isor-
ders, recent fractu res, and rarer cau ses of th rom bosis/ em boli such as sickle cell anem ia
or homocysteinem ia. Do not forget the presence of a patent foram en ovale as a mecha-
n ism for paradoxical em bolu s in som eone w ith long bone fractu res, intravenou s d rug u se
(talc), etc. I obtain a fu ndu s photograph to docu ment the site and extent of the occlu sion.
37
38 Question 10
Fig u re 10 -1. Wh ite ca lcific e m bo li fro m a

ca rdia c va lve lo dge d in th e a rte rio le a t th e
disc a n d a t a bifurca tio n site . No te th e se
e m bo li a re usua lly la rge .
A fluorescein angiogram is not necessary u n less one wants to ru le out other cau ses of
retinal artery occlu sion, such as Su sac synd rome (see below).
If the patient is you ng (ie, less than 30 years of age), the cau se is u n likely to be athero-
m atou s and more likely to be card iac in origin or one of the other rarer cau ses as noted
above. The patient w ill be referred on an u rgent basis to an intern ist/ card iologist for eval-
uation. Both transthoracic and transesophageal echocard iogram s (TEEs) are necessary.
Tran sthoracic echocard iography was positive in 27% cases and TEE in 59% of patients w ith
retinal artery occlu sion.4 Overall, persons w ith an acute retinal arterial occlu sion who are
less than 45 years of age are 3 times more likely to have card iac pathology that requ ires
anticoagu lation or card iac su rgery when compared to those over the age of 45 years.
In patients older than 45 years, the cau se of the em bolu s is more likely to be athero-
m atou s. Therefore, referral to an intern ist for evaluation of carotid or other sou rces of
atherom atou s em boli, such as the aortic arch, for lipid profile stud ies and hyp ertension
evalu ation is ind icated (Table 10-1).
Nonem bolic cau ses of BRAO include id iopath ic recu rrent BRAO (Su sac synd rome), cat
scratch d isease, and toxoplasm a retin itis. Patients w ith Su sac synd rome have recu rrent
episodes of BRAO and characteristic plaques in the vicin ity of the occluded vessel that
can be m istaken for Hollen horst plaques. A fluorescein angiogram is extremely u sefu l in
con firm ing th is d iagnosis where segments of both involved and often u n involved arte-
rioles w ill show fu siform stain ing of the vessel wall (Figu re 10-2). These patients have
hearing deficits and focal neu rological events to encompass the classic triad. Patients w ith
cat scratch d isease and toxoplasm a retin itis w ill u su ally have an active area of retin itis at
the site of the occlu sion.
In a patient w ith a CRAO and no visible em bolu s, temporal (giant cell) arteritis shou ld
be considered as a possible etiology and an enth rocyte sed imentation rate (ESR) shou ld
be obtained promptly (see Table 10-1).
Keep in m ind less com mon cau ses of em boli such as intravenou s d rug u se (talc),
fat em bolism (recent long bone fractu res), am n iotic flu id em bolu s, and pancreatitis.
Asymptom atic persistent foram en ovale is more frequently recogn ized in you ng patients
w ith retinal arterial em boli and occlu sion.
What Should I Do When I See a Patient With a Hollenhorst Plaque? 39
Fig u re 10 -2 . (A) Le ft fun dus o f a 33-ye a r-o ld

A wo m a n with fre sh BRAO. (B) Righ t fundus o f
th e sa m e pa tie n t sh o w in g e vide n ce o f o ld
BRAO. (C) Fluo re sce in a n gio gra m o f th e le ft
e ye sh o win g fusifo rm sta in in g o f se gm e n ts
o f th e re tin a l a rte rio le typica l fo r Susa c syn -
dro m e .
The significance of a Hollen horst plaque or a retinal arteriolar occlu sion in relation to
risk of stroke is u nclear. A study at the Cleveland Clinic of 130 patients w ith a Hollen horst
plaque or a RAO showed a low prevalence of extracranial cerebrovascu lar d isease that
requ ired intervention.5 A review of the literatu re by Wolintz 6 concluded that in patients
w ith asymptom atic plaques, carotid endarterectomy shou ld not be recom mended u nless
the patient develops hem ispheric symptom s or signs referable to carotid circu lation deficit.
40 Question 10
Table 10-1
Recommended Systemic Evaluation of Patients

With Acute Retinal Arterial Occlusion
Study Patients Quality of Evidence
Echocardiog raphy Hig h cardioembolic risk, all young Diag nostic studies
patients
Carotid ultrasonog raphy All adult patients, reg ardless of Diag nostic studies
emboli
Homocysteine level Young patients w ith no other Case series
identifiable cause
Erythrocyte sedimentation rate Older patients w ith a reasonable Case report
to rule out temporal (giant cell) suspicion
arteritis
Anticoag ulation studies (routine Hig her suspicion in young Case series
anticoag ulation studies, protein patients, consider as a secondary
C and S, antithrombin III) screen in older patients
Adapted from Sharma S, Brow n G. Retinal artery obstruction. Retina . Vol 2. St. Louis, MO: Mosby;
1323-1332.
If there is a visible em bolu s in an eye w ith an acute RAO, Nd :YAG laser d isruption of
the visible em bolu s or vitreou s su rgical tech n iques to extract the em bolu s m ay be consid-
ered.7,8
References
1. Hollenhorst RW. The ocular manifestations of internal carotid thrombosis. Med Clin N orth Am. 1960 ;44:897-
908.
2. Hollenhorst RW. Significance of bright plaques in the retinal arterioles. JAMA. 1961;178:23-29.
3. Sharma S, Pater JL, Lam M, Cruess AF. Can different types of retinal emboli be reliably differentiated from one
another? An inter- and intraobserver agreement study. Can J O phthalm ol. 1998;33(3):144-148.
4. Inatomi Y, Hino H, Hashimoto Y, Furuyoshi N, Misumi I, Uchino M. Transesophageal echocardiography for
detection of cardiac diseases in patients with retinal artery occlusion. Intern Med. 2001;40 (6):475-478.
5. Dunlap AB, Kosmorsky GS, Kashyap VS. The fate of patients with retinal artery occlusion and Hollenhorst
plaque. J Vasc Surg. 2007;46 (6):1125-1129.
6. Wolintz RJ. Carotid endarterectomy for ophthalmic manifestations: is it ever indicated? J N euroophthalm ol.
2005;25(4):299-302.
7. García-Arumí J, Martinez-Castillo V, Boixadera A, Fonollosa A, Corcostegui B. Surgical embolus removal in
retinal artery occlusion. Br J O phtham ol. 2006;90 (10):1252-1255.
8. O premcak E, Rehmar AJ, Ridenour CD, Borkowski LM, Kelley JK. Restoration of retinal blood flow via trans-
lumenal Nd:YAG embolysis/embolectomy (TYL/E) for central and branch retinal artery occlusion. Retina.
2008;28 (2):226-235.
11 QUESTION
WHAT SHOULD I DO WHEN I

SEE A PATIENT WITH O NE OR
MORE COTTON-WOOL SPOTS?
Jorge G. Arroyo, MD, MPH
The cotton-wool sp ot (CWS), also called a soft exud ate, is a sm all, wh itish/ grey, cloud-
like, linear or serp entine, slightly elevated lesion w ith fim briated edges. The CWS looks
like it floats w ith in the nerve fiber layer of the retina (Figu re 11-1). Early light m icroscopy
of CWS in the retina revealed the presence of a cytoid body, a rou nd, d ark stain ing “nucle-
u s” w ith in a grossly swollen nerve fiber layer (Figu re 11-2). On electron m icroscopy, a
CWS is an accu mu lation of intracytoplasm ic organelles interrupted in their norm al slow
and fast anterograde and retrograde transport between the ganglion cell bod ies. Focal
ischem ia follow ing occlu sion of precapillary arterioles is p ostu lated to be resp onsible for
the axoplasm ic flow interruption, due to the strong association w ith m icroangiopath ic
d iseases, although other possible etiologies have been suggested.
As you know, d iabetes m ellitu s and system ic hyperten sion are by far the most com mon
cau ses of CWS. In patients w ith a CWS and no know n h istory of d iabetes, an elevated
blood sugar is identified in 20% and an elevated blood pressu re (d iastolic blood pressu re
of 90 m m Hg or greater) in 50%.1 If present, other typical find ings of d iabetic retinopathy,
such as m acu lar edem a, lipid exud ate, flam e or dot/ blot hemorrhages, m icroaneu rysm s,
venou s bead ing, m icrovascu lar abnorm alities, or neovascu larization, can be a clue to the
cau se of the CWS. Conversely, patients w ith system ic hyperten sion wou ld be expected
to demon strate generalized arteriolar narrow ing, arteriovenou s n icking, and, in extreme
cases, optic d isc swelling.
CWS m ay be fou nd in nu merou s other d iseases. I d ivided these d iseases into isch-
em ic, em bolic, in fectiou s, toxic, rad iation-induced, neoplastic, tractional, trau m atic,
im mu ne-med iated, and id iopath ic cau ses (Table 11-1). Ischem ic d isorders that can resu lt
in CWS include retinal vascu lar d iseases, ocu lar ischem ic synd rome, and severe anem ia.
41
42 Question 11
Fig u re 11-1. Th e CWS, a lso ca lle d

a so ft e xuda te , is a sm a ll, wh itish /
gre y, clo ud-like , lin e a r o r se rpe n -
tin e , sligh tly e le va te d le sio n with
fim bria te d e dge s th a t lo o k s like it
flo a ts with in th e n e rve fibe r la ye r
o f th e re tin a .
Fig u re 11-2 . Ea rly ligh t m icro sco -

py o f CWS in th e re tin a reve a le d
th e pre se n ce o f a cyto id bo dy,
a ro un d, da rk sta in in g “n ucle us”
with in a gro ssly swo lle n n e rve
fibe r la ye r.
Em bolic d isorders that can produce focal in ner retinal ischem ia m ay arise from card iac,
carotid, or deep venou s sou rces, as well as intravenou sly injected foreign m aterials such
as talc. In add ition, wh ite cell em boli from long bone fractu res or chest trau m a can occlude
precapillary arterioles, resu lting in CWS and hemorrhages as seen in Pu rtscher’s retinopa-
thy. In fectiou s d iseases such as acqu ired im mu ne deficiency synd rom e (AIDS), Rocky
Mou ntain spotted fever, cat scratch fever (Bartonella henselae), leptospirosis, onchocercia-
sis, or d iseases that resu lt in a bacterem ia or fu ngem ia m ay also produce CWS. Retinal
m icrovascu lar occlu sions m ay occu r after the u se of interferon or other chemotherapeutic
agents and/ or rad iation exposu re, both of wh ich share a sim ilar mechan ism of action on
endothelial cells. Neoplastic d isorders such as lymphom a, leu kem ia, and metastatic car-
cinom a m ay also be associated w ith CWS. Collagen vascu lar/ im mu ne complex d iseases
such as system ic lupu s erythem atosu s or cryoglobu linem ia m ay also resu lt in occlu sion
of precapillary arterioles.
In contrast to a vascu lar occlu sion of the precapillary arteriole, m echan ical d istortion
or trau m atic laceration of the nerve fiber layer can also resu lt in the interruption of axo-
plasm ic flow and the developm ent of a CWS. Severe retinal d istortion due to epiretinal
mem brane contraction (classically seen after cryoretinop exy or scleral buckle placem ent)
can resu lt in CWS (Figu re 11-3). These CWS w ill d isappear w ith in 1 week (as opposed to
What Should I Do When I See a Patient With One or More Cotton-Wool Spots? 43
Table 11-1
Etiologies of Cotton-Wool Spots

Ischemic
a. Ocular ischemic syndrome
b. Retinal vascular occlusion
c. Anemia
d. Increased blood viscosity (eg , multiple myeloma)
e. Diabetic retinopathy
f. Hypertensive retinopathy
Embolic
a. Carotid emboli
b. Cardiac emboli
c. Deep venous emboli
d. White blood cell emboli (Purtscher’s retinopathy)
e. Severe chest compression / long bone fractures
f. Foreig n bod y emboli, often from intravenous drug abuse
Infectious
a. Acquired immune deficiency syndrome (AIDS)
b. Rocky Mountain spotted fever
c. Cat scratch fever (Bartonella henselae )
d. Leptospirosis
e. Onchocerciasis
f. Bacteremia
g. Fung emia
Toxic
a. Interferon
Radiation-induced
N eoplastic
a. Lymphoma/ leukemia
b. Metastatic carcinoma
Tractional
a. Epiretinal membrane
Traumatic
a. Nerve fiber layer laceration
Immune-mediated collagen vascular disease/immune complex disease

a. Systemic lupus erythematosus
b. Dermatomyositis
c. Polyarteritis nodosa
d. Scleroderma
e. Giant cell arteritis
Idiopathic
44 Question 11
Fig u re 11-3 . Seve re re tin a l disto rtio n

due to e pire tin a l m e m bra n e co n tra c-
tio n ca n re sult in CWS.
the 8 to 10 weeks for CWS cau sed by d iabetes or hypertension) after successfu l epiretinal
mem brane p eeling.2 In add ition, trau m atic laceration of the nerve fiber layer w ill sim ilarly
d isrupt axonal transport mechan ism s and produce accu mu lation of intracytoplasm ic
organelles at the edges of the laceration (Figu re 11-4).
Fortu nately, most patients who present w ith CWS have other system ic or ocu lar find-
ings that help narrow the sp ecific etiology. It is, however, the d iscovery of a CWS in the
otherw ise healthy patient that u sually brings w ith it con siderable consternation for both
the patient and the physician. Most of the etiologies described previou sly can be easily
removed from the d ifferential d iagnosis w ith a carefu l patient h istory, review of system s,
blood pressu re measu rement, and ophthalmoscopic exam ination. Special attention shou ld
be paid to identifying su btle signs of hyp ertensive retinopathy such as arteriolar/ venou s
n icking and generalized arteriolar narrow ing or d iabetic retinopathy such as aneu rysm al
capillary changes. Fluorescein angiography can be particu larly helpfu l in identifying
some of these changes as well as sm all m acu lar branch retinal vascu lar occlu sions.
A tru ly isolated CWS in an otherw ise healthy patient requ ires fu rther investigation
before ascribing it to id iopath ic cau ses. First, the patient shou ld be seen by h is or her
prim ary care doctor for a complete h istory and physical exam ination, keeping in m ind
the d ifferential d iagnosis list described previou sly. The prim ary care physician shou ld
be aware of the possible etiologies associated w ith CWS (see Table 11-1). Sp ecial attention
shou ld again be paid to the most com mon etiologies, such as system ic hyp ertension and
d iabetes m ellitu s. Depend ing on the patient’s card iovascu lar risk factors, a more d irected
work-up looking for card iac valvu lar d isease, carotid stenosis, or deep venou s sou rces of
arterial em boli m ay be recom m ended. Though a complete blood cou nt and platelet cou nt
are reasonable in most patients, a fu ll anticoagu lation work-up is u su ally deferred in
patients who do not have other clin ical find ings su spiciou s of hypercoagu lable states. An
eryth rocyte sed im entation rate is advisable in any patient who m ay be at risk for temporal
arteritis, regard less of the accompanying find ings. Fu rther rheu m atologic testing such as
antinuclear antibody titer or rheu m atoid factor testing is reasonable in most p ersons and
w ill help alert the physician to p ossible rheu m atologic d isorders. Finally, if no u nderlying
system ic cau se for the CWS is fou nd, regu lar close follow-up is recom m ended to in su re
that no new clin ical or retinal find ings develop.
What Should I Do When I See a Patient With One or More Cotton-Wool Spots? 45
Fig u re 11- 4 . Tra um a tic la ce ra tio n

o f th e n e rve fibe r la ye r will sim ila r-
ly disrup t a xo n a l tra n spo rt m e ch a -
n ism s a n d pro duce a ccum ula tio n
o f in tra cyto pla sm ic o rga n e lle s a t
th e e dge s o f th e la ce ra tio n .
References
1. Brown GC, Brown MM, Hiller T, Fischer D, Benson WE, Magargal LE. Cotton-wool spots. Retina.
1985;5(4):206-214.
2. Arroyo JG, Irvine AR. Retinal distortion and cotton-wool spots associated with epiretinal membrane contrac-
tion. O phthalm ology. 1995;102(4):662-668.
12 QUESTION
WHAT SHOULD I DO WHEN I SEE

A PATIENT WITH O NE OR MORE
INTRARETINAL HEMORRHAGES?
Christopher N. Singh, MD
The best way to approach the etiology of an intraretinal hemorrhage lies in the iden-
tification of the u nderlying ocu lar and system ic d isease resp onsible for the hemorrhage.
Describing the shap e of the hemorrhage (dot-blot, flam e-shap ed, Roth spot) and con firm-
ing the precise location (subhyaloid, su binternal lim iting mem brane, retinal, subretinal,
or mu ltilayered), laterality, and associated retinal find ings w ill help lead to the d iagno-
sis.
Ocu lar trau m a is a com mon cau se of intraretinal hemorrhage. Hemorrhages associated
w ith com motio retinae are likely due to trau m a if the h istory is con sistent.1 Sclop etaria,
cau sed by h igh velocity m issile inju ries passing close to the globe, is associated w ith intra-
retinal and vitreou s hemorrhages.2 Severe bod ily trau m a is associated w ith Pu rtscher’s
retinopathy, wh ich m ay m an ifest as mu ltiple cotton-wool spots in add ition to intraretinal
hemorrhage.2 Subretinal hemorrhage in the setting of trau m a is likely due to a choroid al
ruptu re.2 In the ped iatric setting, retinal hemorrhages w ith su bdu ral hemorrhages m ay
be a sign of nonaccidental trau m a.2 Retinal and vitreou s hemorrhages m ay be associated
w ith retinal breaks second ary to a closed globe inju ry, necessitating detailed peripheral
retinal exam s and B-scan u ltrasonography if there is no adequate visualization. Posterior
vitreou s detach ment in the absence of trau m a can also cau se intraretinal hemorrhages.2
A detailed m ed ical h istory is important in a patient w ith intraretinal hemorrhages.
Hemorrhages associated w ith hypertension are typically bilateral, flame-shap ed (due
to their location in the nerve fiber layer), and associated w ith narrowed arterioles, arte-
riovenou s crossing changes, and in more severe cases, cotton-wool sp ots, lipid exud ates,
and d isc edem a. Elsch n ig spots and Siegrist streaks m ay also be seen and are sign s of
hyperten sive choroidopathy.1,2 Signs of acute hyp ertensive retinopathy can be seen in
47
48 Question 12
toxem ia of pregnancy and m ay be associated w ith exud ative retinal detach m ent.
Intraretinal hemorrhages associated w ith d iabetes are bilateral, p ostequ atorial, and
dot-blot shap ed (due to their location in the nuclear layers) and can be associated w ith
cotton-wool sp ots, lipid exud ates, intraretinal m icrovascu lar abnorm alities, and venou s
bead ing.2 Sim ilar find ings are noted in rad iation retinopathy as early as 4 months after
external-beam or local plaque therapy. Intraretinal hemorrhages in ocu lar ischem ic syn-
d rom e are typically u n ilateral, located in the m idp eriphery, and dot-blot shap ed. They
m ay be associated w ith narrowed arteries and engorged but not tortuou s veins, m icroan-
eu rysm s, cotton-wool spots, hypotony, iris atrophy, and anterior cham ber cell and flare.1,2
Sickle cell retinopathy can cau se salmon patch hemorrhages, wh ich can be associated w ith
refractile (iridescent) spots, black su nbu rst lesions, and in proliferative d isease, sea fan
neovascu larization in the p eriphery.2 Hu m an im mu nodeficiency viru s (H IV) retinopathy,
the most com mon ocu lar m an ifestation of acqu ired im mu nodeficiency synd rom e (AIDS),
is characterized by intraretinal hemorrhages, cotton-wool spots, and m icroaneu rysm s.1
Intraretinal hemorrhage associated w ith retinal neovascu larization suggests retinal isch-
em ia, wh ich m ay be cau sed by d iabetic retinopathy, rad iation retinopathy, ocu lar ischem ic
synd rome, retinal arterial or venou s occlu sion, sickle cell retinopathy, Eales d isease, reti-
nopathy of prem atu rity, fam ilial exud ative vitreoretinopathy, incontinentia pigm enti, or
ch ron ic uveitis.1
Hem atologic cau ses of intraretinal hemorrhages are frequently associated w ith Roth
sp ots (wh ite-centered hemorrhages). In add ition to Roth spots, anem ia and leu kem ia can
be associated w ith flame- or dot-blot–shap ed hemorrhages. Leu kem ia m ay be associated
w ith vascu lar sheath ing and d isc edem a. Roth spots or hemorrhages in the setting of a
new heart mu rmu r or fever m ay be a sign of em bolic d isease from endocard itis. Other
d iseases associated w ith Roth spots include sickle cell anem ia, collagen vascu lar d iseases,
d iabetes, mu ltiple myelom a, and H IV.1,2
Retinal vascu lar d isease is often associated w ith intraretinal hemorrhages. Venou s
occlu sive d isease is characterized by superficial intraretinal hemorrhages, retinal edem a,
venou s d ilation and tortuosity, and m acu lar edem a.2 These signs vary dep end ing on the
severity and location of the occlu sion. A wedge-shap ed area of cotton-wool spots and
intraretinal hemorrhage m ay represent a branch vein occlu sion. Such areas involving the
superior or in ferior half of the retina represent a hem iretinal vein occlu sion. A central
vein occlu sion w ill involve both superior and in ferior halves. Vein occlu sion s are typically
u n ilateral.2 Hyp erviscosity synd rom es such as mu ltiple myelom a, Waldenström’s m acro-
globu linem ia, polycythem ia vera, and collagen vascu lar d iseases m ay be present in eyes
w ith bilateral central retinal vein occlu sions.1,2
In fectiou s d iseases can be associated w ith intraretinal hemorrhages. Cytomegaloviru s
retinopathy, seen in im mu nocomprom ised patients, is associated w ith vitritis, hemor-
rhagic retin itis, and vascu litis in the acute phase. Th is leaves an atroph ic mottled app ear-
ance once resolved.1,2 Acute retinal necrosis, associated w ith herp etic in fection s, occu rs
in im mu nocompetent patients. It is u su ally typified by the triad of vascu litis, vitritis, and
retin itis, wh ich can be associated w ith intraretinal hemorrhages.1,2 A sim ilar pictu re, pro-
gressive outer retinal necrosis, can be seen in im mu nocomprom ised patients. Vascu litis
and uveitis in progressive outer retinal necrosis are less pronou nced than in acute retinal
necrosis.2
What Should I Do When I See a Patient With Intraretinal Hemorrhages? 49
Intraretinal hemorrhages associated w ith su b- and preretinal hemorrhage can be

cau sed by m acroaneu rysm s, valsalva retinopathy, capillary hem angiom a, neovascu lar
age-related m acu lar degeneration, shaken baby synd rome, or Eales d isease.1,2 Su b-hyaloid
hemorrhages are u su ally boat-shaped, whereas su binternal lim iting m em brane hemor-
rhages are rou nd in appearance. Macroaneu rysm s can also be associated w ith lipid exu-
d ate, m acu lar edem a, and sign s of hyperten sive retinopathy.2 Isolated intraretinal hemor-
rhage close to the fovea can be a sign of early retinal angiom atou s proliferation and be
associated w ith pigm ent epithelial detach ment and choroid al neovascu larization (CN V)
in more advanced stages.2 Ju xtafoveal retinal telangiectasia, a group of id iopath ic vascu-
lar d isorders involving p erifoveal capillaries, is characterized by telangiectatic vessels bet-
ter visu alized on angiography, sm all intraretinal hemorrhages, m acu lar edem a, and lipid
exud ate and can be associated w ith retinal pigm ent epithelial hyp erplasia, crystals, and
rarely CN V. Group I w ith circinate lipid exud ate m ay resem ble Coats’ d isease.1,2
Intraretinal hemorrhages occu r in m any ocu lar and system ic d iseases. Identifying the
cau se of the intraretinal hemorrhage requ ires a carefu l h istory, exam ination, and occa-
sionally add itional ophthalm ic and system ic evalu ation.

Tam er H . Mah moud, MD, PhD
Assistant Professor, Department of Ophthalmology, Vitreoretinal Su rgery
Program Director, Vitreoretin al Fellow sh ip
Kresge Eye In stitute, Wayne State Un iversity, Detroit, MI
References
1. Benson WE, Tasman W. Retina. In: Rhee DJ, Pyfer MF, eds. The W ills Eye Manual. 3rd ed. Philadelphia, PA:
Lippincott Williams & Wilkins; 1999:33-333.
2. Retina and Vitreous. Basic and Clinical Science Course 2004-2005. San Francisco, CA: The Foundation of the
American Academy of O phthalmology; 2004:93-302.
13 QUESTION
HOW DO I WORK-UP AND

MANAGE A PATIENT WITH
A VITREOUS HEMORRHAGE?
Pauline T. Merrill, MD
When evaluating a patient w ith a vitreou s hemorrhage, the first step is to obtain a thor-
ough h istory and perform a complete ophthalm ic exam . Typical presenting symptom s
include new floaters and sudden pain less vision loss. Fu rther h istory w ill help to catego-
rize the etiology of the vitreou s hemorrhage: is it trau m atic, related to a posterior vitreou s
detach m ent, or vascu lar? Wh ile a trau m atic etiology is often obviou s from the h istory, the
clues m ay be more subtle in cases related to vitreou s traction (h istory of photopsias, h igh
myopia) or vascu lar d isease (h istory of d iabetes, hyp ertension, atherosclerosis, m acu lar
degeneration, sickle cell anem ia, other vascu lopath ies, etc).
On exam ination, visu al acu ity m ay correlate w ith the severity of the vitreou s hemor-
rhage and/ or the u nderlying pathology. The intraocu lar pressu re m ay be low due to a
retinal detach ment or h igh due to neovascu lar glaucom a and m ay in fluence the tim ing
of treatment. Important anterior segm ent find ings include neovascu larization of the iris
or angle. Dilated ophthalmoscopic exam ination of both eyes allow s you to evalu ate the
density of the hemorrhage as well as the u nderlying cau se. If there is any view to the
p eripheral retina, look for retinal breaks or detach ment w ith scleral depression. Areas of
the fu ndu s obscu red by hemorrhage shou ld be evalu ated w ith B-scan u ltrasonography.
In add ition to detecting a posterior vitreou s detach m ent, flap tear, or retinal detach ment,
u ltrasonography is also helpfu l in ru ling out u ncom mon but important etiologies such
as intraocu lar tu mors. Fluorescein angiography m ay demon strate abnorm alities such as
neovascu larization, ischem ia, or m acroaneu rysm not on ly in the fellow eye but also in
eyes w ith m ild or moderate vitreou s hemorrhage.
Treatm ent of vitreou s hem orrhage trad itionally has includ ed observation w ith
p osition ing and activity restrictions to allow settling of the hemorrhage or pars plana
51
52 Question 13
vitrectomy. More recently, pharm acologic approaches to clearing vitreou s hemorrhages

have been evalu ated. The tim ing of treatment is dependent on the u nderlying etiology
and co-morbid ities.
When vitreou s hemorrhage is due to trau m a, associated ocu lar inju ries m ay d ictate
the u rgency of vitrectomy. In the absence of sign ificant d irect ocu lar trau m a (eg, Terson’s
synd rome associated w ith intracran ial hemorrhage), in itial observation for spontaneou s
clearing is often appropriate. Becau se m any trau m a patients are often you nger, however,
the risk of am blyopia m ay be an ind ication for early vitrectomy in ch ild ren w ith vitreou s
hemorrhage.
In cases of vitreou s hemorrhage associated w ith p osterior vitreou s detach ment, over
half w ill have a retinal break, w ith almost 90% of the tears located in the sup erior qu ad-
rants.1 When p ossible, bilateral patch ing and bed rest w ith elevation of the head has been
show n to allow visualization of the superior retina w ith in a few d ays in the m ajority of
cases.2 If a retinal break or detach ment is identified and the view is adequ ate, im med iate
laser or cryopexy m ay be perform ed in the office. Otherw ise, prompt vitrectomy and/ or
scleral buckling shou ld be p erformed. In the absence of a retinal break or detach ment,
observation w ith serial B-scan u ltrasonography m ay be continued. Early vitrectomy
shou ld be considered for monocu lar persons or for eyes w ith hemolytic or ghost cell
glaucom a. Vitrectomy has also trad itionally been recom mended for ch ron ic vitreou s hem-
orrhage lasting more than 3 to 6 months. With the faster healing associated w ith sm all-
gauge vitrectomy, there is a trend toward earlier intervention.
The treatm ent approach is sim ilar for vitreou s hemorrhage due to vascu lar cau ses,
such as d iabetic retinopathy or retinal vein occlu sion. In these cases, an important add i-
tional ind ication for early intervention is neovascu lar glaucom a. If the view does not
allow adequ ate pan retinal photocoagu lation, prompt vitrectomy shou ld be considered.
Alternatively, in h ibitors of vascu lar endothelial grow th factor, such as bevacizu m ab, have
been show n to induce rapid regression of neovascu larization.3 Intravitreal bevacizu m ab
injections have also been fou nd to hasten clearing of vitreou s hemorrhage 4 and thu s
p otentially serve a du al role in eyes w ith ru beosis and vitreou s hemorrhage. One mu st
be cautiou s, however, in cases w ith extensive retinal neovascu larization and vitreoretinal
traction. In th is situ ation, intravitreal bevacizu m ab has been associated w ith rapid regres-
sion of neovascu larization and second ary progression of traction retinal detach m ent.5 If
the B-scan u ltrasou nd exam ination demonstrates sign ificant traction, I plan for vitrec-
tomy and consider a bevacizu m ab injection w ith in a few d ays preoperatively. Th is helps
to m in im ize hemorrhage both intra- and postop eratively.
Another potential pharm acologic treatment for vitreou s hemorrhage is intravitreal
hyalu ron id ase. In the Vitrase for Vitreou s Hemorrhage study, h igh ly-pu rified bovine hyal-
u ron id ase demonstrated efficacy in clearing vitreou s hemorrhage su fficiently to allow
treatment of the u nderlying pathology by 3 month s.6 Vitrase has not received Food and
Drug Ad m in istration approval for th is ind ication, however, and has not been adopted for
off-label u se as w idely as bevacizu m ab.
How Do I Work-Up and Manage a Patient With a Vitreous Hemorrhage? 53
Summary
Patients w ith vitreou s hemorrhage shou ld be evaluated carefu lly for the presence of
glaucom a, neovascu larization of the iris or angle, retinal breaks or detach m ent, and intra-
ocu lar tu mors. In the absence of these pathologies, most vitreou s hemorrhages m ay be
safely observed. When ind icated, however, pharm acologic and/ or su rgical intervention
m ay allow treatment of the u nderlying cau se and m ay also speed visu al improvement for
you r patient.
Financial disclosure: Research support from A llergan & N ovartis.

Molly Weiner, MD
Resid ent, Em ergency Med icine
Northwestern Un iversity, Ch icago, IL
References
1. Sarrafizadeh R, Hassan TS, Ruby AJ, et al. Incidence of retinal detachment and visual outcome in eyes pre-
senting with posterior vitreous separation and dense fundus-obscuring vitreous hemorrhage. O phthalm ology.
2001;108 (12):2273-2278.
2. Lincoff H, Stopa M, Kreissig I. Ambulatory binocular occlusion. Retina. 2004;24 (2):246-253.
3. Avery RL, Pearlman J, Pieramici DJ, et al. Intravitreal bevacizumab (Avastin) in the treatment of proliferative
diabetic retinopathy. O phthalm ology. 2006;113(10 ):1695.
4. Spaide RF, Fisher YL. Intravitreal bevacizumab (Avastin) treatment of proliferative diabetic retinopathy compli-
cated by vitreous hemorrhage. Retina. 2006;26(3):275-278.
5. Arevalo JF, Maia M, Flynn HW Jr, et al. Tractional retinal detachment following intravitreal bevacizumab
(Avastin) in patients with severe proliferative diabetic retinopathy. Br J O phthalm ol. 2008;92(2):213-216.
6. Kuppermann BD, Thomas EL, de Smet MD, Grillone LR; Vitrase for Vitreous Hemorrhage Study Groups.
Pooled efficacy results from two multinational randomized controlled clinical trials of a single intravitreous
injection of highly purified ovine hyaluronidase (Vitrase) for the management of vitreous hemorrhage. Am J
O phthalm ol. 2005;140 (4):573-584.
14 QUESTION
HOW CAN O PTICAL COHERENCE

TOMOGRAPHY HELP ME IN MY
PRACTICE AND WHAT CAN IT DETECT?
Elias Reichel, MD
Optical coherence tomography (OCT) has revolution ized the evalu ation of the poste-
rior p ole, particu larly the m acu la. OCT is a reproducible, rapid, non invasive tech n ique for
quantifying the arch itectu ral arrangem ent of the retina and m acu lar structu res. Un like
conventional fu ndu s photography or fluorescein angiography, OCT allow s sagittal view-
ing of the retina. The clin ician can now view the m acu lar arch itectu re from in ner to outer
retina in h igh resolution at the m icron level.
There are 5 com mon situ ations in wh ich OCT is of great utility to the general ophthal-
mologist: 1) preop erative evalu ation of the cataract patient, 2) postoperative evalu ation
of the cataract patient (when th ings aren’t progressing as plan ned), 3) evaluation of the
m acu la in patients w ith d iabetes for determ ination of m acu lar edem a, 4) classification
of m acu lar degeneration as wet or d ry, and 5) identification of m acu lar cau ses of u nex-
plained visu al loss.
A preoperative OCT is u sefu l prior to cataract su rgery to identify epiretinal m em-
branes, m acu lar holes, and cases of vitreom acu lar traction synd rome in eyes that have
m acu las that app ear to be clin ically “norm al.” Preoperative d iagnosis of these cond ition s
can prevent a lot of p ostoperative head aches! Wh ile these d iagnoses m ay not preclude
cataract su rgery, docu mentation of these cond ition s w ill allow for a d iscu ssion about
reasonable exp ectations regard ing visual potential. Add itionally, a staged approach to the
m anagem ent of the m acu lar pathology and cataract can be d iscu ssed.
With resp ect to the m anagement of the postoperative cataract patient, OCT is a qu ick
and non invasive tech n ique for determ in ing the presence of m acu lar edem a in an eye that
either doesn’t improve or where visu al acu ity gets worse. Fluorescein angiography still
rem ain s a u sefu l test in attempting to determ ine the cau se of m acu lar edem a whether it is
55
56 Question 14
Fig u re 14 -1. In tra re tin a l cystic ch a n ge s co n siste n t with cysto id m a cula r e de m a .

Mo de ra te th icke n in g o f th e re tin a is o bse rve d.
related to pseudophakic cystoid m acu lar edem a, retinal venou s obstruction, or choroid al
neovascu larization.
OCT is a u sefu l tool for screen ing patients w ith d iabetes to identify any m acu lar
edem a. Becau se it can be p erformed in an u nd ilated eye, OCT can be u sed im m ed iately
after the in itial work-up to assess whether the patient need s to have fluorescein angiog-
raphy for fu rther characterization of the m acu lar edem a. Wh ile ophthalmologists still
rely on determ in ing the presence of clin ically sign ificant m acu lar edem a u sing slit lamp
biom icroscopy, in practice, we are increasingly relying on OCT to determ ine the presence
of any m acu lar edem a. Though treatm ent protocols do not yet rely on an OCT d iagnosis of
m acu lar edem a in d iabetic patients, OCT allow s u s to determ ine those patients who m ay
be head ing for trou ble and who m ay need intervention sooner rather than later.
In those patients w ith d ry age-related m acu lar degeneration (AMD), detection of cys-
toid m acu lar edem a or su bretinal flu id m ay be best done w ith OCT. Early detection of
conversion from the d ry to the wet form of AMD is important now that effective treatment
strategies exist for wet AMD.
Finally, in patients in whom there exists an u nexplained loss of vision, OCT m ay pro-
vide in sight into processes that are d ifficu lt to detect by conventional means. Cond itions
such as cystoid m acu lar edem a, m acu lar holes, epiretinal mem branes, vitreom acu lar trac-
tion, and foveal atrophy are easily identified w ith OCT and can provide an explanation for
vision loss (Figu res 14-1 to 14-3).
OCT has revolution ized ou r approach to m acu lar pathology and provides an objective
and quantifiable way of observing m acu lar health and d isease.
How Can Optical Coherence Tomography Help Me in My Practice? 57
Fig u re 14 -2 . Epire tin a l m e m bra n e ca usin g th icke n in g o f th e m a cula , lo ss o f n o rm a l

sm o o th fove a l co n to ur a n d co rruga te d a ppe a ra n ce to th e re tin a l surfa ce .
Fig u re 14 -3 . Po ste rio r hya lo id in se rtin g in to th e m a cula re sultin g in bum py a ppe a r-

a n ce to th e ce n tra l fove a co n siste n t with vitre o m a cula r tra ctio n .
15 QUESTION
WITH THE RAPIDLY ADVANCING

TECHNOLOGY OF O PTICAL COHERENCE
TOMOGRAPHY, IS THERE STILL A ROLE
FOR FLUORESCEIN ANGIOGRAPHY?
Asheesh Tewari, MD
As optical coherence tomography (OCT) can now provide near h istologic resolution
of the retina in vivo, the cu rrent role of fluorescein angiography (FA) w ill continue to be
questioned and evolve. FA was first described in 1961, and aside from the advances in
d igital photography and im age captu re tech nology, has essentially rem ained u nchanged
for 48 years. Du ring th is time, FA had become an integral tool in the d iagnosis and m an-
agement of retinal pathology, includ ing all retinal vascu lar d iseases, d iabetic retinopathy,
and age-related m acu lar degeneration (AMD), in add ition to other retinal d iseases.
FA rem ain s a m ainstay in the evalu ation of the retinal vascu lar system . FA is a dynam ic
im aging mod ality that allow s progressive visu alization of the vascu lar components of the
eye and thu s active leakage of the blood-retinal barrier. Many retinal vascu lar d iseases,
includ ing retinal vascu lar occlu sion s, d iabetic retinopathy, and hyp erten sive retinopathy,
have u sefu l d iagnostic find ings on FA. In add ition, in flam m atory d iseases can be visu al-
ized by leakage of fluorescein in areas of vascu litis and cystoid m acu lar edem a. FA is also
an excellent method of docu menting the presence, size, and type of choroid al neovascu-
larization (CN V). In the aforem entioned d iseases, FA not on ly aid s the d iagnosis, but it
also d irectly affects patient m anagement.
FA m aintains an integral role in the d iagnosis and ongoing m anagement in AMD. FA
can d iagnose and docu m ent CN V along w ith mon itoring its activity or regression. Wh ile
most vascu lar endothelial grow th factor (VEGF) in h ibitor PRN treatment algorith m s for
59
60 Question 15
neovascu lar AMD employ OCT parameters, active leakage on FA m ay be u sed to gu ide
treatment decision s.
Optical coherence tomography provides an accu rate m easu rement of intraretinal
and su bretinal flu id along w ith being able to identify pigm ent epithelial detach m ents.
However, OCT is not a dynam ic test and can not demon strate active leakage. In add ition,
m any retina specialists cu rrently feel that a repeat FA shou ld be p erformed before the
cessation of intravitreal anti-VEGF injections even if the OCT does not demonstrate any
flu id. In the u n likely event the fluorescein angiogram demonstrates leakage despite a lack
of accu mu lated flu id on OCT, continued anti-VEGF treatment m ay be warranted. In the
PIER (Phase IIIb, Mu lti-Center, Random ized, Double-Masked, Sham Injection-Controlled
Study of the Efficacy and Safety of Ran ibizu m ab in Su bjects w ith Subfoveal Choroid al
Neovascu larization w ith or w ithout Classic CN V Second ary to AMD) trial, a mu lticenter
study of the anti-VEGF agent, ran ibizu m ab (Lucentis, Genentech, Inc, San Francisco, CA),
in eyes w ith CN V due to neovascu lar AMD, eyes that received qu arterly intravitreal injec-
tions w ith no evidence of flu id on OCT were more likely to improve their visual acu ity if
the FA also demonstrated no evidence of leakage.1
FA is also necessary to docu ment areas of retinal capillary nonperfu sion. OCT can
qu antify m acu lar edem a. However, FA can reveal the area of retinal nonp erfu sion that
is stimu lating VEGF production and that m ay requ ire photocoagu lation to perm anently
decrease the hypoxic stimu lu s. In th is example, the fluorescein angiogram allow s you to
treat the cau se, not the effect.
FA and OCT are cu rrently ind ispen sable and complementary d iagnostic and m anage-
m ent tools for the practicing retina specialist. As tech nology advances, these 2 indep en-
dent im aging mod alities m ay merge into 1 com bined d iagnostic tool. Cu rrently, system s
such as the Sp ectralis H RA+OCT (Heidelberg Engineering, Vista, CA) can captu re FA and
OCT im ages concu rrently and u se the FA im ages as a reference to the OCT scan s. Th is
facilitates precise correlation of the location of retinal pathology between FA and OCT.
Efficiency is m axim ized if the patient on ly has to sit for one test at one m ach ine.
FA w ill rem ain an essential im aging tool in the d iagnosis and m anagement of retinal
pathology. The add ition of OCT along w ith the increasing u se of w ide-field and panoram-
ic im aging tech nology w ill en hance the utility of FA in the modern retina practice.

Ch ristopher P. Majka, MD
Ophthalmology Resid ent, Kresge Eye In stitute
Wayne State Un iversity, Detroit, MI
Reference
1. Regillo CD, Brown DM, Abraham P, Yue H, Ianchulev T, Schneider S, Shams N. Randomized, double-masked,
sham-controlled trial of ranibizumab for neovascular age-related macular degeneration: PIER Study year 1. Am
J O phthalm ol. 2008;145(2):239-248.
16 QUESTION
HOW IMPORTANT IS THE PROMPT

DIAGNOSIS OF MACULAR HOLE?
William E. Smiddy, MD
A prompt and accu rate d iagnosis of a m acu lar hole is important becau se earlier d iag-
nosis and treatm ent m ay lead to a better visu al outcome. From the inception of m acu lar
hole su rgery in 1990, the preoperative du ration of the m acu lar hole was recogn ized as an
important prognostic factor in determ in ing both anatom ic and visu al success. However,
there are other u sefu l covariates that are each important, yet interrelated, prognostic
factors for an eye w ith a m acu lar hole, includ ing m acu lar hole size and the presenting
visu al acu ity. To con fou nd the situ ation, patients are often u naware of exactly when the
symptom s began and of the resu lting visual deficit.
New tech nological advances in ophthalm ic im aging have allowed for an earlier and
more accu rate d iagnosis of a m acu lar hole. The introduction of optical coherence tomog-
raphy (OCT) has been a great improvem ent over the clin ical exam ination for the d iag-
nosis of m acu lar hole. Wh ile OCT is an extremely imp ortant and powerfu l tool, I th in k
that it is important to m aintain a healthy skepticism of its lim itations and interpret the
OCT find ings in the context of the clin ical h istory and exam ination. OCT can detect the
earliest stages of m acu lar hole form ation, even at what app ears to be a pre-hole stage. It
is u nclear at th is time whether earlier su rgical intervention at pre-hole stages of m acu lar
hole form ation is beneficial. Traction in m any such apparently pre-hole cases m ay spon-
taneou sly abort, and there m ay never be progression to a fu ll-th ickness m acu lar hole. The
govern ing principle shou ld be to intervene on ly when the visu al acu ity has had a fairly
d iscern ible and recent deterioration attributable to the m acu lar symptom s or once an
u nequ ivocal fu ll-th ickness m acu lar hole develops.
61
62 Question 16
Are All Macular Holes Treated With Vitrectomy?

I do not th in k there is any other feasible option for the treatment of a fu ll-th ickness
m acu lar hole. Wh ile there have been tech n iques of pneu m atic m acu lopexy reported, these
have not gained w idespread acceptance, probably becau se of their lack of clear efficacy.
A question in my m ind is what featu res of a fu ll-th ickness m acu lar hole render such a
poor prognosis that an operation is not warranted. Associated pathologic featu res such as
m arked retinal pigment epitheliu m (RPE) atrophy, especially w ith h igh myopia, and large
size of the m acu lar hole are poor prognostic featu res that wou ld tend to d iscou rage su rgi-
cal intervention. Ch ron ic m acu lar holes w ith a du ration over 18 months m ight preclude
an attempt at m acu lar hole su rgery, u n less the hole is u nu su ally sm all or the recent visu al
acu ity has been u nu su ally good.
Early stages of m acu lar hole form ation where the hole is not yet fu ll th ickness m ay
or m ay not be treated w ith vitreou s su rgery, and the decision to intervene at these early
stages varies among su rgeon s. Most, however, do not intervene u n less the m acu lar hole
is fu ll th ickness on OCT.
When Should Cataract Surgery Be Considered

After Macular Hole Surgery?
There is no factu al basis for an swering th is question other than the general principle
that by con sensu s we seem to typically wait at least 3 months between retinal interven-
tions and non retinal su rgical interventions, when p ossible. Usually th is is not an issue
since visually sign ificant cataracts take 6 to 12 months to m an ifest follow ing vitreou s
su rgery. There is no way to be assu red of the perm anence of m acu lar hole closu re.
Cataract su rgery has been implicated by some to possibly increase the risk of reop en ing
a previou sly closed m acu lar hole, but th is is probably a more coincidental and anecdotal
find ing rather than true cau se and effect. There is an approxim ately 5% rate of reopen ing
of a m acu lar hole over time. I recom mend cataract extraction whenever it seem s clin ically
ind icated, w ithout concern over reopen ing a previou sly closed m acu lar hole.
17 QUESTION
WHEN SHOULD I REFER A PATIENT

WITH AN EPIRETINAL MEMBRANE
AND WHAT IF THERE IS ASSOCIATED
CYSTOID MACULAR EDEMA?
Gaurav K. Shah, MD
An epiretinal mem brane or m acu lar pucker is a cellu lar avascu lar mem brane overly-
ing the m acu la that m ay lead to central visu al d istu rbances. The Beaver Dam Eye Study
fou nd that about 12% of the popu lation aged 43 to 84 years had epiretinal m em branes.1
Another study by Jah n and colleagues fou nd the prevalence of epiretinal m em branes to
be 9% in those u ndergoing cataract su rgery.2 Wh ile most epiretinal m em branes are id io-
path ic, inciting factors include previou s su rgery or trau m a, retinal break or detach m ent,
uveitis, intraocu lar tu mor, d iabetic retinopathy, and retinal vein occlu sion. The m ajor-
ity of eyes w ith id iopath ic epiretinal mem branes have a posterior vitreou s detach ment.
However, th is is not a prerequ isite for mem brane developm ent. A complete ophthalmo-
logic exam ination is necessary to exclude second ary cau ses such as an intraocu lar tu mor
or retinal break, wh ich requ ire im med iate attention.
Epiretinal mem branes are often asymptom atic in the early stages, and the m ajority
of these mem branes rem ain stable after an in itial p eriod of grow th. Progression, how-
ever, m ay lead to sign ificant visual impairment, metamorphopsia, m acropsia, d iplopia,
and even an iseikon ia (Figu re 17-1). It is especially important to ask these patients about
symptom s of visual d istortion, wh ich can occu r in the setting of norm al visual acu ity. In
m any cases, Snellen visual acu ity m ay not be con sistent w ith the patient’s symptom atic
complaints, and th is m ay be more evident in patients w ith professions that dem and h igh
visu al qu ality (pilots, physician s, nu rses, engineers, etc).
63
64 Question 17
Fig u re 17-1. Th is pa tie n t pre se n te d

with de cre a se d visio n a t 20 / 70 a n d
m e ta m o rp h o p sia . Op h th a lm o sco p ic
e xa m in a tio n re ve a le d a n e pire tin a l
m e m bra n e with va scula r tra ctio n a n d
a sso cia te d m a cula r e de m a .
Fluorescein angiography is a valu able d iagnostic tool in detecting epiretinal mem-

branes, associated m acu lar edem a, and vascu lar leakage from second ary d istortion of
m acu lar vessels. The leakage on fluorescein angiography correspond s to the area of the
retina covered by the mem brane and is u su ally irregu lar (Figu re 17-2). Fluorescein angi-
ography can also be sign ificant in revealing an epiretinal m em brane or pseudom em brane
associated w ith m acu lar hole, m acu lar ischem ia, or choroid al neovascu larization.
Optical coherence tomography (OCT) is a non invasive method for detecting both
epiretinal mem branes and associated m acu lar edem a (Figu re 17-3). Becau se it has been
show n to detect less than half of epiretinal mem branes noted on biom icroscopic exam i-
nation, previou s generation OCTs m ay not be the defin itive d iagnostic tool. The spectral
dom ain OCTs, w ith their better resolution, m ay prove more reliable in detecting epiretinal
m em branes on a consistent basis and m ay allow u s to determ ine and potentially pred ict
postop erative visu al outcom es in patients u ndergoing epiretinal m em brane su rgery.
When Do I Refer a Patient With

an Epiretinal Membrane?
Reassu rance, an Am sler grid for home mon itoring, and period ic follow-up are su fficient
for a patient w ith an id iopath ic epiretinal mem brane and m in im al visual symptom s. If the
patient is troubled by the visual symptom s cau sed by the epiretinal mem brane, referral
to a vitreoretinal su rgeon is the next step. In the past, su rgery was u sually reserved for 2
categories of patients: 1) visu al acu ity of 20/ 60 or less, or 2) those patients that requ ired
excellent visu al acu ity or a h igh degree of stereopsis. Improvem ents in su rgical tech n ique,
intraoperative epiretinal m em brane visu alization, and reduced complications now allow
for earlier vitrectomy and removal of the epiretinal mem brane. Preoperative visu al acu ity
typically correlates w ith postop erative visu al acu ity.3 A th in m em brane, decreased du ra-
When Should I Refer a Patient With an Epiretinal Membrane? 65
Fig u re 17-2 . Fluo re sce in a n gio gra ph y

co n firm e d th e e pire tin a l m e m bra n e
a n d se co n da ry m a cula r e de m a .
Fig u re 17-3 . OCT o f th e sa m e pa tie n t furth e r h igh ligh te d th e e pire tin a l m e m bra n e a n d ta n ge n -
tia l fove a l tra ctio n .
tion of symptom s, and the absence of traction are all factors lead ing to a better postopera-
tive prognosis. In a study of 125 eyes follow ing vitrectomy and peeling of visually sign ifi-
cant epiretinal mem branes, visu al acu ity improved by a mean of 3 lines, and 93% rep orted
improvement in visual fu nction, especially reduction of d istortion.4 Unsu rprisingly,
restoration of norm al foveal arch itectu re is correlated w ith sign ificant improvements in
d istortion as compared w ith patients whose arch itectu re rem ain s u nchanged.
Patients that u ndergo removal of the epiretinal mem brane shou ld be aware that com-
plications of su rgery m ay include increased intraocu lar pressu re, hemorrhage, retinal
detach m ent, and especially cataract progression. Recu rrence of the epiretinal mem brane
has been previou sly reported to occu r in approxim ately 10% of patients. More recently,
however, Schad lu and colleagues fou nd that when the internal lim iting mem brane was
removed in conju nction w ith peeling of the epiretinal mem brane, there was recu rrence in
on ly one of 38 patients at the 20-month follow-up.5
66 Question 17
What if There Is Associated Cystoid Macular Edema?

Retinal traction and vascu lar d istortion cau sed by an epiretinal mem brane m ay lead
to m acu lar edem a. In most cases, su rgical removal of the traction resu lts in resolution of
the m acu lar edem a. Preop erative m acu lar edem a and retinal pigment epithelial changes
occu rring as a resu lt of long-stand ing retinal traction or vascu lar leakage are thought
to be poor prognostic factors for eventu al postoperative visual recovery. Consequently,
earlier su rgical intervention in patients w ith m acu lar edem a m ay lead to better visu al
outcom es. In some patients w ith ch ron ic m acu lar edem a, lipid exud ates are also present.
In an older patient presenting w ith these find ings, it is imperative to exclude the possibil-
ity of m acu lar edem a and lipid exud ates associated w ith choroid al neovascu larization
from age-related m acu lar degeneration. In those patients w ith an epiretinal mem brane
and visu ally sign ificant cataract, nonsteroid al anti-in flam m atory d rops shou ld be pre-
scribed both before and after cataract su rgery to reduce the risk of associated m acu lar
edem a. If m acu lar edem a is noted after cataract su rgery in a patient w ithout preop erative
epiretinal mem brane, fluorescein angiography w ill prove especially helpfu l in d ifferenti-
ating between m acu lar edem a associated w ith postop erative cataract su rgery versu s that
resu lting from vascu lar leakage second ary to the epiretinal mem brane.

Argh avan Almony, MD
Barnes Retin a In stitute
Clin ical In structor, Department of Ophthalmology and Visu al Sciences
Wash ington Un iversity School of Med icine, St. Lou is, MO
References
1. McDonald HR, Johnson RN, Ai E, et al. Macular epiretinal membranes. In: Ryan SJ, Wilkinson CP, eds. Retina.
4th ed. Philadelphia, PA: Elsevier Mosby; 2006:2509-2525.
2. Jahn CE, Minich V, Moldaschel S, et al. Epiretinal membranes after extra-capsular cataract surgery. J Cataract
Refract Surg. 2001;27(5):753-760.
3. Thompson JT. Epiretinal membrane removal in eyes with good visual acuities. Retina. 2005;25(7):875-882.
4. Wong JG, Sachdev N, Beaumont PE, Chang AA. Visual outcomes following vitrectomy and peeling of epireti-
nal membrane. Clin Experim ent O phthalm ol. 2005;33(4):373-378.
5. Schadlu R, Tehrani S, Shah GK, Prasad AG. Long-term follow-up results of ILM peeling during vitrectomy
surgery for premacular fibrosis. Retina. 2008;28 (6):853-857.
18 QUESTION
HOW DO I DIFFERENTIATE A MACULAR

HOLE FROM A LAMELLAR HOLE,
AN EPIRETINAL MEMBRANE WITH A
PSEUDOHOLE, CYSTOID MACULAR
EDEMA, AND VITREOMACULAR
TRACTION, AND WHY DO I CARE?
SriniVas R. Sadda, MD
Optical coherence tomography (OCT) has become a critical tool in d ifferentiating a

m acu lar hole from other d isorders of the vitreoretinal interface.1 In add ition to provid ing
h igh-resolution cross-sectional im ages of the neu rosensory retina, OCT is also capable of
demon strating the relationsh ip between the posterior hyaloid and the retina—a relation-
sh ip that is d ifficu lt to assess by clin ical exam ination alone. In fact, when OCT im aging
was first introduced in the early 1990s, the ability to d istingu ish sim ilar, yet d istinct, con-
d itions such as m acu lar hole, lamellar hole, epiretinal mem brane (ERM) w ith pseudohole,
cystoid m acu lar edem a (CME), and the vitreom acu lar traction synd rome was among the
first really u sefu l clin ical applications of the new tech nology.
Prior to the introduction of OCT, there were m any hypotheses but on ly lim ited evi-
dence regard ing the pathogenesis of m acu lar hole and associated d isorders. Wh ile much
rem ains to be learned, it is now clear that the earliest change in these cond itions app ears
to be perifoveal vitreou s detach m ent and the development of vitreom acu lar traction (ie,
the vitreou s partially separates from the m acu la wh ile rem ain ing attached at the fovea).
Th is traction resu lts in structu ral changes at the foveal center and u ltim ately the form a-
tion of a m acu lar hole or related cond ition.
67
68 Question 18
Vitreomacular Traction Syndrome

and Cystoid Macular Edema
Although vitreom acu lar traction plays a sign ificant role in d isorders of the vitreo-
retinal interface, the vitreom acu lar traction synd rom e is a d istinct clin ical entity. In th is
synd rome, perifoveal vitreou s detach m ent leaves a sm all, focal vitreofoveolar ad hesion,
resu lting in an anteroposterior tractional force and u ltim ately the development of CME
and other structu ral changes.1 Th is synd rome can easily be con fu sed clin ically w ith CME
occu rring in the context of uveitis or in a postoperative setting. The d iagnosis m ay be
much more easily m ade u sing OCT. The p osterior hyaloid u su ally appears hyperreflective
and the tractional attach ment at the foveal center m ay be clearly seen (Figu re 18-1). Several
OCT stud ies have show n that su rgical separation of th is vitreofoveal ad hesion facilitates
resolution of m acu lar th icken ing w ith improvement in visu al acu ity in these patients.
Full-Thickness Macular Holes

Fu ll-th ickness m acu lar hole (FTMH ) form ation is age related, typically occu rring in
p ost-menopau sal wom en. In a sign ificant percentage of cases, the vitreom acu lar traction
in early stage (Stage 1) holes resolves spontaneou sly w ithout hole form ation and, there-
fore, accu rate staging of th is d isorder is crucial to determ in ing the appropriate tim ing
for su rgical intervention. A nu m ber of authors have u sed OCT to classify the sequence
of events in m acu lar hole form ation, begin n ing w ith p erifoveal vitreou s detach ment.1 No
stand ard classification has been firm ly established, although the m ain points of each are
broad ly sim ilar. In Stage I, there is pseudocyst form ation in the in ner retina, wh ich some-
times m ay be seen clin ically as a yellow dot or ring. In Stage II, there is early separation
of the outer photoreceptor layer and, consequently, sm all fu ll-th ickness loss of retinal tis-
sue. Stage III occu rs when the hole is > 400 µm in d iam eter, w ith su rrou nd ing th ickened
retina includ ing intraretinal cystoid spaces. Stage IV holes con sist of Stage III holes w ith
a complete posterior vitreou s detach m ent (Figu re 18-2). Prior to the advent of OCT, the
su bjective d isappearance of a p ortion of slit beam of light (Watzke-Allen test) or a laser
aim ing beam projected into the hole was a u sefu l m ethod for d istingu ish ing a FTMH
from other vitreoretinal interface d isorders.
Lamellar Holes
The term lamellar hole was first suggested by Donald Gass as a complication of ch ron ic
CME follow ing cataract extraction. It is now more com mon ly thought to represent an
aborted process in the spectru m of m acu lar hole form ation (ie, a lamellar hole occu rs when
vitreom acu lar traction cau ses “u n roofing” of a foveal pseudocyst w ithout loss of the outer
photoreceptor layer). OCT im aging shows an irregu lar th in ned foveal contou r, a break in
the in ner fovea, separation of the in ner and outer foveal layers, and the absence of a fu ll
thickness defect (Figu re 18-3).2 Most patients w ith this cond ition complain of m ild symp -
tom s of metamorphopsia and m ay have on ly lim ited central vision loss. Typically, these
patients rem ain stable and on ly rarely progress to more severe levels of visu al deficit.
How Do I Differentiate a Macular Hole From a Lamellar Hole? 69
Fig u re 18 -1. OCT im a ge

o f vitre o m a cu la r tra c-
tio n . No te fo ca l a re a o f
fove a l a dh e sio n o f a pa r-
tia lly de ta ch e d po ste rio r
h ya lo id with a sso cia te d
CME.
Fig u re 18 -2 . OCT im a ge
o f a Sta ge IV FTMH. Th e
de fe ct in vo lve s a ll la ye rs
o f th e re tin a , with cysto id
ch a n ge s in th e swo lle n ,
ro un de d e dge s. Th e re is
a co m ple te po ste rio r vit-
re o us de ta ch m e n t.
Fig u re 18 -3 . OCT im a ge o f a
la m e lla r h o le . No te th e fra g-
m e n t o f “u n ro o fe d ” re tin a
a dh e re n t to th e de ta ch e d po s-
te rio r hya lo id. On ly a ve ry th in
la ye r o f re sidua l re tin a l tissue
re m a in s a t th e ba se o f th e h o le
a n d distin g uish e s th is le sio n
fro m a full-th ickn e ss h o le . No te
th e a bse n ce o f roun de d, swo l-
le n m a rgin s, wh ich a re typica lly
pre se n t in a full-th ickn e ss h o le .
Epiretinal Membrane and Pseudoholes

Centripetal contraction of an ERM can lead to form ation of a d istinct clin ical entity, the
p seudohole.1 Pseudoholes m ay have a sim ilar clin ical app earance to a FTMH or a lamellar
hole. The key d ifference is that there is no loss of foveal tissue in pseudoholes. OCT char-
acteristics of a pseudohole include: steepened foveal pit w ith th ickened edges, reduced
foveal pit d iameter, and norm al or slightly increased foveal th ickness. An epiretinal
mem brane m ay also be observed on OCT (Figu re 18-4). Even w ith the u se of OCT, it m ay
be d ifficu lt to d istingu ish between a pseudohole and a lamellar hole (recent stud ies u sing
u ltrah igh resolution OCT have demon strated a h igh prevalence of ERM accompanying
lamellar holes). Like lamellar holes, pseudoholes tend to be relatively stable cond ition s,
and the visu al acu ity of affected patients m ay be relatively preserved.
70 Question 18
Fig u r e 18 - 4 . Sp e ctra l
do m a in OCT im a ge o f a n
ERM. No te th e co rruga -
tio n o f th e re tin a l surfa ce
in lo ca tio n s wh e re th e
re tin a is fo ca lly de ta ch e d
fro m th e ERM. No pse u-
do h o le is e vide n t o n th is
sca n .
Why Do I Care?
It is imp ortant to d ifferentiate m acu lar holes from lam ellar holes and pseudoholes
becau se the m anagem ent is d ifferent in each case. Vitrectomy su rgery and gas tamp on-
ade are well-established and successfu l treatm ent option s for FTMH s. Sim ilarly, ERM
p eeling m ay be beneficial for the treatment of visu ally sign ificant p seudoholes. On the
other hand, treatment of patients w ith sign ificant visu al loss from a lamellar hole rem ain s
controversial, w ith m any reports noting adverse outcomes such as progression to FTMH
form ation. Although successfu l outcomes after vitrectomy have also been rep orted in
such cases,3 caution is still requ ired.

Pearse Keane, MD
Clin ical Research Fellow, Doheny Eye In stitute, Keck School of Med icine
Un iversity of Southern Californ ia, Los Angeles, CA
References
1. Mirza RG, Johnson MW, Jampol LM. O ptical coherence tomography use in evaluation of the vitreoretinal
interface: a review. Surv O phthalm ol. 2007;52(4):397-421.
2. Witkin AJ, Ko TH, Fujimoto JG, et al. Redefining lamellar holes and the vitreomacular interface: an ultrahigh-
resolution optical coherence tomography study. O phthalm ology. 2006;113(3):388-397.
3. Garretson BR, Pollack JS, Ruby AJ, Drenser KA, Williams GA, Sarrafizadeh R. Vitrectomy for a symptomatic
lamellar macular hole. O phthalm ology. 2008;115(5):884-886.
19 QUESTION
HOW SHOULD PATIENTS WITH

A G AS BUBBLE POSITION THEIR
HEAD AND WHAT PRECAUTIONS
SHOULD THEY FOLLOW?
Abdhish R. Bhavsar, MD
There are 3 m ain types of intraocu lar gas u sed in eyes w ith vitreoretinal pathology:
air, su lfu r hexafluoride (SF6), and perfluoropropane (C 3F8). Wh ile there are no specific
gu idelines for the type or concentration of gas to be u sed or the typ e and du ration of
p ostoperative head position ing to treat variou s vitreoretinal pathologies, there are a few
general principles that gu ide these decision s.
The decision to place a gas bubble into an eye is based on mu ltiple factors. The most
com mon reasons to u se a gas bu bble are to tamponade a retinal break, repair a retinal
detach ment, or repair a fu ll-th ickness m acu lar hole. The time frame requ ired for the gas
bubble to rem ain in contact w ith such pathology depend s on the pathology itself, and th is
in fluences the type and concentration of gas u sed. In general, in a nonvitrectom ized eye
u nd ergoing pneu m atic retinop exy or scleral buckling su rgery, up to 0.6 m L of sterile air,
0.6 m L of 100% SF6, or 0.3 m L of 100% C 3F8 can be u sed. In a vitrectom ized eye, sterile
air can be u sed for a sup erior intraoperative retinal break. For some acute retinal detach-
m ents and m acu lar holes, a nonexpansile concentration of a shorter-acting gas bu bble,
such as 20% SF6, m ay be chosen. For som e m acu lar holes, giant retinal tears, in ferior reti-
nal detach m ents, or complex or ch ron ic retinal detach ments, particu larly w ith prolifera-
tive vitreoretinopathy, a nonexpansile concentration of a longer-acting gas bu bble, such
as 14% C 3F8, is preferred by most su rgeons. The gas bu bble need s to appose the specific
pathology.
71
72 Question 19
The longevity of gas w ith in the vitreou s cavity varies and dep end s on the type of gas
u sed, concentration of gas, volu m e of gas, volu me of the vitreou s cavity, and lens statu s of
the eye.1 Phakic eyes allow the gas to rem ain longer in side the vitreou s cavity. The longev-
ity of the gas bu bble varies among patients but is typically about 3 to 5 d ays for air, 2 to 3
weeks for SF6, and approxim ately 6 to 8 weeks for C 3F8.
Optim izing the gas fill in eyes u ndergoing vitreou s su rgery requ ires a complete vitrec-
tomy. If su bstantial vitreou s rem ains in the eye at the end of the su rgery, then there cou ld
be an u nder fill of gas p ostop eratively becau se the rem ain ing vitreou s occupies volu m e in
the vitreou s cavity. After the air-flu id exchange, con sider waiting several m inutes for the
rem ain ing flu id w ith in the residu al vitreou s to trickle dow n to the posterior pole so that
it can be removed. Th is cou ld improve the postoperative gas fill by as much as 15%.2
Patient Positioning
RETINAL DETACHMENT
In nonvitrectom ized eyes w ith a retinal detach m ent repaired by pneu m atic retinop exy
or scleral buckling su rgery w ith gas bubble placem ent, I prefer that the m acu la reat-
taches im med iately postoperatively. Position ing the patient face-dow n im m ed iately after
su rgery for at least several hou rs allow s any residu al subretinal flu id to flow out of the
su bm acu lar space (Figu re 19-1). Follow ing th is, position ing the head opposite the location
of the retinal break w ill allow the gas to d irectly tamponade that break. Patients shou ld
m aintain th is position for 90% of the tim e for the ensu ing 3 to 5 d ays.
In retinal detach ment cases repaired by vitreou s su rgery w ith breaks in the sup erior
6 clock hou rs, I typically instruct patients to m aintain face-dow n position ing for at least
1 d ay after su rgery. As long as there is a relatively complete gas fill, upright position-
ing allow s the gas bu bble to tamponade the retinal breaks after the first d ay. For retinal
detach m ents w ith breaks in the in ferior 6 clock hou rs, I typically in struct patients to
m aintain face-dow n position ing for 5 d ays. Th is is esp ecially important for in ferior retinal
breaks. Between 1 and 2 weeks, the retinopexy has an increased retina-retinal pigment
epitheliu m ad hesion due to scar m atu ration, therefore m aintain ing some sort of face-
dow n p osition for at least 5 to 7 d ays is advisable. If the gas fill were su boptim al (less than
85%), then I wou ld recom m end extend ing the p ostoperative position ing requ irements by
several d ays. Add itional gas can be placed into the vitreou s cavity in the office. I instruct
patients not to sleep on their backs but rather to sleep on their sides u ntil the bu bble is
gone so it does not contact the crystalline lens and lead to increased cataract form ation or
a potentially h igher risk of peripheral anterior synech iae in pseudophakic patients.
MACULAR HOLE
Among vitreoretinal su rgeons, the face-dow n position ing recom mended after m acu lar
hole su rgery has varied substantially from no position ing at all to 3 weeks of face-dow n
p osition ing. In most series, 1 week of face-dow n position ing yield s m acu lar hole closu re
rates well over 90%.3 With less than 1 week of position ing and particu larly w ith less than
3 d ays of position ing, the closu re rates can fall below 90%. Without any face-dow n posi-
tion ing but w ith the long-acting C 3F8 gas, closu re rates have been rep orted between 64%
and 79%.4,5 With respect to the choice of gas, I u se 28% SF6. Th is yield s an 85% to 90%
How Should Patients With a Gas Bubble Position Their Head? 73
Fig u re 19 -1. Cro ss se ctio n th ro ugh th e

e ye de m o n stra tin g th e a ppo sitio n o f
th e ga s bubble a ga in st th e po ste rio r
po le a n d m a cula durin g fa ce do w n
po sitio n in g.
bubble on the first postop erative d ay. By the first postoperative week, there is typically a
60% gas bu bble rem ain ing. By 2 weeks postoperatively, the bubble is gone.
I requ ire patients to m aintain face-dow n position ing for 1 week. I believe a critical point
is that patients shou ld not be in the supine p osition du ring the first postoperative week.
Du ring the second p ostoperative week, I typically ask patients to sleep on their sides at
bedtime and look dow n for 15 m inutes 4 tim es a d ay to allow the aqueou s to bathe the
posterior su rface of the lens. However, they are free to be upright and to resu m e norm al
activities after the first postoperative week.
EPIRETINAL MEMBRANE SURGERY

The are a few circu m stances when I wou ld requ ire position ing after epiretinal m em-
brane su rgery. Those include patients w ith severe CME preoperatively, those w ith a partial
or fu ll-th ickness m acu lar hole that form s du ring epiretinal or internal lim iting mem brane
(ILM) d issection, or those w ith an iatrogen ic tear du ring the vitrectomy su rgery. I am
concerned that patients w ith pre-existing severe CME m ay be pred isposed to developing
a m acu lar hole postoperatively after epiretinal mem brane su rgery w ith or w ithout ILM
d issection. My hope is that the placement of a 28% SF6 gas bubble w ith some position ing
face dow n, for about 3 d ays, w ill help to reduce the risk of m acu lar hole form ation. In
the case of a partial or fu ll-th ickness m acu lar hole that develop s at the time of epiretinal
m em brane or ILM d issection, I treat the case as if there was a pre-existing m acu lar hole.
DIABETIC RETINOPATHY
I typically place a gas bu bble in eyes that have u ndergone vitrectomy w ith mem brane
d issection for tractional retinal detach ments, even if I do not believe that there is a tear
in the retina. I u su ally find that there seem s to be a benefit of gas tamponade of any m ild
oozing of red blood cells from the fragile retinal vessels after su rgery. In add ition, the gas
bubble w ill prevent retinal detach ment from any occu lt iatrogen ic retinal tears or holes
that m ay have been created du ring the m em brane d issection or delam ination. I u su ally
u se 28% SF6 gas for relatively lim ited tractional detach m ents and 14% to 18% C 3F8 for
complex tractional, com bined tractional/ rhegm atogenou s detach ments, or cases w ith
74 Question 19
mu ltiple in ferior retinal tears. I u sually recom mend face-dow n p osition ing for 3 d ays fol-
lowed by upright position ing du ring the d ay and sleeping on the sides at bedtime.
Th is position ing on the sides wh ile sleeping continues for about 1.5 weeks for eyes w ith
SF6 and for about 6 weeks for eyes w ith C 3F8. I u su ally try to tamp onade the side where
the m ajority of the retinal tears are located du ring the bedtime position ing.
I ATROGENIC RETINAL TEARS

I always place a gas bu bble in eyes w ith iatrogen ic retinal tears and if the iatrogen ic
tear is the on ly pathology requ iring the gas bu bble, then I always u se 20% to 28% SF6.
For superior retinal tears w ithout a retinal detach ment, I requ ire patients to position
face-dow n or on the opposite side of the retinal tear overn ight after su rgery followed by
norm al position ing du ring the d aytime. At bedtime, I requ ire patients to position so that
the bu bble tamponades the m ajor iatrogen ic tear(s). Th is continues at bedtim e for 1 to 2
weeks postop eratively.
Precautions With an Intraocular Gas Bubble

Wh ile a gas bu bble is present in side the eye, patients shou ld avoid traveling to h igh
altitudes and flying in airplanes. The lower atmospheric pressu re at h igh altitudes
lead s to expansion of the intraocu lar gas bu bble in the closed globe, wh ich can resu lt in
extremely elevated intraocu lar pressu re and central retinal artery occlu sion.6 Even sm all
gas bu bbles can resu lt in substantial intraocu lar pressu re rise and loss of vision.7,8 I rec-
om m end against air or h igh-altitude travel for any patient w ith an intraocu lar gas bubble.
Diving, particu larly scuba d iving, cou ld also resu lt in contraction and expan sion of the
intraocu lar gas bu bble and is not recom m ended. Cataract su rgery in eyes w ith an intra-
ocu lar gas bubble is not advised since the gas bu bble can exert flotation forces against the
lens and cou ld complicate cataract su rgery.
References
1. Michels RG, Wilkinson CP, Rice TA. Retinal Detachm ent. St. Louis, MO : Mosby; 1990.
2. Rubin JS, Thompson JT, Sjaarda RN, Pappas SS Jr., Glaser BM. Efficacy of fluid-air exchange during pars plana
vitrectomy. Retina. 1995;15(4):291-294.
3. Bhavsar AR, Gilbert HD. Macular hole surgery with 1 week face down positioning and SF6 gas. Poster pre-
sented at the Vitreous Society 17th Annual Meeting, Rome, Italy, September, 1999.
4. Szurman P, Di Tizio FM, Lafaut B, et al. Stellenwert der postoperativen Positionierung in der Chirurgie des
idiopathischen Makulaforamens--Kontrollierte konsekutive Studie [Significance of postoperative face-down
positioning after surgery for idiopathic macular holes: consecutive case-control study]. Klin Monatsbl
Augenheilkd. 2000 ;217(6):351-355.
5. Tornambe PE, Poliner LS, Grote K. Macular hole surgery without face-down positioning: a pilot study. Retina.
1997;17(3):179-185.
6. Fang IM, Huang JS. Central retinal artery occlusion caused by expansion of intraocular gas at high altitude.
Am J O phthalm ol. 2002;134 (4):603-605.
7. Gandorfer A, Kampik A. Expansion intraokularer gase infolge reduktion des atmospharendruckes. kasuistik
und literaturubersicht [Expansion of intraocular gas due to reduced atmospheric pressure]. O phthalm ologe.
2000 ;97(5):367-370.
8. Mills MD, Devenyi RG, Lam WC, Berger AR, Beijer CD, Lam SR. An assessment of intraocular pressure rise
in patients with gas-filled eyes during simulated air flight. O phthalm ology. 2001;108 (1):40 -44.
20 QUESTION
WHAT IS CENTRAL SEROUS RETINOPATHY

AND HOW DO I TREAT IT?
Jennifer I. Lim, MD
Central serou s retinopathy (CSR) is classically described as a relatively ben ign d isease
of the m acu la that predom inantly affects you nger persons, often m ales. However, in my
experience, there are a fair nu m ber of older ind ividu als who present w ith CSR as well as
some person s w ith more severe form s of the d isease. The clin ical find ings shou ld gu ide
d iagnosis, and the d isease severity shou ld gu ide the decision to treat.
Several cond itions associated w ith the d iagnosis of CSR include exogenou s cortico-
steroid u sage, pregnancy, and type A personality. Thu s, the clin ical setting is helpfu l.
Laboratory stud ies have show n h igh levels of catecholam ines in CSR patients.1 Typ e
A p ersonality has been associated w ith th is cond ition and perhaps such patients have
increased catecholam ine and other stress steroid output. When I see a patient w ith CSR, I
inqu ire about h is or her u se of corticosteroid s (oral, nasal, topical), state of pregnancy (in
fem ales), and recent stresses or m ajor life events.
Alteration s in choroid al p ermeability m ay resu lt in retinal pigm ent epithelial (RPE)
d am age. The clin ical pictu re in an acute case is typically one of serou s retinal detach-
m ent, pigm ent epithelial detach ment (PED), or a com bination of both find ings. Rarely,
eyes w ith CSR m ay have an exud ative retinal detach m ent. In the latter case, the scenario
is u su ally a patient w ith no other risk factors for an exud ative detach m ent who m ay be on
corticosteroid s. In those cases, an u ltrasou nd evalu ation is helpfu l to ru le out choroid al
th icken ing or a m ass lesion.
Ch ron ic CSR is one su bset of CSR and can resu lt in d iffu se RPE mottling w ith sign ifi-
cant pigm entary d isruption. Th is can occu r bilaterally. Clues to the d iagnosis are gutter-
like areas of pigmentation (Figu re 20-1), d iffu se areas of RPE d isruption, and tracks of
pigment or mottling in the perifoveal area. Extensive areas of leakage m ay cau se sign ifi-
cant visu al morbid ity.
75
76 Question 20
Fig u re 2 0 -1. Gutte r-like a re a s o f

pigm e n ta tio n m a y de ve lo p in
a n eye with CSR.
Another subset of CSR is that of a su bretinal fibrinoid reaction (Figu re 20-2). Th is is

typically fou nd in CSR associated w ith pregnancy and corticosteroid s.2 Pregnancy associ-
ated CSR is a self-lim ited cond ition that resolves arou nd the tim e of delivery. In ch ron ic
corticosteroid u sage, severe CSR w ith subretinal fibrosis and band ing can occu r. Ideally,
corticosteroid s shou ld be tap ered before th is occu rs.
The typical cou rse of id iopath ic CSR is self-lim ited. Patients u sually note blu rred
central or pericentral vision, m etamorphop sia, or central relative scotom as. Sometimes a
patient presents when the su bretinal flu id is clin ically gone, but the vision is still blu rred
or colors app ear less satu rated. In these situ ation s, I reassu re the patient that resolution
of visu al symptom s lags beh ind the clin ical resolution. Optical coherence tomography
(OCT) im aging m ay show subclin ical flu id in some eyes when the clin ical exam show s
no flu id. I let the patient know that h is or her vision m ay not completely retu rn to norm al
and that colors m ay perm anently appear less vivid. In fact, contrast sensitivity testing
is abnorm al du ring the d isease and does not always recover, even w ith fu ll recovery of
visu al acu ity.
Recu rrences of CSR occu r in about 30%. When a patient has had several episodes of
CSR, one shou ld con sider treatment. Laser photocoagu lation has been show n to decrease
the nu m ber of recu rrences but does not affect the final visu al acu ity. When a patient has
a second affected eye, it m ight be more reasonable to treat if the vision has been sign ifi-
cantly affected by severe CSR in the fellow eye. More recently, retinal specialists are sh ift-
ing to the u se of photodynam ic therapy for ch ron ic leakage from CSR.
Prior to any treatm ent, a fluorescein angiogram is u sefu l. One m ay also get an OCT
to more objectively follow resolution of su bretinal flu id. An OCT is also helpfu l when
explain ing the find ings and su bsequent cou rse to the patient.
I reserve laser photocoagu lation therapy for leaks that are outside of the perifoveal area
and are cau sing visu al morbid ity for more than 4 months. The patient shou ld be in formed
of the risks for choroid al neovascu larization (CN V) development at the laser photoco-
agu lation site. Photodynam ic therapy w ith verteporfin is an acceptable therapy for foveal
What Is Central Serous Retinopathy and How Do I Treat It? 77
Fig u re 2 0 -2 . A subre tin a l fibrin o id

re a ctio n m a y de ve lo p in a n e ye with
CSR.
leaks and ch ron ic CSR.3 Frequently, on ly one treatment is ind icated. After treatment, the
patient can exp ect an improvem ent in visu al acu ity. Note that visu al acu ity improvement
m ay lag beh ind flu id resolution.
An exp erimental area of treatment for CSR is the u se of antagon ists of glucocorticoid
receptors, such as oral m ifepristone (RU-486).4 Of cou rse, the u se of RU-486 in pregnant
wom en or in a wom an trying to get pregnant is contraind icated.
Choroid al neovascu larization can occu r in som e eyes w ith CSR w ith or w ithout treat-
ment. Fluorescein angiography shou ld be performed when CN V is su sp ected, and treat-
ment of CN V cou ld include laser photocoagu lation, photodynam ic therapy (PDT), or an
anti-vascu lar endothelial grow th factor (anti-VEGF). There has been no study comparing
these mod alities for CN V in eyes w ith CSR.
Summary
CSR is u su ally a self-lim ited d isease that can affect you nger as well as older patients.
There are variants of the d isease, wh ich includes fibrinoid CSR, bu llou s exud ative retinal
detach m ent, and ch ron ic CSR w ith pigmentary alteration. Treatment is rarely needed and
does not resu lt in better visu al acu ity than what wou ld natu rally occu r. However, CN V
associated w ith CSR does requ ire treatm ent.
References
1. Haimovici R, Rumelt S, Melby J. Endocrine abnormalities in patients with central serous chorioretinopathy.
O phthalm ology. 2003;110 (4):698-703.
2. Gass JD. Central serous chorioretinopathy and white subretinal exudation during pregnancy. Arch O phthalm ol.
1991;109 (5):677-681.
78 Question 20
3. O ber MD, Yannuzzi LA, Do DV, et al. Photodynamic therapy for focal retinal pigment epithelial leaks second-
ary to central serous chorioretinopathy. O phthalm ology. 2005;112(12);2088-2094.
4. Nielsen JS, Weinreb RN, Yannuzzi L, Jampol LM. Mifepristone treatment of chronic central serous chorioreti-
nopathy. Retina. 2007;27(1):119-122.
21 QUESTION
WHAT IS THE TREATMENT

PARADIGM FOR POSTOPERATIVE
CYSTOID MACULAR EDEMA?
Rishi P. Singh, MD
Postoperative cystoid m acu lar edem a (CME) typically occu rs after complicated cata-
ract su rgery but has been noted in 1% to 6% of u ncomplicated cataract su rgery cases. It
has been termed Irvine-Gass syndrome and is the most frequent cau se of decreased visu al
acu ity follow ing cataract su rgery, classically presenting 4 to 6 weeks p ostoperatively, w ith
a rep orted range of 3 weeks to 6 months.
Su rgical risk factors for CME are intra- or extracap su lar cataract extraction w ithout
phacoemu lsification, ruptu red posterior capsu le (11.5% to 20% incidence), retained lens
fragments (29% incidence), vitreou s strand s to the su rgical wou nd, and recent Nd :YAG
capsu lotomy (1% incidence). Certain pre-existing eye cond itions such as d iabetes, uveitis,
and retinal venou s occlu sive d isease are associated w ith a h igher likelihood of develop -
ing CME after cataract extraction. I choose to begin treatment w ith a topical nonsteroid al
anti-in flam m atory d rug (NSAID) either prior to su rgery or when the operative risk factor
is identified to m in im ize or even attempt to prevent the development of postop erative
CME.
In the d iagnosis of postoperative CME, it is essential to assess the patient w ith fluores-
cein angiography and optical coherence tomography (OCT) (Figu re 21-1). Con firm ation of
a “hot” optic nerve head in late frames of the angiogram helps to better secu re the d iag-
nosis of postcataract extraction CME and d ifferentiate p ostoperative CME from alterna-
tive cau ses for CME such as m acu lar degeneration, d iabetes, or venou s occlu sive d isease.
OCT is u sefu l in elim inating subclin ical vitreom acu lar traction synd rome and epiretinal
mem branes, wh ich m ay also be associated w ith CME and can con fu se the d iagnosis.
There are several d ifferent therap eutic approaches to treating postoperative CME.
My typical approach is a topical N SAID (ketorolac, d iclofenac, nepafenac, brom fenac) in
79
80 Question 21
Fig u re 21-1. OCT de m o n stra te s cysto id m a cula r e de m a (CME).
com bination w ith a topical steroid (pred n isolone), u n less contraind icated. It can typically
take 6 to 12 weeks to ach ieve improvem ent in visu al acu ity, contrast sensitivity, and leak-
age on fluorescein angiography. There is som e evidence that com bination therapy show s
greater improvement than treatm ent w ith a topical steroid or NSAID alone (+3.8 lines
versu s +1.1 to +1.6 lines).1 Su bsequent investigation has yet to show the benefit of one
particu lar NSAID over another. Even cases of long-term CME (greater than 6 months) can
resp ond to topical therapy alone.2 Therefore, even in ch ron ic cases, I start w ith topical
com bination treatm ent of NSAID and steroid.
The u se of periocu lar and intraocu lar triamcinolone injections for postoperative CME
is supported by several case series. Several d ifferent triamcinolone preparation s are u sed
by ophthalmologists. Typically, there is a good treatm ent effect w ith a reduction in edem a
on OCT and improvement in leakage on fluorescein angiography. Th is has to be balanced
again st the h igh rate of side effects (up to 50%), the most com mon being an increase in
intraocu lar pressu re as well as the possibility of recu rrent edem a when the steroid effect
wears off. Generally, I defer intravitreal steroid injection in favor of topical therapy for at
least 6 weeks. I obtain follow-up OCTs to docu ment change. If no improvement is ach ieved
w ith in 6 weeks, I then typically ad m in ister 40 m g of triamcinolone as a posterior subtenon
injection. Proper placement in the su btenon’s space nearest to the m acu la is essential in
ach ieving resp on se from th is therapy. I perform up to 3 injections over time before trying
intravitreal steroid s. When an intravitreal steroid is u sed, most su rgeons recom m end 4 mg
(0.1 m L) of triamcinolone. Neither posterior subtenon nor intravitreal steroid is approved
by the Food and Drug Ad m in istration (FDA) for the treatment of CME.
Intravitreal injection of bevacizu m ab (Avastin [Genentech, Inc]), an anti-vascu lar
endothelial grow th factor antibody (anti-VEGF), has also been u sed for m acu lar edem a
(not approved by the FDA). Wh ile it is not com mon ly u sed to treat postoperative CME,
a recent interventional case series demonstrated improvement in both visu al acu ity and
OCT find ings.3 I on ly u se intravitreal anti-VEGF agents in patients w ith glaucom a or w ith
a previou s h istory of steroid-induced ocu lar hyperten sion. In the m inority of cases when
p ostop erative CME does not resolve spontaneou sly or after topical or intravitreal treat-
m ent, I have p erformed pars plana vitrectomy w ith elevation of the posterior hyaloid as
a last resort.4
What Is the Treatment Paradigm for Postoperative Cystoid Macular Edema? 81
There are nu merou s strategies for the treatm ent of postop erative CME ranging from
com bination topical steroid and NSAID treatment, subtenon or intravitreal steroid s, intra-
vitreal anti-VEGF agents, to vitreou s su rgery. Identifying patients w ith preop erative and
p ostoperative risk factors for developing CME allow s for prompt treatment w ith any of
these strategies alone or in com bination.

Mark R. Barakat, MD
Retina Fellow, Cole Eye In stitute
Cleveland Clin ic Fou nd ation, Cleveland, OH
References
1. Heier JS, Topping TM, Baumann W, Dirks MS, Chern S. Ketorolac versus prednisolone versus combination
therapy in the treatment of acute pseudophakic cystoid macular edema. O phthalm ology. 2000 ;107(11):2034-
2038.
2. Weisz JM, Bressler NM, Bressler SB, Schachat AP. Ketorolac treatment of pseudophakic cystoid macular edema
identified more than 24 months after cataract extraction. O phthalm ology. 1999;106 (9):1656-1659.
3. Arevalo JF, Garcia-Amaris RA, Roca JA, et al. Primary intravitreal bevacizumab for the management of pseudo-
phakic cystoid macular edema: pilot study of the Pan-American Collaborative Retina Study Group. J Cataract
Refract Surg. 2007;33(12):2098-2105.
4. Pendergast SD, Margherio RR, Williams GA, Cox MS Jr. Vitrectomy for chronic pseudophakic cystoid macular
edema. Am J O phthalm ol. 1999;128 (3):317-323.
22 QUESTION
WHICH PATIENTS WITH UNILATERAL OR

BILATERAL DRY AGE-RELATED MACULAR
DEGENERATION ARE AT HIGHEST RISK
FOR DEVELOPING WET AGE-RELATED
MACULAR DEGENERATION?
Karl G. Csaky, MD, PhD
In the past decade, there have been exciting advances in the u se of vascu lar endothelial
grow th factor (VEGF) in h ibitor agents for the treatm ent of choroid al neovascu larization
(CN V) associated w ith neovascu lar age-related m acu lar degeneration (AMD).1 However,
these agents are not successfu l in all eyes, w ith the size of the CN V being one of the
risk factors for poor resp onse.2 In add ition, the treatments do not app ear to be cu rative
but rather ameliorative w ith respect to the progression of CN V. Therefore, identifying
patients w ith h igh-risk factors for CN V development is critical both to in itiate preventive
m easu res and to ensu re early treatm ent shou ld CN V develop.
The Age-Related Eye Disease Study (AREDS) 3 demon strated that h igh-dose antioxi-
d ant vitam in supplem entation reduced the risk of developing advanced AMD in patients
w ith interm ed iate or advanced AMD. In in itial reports, AREDS defined 4 categories of
AMD based prim arily on size of d ru sen in one or both eyes:
1. A REDS category 1: no AMD (few sm all d ru sen [< 63 µm])
2. A REDS category 2: m ild AMD (exten sive sm all d ru sen, pigment abnorm alities, or
few intermed iate d ru sen [63 to 124 µm])
83
84 Question 22
3. A REDS category 3: intermed iate AMD (extensive interm ed iate d ru sen, few large
d ru sen [> 124 µm] or non foveal geograph ic atrophy)
4. A REDS category 4: advanced AMD (one eye w ith visu al loss from either geograph ic
atrophy [GA] or CN V)
AREDS researchers fou nd that eyes classified in category 3 or 4 develop ed advanced
AMD at rates of 27% and 43%, respectively, over 5 years. These patients lowered their
risk by about 25% when treated w ith a h igh-dose com bination of vitam in C, vitam in
E, beta-carotene, and zinc. For those study participants who had either category 1 or 2
AMD, the supplements d id not provide an apparent benefit. For th is reason, the AREDS
supplem ents are on ly recom mended for patients classified in AREDS category 3 or 4.
The AREDS formu lation is contraind icated in cigarette smokers becau se beta-carotene
increased the incid ence of lu ng cancer in smokers. These patients shou ld be offered zinc
and antioxid ant supplementation w ithout beta-carotene. The add ition of cupric oxide to
the AREDS vitam in formu lation prevents a copp er deficiency anem ia in p ersons taking
h igh-dose zinc.
Although these AREDS category designation s demon strated sign ificant prognostic
find ings, they are somewhat cu m bersome to u se in a clin ical setting. In 2005, AREDS
Report No. 18 add ressed th is issue by describing a Simplified Severity Scale that defines
risk categories based up on 2 risk factors: 1) the presence of at least 1 large d ru se (d iameter
≥ that of a large vein at the d isc m argin) and 2) the presence of any pigment abnorm ality.4
Each eye receives a score of 0, 1, or 2 points, for a total of 4 possible points for the 2 eyes.
Patients who have no large d ru sen but instead have intermed iate-sized d ru sen (≥ half the
d iameter of a large d ru se) in both eyes are assigned 1 p oint.
Cou nting these risk factors and su m m ing their presence in both eyes provides a clin i-
cally u sefu l estim ate of the risk that a patient w ill develop advanced AMD over the next
5 years. Thu s, the AREDS Simplified Severity Scale provides a conven ient way to fol-
low patients and pred ict outcomes. For clin ical pu rp oses, as the nu m ber of risk factors
increases from 0 to 4, the global 5-year risk of advanced AMD in at least 1 eye increases in
the sequence of approxim ately 0.5%, 3%, 12%, 25%, and 50%. It is important to recogn ize
that in patients w ithout advanced AMD in either eye, the estim ated risk for each eye is
on ly approxim ately one-half of th is percentage.
The single most important pred ictor of progression to advanced AMD for an ind i-
vidu al eye is the presence of advanced AMD in the fellow eye. In these cases, u sing the
Simplified Severity Scale, the eye w ith advanced d isease receives 2 risk factors and the
fellow eye receives an add itional 0, 1, or 2 points based on the presence of large d ru sen or
pigment abnorm alities. Wh ile the pred ictive value for the overall risk factor level and the
rates of progression rem ain the same as for patients w ithout advanced AMD, these rates
reflect progression for the rem ain ing eye. Thu s, the presence of advanced d isease in 1 eye
dou bles the risk for progression to advanced d isease in the 1 rem ain ing eye. Therefore, it
appears reasonable to recom m end AREDS vitam in supplementation for su bjects w ith 3 or
more risk factors and for su bjects w ith on ly 2 risk factors if the fellow eye has advanced
AMD.
These fu ndu s characteristics represent the ocu lar risk factors that are the most pre-
d ictive factors for the developm ent of advanced AMD. However, other mod ifiable and
non mod ifiable risk factors such as smoking, fam ily h istory, and antioxid ant u se also alter
an ind ividu al’s risk. As a resu lt, it is important to cou n sel patients on m aking lifestyle
Which Patients With Dry AMD Are at Highest Risk for Developing Wet AMD? 85
mod ifications that m ay help prevent visu al loss associated w ith advanced AMD. Smoking
is the most important mod ifiable risk factor, and patients shou ld be educated about blind-
ness as a smoking-related cond ition.
No clear gu idelines exist for follow ing patients w ith h igh-risk AMD. Given the yearly
10% risk for developing advanced AMD, it seem s prudent to follow patients w ith 4 risk
factors or w ith AREDS category 4 d isease on a bian nu al basis and to instruct all patients
on the u se of at-home vision testing, includ ing the Am sler grid.
Acknow ledgments of Significant Collaboration

Kath leen Urban, MD
Department of Ophthalmology, Du ke Un iversity, Du rh am, NC
Fred erick L. Ferris, MD
Scientific Director, National Eye In stitute
National In stitutes of H ealth, Bethesd a, MD
References
1. Rosenfeld PJ, Rich RM, Lalwani GA. Ranibizumab: phase III clinical trial results. O phthalm ol Clin N orth Am.
2006;19 (3):361-372.
2. Kaiser PK, Brown DM, Zhang K, et al. Ranibizumab for predominantly classic neovascular age-related macular
degeneration: subgroup analysis of first-year ANCHO R results. Am J O phthalm ol. 2007;144 (6):850 -857.
3. Age-Related Eye Disease Study Research Group. A randomized, placebo-controlled, clinical trial of high-dose
supplementation with vitamins C and E, beta carotene, and zinc for age-related macular degeneration and
vision loss: AREDS report no. 8. Arch O phthalm ol. 2001;119 (10 ):1417-1436.
4. Ferris FL, Davis MD, Clemons TE, et al. A simplified severity scale for age-related macular degeneration:
AREDS Report No. 18. Arch O phthalm ol. 2005;123(11):1570-1574.
23 QUESTION
HOW DO I COUNSEL MY PATIENTS

WITH DRY AGE-RELATED MACULAR
DEGENERATION AND WHAT ABOUT
VARIOUS VITAMIN SUPPLEMENTS?
Catherine B. Meyerle, MD
Let u s first review the rationale and resu lts of the Age-Related Eye Disease Study
(AREDS). Begin n ing in the early 1990s, several observational stud ies showed the protec-
tive association of increased antioxid ant vitam in intake w ith decreased risk of age-related
m acu lar degeneration (AMD) developm ent. AREDS, begin n ing in 2001 u nder National
Eye Institute spon sorsh ip, was designed to evalu ate th is suggested nutritional role in
a prospective clin ical trial. Sp ecifically, AREDS was a 5-year random ized study of 4757
participants evalu ating whether h igh-dose antioxid ants and zinc cou ld reduce progres-
sion of both AMD and age-related len s opacity. Participants were categorized accord ing
to d ru sen size. Sm all d ru sen were defined as ≤ 63 µm, intermed iate d ru sen as between
63 µm and 125 µm, and large d ru sen as > 125 µm. Patients were then random ized to oral
supplem entation of h igh dose antioxid ant vitam in s and/ or zinc. Final analysis demon-
strated that 500 mg vitam in C, 400 IU vitam in E, 15 mg beta-carotene, and 80 mg zinc
oxid e (along w ith 2 mg cupric oxide to prevent copp er-deficiency anem ia) reduced the
risk of advanced AMD, defined as neovascu lar AMD or geograph ic atrophy involving
the fovea, by 25% in participants w ith intermed iate AMD (exten sive interm ed iate d ru sen
and/ or any large d ru sen) or w ith advanced AMD in one eye.1
Now that we have reviewed the basic principles of AREDS, let’s d iscu ss how I apply
these clin ical trial resu lts to my everyd ay practice. First, I recom m end the original AREDS
formu lation to all non smoking patients w ith intermed iate or advanced AMD in one eye.
87
88 Question 23
However, if my patient smokes, I do not in itiate the original antioxid ant formu la becau se
beta-carotene supplementation du ring AREDS increased the risk of lu ng cancer and its
associated mortality in smokers. Second, I do not recom mend the original AREDS formu-
lation for patients w ith early AMD defined as mu ltiple sm all d ru sen or som e intermed iate
d ru sen. These patients had a low rate of developing advanced AMD du ring AREDS, 1.3%
in 5 years, so antioxid ant supplem entation wou ld not resu lt in a sign ificant benefit for
these early AMD patients. Instead, I cou n sel these early AMD patients about the impor-
tance of regu lar d ilated eye exam inations and a healthy d iet, includ ing lots of fru its and
vegetables. Th ird, for reasons d iscu ssed above, I do not recom mend antioxid ant therapy
for the offspring of affected ind ividu als u n less they personally m eet the criteria for inter-
med iate or advanced AMD in one eye. And finally, when patients w ith bilateral advanced
AMD inqu ire if they shou ld take the original AREDS formu la, I tell them that the study
was not designed to determ ine whether nutritional therapy cou ld prevent scotom a exten-
sion.
Wh ile the AREDS antioxid ant formu la has had a m ajor pu blic health impact in term s of
preventing advanced AMD, it can have side effects ju st like all med ication s. Fortu nately,
the adverse event rate is low. As mentioned previou sly, beta-carotene supplementation is
contraind icated in smokers. Beta-carotene add itionally cau sed yellow d iscoloration of the
skin in both smokers and non smokers, but th is was of no health consequence. Also, there
were complication s associated w ith zinc supplementation in AREDS. Patients random ized
to zinc had an increase in gen itou rinary complications such as u rinary tract in fection,
prostate hyp erplasia, and stress incontinence. However, an analysis of zinc versu s no zinc
fou nd a statistically sign ificant reduction in mortality in the zinc group (relative risk 0.73,
95% CI: 0.61 to 0.89). So overall, the original AREDS formu lation is safe.
Recently, some meta-analyses have warned of increased morbid ity and mortality asso-
ciated w ith vitam in therapy.2,3 The methodology of these m eta-analyses, however, is ques-
tionable. Som e u se flawed su bgroup analysis to support their claim s. Others focu sed on
large beta-carotene trials that included smokers, thu s creating a negative bias. Due to the
low adverse event rate in AREDS, I feel that antioxid ant therapy is safe in nonsmokers.
One last point I wou ld like to d iscu ss is the cu rrent AREDS2 project (w w w.ared s2.
org). The impetu s for th is random ized clin ical trial of 4000 participants was d ietary d ata
obtained from the original AREDS participants and nutritional concerns regard ing beta-
carotene and zinc. Du ring AREDS, all participants completed a food frequency question-
naire. Baseline intake of lutein/ zeaxanth in, carotenoid s fou nd in green leafy vegetables
such as spinach, collard greens, and kale, was inversely associated w ith neovascu lar
AMD, geograph ic atrophy, and large or extensive interm ed iate d ru sen.4 Th is is notewor-
thy becau se m acu lar pigment is composed prim arily of lutein/ zeaxanth in, and these
carotenoid s are postu lated to protect the retina from oxid ative stress. A second d ietary
factor that was sign ificant was the baseline intake of omega-3 long chain polyu n satu rated
fatty acid s (LCPUFAs) derived from fish. H igher fish consu mption was inversely associat-
ed w ith neovascu lar AMD.5 Based on these nutritional associations, AREDS2 en rollment
began in 2006 to determ ine whether oral supplementation w ith m acu lar xanthophylls
(lutein 10 mg/ d ay and zeaxanth in 2 m g/ d ay) and/ or omega-3 LCPUFAs (docosahexae-
noic acid [DH A] 350 mg/ d ay and eicosap entaenoic acid [EPA] 650 m g/ d ay, for a total of
1g of fish oil p er d ay) w ill decrease the risk of progression to advanced AMD as compared
to placebo. Given the risks of beta-carotene and the fact that beta-carotene is not fou nd
How Do I Counsel My Patients With Dry Age-Related Macular Degeneration? 89
w ith in hu m an ocu lar tissue u n like lutein/ zeaxanth in, a second ary random ization in
AREDS2 w ill evalu ate whether it wou ld be worthwh ile to remove beta-carotene from
the AREDS2 formu la. The last research question w ill involve zinc. Although there was
decreased mortality overall in the zinc group, nutritional experts are increasingly con-
cerned about h igh dose zinc u sage. In add ition to the increased gen itou rinary complica-
tions fou nd in AREDS, zinc has been reported to have an adverse effect on cholesterol
levels in other stud ies. Based on these concerns, the second random ization in AREDS2
w ill investigate whether lowering zinc to 20 mg/ d ay wou ld be beneficial.
The AREDS2 resu lts w ill not be available u ntil after 2011, but the final analysis m ay
affect my recom m end ations for non neovascu lar AMD m anagement. But for now, preven-
tion w ith the original AREDS formu lation in the appropriate patient popu lation is the
cornerstone of my AMD cou n seling.
References
1. Age-Related Eye Disease Study Research Group. A randomized, placebo-controlled, clinical trial of high-dose
supplementation with vitamins C and E, beta carotene, and zinc for age-related macular degeneration and
vision loss: AREDS report no. 8. Arch O phthalm ol. 2001;119 (10 ):1417-1436.
2. Bjelakovic G, Nikolova D, Gluud LL, Simonetti RG, Gluud C. Mortality in randomized trials of anti-
oxidant supplements for primary and secondary prevention: systematic review and meta-analysis. JAMA.
2007;297(8):842-857.
3. Miller ER 3rd, Pastor-Barriuso R, Dalal D, Riemersma RA, Appel LJ, Guallar E. Meta-analysis: high-dosage
vitamin E supplementation may increase all-cause mortality. Ann Intern Med. 2005;142(1):37-46.
4. Age-Related Eye Disease Study Research Group, SanGiovanni JP, Chew EY. The relationship of dietary carot-
enoid and vitamin A, E, and C intake with age-related macular degeneration in a case-control study: AREDS
Report No. 22. Arch O phthalm ol. 2007;125(9):1225-1232.
5. SanGiovanni JP, Chew EY, Clemons TE, et al. The relationship of dietary lipid intake and age-related macular
degeneration in a case-control study: AREDS Report No. 20. Arch O phthalm ol. 2007;125(5):671-679.
24 QUESTION
WHAT DRUGS CAN BE INJECTED

INTRAVITREALLY AND WHAT O CULAR
AND SYSTEMIC COMPLICATIONS
SHOULD I LOOK FOR?
Diana V. Do, MD
Over the past several years, an increased u nderstand ing of angiogenesis has led to
the development of pharm acologic therapy via intravitreal injection for choroid al and
retinal vascu lar cond itions such as neovascu lar age-related m acu lar degeneration (AMD),
d iabetic m acu lar edem a, and m acu lar edem a second ary to vein occlu sion (Figu re 24-1).
Th is bu ild s upon the decades-long exp erience of intravitreal injection of anti-in fectives
for endophthalm itis and gases for treatment of retinal detach m ent.
Several d rugs have been sp ecifically approved by the Un ited States Food and Drug
Ad m in istration (FDA) for intravitreal injection and provide level I med ical evidence for
safety and efficacy. Two vascu lar endothelial grow th factor (VEGF) in h ibitors have been
approved for intravitreal injection for the treatment of neovascu lar AMD. Pegaptan ib
sod iu m (Macugen [Eyetech, Inc, Palm Beach Gardens, FL]), an aptamer that bind s one
isoform of VEGF, was the first intravitreal treatment for neovascu lar AMD approved by
the FDA.1,2 Ran ibizu m ab (Lucentis), a monoclonal antibody d irected again st all isoform s
of VEGF, was approved in 2006 for the treatm ent of neovascu lar AMD.3,4 Both ran ibi-
zu m ab and pegaptan ib sod iu m u nderwent rigorou s preclin ical an im al testing and clin i-
cal hu m an testing in phase I, II, and III clin ical trials to en su re the safety and bioactivity
of each agent.
In add ition to VEGF in h ibitors, the FDA has also approved 2 sp ecific formu lations of
the steroid triamcinolone aceton ide (Trivaris [Allergan, Irvine, CA] and Triesence [Alcon,
Hü nenberg, Sw itzerland]) for intraocu lar u se. Both agents were sp ecifically tested and
91
92 Question 24
Fig u re 2 4 -1. In tra vitre a l in je ctio n s

h a ve be co m e a po pula r ro ute o f
drug de live ry fo r a va rie ty o f po s-
te rio r se gm e n t diso rde rs.
designed for intravitreal injection. However, both form s of triamcinolone aceton ide still
have the p otential risks of elevated intraocu lar pressu re, cataract progression, and intra-
ocu lar in fection.
What Other Drugs Can Be Injected Into the Eye?

Many pharm acologic agents that ophthalmologists inject intravitreally constitute the
“off-label” u se of d rugs. For example, intravitreal antibiotics u sed to treat endophthal-
m itis are considered as “off-label” ind ications of these m ed icines. Antibiotics were never
designed for intraocu lar u se; rather, they were approved for intravenou s u se. However,
intravitreal antibiotics have been accepted by the ophthalmology com mu n ity as part
of the stand ard of care for patients w ith endophthalm itis becau se of their efficacy in
erad icating in fectiou s organ ism s.5 Although u sefu l for the treatment of endophthalm itis,
antibiotics have the potential to cau se sign ificant ocu lar toxicity. The am inoglycosides, in
particu lar gentam icin, are the most toxic of the antibiotics com mon ly u sed in ophthalmol-
ogy.6 Extreme caution shou ld be u sed when ad m in istering a periocu lar injection of an
am inoglycoside for the treatm ent or prophylaxis of in fection, and am inoglycosides shou ld
never be ad m in istered as an intraocu lar injection u n less there is no alternative.
Other d rugs that have been u sed “off-label” include steroid s (ie, triamcinolone ace-
ton ide, dexam ethasone, etc.), recom binant tissue plasm inogen activator (rt-PA (Activase
[Genentech, Inc]), and bevacizu m ab (Avastin), a monoclonal antibody d irected again st
VEGF. These pharm acologic agents were never designed to be u sed inside of the eye.
However, case rep orts and retrospective stud ies have ind icated bioactivity and short-term
safety of these agents, wh ich has led to w ide-spread off-label u se for a variety of posterior
segment d iseases includ ing neovascu lar AMD and m acu lar edem a.
Although “off-label” u se of m ed icines is com mon in ophthalmology, I wou ld not rec-
om mend attempting to inject agents that have not been rigorou sly tested th rough preclin i-
cal and hu m an stud ies. Injecting new agents into the hu m an eye can lead to potentially
blind ing complication s, especially since m any med icines are not formu lated for intraocu-
lar u se and m ay contain potentially hazardou s preservatives.
What Drugs Can Be Injected Intravitreally? 93
Fig u re 2 4 -2 . En do ph th a lm itis is
a se rio us but ra re o cula r a dve rse
e ve n t th a t ca n o ccur a fte r in tra -
vitre a l in je ctio n .
What Adverse Events Can Occur?

Adverse events can be separated into 2 broad categories: 1) ocu lar and 2) system ic.
Potential seriou s ocu lar adverse events include endophthalm itis (Figu re 24-2), in flam-
m ation (uveitis), elevated intraocu lar pressu re, cataract form ation, retinal ischem ia, optic
nerve in flam m ation, vitreou s hemorrhage, and retinal detach m ent. Endophthalm itis
can occu r after any intravitreal injection. Although th is is a very rare complication, it is
u sually related to the injection procedu re and not associated w ith the particu lar m ed ica-
tion being injected (u n less the med ication is contam inated). To avoid potential in fection
after intravitreal injection, u se of a sterile lid specu lu m and application of betad ine to the
ocu lar su rface is recom mended. To m in im ize ocu lar toxicity, rigorou s pre-clin ical testing
shou ld also be p erformed on pharm acologic agents before u se in hu m ans.
Potential seriou s system ic adverse events can also occu r w ith intravitreal d rugs if
the d rug reaches the seru m and has a physiologic effect in the system ic circu lation. The
range of system ic adverse events is dependent on the particu lar d rug and its mechan ism
of action. Although system ic adverse events m ay occu r, these events are more d ifficu lt
to d iagnose than local ocu lar adverse events becau se of u nderlying risk factors that m ay
pred ispose patients to comorbid cond itions and system ic problem s. For example, intra-
venou s VEGF in h ibitors have been associated w ith an increased risk of th rom boem bolic
events. Although phase III clin ical trials of VEGF in h ibitors d id not show a statistically
sign ificant increase in th rom boem bolic events in subjects assigned to intravitreal VEGF
in hibition,4 these ophthalm ic clin ical trials were not su fficiently p owered to evalu ate th is
safety issue. Large random ized clin ical trials en rolling ten s of thou sand s of participants
wou ld be needed to add ress these safety concerns. Since such stud ies are challenging and
m ay be impossible to complete, physician s shou ld have a heightened clin ical awareness
of potential system ic side effects. In add ition, vigilant postm arket safety su rveillance is
recom mended for intraocu lar pharm acologic agents that have the potential for cau sing
seriou s system ic and ocu lar adverse events.
Financial disclosure: Research support from Genentech and Regeneron.

94 Question 24
References
1. Gragoudas ES, Adamis AP, Cunningham ET Jr, et al. Pegaptanib for neovascular age-related macular degenera-
tion. N Engl J Med. 2004;351(27):2805-2816.
2. VEGF Inhibition Study in O cular Neovascularization (V.I.S.I.O.N.) Clinical Trial Group, D’Amico DJ,
Masonson HN, et al. Pegaptanib sodium for neovascular age-related macular degeneration: two-year safety
results of the two prospective, multicenter, controlled clinical trials. O phthalm ology. 2006;113(6):992-1001.
3. Ferrara N, Damico L, Shams N, Lowman H, Kim R. Development of ranibizumab, an anti vascular endothelial
growth factor antigen binding fragment, as therapy for neovascular age-related macular degeneration. Retina.
2006;26 (8):859-870.
4. Brown DM, Kaiser PK, Michels M, et al. Ranibizumab versus verteporfin for neovascular age-related macular
degeneration. N Engl J Med. 2006;355(14):1432-1444.
5. Results of the Endophthalmitis Vitrectomy Study. A randomized trial of immediate vitrectomy and of intra-
venous antibiotics for the treatment of postoperative bacterial endophthalmitis. Endophthalmitis Vitrectomy
Study Group. Arch O phthalm ol. 1995;113(12):1479-1496.
6. Thomas T, Galiani D, Brod RD. Gentamicin and other antibiotic toxicity. O phthalm ol Clin N orth Am.
2001;14 (4):611-624.
25 QUESTION
WHY ARE THERE SO MANY DIFFERENT

TRIAMCINOLONE PREPARATIONS AND
WHICH O NE SHOULD I USE?
Sophie J. Bakri, MD
Triamcinolone aceton ide (TA) is a synthetic corticosteroid that can be ad m in istered

orally, parenterally, topically, by in halation, p eriocu larly, and intraocu larly. The u se of the
intravitreal route to deliver corticosteroid to the eye was first described for endophthal-
m itis in 1974.1 Tano et al2 rep orted intravitreal TA for experimental proliferation in 1980,
followed by descriptions for treatment of d iabetic m acu lar edem a m any years later by
Martid is et al3 and Jonas et al.4 Since then, every possible ind ication has been explored,
includ ing choroid al neovascu larization, uveitis, retinal vascu lar d isease, cystoid m acu lar
edem a, and proliferative retinopath ies, among others.
There are several d ifferent preparations of TA that are u sed in ophthalmology:
Kenalog-40 (Bristol-Myers Squ ibb, New York, N Y), preservative-free TA (prepared by a
compou nd ing pharm acy), Triesence, and Trivaris. H istorically, Kenalog-40 has been u sed
for intravitreal injection of TA as it was the on ly formu lation that was available. Becau se
Kenalog-40 is not approved by the FDA for intraocu lar u se, its u se via intravitreal injec-
tion is off-label and questions arose about its sterility, toxicity, and overall safety profile
when u sed intravitreally.
Despite its well-demonstrated efficacy in a variety of retinal cond ition s, the u se of
intravitreal triamcinolone has been associated w ith several safety issues. In add ition to
the increased incidence of cataract and glaucom a follow ing intravitreal triamcinolone
ad m in istration, there have been docu mented cases of sterile in flam m ation (“p seudo-
endophthalm itis”) occu rring w ith in a week after injection, w ith an estim ated incidence
of 1%.5 The exact reasons for th is in flam m ation are u nclear, but it is thought that the
preservative ben zyl alcohol in Kenalog-40 m ay be responsible.
95
96 Question 25
Ou r Eu ropean cou nterparts have been u sing preservative-free TA for m any years;
however, th is has not been w idely available in the Un ited States u ntil recently. In order
to obtain a preservative-free triamcinolone preparation, ophthalmologists in the Un ited
States order it from compou nd ing pharm acies that prepare it at a cost of approxim ately
$15 per 40-mg (1-m L) vial, w ith a shelf life of 45 d ays. Th ree-month d ata suggest that
intravitreal preservative-free TA m ay be a good alternative to the stand ard formu lation
of Kenalog-40 and that the sm aller particle size can allow for greater d ispersion in the
vitreou s cavity.6
More recently, the FDA has approved 2 com m ercially available, preservative-free
triamcinolone preparations, Triesence and Trivaris. The package inserts state that they
are approved for intraocu lar u se and also ind icated for visu alization du ring vitrectomy
and treatment of sympathetic ophthalm ia, uveitis, and ocu lar in flam m atory cond itions
u nresponsive to topical corticosteroid s. Triesence is packaged in a 1-m L vial contain ing a
su sp en sion of TA w ith a concentration of 40 mg/ m L. Recom mended dosing is 1 mg to 4
m g (0.025 m L to 0.1 m L) ad m in istered intravitreally. Trivaris is a gel that comes in a blister
pack as a 0.1-m L vial w ith 8 mg of TA, therefore the concentration is 80 mg/ m L. To inject
4 mg of TA requ ires either 0.1 mL of Triesence or 0.05 m L of Trivaris.
Summary
The proliferation of TA preparations is a byproduct of their h istorical development and
u se in ophthalmology as opposed to d ifferences in therap eutic effects. For intravitreal
injection, I u se a preservative-free triamcinolone preparation. Wh ile the com m ercially
available ones are more expen sive (approxim ately $120 per vial), they have a more stan-
d ard preparation. For su btenon injection, any of the preparations m ay be u sed, as sterile
endophthalm itis is not a concern. There are recent reports of sterile endophthalm itis
occu rring after Triesence injection,7 but its frequency compared to that of Kenalog-40 or
a compou nded preparation is not know n. Since sterile endophthalm itis is so rare, deter-
m ination of its incidence w ith the variou s preparations wou ld requ ire carefu l observation
w ith in a rigorou s study to determ ine its frequency. At th is time, Trivaris is not yet avail-
able for u se.
References
1. Graham RO, Peyman GA. Intravitreal injection of dexamethasone. Treatment of experimentally induced endo-
phthalmitis. Arch O phthalmol. 1974 Aug;92(2):149-54
2. Tano Y, Chandler D, Machemer R. Treatment of intraocular proliferation with intravitreal injection of triam-
cinolone acetonide. Am J O phthalm ol. 1980 ;90 (6):810 -6.
3. Martidis A, Duker JS, Greenberg PB, et al. Intravitreal triamcinolone for refractory diabetic macular edema.
O phthalm ology. 2002;109 (5):920-7.
4. Jonas JB, Hayler JK, Söfker A, Panda-Jonas S. Intravitreal injection of crystalline cortisone as adjunctive treat-
ment of proliferative diabetic retinopathy. Am J O phthalm ol. 2001;131(4):468-71.
5. Moshfeghi DM, Kaiser PK, Bakri SJ, et al. Presumed sterile endophthalmitis following intravitreal triamcinolone
acetonide injection. O phthalm ic Surg Lasers Im aging. 2005;36 (1):24-29.
6. Bakri SJ, Shah A, Falk NS, Beer PM. Intravitreal preservative-free triamcinolone acetonide for the treatment of
macular oedema. Eye. 2005;19 (6):686-688.
7. Bakri SJ, Edwards AO, Couch SM. Noninfectious Endophthalmitis O ccurring After Intravitreal Triesence
Injection. Retinal Cases & Brief Reports. 2009:3(3):316-318.
26 QUESTION
HOW DO I MONITOR FOR

STEROID-INDUCED GLAUCOMA
AFTER AN INTRAVITREAL INJECTION?
Baruch D. Kuppermann, MD, PhD
Steroid-induced glaucom a is a form of op en-angle glaucom a that u su ally is associated

w ith topical steroid u se, but it m ay develop w ith in haled, oral, intravenou s, periocu lar,
or intravitreal steroid ad m in istration. Intravitreal steroid injections, particu larly w ith
triamcinolone aceton ide, have become very popu lar to treat eyes w ith visu al loss from
variou s retinal d iseases. The prevalence of corticosteroid-induced glaucom a has and m ay
continue to rise.
A nu m ber of risk factors have been identified for the development of corticosteroid-
induced glaucom a, includ ing age, personal or fam ily h istory of glaucom a, type 1 d iabetes
m ellitu s, con nective tissue d isease, and h igh myopia. Clin ician s shou ld be aware that an
elevated intraocu lar pressu re (IOP) m ay occu r in nearly half of all cases follow ing a single
injection of intravitreal triamcinolone (IVTA)1 and m ay be acute or ch ron ic. Intraocu lar
injection in aphakic or pseudophakic eyes m ay resu lt in d isp ersion of triamcinolone par-
ticles, not on ly th roughout the vitreou s, but also into the anterior cham ber. On occasion,
th is particu late m atter m ay settle out in the anterior cham ber (pseudohyp opyon) and
cou ld block the trabecu lar spaces, cau sing an acute pressu re rise.2 Generally, the d isp er-
sion of triamcinolone into the anterior cham ber does not need any d irect intervention,
and the m ed ication d issipates qu ickly.
The u se of the fluocinolone implant (Retisert [Bau sch & Lom b, Rochester, N Y]), wh ich
releases 0.59 mg of d rug over 36 months, has show n that even very low levels of d rug
ad m in istered over a prolonged period can be associated w ith a h igh incidence of steroid-
induced glaucom a. In that study, 40% of eyes requ ired su rgical control of IOP, frequently
w ith the u se of tube shu nt implants rather than trabecu lectomy.3 Other steroid-based
97
98 Question 26
d rug delivery system s are being investigated, includ ing the Alim era Iluvien implant
(Alpharetta, GA), wh ich contain s half as much fluocinolone as the Retisert; Su rMod ics I-
Vation implant (Ed en Prairie, MN ), wh ich delivers triamcinolone for 18 months; and the
Allergan Ozu rdex implant, wh ich delivers dexamethasone over 6 month s. The Ozu rdex
implant was approved by the FDA in 2009 for the treatment of m acu lar edem a cau sed by
retinal vein occlu sion and is u nder investigation for other cau ses of m acu lar edem a. Large
scale trials are u nderway for these steroid-based d rug delivery system s and w ill help
clarify the risk of steroid-induced glaucom a for each of the devices.4,5
After an intravitreal triamcinolone injection, eyes shou ld be followed carefu lly.
Postinjection exam inations shou ld occu r approxim ately 1 week later for all patients and
1 d ay in h igh risk patients. Regu lar follow-up exam ination s shou ld continue for greater
than 6 months. Exam ination s shou ld include mon itoring of the IOP and, as ind icated,
gonioscopy for pseudophakic eyes or those eyes that have had a prior vitrectomy to detect
m echan ical obstruction of the trabecu lar m eshwork by the m ed ication.
The first step in m anaging steroid-induced glaucom a is to d iscontinue the d rug,6
but th is can not be done when the triamcinolone is in the intravitreal space. I wou ld not
recom mend pars plana vitrectomy for removal of the intravitreal triamcinolone crystals
since it is tech n ically d ifficu lt to remove all crystals from the vitreou s base w ithout risk
to the crystalline lens. Fu rthermore, there m ay still be some rem ain ing steroid effect after
the su rgical removal of the triamcinolone crystals. Topical IOP-lowering agents are the
first line of treatment begin n ing w ith a beta-blocker if no contraind ications exist. I wou ld
avoid u sing prostagland ins in itially for fear of worsen ing the m acu lar edem a but wou ld
u se them w ithout hesitation if facing oral, laser, or su rgical treatment option s for the ele-
vated IOP. My next step of treatm ent wou ld be to add an oral carbon ic an hyd rase in h ibi-
tor. As is often true in second ary glaucom as, argon and selective laser trabecu loplasties
are ineffective. Filtering su rgery m ay be necessary to control IOP follow ing intravitreal
steroid injection.7 The constant level of steroid in these eyes m ay produce excellent su rgi-
cal resu lts w ith m in im al scarring. One of the best-looking blebs in my practice is in an
eye w ith the Retisert intravitreal steroid d rug delivery device. Becau se m any eyes w ith
steroid-induced glaucom a are also at risk for posterior su bcapsu lar cataract, approaches to
m anage IOP that include the m anagement of cataract mu st be considered. These include
stand ard trabecu lectomy w ith an antifibrosis agent alone and phacoemu lsification alone,
or com bined trabecu lectomy w ith an antifibrosis agent and phacoemu lsification. In some
eyes, tu be shu nt placement is perform ed in place of trabecu lectomy. Controlling IOP and
restoring visual fu nction w ithout jeopard izing the retinal pathology that was treated w ith
the intravitreal steroid can present a d istinct challenge.
The first treatment option for steroid-induced glaucom a is frequently con servative
observation w ith topical IOP-lowering d rops. Most affected eyes w ill eventu ally retu rn
to norm al IOP follow ing cessation of the steroid exposu re. Most of the available rep orts
regard ing steroid-induced glaucom a are u ncontrolled, w ith few patients and relatively
short follow-up of 6 months or less. The reported rates of pressu re elevation after intravit-
real steroid injections range from 20% to 80%, and the time between injection and nota-
tion of the pressu re elevation ranges from 2 d ays to 6 weeks. In most eyes, the pressu re
How Do I Monitor for Steroid-Induced Glaucoma After an Intravitreal Injection? 99
elevations respond well to topical med ication s. Of those patients w ith IOP elevation, the
du ration of the rise has not been well described, due to the relatively short longitud inal
assessm ent. One study w ith a m ean follow-up of 6 months fou nd that the IOP had begu n
to retu rn to preinjection levels by the end of the follow-up period. Clearly, a dose-respon se
trial and controlled trials w ith longer follow-up mu st be pu rsued. The National Eye
Institute’s ongoing DRCRnet and SCORE study w ill provide fu rther gu id ance.8,9
Financial disclosure: Consulting fees from A llergan, CoM entis, Genentech, Glaukos, N eovista, N ovagali
Pharma, SurM odics, and TargeGen. Contracted research from A limera Sciences, A llergan, Eli Lilly,
Genentech, and Regeneron Pharmaceuticals. Ownership interest in Ophthotech, OPKO Health, and
Vitreoretinal Technologies.

Ash ish Sharm a, MD
Glau com a Fellow, Department of Ophthalmology,
Bascom Palm er Eye In stitute, Un iversity of Miam i, Miam i, FL
References
1. Jones R 3rd, Rhee DJ. Corticosteroid-induced ocular hypertension and glaucoma: a brief review and update
of the literature. Curr O pin O phthalm ol. 2006;17(2):163-167.
2. Jonas JB. Intraocular availability of triamcinolone acetonide after intravitreal injection. Am J O phthalm ol.
2004;137(3):560-562.
3. Bausch & Lomb. Retisert package insert. 2007. Available at: http://www.bausch.com /en_US/downloads/ecp /
pharma /general/retisert_pkginsert.pdf. Accessed May 27, 2008.
4. US National Institutes of Health. Study of the safety and efficacy of a new treatment for noninfectious inter-
mediate or posterior uveitis. 2006. Available at: http://clinicaltrials.gov/ct2/show/NCT00333814?term =posur
dex&rank=1. Accessed May 27, 2008.
5. US National Institutes of Health. The MAP study: FA/Medidur(TM)for AMD Pilot. 2008. Available at: http://
clinicaltrials.gov/ct2/show/NCT00605423?term =medidur&rank=1. Accessed May 27, 2008.
6. Meyer CH, Mennel S, Schmidt JC. Intravitreal triamcinolone acetonide may increase the intraocular pressure
even in vitrectomized eyes after more than 3 months. Am J O phthalm ol. 2005;140 ;766-767; author reply 767-
768.
7. Jonas J, Heatley G, Spaide R, Varma R. Intravitreal triamcinolone acetonide and secondary ocular hyperten-
sion. J Glaucom a. 2005;14 (2):168-171.
8. Ip MS, Bressler SB, Antoszyk AN, et al. A randomized trial comparing intravitreal triamcinolone and laser
photocoagulation for diabetic macular edema. Retina. 2008;28 (7):919-930.
9. Diabetic Retinopathy Clinical Research Network, Chew E, Strauber S, et al. Randomized trial of peribulbar
triamcinolone acetonide with and without focal photocoagulation for mild diabetic macular edema: a pilot
study. O phthalm ology. 2007;114 (6):1190-1196.
27 QUESTION
WHAT SYSTEMIC MEDICATIONS REQUIRE

PERIODIC FUNDUS EVALUATION,
WHAT AM I LOOKING FOR,
AND WHAT TESTS DO I DO?
William F. Mieler, MD
Every ophthalmologist shou ld be fam iliar w ith med ication-induced ocu lar side effects.
There are 4 particu lar med ications that are associated w ith well-described ocu lar effects
even when u sed at “safe” d aily dosages. These med ications include hyd roxych loroqu ine,
th iorid azine, tamoxifen, and sildenafil.
Hyd roxych loroqu ine (Plaquen il [Sanofi-aventis US, Bridgewater, N J]) is prescribed to
treat con nective tissue d iseases. Although clin ically identical to the retinopathy of ch loro-
qu ine, the incidence is much lower since hyd roxych loroqu ine does not as read ily cross the
blood-retinal barrier. Risk factors for the development of retinopathy include d aily dosage
exceed ing 6.5 m g/ kg/ d ay, age greater than 60 years, liver or renal d isease, concom itant
retinal d isease, treatment du ration greater than 5 years, and obesity. The d aily dosage
shou ld be calcu lated based on ideal rather than actu al body weight since the med ication
is preferentially stored in leaner tissues rather than ad ip ose cells.
The Am erican Academy of Ophthalmology gu idelines suggest that a d ilated fu n-
du s exam ination be perform ed at the com mencem ent of hyd roxych loroqu ine therapy.1
Retinal photography, fluorescein angiography (FA), and Am sler grid testing are optional,
although they m ay be helpfu l for patients w ith pre-existing m acu lar pathology. Perimetry
(Hu mph rey 10-2) w ith or w ithout a red test object m ay detect an early paracentral scotom a,
as fu nctional changes u su ally occu r before visible retinal abnorm alities. Sequential mon i-
toring m ay be requ ired to detect a change. Fu ndu s autofluorescence m ay d isclose early
pigment changes, though th is has yet to be correlated w ith long-term fu nctional changes.
101
102 Question 27
Color vision testing is not sen sitive enough as a screen ing tool. Mu ltifocal electroretinog-
raphy m ay provide an objective m ethod to detect early toxicity and mon itor recovery after
med ication d iscontinu ation; however, th is test is not u n iform ly available.2 Recent stud ies
w ith spectral dom ain optical coherence tomography (SD-OCT) have show n th in n ing of
the in ner retina wh ich m ay precede clin ically apparent d am age. However, th is test can not
yet be utilized as a defin itive screen ing tool.3
An nu al screen ing exam ination s shou ld be performed. Signs of toxicity include changes
in visual acu ity, paracentral scotom a, m acu lar pigmentary changes (Figu re 27-1), and
Am sler grid or visu al field defects. Concerns of toxicity warrant more frequent exam ina-
tions, and the med ication shou ld be d iscontinued when there are defin ite signs of toxicity.
There is no treatm ent once retinopathy develop s.
Th iorid azine (Mellaril [Novartis, Basel, Sw itzerland]) is an antipsychotic d rug that was
once w idely utilized. However, over concern of sign ificant retinal toxicity, it cu rrently is
less frequently prescribed. From the retinal standp oint, it is one of the most toxic med ica-
tions. A d aily dosage greater than 800 mg/ d ay can cau se con siderable retinal d am age,
though when utilized at a lower dosage, there is less concern. Acute toxicity m ay present
w ith a variable degree of central visual impairment, dysch rom atopsia, and nyctalopia.
Early fu ndu s abnorm alities include granu lar pigment stippling, wh ich can read ily prog-
ress to circu m scribed areas of retinal pigm ent epithelial (RPE) and choriocapillaris loss,
often in a nu m mu lar pattern (Figu re 27-2). Photography and FA are u sefu l ancillary
tests. FA h igh lights the irregu larity of the RPE in itially followed by severe atrophy of the
choriocapillaris in the late stage of d isease (Figu re 27-3). Visu al field s, electroretinogra-
phy, and electro-ocu lography m ay be variably effective but are not good screen ing tools.
If evidence of toxicity is d iscovered, the med ication shou ld be promptly d iscontinued.
Progressive retinal changes m ay occu r for years since the m ed ication is indefin itely stored
in the eye.4,5
Tamoxifen (Nolvadex [AstraZeneca, Wilm ington, DE]), an estrogen antagon ist, is
u sed in the treatment of metastatic breast adenocarcinom a and advanced glioblastom a
mu ltiform e. Despite early descriptions of crystalline retinopathy at h igh doses (60 to 100
m g/ d ay), toxicity is u nu su al at doses of 20 mg/ d ay or less, although crystals m ay still be
seen in 3% to 4% w ith ch ron ic ad m in istration. Early symptom s include m ild decreased
vision and dysch rom atopsia. Clin ically, yellow ish-wh ite intraretinal crystalline deposits
are noted, u su ally in a parafoveal d istribution (Figu re 27-4). Other associated featu res
include corneal verticillata, retinal pigmentary changes, and cystoid m acu lar edem a
(CME). Optical coherence tomography m ay reveal hyperreflective in ner plexiform depos-
its, photoreceptor d isruption, and CME. Despite crystalline deposition, there is typically
little change in retinal fu nction, and asymptom atic patients are generally mon itored, as
the med ication is tru ly needed for treatment of the u nderlying m alignancy.4,5 However, if
there is con firm ed evidence of retinal dysfu nction, then consideration shou ld be given to
d iscontinue the med ication.
Sildenafil (Viagra [Pfizer, New York, N Y]), as well as other selective phosphod iesterase-
5 in h ibitors, are com mon ly prescribed for m ale erectile dysfu nction. Although selective
for the p en ile corpora cavernosa, cross-reactivity w ith phosphod iesterase-6 in photo-
receptors occu rs and m ay be resp onsible for ocu lar changes. Blu ish dyschrom atopsia,
halos, and increased light sensitivity lasting m inutes to hou rs m ay occu r after ingesting
the lowest therapeutic dose (25 mg/ d ay).4,5 These symptom s occu r even more frequently
What Systemic Medications Require Periodic Fundus Evaluation? 103
Fig u re 27-1. Hydroxych lo ro quin e m a culo pa thy in a typica l bull’s eye pa tte rn in a pa tie n t be in g
tre a te d fo r rh e um a to id a rth ritis. Th e FA do cum e n ts a tra n sm issio n de fe ct co rre spo n din g to th e
a re a o f pigm e n t a lte ra tio n . Th e visua l a cuity wa s 20 / 6 0.
Fig u re 27-2 . Th io rida zin e toxicity in a ge o gra ph ic n um m ula r pa tte rn . Th e FA do cum e n ts ge o -

gra ph ic lo ss o f RPE a n d ch o rio ca pilla ris. Th e visua l a cuity wa s 20 / 20 0.
Fig u re 27-3 . Adva n ce d th io rida zin e toxicity re se m blin g re tin itis pigm e n to sa . Th e FA re ve a ls vir-
tua lly co m ple te lo ss o f RPE, a llo win g fo r visua liza tio n o f th e la rge r ch o ro ida l ve sse ls. Th e visua l
a cuity wa s h a n d m o tio n .
104 Question 27
Fig u re 27-4 . Ta m oxife n crysta llin e re tin o pa thy. Ye llo wish -wh ite in tra re tin a l crysta llin e de po sits
a re se e n in a pe rifove a l distributio n . Th e re is pro m in e n t un de rlyin g CME. (Co urte sy o f Da vid
Sa rra f.)
at h igher doses (100 mg/ d ay). Reports describing objective find ings are u ncom mon but
include retinal hemorrhages, retinal vascu lar occlu sion, acceleration of proliferative d ia-
betic retinopathy, ocu lomotor nerve palsy, central serou s choroidopathy, and nonarteritic
anterior ischem ic optic neu ropathy (NAION ). There are no recogn ized long-term toxic
effects follow ing ch ron ic ad m in istration of the med ication, although patients who have
exp erienced NAION in one eye shou ld be d iscou raged from u sing sildenafil, as it m ay
potentiate the risk of developing NAION in the fellow eye. There is no ind ication for
routine screen ing at th is time, but cou n seling patients w ith pre-existing retinal d isease is
prudent, and symptom atic patients shou ld be evalu ated for retinal pathology.
Summary
Although there are thou sand s of system ic med ication s, on ly a sm all nu m ber of these
agents produce retinal changes. Retinal toxicity can occu r when agents are u sed at stan-
d ard therap eutic dosages or when utilized for non-approved ind ications. With mu ltiple
new d rugs reach ing the m arket an nu ally, ophthalmologists need to m aintain a h igh index
of su spicion that patients’ clin ical find ings and visu al symptom s m ay be related to one or
more of their system ic m ed ications.

Dan iel F. Kiernan, MD
Vitreoretin al Fellow, Un iversity of Illinois at Ch icago, Ch icago, IL
What Systemic Medications Require Periodic Fundus Evaluation? 105
References
1. Marmor MF, Carr RE, Easterbrook M, Farjo AA, Mieler WF, American Academy of O phthalmology.
Recommendations on screening for chloroquine and hydroxychloroquine retinopathy: a report by the
American Academy of O phthalmology. O phthalm ology. 2002;109 (7):1377-1382.
2. Mittra RA, Mieler WF. Drug toxicity of the posterior segment. In: Ryan SJ, ed. Retina. 4th ed. St. Louis, MO :
CV Mosby-Year Book, Inc; 2006:1779-1796.
3.. Pasadhika S, Fishman GA: Effects of chronic exposure to hydroxychloroquine or chloroquine on inner retinal
structures. Eye. 2009. (Epub ahead of print).
4. Lyons JS, Severns ML. Using multifocal ERG ring ratios to detect and follow Plaquenil retinal toxicity: a review:
review of mfERG ring ratios in Plaquenil toxicity. Doc O phthalm ol. 2009;118 (1):29-36.
5. Lim RT, Mieler WF. Retinal toxicity of xystemic (and topical) medications. In: Nguyan Q D, Farah M, Rodrigues
E, Mieler WF, eds. Retinal Pharm acotherapy, London, England: Elsevier; In press.
28 QUESTION
HOW SHOULD I FOLLOW

DIABETIC MACULAR EDEMA AND
WHEN DOES IT REQUIRE TREATMENT?
Peter K. Kaiser, MD
Diabetic m acu lar edem a (DME) is present in approxim ately 20% of type 1 and typ e 2
d iabetics.1 DME can be seen in all stages of d iabetic retinopathy, and it is the most com-
mon cau se of vision loss in eyes w ith nonproliferative d iabetic retinopathy (N PDR). Early
retinal edem a occu rs in areas w ith capillary m icroaneu rysm s (MA), a hallm ark of N PDR.
Retinal hard exud ates (H E) are also fou nd in association w ith DME and represent leakage
of lipoprotein plasm a con stituents. Leakage occu rs becau se of increased vascu lar perme-
ability arou nd MA, wh ich likely reflects a breakdow n of the m icrovascu lar blood-retinal
barrier second ary to capillary pericyte loss and basem ent mem brane abnorm alities from
ch ron ic hyp erglycem ia, and the release of perm eability factors such as vascu lar endothe-
lial grow th factor (VEGF) due to m acu lar ischem ia.
We approach the m anagement of early DME largely accord ing to the gu idelines estab-
lished by the land m ark Early Treatment Diabetic Retinopathy Study (ETDRS),2,3 wh ich
exam ined the role of argon laser photocoagu lation in these patients. The location and
extent of d iabetic m acu lar edem a are imp ortant in th is decision-m aking process. The
ETDRS reported that clin ically sign ificant m acu lar edem a (CSME) was likely to resu lt
in vision loss, and defined CSME as retinal th icken ing w ith in 500 µm of the fovea, H E
w ith in 500 µm of the fovea in association w ith adjacent retinal th icken ing, or 1 d isc area
or more of retinal th icken ing w ith in 1 d isc d iameter of the fovea. Based on these criteria,
eyes w ith ME that do not m eet the CSME defin ition are mon itored every 3 to 4 months,
and eyes w ith more d istant edem a are followed every 4 to 6 months. In contrast, patients
w ith CSME are treated regard less of their visual acu ity.
107
108 Question 28
Ou r first-line treatm ent for CSME is focal/ grid m acu lar laser photocoagu lation applied
to areas of MA and leakage. Although CSME is a clin ical d iagnosis, fluorescein angiog-
raphy (FA) can help outline areas of MA and leakage and gu ide treatment targeting, par-
ticu larly in those eyes need ing repeat laser treatment. In add ition, FA can delineate areas
of ischem ia. The ETDRS showed that application of focal/ grid laser to areas of leakage
reduced the rate of moderate visu al acu ity loss (≥ 3 lines or 15 letters) by 50% (12% over
3 years compared to 24% among u ntreated eyes). We apply a light grid of 50 to 100 µm
bu rns over d iffu se areas of th icken ing and focally target MA w ith 50 µm spots, always
staying at least 500 µm from the fovea. We u se very light bu rns that produce a barely
p erceptible change in color. In special cases, such as H E th reaten ing the fovea, we w ill
treat earlier and mon itor more frequently. When applying m acu lar laser, it is important to
in form the patient of u ncom mon but sign ificant potential complications, includ ing para-
central scotom as and choroid al neovascu larization (CN V).
Sp ecial challenges are p osed in cases w ith d iffu se leakage or those w ith a prom inent
cystic component w ith extensive foveal involvement. These cases are often refractory
to laser therapy, particu larly since laser can not be applied d irectly to foveal th icken ing.
The recent Diabetic Retinopathy Clin ical Research Network (DRCR.net) h igh lighted the
imp ortance of a trial of focal/ grid laser as the appropriate first-line approach in these
cases.4 However, despite the DRCR.net find ings, we have a lower th reshold for trying
alternative approaches, either alone or in conju nction w ith rep eat laser therapy, if there
is no improvement in the edem a after 2 or more laser sessions over a 6-month period. In
these cases, FA is u sefu l in identifying new areas of leakage and determ in ing the extent
of m acu lar ischem ia, wh ich can be exacerbated by excessive laser.
The u se of corticosteroid s for d iabetic m acu lar edem a has been evaluated in variou s
settings, and we have fou nd that triamcinolone aceton ide, ad m in istered locally via either
an intravitreal or subtenon route, to be an effective alternative and/ or adju nct to laser in
refractory cases of DME.5 In ou r experience, visual acu ity often improves by 1 to 2 lines
and optical coherence tomography (OCT) central retinal th ickness decreases w ith in 3
month s of injection, and the effects are at least partially m aintained at 6 and even 12
month s. However, it is important to note that the recent DRCR.net d id not find a benefit
in the u se of intravitreal steroid s in patients who have CSME at baseline.4 We wou ld not
have performed steroid injections in the m ajority of patients who were entered in the
DRCR.net trial, instead reserving steroid s for those who have failed laser or have persis-
tent d iffu se DME. We reassess the DME every 3 to 4 month s, and in cases w ith recu rrent
edem a, rep eat injections m ay be requ ired. It is imp ortant to follow patients closely for the
well-docu mented side effects of cataract progression and intraocu lar pressu re (IOP) eleva-
tion after steroid injection s that m ay requ ire add itional m anagement but typically do not
resu lt in perm anent vision loss. Most cases of IOP elevation can be m anaged m ed ically. It
is possible that com bination treatment w ith steroid s and focal/ grid laser m ay work even
better, and th is is being evaluated in clin ical trials. If these stud ies show a benefit to ste-
roid s in DME, the u se of long-term steroid implants cou ld be utilized to avoid repeated
injection s. If th is is the case, off-label u se of non-erod able su stained release steroid
implants such as the fluocinolone aceton ide implants (Retisert, Med idu r [Alim era]) and
injectable biodegrad able dexamethasone steroid implants (Ozu rdex) m ay find a role in
futu re DME m anagement.
How Should I Follow Diabetic Macular Edema? 109
Given the resu lts of the DRCR.net study, where there was no benefit of intravitreal
steroid injection s over laser treatm ent at the 2- and 3-year follow-up tim e points, there
has been a renewed interest in the u se of in h ibitors of vascu lar endothelial grow th fac-
tor (VEGF) such as p egaptan ib sod iu m (Macugen), ran ibizu m ab (Lucentis), and bevaci-
zu m ab (Avastin) to reduce vascu lar permeability. Their u se in DME has been rep orted
in lim ited stud ies w ith som e short-term benefit.6 Like steroid s, rep eated injections are
typically requ ired. The u se of intravitreal anti-VEGF agents are cu rrently being stud ied in
random ized phase 3 clin ical trials. We do not feel that com bination of an anti-VEGF agent
w ith triamcinolone offers add itional benefit over triamcinolone alone; however, com bina-
tion therapy w ith laser is being stud ied.
OCT is a u sefu l mon itoring tool for assessing serial changes in central m acu lar th ick-
ness in response to treatment and for evaluating the extent of the m acu lar edem a and
cysts. It is also u sefu l in evalu ating the presence of posterior hyaloid al traction that cou ld
be exacerbating the DME. These cases wou ld benefit from pars plana vitrectomy w ith
p eeling of the posterior hyaloid, and in som e cases, peeling of the internal lim iting mem-
brane, in stead of m ed ical therapy.
Summary
It is important to com mu n icate regu larly w ith prim ary care providers and to explain
to patients the importance of good glycem ic, blood pressu re, and seru m lipid control
since all of these variables affect visu al prognosis. Focal/ grid laser photocoagu lation is
the m ainstay of m anagement of d iabetic m acu lar edem a. In d ifficu lt and d iffu se leak-
age cases, although not proven in a large random ized clin ical trial, the u se of anti-VEGF
agents and steroid s have been u sed w ith good success. Thu s, a w ide sp ectru m of DME
can be m anaged successfu lly u sing a com bination of treatm ents, carefu l follow-up at 3- to
6-month intervals w ith appropriate d iagnostic im aging, and optim ization of system ic risk
factors.
Financial disclosure: Dr. Kaiser has received honoraria from A lcon, A llergan, Bausch & Lomb, Eli Lilly,
Genentech, Occulogix, N ovartis, TargeGen, and QLT that have been disclosed to the Conflict of Interest
Committee of the Cleveland Clinic. The Cole Eye Institute, Dr. Kaiser’s employer, has received research grant
support on his behalf from A lcon, A llergan, Bausch & Lomb, OSI-Eyetech, Genentech, N ational Eye Institute,
N ovartis, Regeneron, SIRNA , TargeGen, Quark, and QLT.

H ajir Dad gostar, MD, PhD
Vitreoretin al Fellow, Cole Eye In stitute, Cleveland, OH
110 Question 28
References
1. Ryan SJ, Puliafito CA, Davis JL, Parel JM, Milne P. Retina. 4th ed. Philadelphia, PA:Elsevier Mosby; 2006:3
v.(xxiv, 2654, 64).
2. Photocoagulation for diabetic macular edema. Early Treatment Diabetic Retinopathy Study Report Number 1.
Early Treatment Diabetic Retinopathy Study research group. Arch O phthalm ol. 1985;103(12):1796-1806.
3. Treatment techniques and clinical guidelines for photocoagulation of diabetic macular edema. Early Treatment
Diabetic Retinopathy Study Report Number 2. Early Treatment Diabetic Retinopathy Study Research Group.
O phthalm ology. 1987;94(7):761-774.
4. Diabetic Retinopathy Clinical Research Network (DRCR.net) Study Group. Three-year follow-up of a random-
ized trial comparing focal/grid photocoagulation and intravitreal triamcinolone for diabetic macular edema.
Archives of O phthalm ology. 2009;127(3):245-51.
5. Bakri SJ, Kaiser PK. Posterior subtenon triamcinolone acetonide for refractory diabetic macular edema. Am J
O phthalm ol. 2005;139 (2):290-294.
6. Fraser-Bell S, Kaines A, Hykin PG. Update on treatments for diabetic macular edema. Curr O pin O phthalm ol.
2008;19 (3):185-189.
29 QUESTION
WHEN SHOULD EYES

WITH PROLIFERATIVE DIABETIC
RETINOPATHY RECEIVE PANRETINAL
LASER PHOTOCOAGULATION?
Ivan J. Suñer, MD
Diabetic retinopathy is the lead ing cau se of blind ness in patients 20 to 64 years old in
developed cou ntries. In eyes w ith proliferative d iabetic retinopathy (PDR), fibrovascu lar
proliferation resu lts from ischem ia and release of vasoproliferative factors. Proliferation
extend s beyond the internal lim iting mem brane (ILM) and resu lts in vitreou s hemor-
rhage (VH ) and vitreoretinal traction, wh ich can lead to severe vision loss (SVL).
Pan retinal photocoagu lation (PRP) is effective in stabilizing PDR and reducing the risk
of severe visu al loss. The goal of PRP is to induce regression of existing neovascu lar tissue
and to prevent progressive neovascu larization.
The Diabetic Retinopathy Study (DRS) was a random ized, controlled, prosp ective
clin ical trial evaluating PRP in eyes w ith clear med ia and h igh-risk PDR. H igh-risk PDR
was defined as neovascu larization of the d isc (N VD) involving 25% to 33% of a d isc area
w ith VH, moderate to severe N VD w ith or w ithout VH, or neovascu larization elsewhere
(N VE) on the retina w ith VH.1 The DRS demon strated a 50% reduction in the risk of
severe visual loss, defined as visu al acu ity of less than 5/ 200 on 2 consecutive follow-up
exam inations 4 months apart.2 If there is concu rrent clin ically sign ificant m acu lar edem a
(CSME), th is shou ld be treated either before or at the same tim e as the first session of
PRP.
When d iscu ssing the procedu re w ith the patient, I emphasize that we are trying to
prevent SVL and other neovascu lar complication s. I also d iscu ss the p otential second ary
111
112 Question 29
effects such as decrease in peripheral vision, n ight vision, color vision, as well as possible
temporary loss of accom mod ation. I also d iscu ss the potential for the developm ent or
progression of CSME, choroid al effu sions, and VH and the possible need for vitrectomy
su rgery.
In add ition to h igh-risk PDR, neovascu larization of the iris (N VI) is another ind ication
for PRP. These eyes have a sign ificant risk of developing neovascu lar glaucom a due to
neovascu larization of the angle (N VA). PRP is ind icated prior to glaucom a su rgery as it
can control progression of N VA and, in some cases, avoid the need for glaucom a filtration
su rgery. Fu rthermore, PRP helps decrease postop erative VH after glaucom a su rgery.
I perform PRP as soon as possible, preferably on the d ay of d iagnosis (esp ecially in
eyes w ith anterior segment neovascu larization) and u su ally no later than 1 week. The key
is to m ake the patient as com fortable as possible and m ake it as conven ient as possible.
I try to p erform it u nder topical anesthesia in 2 sessions if the patient can tolerate it. I
prefer slit-lamp delivery u sing a contact lens, typically a w ide-field len s. The laser p ower
settings vary dep end ing on the lens, laser, degree of fu ndu s pigmentation, and degree of
vitreou s hemorrhage. In cases where VH m akes treatm ent d ifficu lt, krypton or red laser
wavelength s m ay allow better p enetration th rough hemorrhage. In general, I begin w ith a
spot size of at least 300 µm and du ration of 0.1 second. The goal is to ach ieve wh ite bu rns
approxim ately 1 to 1.5 bu rn-w idths apart.
In the first session, I ideally treat the posterior retina from the area arou nd the arcades,
approxim ately 1 d isc d iameter from the optic nerve and from the temporal m acu la, and
anteriorly into the in ferior one-half of the retina. I do the latter, if p ossible, in case there
is su bsequent VH, wh ich u sually settles in feriorly. I w ill then complete the laser treat-
ment in a second session. The more anterior retina can be accessed w ith a 3-m irror lens
or ind irect laser.
If the patient can not tolerate the procedu re w ith a topical anesthetic, I w ill ad m in ister
a peribu lbar or retrobu lbar block w ith 2% lidocaine. In these eyes, I w ill try to do as much
as possible in the first session and in some cases, the entire treatment if possible. Com fort
is of the utmost importance as one wants the patient to retu rn for futu re exam s and treat-
ments.
Vascu lar endothelial grow th factor (VEGF) antagon ists have become a u sefu l adju nct
in select, d ifficu lt circu m stances. VEGF is believed to be the prom inent factor released in
resp on se to retinal ischem ia and d rives angiogenesis. In eyes where there is N VA cau s-
ing corneal edem a from elevated intraocu lar pressu re, d isp ersed red blood cells in the
anterior or p osterior segment, or su boptim al d ilation due to N VI, intraocu lar injection
of VEGF antagon ists m ay provide qu ick resolution (1 to 2 d ays) to facilitate PRP. These
agents effectively “tu rn back the clock” in PDR and temporarily induce regression of neo-
vascu larization. VEGF antagon ists do not ach ieve perm anent regression and therefore are
not a su bstitute for PRP in these cases. Another caution is the p otential for progression
of traction retinal detach ment due to rapid conversion of active fibrovascu lar tissue to
contracting fibrotic tissue. Un less vitreoretinal su rgical intervention is plan ned w ith in 1
week, injection of these agents shou ld be u sed w ith caution in eyes w ith traction or large
patches of N VE due to PDR.
PRP for PDR is a time-tested, effective therapy and rem ains the cornerstone of treat-
ment for PDR and its neovascu lar complication s.
When Should Eyes With PDR Receive Panretinal Laser Photocoagulation? 113
References
1. Four risk factors for severe visual loss in diabetic retinopathy. The third report from the Diabetic Retinopathy
Study. The Diabetic Retinopathy Study Research Group. Arch O phthalm ol. 1979;97(4):654-655.
2. Photocoagulation treatment of proliferative diabetic retinopathy: clinical application of diabetic retinopathy
study (DRS) findings. DRS Report 8. The Diabetic Retinopathy Study Research Group. O phthalm ology.
1981;88 (7):583-600.
30 QUESTION
WHEN SHOULD I REFER

A PATIENT WITH DIABETIC
RETINOPATHY FOR A VITRECTOMY?
Ronald C. Gentile, MD
Diabetic retinopathy is the lead ing cau se of blind ness in American s of working age.1
Vitrectomy is reserved for eyes w ith complications of d iabetic retinopathy not amenable
to laser or intravitreal injection s and is necessary in at least 5% of patients.2 The goals
of vitrectomy in patients w ith d iabetes are to clear the vitreou s of opacities, release and
remove tractional elem ents from the retinal su rface, and complete pan retinal photoco-
agu lation (PRP), if needed.
The 2 predom inant classic ind ication s for vitrectomy in d iabetes (ie, nonclearing vit-
reou s hemorrhage and tractional retinal detach ment [RD] involving the m acu la) have
rem ained the same since the resu lts of the Diabetic Retinopathy Vitrectomy Study (DRVS)
were pu blished.3,4 Since the DRVS, the ind ications for vitrectomy in patients w ith d iabetes
have not on ly expanded, but the waiting time to su rgery in som e cases has been reduced
due to advances in in stru m entation, view ing system s, and su rgical tech n iques.
I have d ivided ind ications for d iabetic vitrectomy into 5 broad categories (Table 30-1).
These categories include RD, hemorrhage, severe retinal neovascu larization, m acu lar
pathology, and postvitrectomy. Since it is not u ncom mon for eyes to have more than one
ind ication for d iabetic vitrectomy, a sixth category was added to include com binations of
1 to 5. Most ind ications are for complications of proliferative d iabetic retinopathy (PDR)
w ith a sm all percentage due to complications of nonproliferative d iabetic retinopathy.
115
116 Question 30
Table 30-1
Indications for Vitrectomy in Eyes With Diabetic Retinopathy

1. Retinal Detachment
• Tractional RD involving or threatening the macula
• Combined tractional-rhegmatogenous RD
• Rhegmatogenous RD
2. Hemorrhage
• Nonclearing vitreous hemorrhage
• Subhyaloid premacular loculated hemorrhage
• Vitreous hemorrhage with anterior segment neovascularization
• Bilateral vitreous hemorrhage
• Subretinal hemorrhage
• Other vitreous opacities (ie, asteroid hyalosis, amyloidosis, inflammatory cells)
3. Severe Retinal N eovascularization and Fibrovascular Proliferation

4. M acular Pathology
• Taut hyaloid
• Premacular fibrosis with or w ithout neovascularization
• Vitreomacular traction syndrome
• Macular hole
• Persistent macular edema
5. Postvitrectomy
• All of the above that apply
• Anterior hyaloidal fibrovascular proliferation
• Fibrinoid syndrome
• Re-proliferation of preretinal membranes
• Ghost cell glaucoma
6. Combinations of 1 through 5
Retinal Detachment
Tractional RD (Figu re 30-1) involving or th reaten ing the m acu la is one of the classic
ind ications for d iabetic vitrectomy. Th is occu rs when retinal neovascu larization becomes
fibrotic and contracts, cau sing the u nderlying retina to elevate and detach. Th is can occu r
w ith or w ithout a partial posterior vitreou s detach ment. Tractional RD not involving the
m acu la is u su ally not an ind ication for su rgery since it often rem ains stable. Th is is espe-
cially true when the retinal neovascu lar comp onent is inactive and fibrotic, most often
seen after fu ll prior PRP has been p erformed.
Com bined tractional-rhegm atogenou s RD (Figu re 30-2) u su ally is a resu lt of a
partial p osterior vitreou s detach ment in the setting of cicatricial proliferative d ia-
betic retinopathy. These detach ments can progress very qu ickly and requ ire extensive
When Should I Refer a Patient With Diabetic Retinopathy for a Vitrectomy? 117
A B
Fig u re 3 0 -1. (A) Tra ctio n a l dia be tic re tin a l de ta ch m e n t invo lvin g th e m a cula . (B) Two m o n th s
po stdia be tic vitre cto m y with re m ova l o f th e fibro tic n e ova scula riza tio n a n d re le a se o f th e tra c-
tio n . Th e re tin a subse que n tly re a tta ch e d with re sidua l m a cula r tra n spo sitio n .
A B
Fig u re 3 0 -2 . (A) Co m bin e d dia be tic tra ctio n a l-rh e gm a to ge n o us re tin a l de ta ch m e n t. No te th e

o ute r re tin a l hydra tio n lin e ove r a pa rtia lly co nve x surfa ce supe rio rly. (B) Th re e m o n th s po st-
dia be tic vitre cto m y a fte r re m ova l o f th e fibro tic n e ova scula riza tio n , re le a se o f a ll tra ctio n ,
dra in a ge o f fluid fro m th e re tin a l bre a k, e n do la se r, a n d re a tta ch m e n t o f th e re tin a usin g ga s
ta m po n a de .
hyaloid and mem brane removal to repair. Breaks in tractional-rhegm atogenou s retinal
detach m ents are u su ally sm all and located adjacent to tractionally elevated epicenters
posterior to the equ ator, but sometim es they occu r at old chorioretinal laser scars. These
detach m ents can also originate from a m acu lar hole, although th is is much less com mon.
Many breaks are on ly identified intraoperatively since fibrosis and hemorrhage m ay pre-
vent preoperative visu alization. Un like solely tractional RDs, tractional-rhegm atogenou s
RDs are more mobile and have outer retinal hyd ration lines and a partially convex su r-
face.
118 Question 30
Hemorrhage
Vitreou s hemorrhage is the other classic ind ication for d iabetic vitrectomy when it is
visu ally sign ificant and nonclearing. The hemorrhage is considered visually sign ificant
if it is resp onsible for visu al acu ity less than 20/ 400 and the opacity prevents laser pho-
tocoagu lation. To determ ine if the hemorrhage is nonclearing, the patient wou ld have to
give a h istory of no improvement or be exam ined w ithout improvement for at least 3 to 4
weeks. When exam in ing a patient w ith vitreou s hemorrhage, it is important to determ ine
the density of the hemorrhage, the statu s of the u nderlying retina by B-scan echography,
the presence or absence of anterior segment neovascu larization, the statu s of the fellow
eye, the extent of prior PRP, and the type of d iabetes. Ultrasonography shou ld ru le out
a rhegm atogenou s RD or tractional RD involving the m acu la that wou ld supersede the
vitreou s hemorrhage as an ind ication for su rgery. Reasons for an earlier vitrectomy also
include vitreou s hemorrhage in the fellow eye, type 1 d iabetes w ithout prior PRP and very
active PDR, and anterior segment neovascu larization (Table 30-2.) Preop erative intravit-
real bevacizu m ab (Avastin [Genentech USA, Inc, San Francisco, CA]) m ay play a role in
the m anagem ent of such patients, especially in the latter 2 cond ition s. However, the devel-
opment or progression of a tractional retinal detach ment after its u se m ay requ ire earlier
vitreoretinal su rgical intervention that plan ned.
A su bhyaloid, prem acu lar, locu lated hemorrhage (Figu re 30-3), especially when th ick
and not associated w ith any progressive detach ment of the hyaloid, is an ind ication for
vitrectomy. Den se prem acu lar hemorrhage can lead to progressive fibrosis and, even if it
clears, can resu lt in a prem acu lar fibrotic mem brane. Su bretinal hemorrhage is rare and
generally associated w ith a very ad herent and attached hyaloid or retinal break. Th is is
most often d iagnosed intraoperatively.
Other vitreou s opacities, when severe enough to prevent treatment of d iabetic retinopa-
thy, are an ind ication for vitrectomy. These include the rare case of den se asteroid hyalosis,
amyloidosis, or in flam m atory cells.
Severe Retinal Neovascularization

and Fibrovascular Proliferation
Retinal neovascu larization (Figu re 30-4), in the presence of an attached hyaloid, can
sometimes cover the optic nerve and the entire temporal arcade. The natu ral h istory of
th is process is fibrosis and traction as the neovascu larization cicatrizes and resu lts in a
broad-based tractional RD w ith tabletop con figu ration. Vitrectomy in these cases can be
d ifficu lt but can prevent inevitable blind ness when successfu l.
Macular Pathology
In a sm all su bset of eyes, a taut hyaloid resu lts in persistent m acu lar edem a due to tan-
gential vitreou s traction. The m acu lar edem a is u su ally d iffu se, and removing the hyaloid
is very effective in decreasing the edem a. Prem acu lar fibrosis can occu r and con sists of
Table 30-2
Vitreous Hemorrhage:
Factors That Support Diabetic Vitrectomy
Vitrectomy No Yes
Vitreous hemorrhag e Not severe Severe
Density PRP possible PRP not possible
B-scan ultrasound PVD No PVD
Tractional macular elevation
Anterior seg ment neovascularization Absent Present
Fellow eye Stable Active PDR
Good vision Poor vision
Diabetes type ll l
Prior PRP Yes No
Severe retinal neovascularization No Yes
Inactive Active
PRP = panretinal photocoag ulation, PVD = posterior vitreous detachment
Fig u re 3 0 -3 . Subhya lo id, pre m a cu-

la r, lo cula te d h e m o rrh a ge .
a fibrotic and contracted hyaloid m em brane in front of the m acu la, cau sing obscu ration
of the fovea. Vitreom acu lar traction synd rome (Figu re 30-5) can resu lt when the hyaloid
rem ains attached to the fovea or perifoveal region and resu lts in edem a and loss of vision.
OCT has been very helpfu l in identifying th is cond ition. Macu lar holes can occu r in the
setting of d iabetic retinopathy and can resu lt in a posterior RD when associated w ith
exten sive fibrovascu lar proliferation. Persistent m acu lar edem a can be an ind ication for
vitrectomy when all other treatment mod alities fail. Most su rgeons con sider th is a treat-
ment of last resort.
120 Question 30
Fig u re 3 0 -4 . Se ve re re tin a l n e ova s-

cula riza tio n a n d fibrova scula r pro -
life ra tio n de spite PRP.
Fig u re 3 0 -5 . Vitre o m a cula r tra ctio n

syn dro m e co m plica tin g dia be tic re t-
in o pa thy im a ge d by o ptica l co h e r-
e n ce to m o gra phy.
Postvitrectomy
Ind ications u n ique to the postd iabetic vitrectomy eye include anterior hyaloid al fibro-
vascu lar proliferation and fibrinoid synd rome (Figu re 30-6). These complications occu r
in the most severely affected eyes and, w ithout rep eat su rgery, these eyes w ill u su ally lose
light perception and progress to phth isis bu lbi. Anterior hyaloid al fibrovascu lar prolifera-
tion presents most com mon ly w ith early p ostoperative vitreou s and/ or retrolental hem-
orrhage w ith extraretinal neovascu larization extend ing from the anterior retina to the
anterior hyaloid face and beh ind the len s. Fibrinoid synd rome is the most severe form of
p ostvitrectomy fibrin form ation w ith fibrin strand s and sheets across the vitreou s cavity.
Tissue plasm inogen activator can be u sed to break dow n the fibrin in the p erioperative
p eriod. The u se of intraoperative and postop erative triamcinolone, bevacizu m ab, and
extensive endolaser m ay decrease the incidence of these complication s.
Preretinal reproliferation of m em branes can occu r after d iabetic vitrectomy and is
m anaged su rgically, if severe. In some cases, these mem branes can be very ad herent and
can exert traction or act as a scaffold for neovascu larization in the absence of an intact
posterior hyaloid. Ghost cell glaucom a is m anaged su rgically if the intraocu lar pressu re
can not be controlled m ed ically.
Fig u r e 3 0 - 6 . Fibrin o id syn d ro m e

co m plica tin g d ia b e tic vitre cto m y.
No te th e fibrin stra n ds be h in d a n
in ta ct pe riph e ra l ca psule .
Combinations of 1 Through 5
Many eyes have mu ltiple ind ications for d iabetic vitrectomy. Progressive RDs and
m ed ically u ncontrollable neovascu lar glaucom a w ith med ia opacities take priority over
the other ind ication s.
Summary
The ind ications for vitrectomy in patients w ith d iabetes continue to expand. In add ition
to advances in su rgical instru mentation, the more recent u se of p eriop erative intravitreal
pharm acologic therapies has resu lted in improved outcomes for patients w ith d iabetes.
Financial disclosure: Research support from IRIS M edical Instruments, Controlled Delivery Systems,
Ely Lilly, Genentech, Eyetech, Immusol, Columbia University, N otal Vision, A lcon, Ciba Vision, Regeneron,
Inspire, OPKO, A llergan, RegeneRx, N ational Eye Institute, and N ew York Eye and Ear Infirmary..
References
1. Centers for Disease Control and Prevention. National diabetes fact sheet: general information and national
estimates on diabetes in the United States, 2005. Atlanta, GA: US Department of Health and Human Services,
Centers for Disease Control and Prevention; 2005.
2. Flynn HW Jr, Chew EY, Simons BD, Barton FB, Remaley NA, Ferris FL 3rd. Pars plana vitrectomy in the Early
Treatment Diabetic Retinopathy Study. ETDRS report number 17. The Early Treatment Diabetic Retinopathy
Study Research Group. O phthalm ology. 1992;99 (9):1351-1357.
3. Early vitrectomy for severe vitreous hemorrhage in diabetic retinopathy: four-year results of a randomized trial:
Diabetic Retinopathy Vitrectomy Study Report Number 5. Arch O phthalm ology. 1990 ;108 (7):958-964.
4. Early vitrectomy for severe proliferative diabetic retinopathy in eyes with useful vision: results of a randomized
trial—Diabetic Retinopathy Vitrectomy Study Report 3. The Diabetic Retinopathy Vitrectomy Study Research
Group. O phthalm ology. 1988;95(10 ):1307-1320.
31 QUESTION
HOW DO I FOLLOW A PATIENT

WHO HAS DIABETES AND BECOMES
PREGNANT? WHAT TESTS CAN I DO?
Judy E. Kim, MD
Pregnancy complicates the ophthalm ic m anagement of patients w ith pre-existing

d iabetes. Patients who develop gestational d iabetes do not develop d iabetic retinopathy.
Hence th is chapter is on ly concerned w ith the first group. H istorically, stud ies have
focu sed on patients w ith type 1 d iabetes, though we presu me resu lts can be extrapolated
to patients w ith typ e 2 d iabetes.
Diabetic retinopathy is know n to worsen du ring pregnancy, although the reason s why
are still debated. Up to 27% of d iabetic pregnancies w ill have d iabetic retinopathy at som e
p oint.1 Both clin ically sign ificant m acu lar edem a and proliferative d isease m ay worsen
or develop de novo.2,3 Though proliferative d isease attracts the most attention, case series
have reported severe vision loss from m acu lar edem a.3 Rates of progression toward pro-
liferative d isease have varied w idely in the literatu re. In case series, anywhere from 8% to
70% of patients have show n progression.1 There is a com mon perception that retinopathy
progression to proliferative d isease du ring pregnancy reverses after delivery. However,
the facts do not bear th is out. One study fou nd that on ly 54% of patients improve post-
partu m,2 and another detailed a series of patients who demonstrated severe progression
follow ing delivery.4 The risk of progression has been correlated w ith the du ration of
d iabetes prior to pregnancy, rapid glucose norm alization du ring pregnancy, severity of
d iabetes at baseline, baseline glycem ic control, hyp ertension, and eclampsia.2 Therefore,
patients w ith these risk factors shou ld be closely mon itored and educated.
123
124 Question 31
How Do I Examine These Patients?

Du ring a clin ic visit, the same clin ical exam ination as recom mended for a nonpreg-
nant patient shou ld be followed. Th is includes visu al acu ity, intraocu lar pressu re, and
a thorough anterior segment exam ination w ith gon ioscopy if necessary to evaluate for
neovascu larization of the iris and angle. A d ilated fu ndu s exam is necessary, as u nd ilated
exam s m ay fail to d iagnose up to 50% of cases of d iabetic retinopathy.5 Cyclop entolate,
tropicam ide, and phenyleph rine are all category C for u se in pregnancy, although they are
presu m ably u sed frequently w ith no reported evidence of d anger. Nonetheless, it m ay be
prudent to u se the lower concentrations of these d rugs that are available com mercially or
to consider pu nctal occlu sion to reduce system ic circu lation of these d rugs.
Regard ing ancillary tests, fu ndu s photography and optical coherence tomography
clearly have no adverse effects or risk to the pregnancy. The real question, however, is
whether to get a fluorescein angiogram (FA). Angiography is not ben ign, and even in
nonpregnant patients there is a one in 200,000 chance of a seriou s adverse event or death.5
Fluorescein has no know n risks or teratogen ic effects in pregnancy, but neither has it
been conclu sively determ ined to be safe. Becau se of th is u ncertainty and med ico-legal
concerns, we don’t typically get an FA in pregnant patients. Fortu nately, an FA is not
absolutely necessary to d iagnose or treat d iabetic retinopathy. Fu rthermore, im ages from
optical coherence tomography, includ ing the topography m ap, are helpfu l to determ ine
the areas and the m agn itude of retinal th icken ing. These im ages can gu ide areas to be
treated and can help assess treatment effect or progression of retinal th icken ing.
How Often Do I Follow These Patients?

Id eally, d iabetic patients plan n ing to become pregnant shou ld have a baseline exam ina-
tion w ith any necessary care instituted. After conception, an eye exam ination shou ld be
done in the first trimester.5 Su bsequent follow-up schedu les dep end on the assessed risk
of progression, wh ich requ ires the judgment of the exam iner; d ifferent schedu les have
been suggested in the literatu re. The most recent pu blication of the Preferred Practice
Patterns by the American Academy of Ophthalmology for d iabetic retinopathy suggests
the same follow-up schedu le one wou ld u se for other d iabetic patients.5 Other sou rces
suggest at least every 3 month s 6 or every 4 to 6 weeks.7 Interestingly, som e case series in
the literatu re have followed patients every 2 weeks.1,4 Follow-up shou ld continue in the
p ostpartu m period as well, as ind icated by retinopathy severity.1
The recent case series by Rah m an and colleagues1 and Chan and coworkers 4 certain ly
suggest a cautiou s follow-up schedu le. Patients w ith more severe d isease at baseline were
more likely to progress and progressed earlier in their pregnancy. Those w ith no retinopa-
thy had a 9% rate of progression, wh ile those w ith early, u ntreated proliferative d isease
had a 75% rate of progression. Prior treatment d id not completely safegu ard against pro-
gression; 1 of 4 patients w ith prior pan retinal photocoagu lation requ ired fu rther treat-
ment. Rah m an and coworkers1 do concede that their study popu lation probably had a
lower rate of treatment compliance and pregestational d iabetic care than other groups in
the literatu re, accou nting for their h igh rates of progression. Chan and colleagues 4 specifi-
cally looked at patients whose retinopathy progressed. The key conclu sion one can d raw
How Do I Follow a Patient Who Has Diabetes and Becomes Pregnant? 125
from their paper is to treat pregnant patients w ith proliferative d isease and even those
w ith severe nonproliferative d isease, as it w ill likely progress du ring pregnancy and m ay
not regress after delivery.
How Do I Treat These Patients?

As in all d iabetic patients, optim al control of system ic hyperglycem ia and hyperten-
sion is essential. The m ain stay of treatm ents for d iabetic retinopathy in pregnancy is laser
photocoagu lation, w ith the sam e protocols and ind ications as for non-pregnant d iabetic
patients.5 Wh ile triamcinolone acetate (Kenalog), bevacizu m ab (Avastin), and ran ibizu m-
ab (Lucentis) cou ld also be theoretically u sed for treatment of m acu lar edem a or prolifera-
tive d isease in these patients, they are all category C for u se du ring pregnancy. Pregnancy
is a contraind ication for bevacizu m ab and ran ibizu m ab in ou r practice becau se the safety
of these med ications du ring pregnancy is not well know n at th is time. However, th is m ay
change w ith fu rther reports on u se of these d rugs du ring pregnancy. Su rgery is also an
option in select cases, but the risk to the mother and fetu s mu st be considered. Natu rally,
su rgery wou ld always be viewed as the last option and, if possible, avoided u ntil after
delivery.
Financial disclosure: Research funding from Genentech, N ovartis, A llergan, and N IH. Consulted for
A llergan.

Dara D. Koozekanan i, MD, PhD
Assistant Professor, Department of Ophthalmology
Un iversity of Min nesota, Min neap olis, MN
References
1. Rahman W, Rahman FZ, Yassin S, Al-Suleiman SA, Rahman J. Progression of retinopathy during pregnancy in
type 1 Diabetes Mellitus. Clin Experim ent O phthalm ol. 2007;35(3):231-236.
2. Soubrane G, Coscas G. Influence of pregnancy on the evolution of diabetic retinopathy. Int O phthalm ol Clin.
1998;38 (2):187-194.
3. Sinclair SH, Nesler C, Foxman B, Nichols CW, Gabbe S. Macular edema and pregnancy in insulin-dependent
diabetes. Am J O phthalm ol. 1984;972):154-167.
4. Chan WC, Lim LT, Q uinn MJ, Knox FA, McCance D, Best RM. Management and outcome of sight-threatening
diabetic retinopathy in pregnancy. Eye. 2004;18 (8):826-832.
5. American Academy of O phthalmology Retina Panel. Preferred Practice Pattern: Diabetic Retinopathy. San
Francisco, CA: American Academy of O phthalmology; 2003.
6. American Academy of O phthalmology. Basic and Clinical Science Course: Retina and Vitreous. Section 12.
2004-2005. San Francisco, CA: American Academy of O phthalmology; 2004.
7. Chatterjee S, Tsaloumas MD, Gee H, Lipkin G, Dunne FP. From minimal background diabetic retinopathy to
profuse sight-threatening vitreoretinal haemorrhage: management issues in a case of pregestational diabetes
and pregnancy. Diabet Med. 2003;20 (8):683-685.
32 QUESTION
WHEN DO I REFER A PATIENT WITH A

BRANCH RETINAL ARTERY O CCLUSION OR
CENTRAL RETINAL ARTERY O CCLUSION,
WHAT IS THE WORK-UP, AND WHAT
ARE THE TREATMENT O PTIONS?
Todd Klesert, MD, PhD
Although a thorough system ic work-up m ay not always yield a cau sative etiology,
there is no such th ing as an “id iopath ic” retinal artery occlu sion. Retinal artery occlu-
sions are often the sign of a larger, potentially seriou s system ic cond ition that m ay requ ire
treatm ent and therefore always warrant fu rther evalu ation. In my opin ion, th is is always
best accomplished in conju nction w ith the patient’s prim ary physician.
The cau ses of retinal artery occlu sion are m any but can be grouped into a few pri-
m ary categories: em bolic, coagu lopath ic, in flam m atory, trau m atic, in fectiou s, and struc-
tu ral. Overall, em bolism is the most frequent cau se of arterial occlu sion (Figu re 32-1).
For patients over age 40, atherosclerotic plaques in the carotid, aortic, and ophthalm ic
arteries are the most frequent sou rce of em boli. In patients u nder age 40, card iac sou rces
predom inate: valvu lar d isease (includ ing septic em boli from endocard itis), arrhyth m ia,
wall motion abnorm alities, myxom a, and septal defects such as patent foram en ovale in
the setting of deep venou s th rom bosis.
The su spicion for coagu lopathy shou ld increase as patient age decreases. Many of
the know n genetic risk factors for venou s th rom bosis (ie, Factor V Leiden mutation,
proth rom bin gene mutation, protein C and S deficiency, activated protein C resistance,
and antith rom bin III deficiency) have not been established as risk factors for arterial
127
128 Question 32
Fig u r e 3 2 -1. “Ho rse ”—Su p e ro te m p o ra l

BRAO in a 78 -ye a r-o ld fe m a le w ith a
Ho lle n h o rst pla que (a rro w ).
th rom bosis1, and I therefore do not routinely order these tests. Procoagu lant factors that
have been associated w ith arterial occlu sion include antiphospholipid antibod ies, anticar-
d iolipin antibod ies, elevated homocysteine levels, and sickle cell d isease.
One system ic in flam m atory d isease that shou ld always be forefront in the m ind of any
clin ician w ith an older patient w ith retinal artery occlu sion is giant cell arteritis (GCA).
Although less than 2% of artery occlu sion s are a resu lt of GCA, the con sequences of
m issing th is d iagnosis can be catastroph ic. Therefore, I routinely order a stat eryth rocyte
sed im entation rate (ESR) and C-reactive protein (CRP) on all patients over age 55 w ithout
an obviou s em bolu s on clin ical exam, even though they m ay not have any other signs or
symptom s of GCA. Patients for whom the su spicion for GCA is h igh shou ld be im me-
d iately started on h igh-dose oral steroid s. Other in flam m atory d iseases that shou ld be
con sidered, particu larly in you nger patients, are system ic lupu s erythem atosu s, polyar-
teritis nodosa, Wegener’s granu lom atosis, Behçet’s, derm atomyositis, and Su sac synd rome
(Figu re 32-2).
Trau m atic cau ses of retinal artery occlu sion are u su ally evident from the h istory.
Inadvertent, prolonged pressu re on the eye du ring su rgery or du ring a d rug- or alcohol-
induced stupor is one cau se. Occlu sions can also be seen after retrobu lbar anesthetic injec-
tions, periorbital steroid injections, and injections of cosm etic soft tissue fillers arou nd the
face. In fectiou s cau ses include Lyme d isease, syph ilis, tu bercu losis, and toxoplasmosis.
Finally, structu ral abnorm alities that can lead to retinal artery occlu sion include carotid
d issection, vasospasm du ring m igraine, and optic nerve head d ru sen.
When Do I Refer a Patient With a Branch Retinal Artery Occlusion? 129
Fig u r e 3 2 -2 . “Ze b ra ”—In fe ro te m p o ra l

BRAO in a 25 -ye a r-o ld m a le w ith Susa c
syn dro m e . Fo ca l a rte rio la r sta in in g (a rro w )
o utside th e a re a o f o cclusio n .
How Do I Approach a Patient With Newly

Diagnosed Central Retinal Artery Occlusion
or Branch Retinal Artery Occlusion?
The first th ing I do is in itiate u rgent treatment if the occlu sion is less than 24 hou rs old.
For me, th is includes an im med iate anterior cham ber paracentesis, followed by ocu lar
m assage and ad m in istration of topical aqueou s suppressants. These m aneuvers m ay help
d islodge and force an em bolu s dow nstream by increasing the effective perfu sion “pres-
su re head” beh ind the em bolu s. Although there is no proven benefit to these treatm ents,
they are easily ad m in istered, and the risk is m in im al.
I am fortu nate to work in a large academ ic center w ith fu ll-service interventional rad i-
ology available 24 hou rs a d ay. Th is m akes local intra-arterial fibrinolysis an option for
acute m anagement of central retinal artery occlu sion (CRAO). I believe th is is an effec-
tive treatment for a subset of patients, and I have personally w itnessed a few rem arkable
130 Question 32
recoveries after th is procedu re. Several case series have ind icated that th is is a relatively
safe procedu re w ith outcomes sup erior to conservative m anagement,2,3 although random-
ized controlled clin ical trial evidence is cu rrently lacking. Th is procedu re is not w idely
available, however, and treatm ent shou ld be ad m in istered w ith in 8 to 12 hou rs of symp -
tom on set for any reasonable hope of visu al recovery.
Once I have ad m in istered acute treatment, I m ake a decision about fu rther work-up.
A fluorescein angiogram is u sually not needed to m ake the d iagnosis, but I som etim es
find it helpfu l in delineating a cau se. Delayed choroid al filling m ight ind icate sign ificant
carotid or ophthalm ic artery stenosis, or focal vascu lar leakage m ay ind icate u nderlying
in flam m atory etiology. Any patient over age 55 w ithout a visible em bolu s gets a stat ESR
and CRP to ru le out GCA. I then refer patients to their prim ary physician for fu rther
evalu ation.
A thorough h istory and review of system s is crucial in gu id ing the work-up, wh ich can
be done in a stepw ise fash ion. For patients over age 40, I recom mend a thorough physi-
cal exam, complete blood cou nt (CBC), cholesterol panel, carotid doppler or m agnetic
resonance im aging/ angiogram (MRI/ MRA), and electrocard iogram. If there are any clear
risk factors for card iac em boli, such as a h istory of myocard ial in farction, rheu m atic fever,
endocard itis, know n valvu lar d isease, or heart mu rmu r, I also recom m end a transesopha-
geal echocard iogram . For patients u nder age 40, I recom mend everyone in itially have a
thorough physical exam, transesophageal echocard iogram (w ith bu bble study to look for
septal defects), electrocard iogram (EKG), CBC, fasting homocysteine level, antiphospho-
lipid antibod ies, anticard iolipin antibod ies, and antinuclear antibod ies.
If these in itial tests fail to identify a cau se for the artery occlu sion, or if som eth ing
su spiciou s is elicited on the h istory or review of system s, other relevant tests m ight
include: hemoglobin electrophoresis (sickle cell), Holter mon itoring for occu lt arrhyth m ia,
p -ANCA/ c-ANCA for Wegener’s, FTA-ABS for latent syph ilis, Lyme antibod ies, PPD,
toxoplasmosis antibod ies and H LA-B51 (Behçet’s).
Iris neovascu larization occu rs in 18% of CRAO patients and tend s to occu r earlier than
in branch retinal artery occlu sion (BRAO) (as early as 2 weeks).4 Therefore, CRAO patients
shou ld be seen for a follow-up 2 to 3 weeks after the in itial occlu sion and regu larly there-
after for the first 4 to 6 month s.
References
1. Feinbloom D, Bauer KA. Assessment of hemostatic risk factors in predicting arterial thrombotic events.
Arterioscler Throm b Vasc Biol. 2005;25(10 ):2043-2053.
2. Aldrich EM, Lee AW, Chen CS, et al. Local intraarterial fibrinolysis administered in aliquots for the treatment
of central retinal artery occlusion: the Johns Hopkins Hospital experience. Stroke. 2008;39 (6):1746-1750.
3. Noble J, Weizblit N, Baerlocher MO, Eng KT. Intra-arterial thrombolysis for central retinal artery occlusion: a
systematic review. Br J O phthalm ol. 2008;92(5):588-593.
4. Duker JS, Sivalingam A, Brown GC, Reber R. A prospective study of acute central retinal artery obstruction:
the incidence of secondary ocular neovascularization. Arch O phthalm ol. 1991;109 (3):339-342.
33 QUESTION
WHAT DO I DO WHEN I SEE

SOMEONE WITH A BRANCH RETINAL
VEIN O CCLUSION AND WHAT ARE
THE TREATMENT O PTIONS?
Michael S. Ip, MD
Most ind ividu als w ith a branch retinal vein occlu sion (BRVO) are in their seventh
decade of life. The Beaver Dam Eye Study reported a 5-year incidence of 0.6% in the 43-
to 84-year-old popu lation group.1 As m any as two-th ird s of patients w ith a BRVO have
system ic hypertension. Other risk factors include a h istory of card iovascu lar d isease,
smoking, and increased body m ass. Extensive work-up follow ing BRVO is typically not
necessary in older ind ividu als as the com mon risk factors are often elicited on basic
screen ing. However, in persons u nder age 50 and w ith no risk factors, a work-up m ay be
warranted, and the prim ary care physician consu lted.
Eyes w ith a BRVO (Figu re 33-1) u sually present w ith variable amou nts of intraretinal
hemorrhage, cotton-wool sp ots, intraretinal edem a, and a tortuou s d ilated branch retinal
vein in the affected segm ental d istribution d istal to the site of the occluded vein. The
occlu sion is u su ally fou nd at the ju nction of where a branch retinal artery crosses over an
u nderlying branch retinal vein, most com mon ly in the superotemp oral quad rant.
Macu lar edem a (Figu re 33-2) is the most com mon complication and the lead ing cau se
of vision loss in eyes w ith BRVO. If the m acu lar edem a is perfu sed on fluorescein angi-
ography (FA), approxim ately one-th ird of patients w ith m acu lar edem a w ill regain some
vision over time w ithout treatm ent. Intraretinal hemorrhages typically resolve over weeks
to months but can persist for years. A p erfu sed BRVO m ay become nonp erfu sed du ring
the ensu ing weeks, months, and even years after the in itial in su lt. Large areas of retinal
nonp erfu sion can lead to neovascu larization of the retina or d isc. Nu merou s stud ies have
131
132 Question 33
Fig u re 33 -1. Co lo r fun dus ph o to gra ph

o f a bra n ch re tin a l ve in o cclusio n with
in tra re tin a l h e m o rrh a ge s.
Fig u r e 3 3 -2 . Op tica l
Co h e re n ce to m o g ra m
o f m a cula r e de m a se c-
o n da ry to bra n ch re tin a l
ve in o cclusio n .
reported a 20% to 30% incidence of retinal neovascu larization that typically develop s
at the border between perfu sed and nonp erfu sed retina; it rarely develops outside the
area of nonp erfu sion. Eyes w ith a BRVO m ay also develop a vitreou s hemorrhage from
neovascu larization, traction retinal detach ment, and (less com mon ly) anterior segm ent
neovascu larization and neovascu lar glaucom a. Rhegm atogenou s retinal detach ment is
a rare complication of BRVO and m ay form follow ing development of a posterior retinal
break cau sed by traction on areas of fibrovascu lar proliferation.
In an eye w ith a longstand ing BRVO, the fu ndu s find ings m ay become less obviou s
and sim ilar to what m ay be seen in an eye w ith hypertensive retinopathy, rad iation reti-
nopathy, or nonproliferative d iabetic retinopathy. Ophthalmoscopic clues suggestive of a
ch ron ic BRVO include segmental m icrovascu lar abnorm alities and intraretinal collateral
vessels d rain ing across the m ed ian raphe. In a ch ron ic, nonp erfu sed BRVO, sclerosis and
sheath ing of the retinal veins in the d istribution of the occlu sion m ay be observed.
On FA, a delay in venou s filling is often noted in the area d rained by the occluded vein.
Affected venou s tributaries m ay appear narrowed. Occasionally, early hyperfluorescence
ju st proxim al to the site of occlu sion is observed. Collateral vessels are flat and do not leak
fluorescein, wh ile neovascu lar vessels leak and m ay be elevated. When evaluating the
m acu lar edem a, it is important to note whether capillary perfu sion is present (Figu re 33-3).
What Do I Do When I See Someone With a Branch Retinal Vein Occlusion? 133
Fig u re 33 -3 . Fun dus fluo re sce in a n gio -

gra m o f a bra n ch re tin a l ve in o cclusio n
sh o win g to rtuo us re tin a l ve in s a n d ca pil-
la ry dro po ut.
If greater than 5 d isc d iameters of retinal capillary nonp erfu sion are present on angiogra-
phy, the BRVO is classified as nonp erfu sed or ischem ic. Otherw ise, the BRVO is considered
perfu sed or non ischem ic.
Vision loss can vary from m in im al to severe; however, the visu al prognosis of BRVO
is qu ite favorable. Approxim ately 50% to 60% of u ntreated eyes w ill have a final visual
acu ity of 20/ 40 or better. However, severe vision loss is not u ncom mon, w ith 20% to 25%
having a visu al acu ity of 20/ 200 or worse. A final visu al acu ity of hand motion or worse
is u ncom mon.2,3
Treatment for perfu sed m acu lar edem a typically con sists of grid pattern laser photo-
coagu lation as demon strated by the Branch Vein Occlu sion Study (BVOS). Treatment was
delayed for at least 3 months after the development of a BRVO in the BVOS popu lation
and in those w ith visual acu ity of better than 20/ 40. Suggested treatm ent parameters
include argon green laser w ith a spot size of 50 to 100 µm, a 0.1-second du ration, and a
p ower setting su fficient to produce a light to med iu m wh ite bu rn. The fovea shou ld be
avoided. Typical follow-up is approxim ately 12 weeks after treatm ent w ith repeat laser
treatment as needed. Intravitreal preservative-free triamcinolone aceton ide (4 m g in 0.1
m L) or intravitreal antivascu lar endothelial grow th factor agents, such as off-label bevaci-
zu m ab (Avastin, 1.25 mg in 0.05 m L), m ay also be u sed for m acu lar edem a in BRVO. In itial
reports of intravitreal treatm ent w ith these agents have been prom ising. Intravitreal tri-
amcinolone is cu rrently being evaluated in the Stand ard Care Versu s Corticosteroid for
Retinal Vein Occlu sion (SCORE) study, a mu lticenter, random ized, Phase III National Eye
In stitute-sponsored study that is investigating the efficacy and safety of stand ard care
versu s intravitreal injection (s) of triamcinolone for m acu lar edem a in these eyes. A dexa-
m ethasone su stained release pellet (Ozu rdex) for retinal vein occlu sion is also cu rrently
being evaluated in Phase III testing. Ran ibizu m ab (Lucentis) is cu rrently being evaluated
in Phase III testing.
The BVOS also demonstrated that in eyes w ith retinal neovascu larization, sectoral
scatter laser photocoagu lation reduces the risk of vitreou s hemorrhage from 60% to 30%.4
134 Question 33
Scatter laser photocoagu lation is applied w ith the argon green laser to ach ieve “med iu m”
wh ite bu rns (200 to 500 µm in d iameter, 0.1-second du ration) spaced 1 bu rn-w idth apart
and covering the entire area of capillary nonp erfu sion. Treatment shou ld extend no closer
than 2 d isc d iameters from the center of the fovea.
Ind ications for pars plana vitrectomy include vitreou s hemorrhage, tractional/ rheg-
m atogenou s retinal detach m ent involving or th reaten ing the m acu la, epiretinal mem-
brane, second ary glaucom a, and, less com mon ly, m acu lar edem a. Arteriovenou s adventi-
tial sheathotomy at the arteriovenou s crossing site in an attempt to decompress the venou s
occlu sion is rarely p erformed.
Consu ltation w ith a retina specialist m ay be beneficial to evalu ate the extent of retinal
nonp erfu sion on fluorescein angiography and/ or to assist w ith the tim ing and ad m in istra-
tion of the above-mentioned treatm ent options. Consu ltation typically is obtained w ith in
1 to 3 weeks of d iscovery of the BRVO w ith more u rgency suggested for the presence of
neovascu larization, vitreou s hemorrhage, or traction from fibrovascu lar proliferation.

Jonathan B. Gu nther, MD
Excel Eye Center, Provo, UT
References
1. Klein R, Klein BE, Moss SE, Meuer SM. The epidemiology of retinal vein occlusion: the Beaver Dam Eye Study.
Trans Am O phthalm ol Soc. 2000;98:133-141.
2. Gutman FA, Zegarra H. The natural course of temporal retinal branch vein occlusion. Trans Am Acad
O phthalm ol O tolaryngol. 1974;78 (2):O P178-192.
3. Michels RG, Gass JD. The natural course of retinal branch vein obstruction. Trans Am Acad O phthalm ol
O tolaryngol. 1974;78 (2):O P166-177.
4. Argon laser photocoagulation for macular edema in branch vein occlusion. The Branch Vein O cclusion Study
Group. Am J O phthalm ol. 1984;98 (3):271-282.
34 QUESTION
WHEN DO I REFER A PATIENT WITH

A CENTRAL RETINAL VEIN O CCLUSION,
WHAT IS THE WORK-UP, AND WHAT
ARE THE TREATMENT O PTIONS?
Richard Spaide, MD
Central retinal vein occlu sion (CRVO) is a cond ition m an ifested by d ilated and tortu-
ou s veins in all 4 qu ad rants of the fu ndu s, intraretinal hemorrhages, m acu lar edem a,
and variable areas of capillary nonp erfu sion. In the past, grad ing of the areas of capil-
lary nonp erfu sion was u sed to separate CRVO cases into 2 d istinct typ es: 1) ischem ic or
nonp erfu sed, and 2) p erfu sed or non ischem ic.1,2 The field of retina is one of the few in
med icine in wh ich a vascu lar occlu sion cau sing pathological changes is thought to be
“non-ischem ic,” h igh lighting the poor term inology u sed to describe d isease. Cu riou sly,
it is not a simple m atter to generate an im al models that faith fu lly m im ic CRVO. Simple
ligation of the central retinal vein in prim ates d id not produce a pictu re consistent w ith
what is termed CRVO in adu lt hu m an s. To produce a better approxim ation of the fu ndu s
pictu re of CRVO, simu ltaneou s arterial occlu sion had to be induced as well. Becau se of
the lack of su itable an im al models, much of what we have learned about CRVO is based
on carefu l clin ical stud ies of hu m an su bjects.
Patients developing CRVO note decreased vision, scotom ata, and visu al d istortion. In
the Central Vein Occlu sion Study (CVOS), a collaborative study of 725 affected eyes, the
baseline acu ity was reported in 3 categories: 1) 29% had a good visu al acu ity of 20/ 40
or better; 2) in 43%, the visu al acu ity was classified as intermed iate (20/ 50 to 20/ 200);
and 3) in 28%, it was classified as poor (less than 20/ 200).1 Approxim ately two-th ird s of
patients who presented w ith good visu al acu ity had good visu al acu ity after 3 years of
135
136 Question 34
follow-up.1 Of those presenting w ith intermed iate visu al acu ity, 19% improved to good,
44% stayed in the intermed iate group, and 37% decreased to p oor acu ity.1 Those in itially
in the poor visu al acu ity group had an 80% chance of rem ain ing in that group at the end
of follow-up.1 Eyes w ith 10 or more d isc areas of capillary nonp erfu sion were classified as
being nonp erfu sed, wh ile those w ith 10 d isc areas or less were considered to be p erfu sed.
In some patients, the classification was not in itially p ossible and these eyes were term ed
indeterm inate. Over the follow-up p eriod, about one-th ird of eyes that were in itially clas-
sified as p erfu sed becam e nonp erfu sed, wh ile 5/ 6 of the eyes that were indeterm inate
were later classified as nonp erfu sed. The classification of perfu sed versu s nonp erfu sed
principally was m ade to pred ict wh ich eyes wou ld progress to neovascu larization of the
iris and neovascu lar glaucom a. In the CVOS, 10% of eyes classified as p erfu sed developed
neovascu larization of the iris or angle, as d id 35% of the nonperfu sed eyes. Therefore, the
classification of perfu sed versu s nonp erfu sed was neither very sen sitive nor sp ecific for
the development of neovascu larization.
In itial treatm ent of decreased visu al acu ity in CRVO focu sed on m acu lar edem a. The
CVOS stud ied the effect of grid laser photocoagu lation on perfu sed eyes w ith m acu lar
edem a and visu al acu ity of 20/ 50 or worse and demonstrated a beneficial anatom ic effect
(decreased edem a on angiography), but visu al acu ity d id not improve.3 Th is study was
done prior to the era of optical coherence tomography so it is d ifficu lt to estim ate the
reduction in retinal th ickness ach ieved w ith laser.
Su bsequent therapies attempted to increase collateral vessel development becau se
collateral vessels red irect blood flow arou nd an area of occlu sion. These therapies were
largely u nsuccessfu l, had low rates of adoption, and are now interesting h istorical anec-
dotes: h igh-power laser photocoagu lation and vitreou s su rgical tech n iques for chorioreti-
nal venou s anastomosis form ation and rad ial optic neu rotomy.
More recently, intravitreal injections of variou s pharm acologic agents have been stud ied
in eyes w ith CRVO and m acu lar edem a. Intravitreal triamcinolone was associated w ith
cataract form ation; elevated intraocu lar pressu re; and a rapid, often large improvement
in visual acu ity in eyes w ith CRVO that was u n fortu nately short-lived, all of wh ich was
compou nded by the need for continu ing injections. Becau se of lim ited d ata on long-term
safety and efficacy, a random ized trial sp on sored by the National Eye Institute, called the
SCORE Study, has now been completed and the resu lts are forthcom ing.
Su bsequently, vascu lar endothelial grow th factor (VEGF) in h ibitors were noted anec-
dotally to have a beneficial effect up on m acu lar edem a and visual acu ity, w ithout the risk
of cataract or elevated intraocu lar pressu re associated w ith triamcinolone aceton ide. We
know the retina has a con stitutive secretion of VEGF and that a decrease in blood flow
th rough tissue resu lts in an increase in VEGF production. Wh ile it is beyond the con fines
of th is chapter to d iscu ss in detail, the follow ing are know n to be true about VEGF in
these eyes: 1) vitreou s samples from patients w ith CRVO are elevated ; 2) severity of CRVO
is related to VEGF level; and 3) intravitreal VEGF injection in prim ate eyes produced find-
ings that m im ic CRVO, includ ing venou s d ilation and tortuosity, intraretinal hemorrhage,
telangiectasis, and areas of nonp erfu sion. Subsequent h istopathologic analysis demon-
strated endothelial cell proliferation w ith in capillaries and venu les, lead ing to occlu sion
When Do I Refer a Patient With a Central Retinal Vein Occlusion? 137
and nonp erfu sion. What app ears to be nonp erfu sion in ischem ic eyes seem s to resu lt as
a con sequence of elevated VEGF levels, not someth ing intrin sic to the central retinal vein
occlu sion itself. These find ings point to an alternate hypothesis implicating VEGF as an
imp ortant cau se of the clin ical find ings associated w ith CRVO, as opposed to simply
being a sequelae of the occlu sion itself.
Becau se VEGF m ay be a d riving force in producing pathologic m an ifestations of CRVO,
it is logical to u se anti-VEGF agents to treat CRVO. Intravitreal bevacizu m ab (Avastin),
an anti-VEGF agent, was evalu ated in an off-label fash ion and benefits included visu al
acu ity improvement; a decrease in m acu lar edem a; and a reduction in venou s tortuosity,
venou s d ilation, nerve swelling, and intraretinal hemorrhage in selected patients. A recent
prosp ective study u sing intravitreal ran ibizu m ab (Lucentis) fou nd that improvement in
visu al acu ity was not correlated w ith the change in m acu lar edem a. Norm alization of the
mean m acu lar th ickness occu rred in 1 week but continu ing visu al acu ity improvement
occu rred over 1 year of near-month ly injections.4
What Is the Best Way to Handle

Central Retinal Vein Occlusion?
CRVO cau ses d ilated tortuou s veins, intraretinal hemorrhages, and retinal edem a. The
d iagnosis of CRVO is often con firmed w ith fluorescein angiography. Optical coherence
tomography is helpfu l in provid ing objective m easu rement of m acu lar th ickness and for
the detection of any su bretinal flu id. Hypertension and glaucom a are well-know n risk fac-
tors. Med ical work-up is not ind icated for the vast m ajority of patients becau se the resu lts
of hem atologic or clotting evalu ations often produce the same frequency of abnorm alities
as an age-adju sted norm al p opu lation. Referral to an appropriate sp ecialist is ind icated for
the rare older patient w ithout regu lar m ed ical care and for you ng patients, bilateral cases,
and patients in whom a review of system s ind icates other potential med ical problem s.
Anti-VEGF therapy is a good first-choice treatment given the low risk of side effects.
The keys to successfu l anti-VEGF treatment are early and frequent treatment.4 Anecdotal
evidence suggests that early treatment is better than waiting. Giving a single injection to
“see what happ ens” w ill likely u nderestim ate the benefit compared w ith rep eat injection s.
Getting a response is easy; the question is how long w ill treatm ent be requ ired? Although
th is type of in form ation has not been pu blished, it app ears that the need for injection
declines du ring the second and th ird years (Figu re 34-1).
Financial disclosure: Research support from Genentech.

138 Question 34
Fig u re 3 4 -1. (A) To p le ft: On pre se n ta tio n , th e visua l a cuity wa s 20 / 20 0. Th e

pa tie n t h a d w ide spre a d in tra re tin a l h e m o rrh a ge , ve n o us dila tio n , to rtuo sity,
a n d (to p righ t) m a ssive e de m a a n d subm a cula r fluid. Middle le ft: On e we e k
a fte r a n in tra vitre a l in je ctio n o f 0.5 m g ra n ibizum a b, th e re a lre a dy is m uch
le ss h e m o rrh a ge , a n d (m iddle righ t) th e re is a lm o st co m ple te re so lutio n o f th e
e de m a . Th e visua l a cuity wa s 20 / 10 0. (B) Le ft: On e ye a r la te r, th e eye lo o ks
n e a rly n o rm a l. Th e re a re so m e subtle pigm e n ta ry ch a n ge s in th e ce n tra l
m a cula fro m pre vio us e de m a , but n o te th e la ck o f h e m o rrh a ge, to rtuo sity,
a n d n o ta bly th e la ck o f a ny co lla te ra l ve sse ls. Th e m a cula sh o we d n o e de m a
o r subre tin a l fluid (righ t); it is co m m o n fo r po st-tre a tm e n t OCT sca n s to sh o w
bo th lo ss o f th e in n e r re tin a a n d re duce d diffe re n tia tio n o f th e n o rm a lly se e n
re tin a l la ye rs. Th e pa tie n t h a d 9 in tra vitre a l in je ctio n s ove r th e pre vio us ye a r,
a n d th e visua l a cuity wa s 20 / 63. At la st fo llo w-up 15 m o n th s a fte r pre se n ta -
tio n , th e visua l a cuity re m a in e d 20 / 63, a n d th e pa tie n t did n o t n e e d a n in tra -
vitre a l ra n ibizum a b in je ctio n fo r 4 m o n th s.
When Do I Refer a Patient With a Central Retinal Vein Occlusion? 139
References
1. Natural history and clinical management of central retinal vein occlusion. The Central Vein O cclusion Study
Group. Arch O phthalm ol. 1997;115(4):486-491.
2. A randomized clinical trial of early panretinal photocoagulation for ischemic central vein occlusion. The
Central Vein O cclusion Study Group N report. O phthalm ology. 1995;102(10):1434-1444.
3. Evaluation of grid pattern photocoagulation for macular edema in central vein occlusion. The Central Vein
O cclusion Study Group M report. O phthalm ology. 1995;102(10):1425-1433.
4. Spaide RF, Chang LK, Klancnik JM, et al. Prospective study of intravitreal ranibizumab as a treatment for
decreased visual acuity secondary to central retinal vein occlusion. Am J O phthalm ol. 2009;147(2):298-306.
35 QUESTION
IF I SEE A PATIENT WITH SHAKEN BABY

SYNDROME, WHAT ARE THE NEXT STEPS?
Christine R. Gonzales, MD
Ophthalmologists are occasionally consu lted to see a child in whom nonaccidental trau-
m a is su spected. By helping to d iagnose shaken baby synd rome (SBS), the ophthalmologist
can often protect the ch ild from fu rther abu se. The classic triad of clin ical find ings in the
SBS includes encephalopathy; subdu ral hemorrhage; and pre-, intra-, and subretinal hem-
orrhages (Figu re 35-1).1 Controversy exists regard ing the exact cau se of the pre-, intra-, and
subretinal hemorrhages. While the classic teaching suggests that the hemorrhages occu r
from trau m atic shearing of the retinal vessels, other hypotheses have suggested that the
hemorrhages resu lt from a com bination of cerebral hypoxia, raised intracranial pressu re
from cerebral edem a, raised arterial pressu re, and raised central venou s pressu re.
The first step in evalu ating these ch ild ren is taking a detailed h istory from any and
all involved people. Often it is d ifficu lt to ascertain an accu rate h istory as these ch ild ren
m ay have been removed from their prim ary caretakers. The incident m ay have occu rred
outside the home or w ith caregivers who are not present when eliciting the h istory.
Witnesses to the trau m a m ay not be present, and parents or caregivers m ay not give an
accu rate accou nt of the details su rrou nd ing the inju ry. It is not u ncom mon for a h istory
of trau m a to be lacking altogether. It is very important to approach each situ ation in an
u nbiased, objective m an ner and to reserve judgment when docu menting the find ings.
Exten sive retinal hemorrhages that extend to the ora serrata and involve mu ltiple
layers (ie, pre-, intra-, and su bretinal) are h igh ly sp ecific for SBS; however, w ith less
extensive hemorrhages, other d iagnoses shou ld be con sidered. These include nontrau-
m atic cau ses of increased intracran ial pressu re (eg, tu mor, arteriovenou s m alform ation,
intracran ial aneu rysm, etc.), accidental trau m a, Pu rtscher’s synd rome, anem ia, leu ke-
m ia, central nervou s system in fection, coagu lopath ies, card iopu lmonary resu scitation,
and seizu res. Intraretinal hemorrhages can also occu r du ring ch ild birth, especially if a
vacuu m-assisted vaginal delivery was performed, and these hemorrhages tend to resolve
by 6 weeks of age.
141
142 Question 35
Fig u re 35 -1. Multila ye re d re tin a l

h e m o rrh a ge s in a n in fa n t w ith
sh a ke n ba by syn dro m e .
Although the retinal hemorrhages in SBS tend to be bilateral, u n ilateral retinal hemor-
rhages have been reported in 10% to 16% of SBS cases.2-4 Other possible retinal find ings in
SBS include perim acu lar retinal fold s, retinal d ialysis, tractional or rhegm atogenou s reti-
nal detach m ent, retinosch isis, vitreou s base avu lsion, and m acu lar hole. Pupillary reactiv-
ity shou ld be docu mented, as trau m a to the optic nerve or brainstem m ay be present.
Carefu l docu m entation of find ings is extremely important as you m ay be called up on
to testify in cou rt. Detailed retinal d raw ings w ith carefu l attention to the location, extent,
and level of hemorrhages are critical. If possible, fu ndu s photographs shou ld be taken.
A CT scan of the brain is important to identify intracran ial hemorrhages, and skeletal
X-rays are often obtained to ru le out fractu res of the long bones and/ or ribs. Carefu l
evalu ation and m anagem ent by the p ed iatric team of the system ic m an ifestations of the
SBS is important. In the futu re, biom arkers that are released into the blood stream after
head trau m a, such as neu ron-sp ecific enolase, S100B, and myelin-based protein, m ay be
u sefu l as a screen ing tool for SBS.5
Most of the time the ch ild ren do not present to the ophthalmologist. Rather, they are
brought to the emergency room becau se of lethargy, vom iting, mental statu s change, or
altered states of con sciou sness; therefore, ch ild protective services (CPS) m ay already be
involved. If they are not, however, and you su spect ch ild abu se, CPS shou ld be contacted
im m ed iately.
Management
The most com mon cau se of blind ness in SBS is cortical blind ness, but vitreou s hemor-
rhage, prem acu lar hemorrhage, retinal detach m ent, and/ or m acu lar hole m ay also resu lt
in vision loss and m ay requ ire su rgical intervention. If the on ly ocu lar find ings are retinal
hemorrhages w ithout involvement of the central m acu la, observation is u su ally the best
cou rse of m anagement. The hemorrhages tend to resolve spontaneou sly over the cou rse
of 4 to 6 weeks. Vitreou s hemorrhage obstructing the visu al axis or m acu lar su bhyaloid
hemorrhage m ay requ ire vitrectomy su rgery. Visu al outcomes in these cases are variable
If I See a Patient With Shaken Baby Syndrome, What Are the Next Steps? 143
and dep end on the severity of the find ings. Sm all m acu lar subhyaloid hemorrhage and
m ild vitreou s hemorrhage can often be observed for sp ontaneou s resolution, but am blyo-
pia shou ld always be kept in m ind. Lengthy delays in intervention shou ld be avoided.
The m anagement of these in fants involves a team approach. The fam ily’s emotions
shou ld be kept in m ind, and care shou ld be taken to avoid accu satory statem ents. The
tim ing of ocu lar su rgery shou ld be coord inated w ith other med ical intervention s and m ay
be affected by the patient’s neu rologic state. The social and emotional issues of the patient
and fam ily are u su ally best m anaged by social workers and CPS.
References
1. Gerber P, Coffman K. Nonaccidental head trauma in infants. Childs N erv Syst. 2007;23(5):499-507.
2. Kivlin JD, Simons KB, Lazoritz S, Ruttum MS. Shaken baby syndrome. O phthalm ology. 2000;107(7):1246-
1254.
3. Morad Y, Kim YM, Armstrong DC, Huyer D, Mian M, Levin AV. Correlation between retinal abnormalities and
intracranial abnormalities in the shaken baby syndrome. Am J O phthalm ol. 2002;134(3):354-359.
4. Arlotti SA, Forbes BJ, Dias MS, Bonsall DJ. Unilateral retinal hemorrhages in shaken baby syndrome. J AAPO S.
2007;11(2):175-178.
5. Berger RP, Dulani T, Adelson PD, Leventhal JM, Richichi R, Kochanek PM. Identification of inflicted trau-
matic brain injury in well-appearing infants using serum and cerebrospinal markers: a possible screening tool.
Pediatrics. 2006;117(2):325-332.
36 QUESTION
WHAT TYPE OF INTRAOCULAR LENS

SHOULD BE CONSIDERED IN AN EYE
WITH VITREORETINAL DISEASE?
Kourous A. Rezaei, MD
It is well docu mented that vitreou s su rgery lead s to the progression of crystalline lens
opacity. Th is is not a complication of vitrectomy su rgery, but rather a pathophysiologic
resp onse to removal of the vitreou s gel. The etiology of cataract form ation after vitreo-
retinal su rgery is not completely know n; however, it has been reported that vitreou s
removal lead s to increased oxygen tension in the posterior segment, resu lting in oxid a-
tive d am age and cataract progression.1 I in form all of my phakic patients that they w ill
need cataract su rgery after vitrectomy su rgery. The tim ing of the cataract su rgery varies
accord ing to the degree of pre-existing cataract, the ind ication for vitrectomy, the u se of
temporary tamponade (typ e of gas bubble or silicone oil), and the patient’s compliance
w ith position ing.
Becau se cataract su rgery is practically inevitable after vitreou s su rgery, I generally rec-
om m end cataract su rgery prior to vitrectomy su rgery for nonu rgent vitreoretinal cases.
Th is is based on the follow ing reason s:
• Cataract su rgery is generally considered easier to perform prior to vitrectomy
su rgery becau se the vitreou s gel provides intraoperative support of the crystalline
lens.
• If a complication develops du ring cataract su rgery, the patient is already schedu led
for vitrectomy su rgery.
• An intraocu lar lens (IOL) generally offers a superior view du ring vitreou s su rgery
when compared to an aged natu ral lens.
• Pseudophakic statu s allow s for more thorough p eripheral vitreou s shaving.
145
146 Question 36
• Inspection of the peripheral retina du ring vitreoretinal su rgery is easier when the
eye is pseudophakic.
• My preference is to wait at least 3 weeks between cataract su rgery and vitrectomy
su rgery to perm it the eye to qu iet dow n and the cataract wou nd to heal.
There are a large variety of IOL implants cu rrently available. The m ain factors that d if-
ferentiate the implants are the follow ing:
• The type of implant material: polymethylm ethacrylate (PMMA), silicone, acrylic
• The light-filtering characteristics: blue light-filtering yellow lens versu s clear lens
• The focusing characteristics: presbyopia-correcting lenses (Crystalen s [Bau sch &
Lom b], ReSTOR [Alcon], ReZoom [Abbott Med ical Optics, Santa Ana, CA]) versu s
regu lar fixed-d istance len ses
PMMA was the first m aterial to be u sed for an IOL. Cu rrent IOLs are m ade m ain ly
from silicone or acrylic. In eyes w ith retinal pathology, cataract su rgeons shou ld avoid
u sing silicone len ses becau se som e complex retinal procedu res m ay requ ire the u se of
a silicone oil tamponade. Silicone oil d roplets ad here irreversibly to silicone IOLs (in
the presence of a posterior capsu lar open ing), lead ing to a poor view of the fu ndu s and
reduced visu al acu ity.2 These patients likely wou ld need to have the IOL exchanged. Th is
contraind ication for silicone IOLs hold s true for eyes w ith a h istory of silicone oil removal
becau se silicone oil removal is never complete. The emu lsified oil d roplets that rem ain in
a flu id-filled eye m ay ad here to the second arily implanted silicone IOL, likely lead ing to
reduced vision. I prefer the implantation of acrylic lenses in eyes that m ay need a retinal
procedu re in the futu re.
We and others have rep orted that a blue light-filtering IOL m ay be beneficial in eyes
that are at h igh risk for progression of age-related m acu lar degeneration.3-5 A recent ran-
dom ized controlled clin ical trial evalu ated whether the yellow tint of a blue light-filter-
ing IOL interferes w ith the su rgeon’s ability to perform vitreoretinal m aneuvers.6 Sixty
patients were en rolled in th is trial, and none of the su rgeon s reported any adverse events
due to the presence of the blue light-filtering lens. No mod ification of the su rgical set-up
or procedu re was requ ired du ring vitreoretinal su rgery.
IOLs that correct for presbyopia also correct for d istance. A recent article reviewed
the 3 com mon ly u sed presbyopia-correcting len ses: Crystalen s, ReSTOR and ReZoom.7
Crystalens is an accom mod ating silicone monofocal lens and is therefore contraind icated
in eyes that m ay need vitrectomy su rgery. The ReSTOR len s is an acrylic mu ltifocal lens.
The accu rate position ing of th is len s in the bag (single piece) or su lcu s (3 piece) is crucial
for its optim al perform ance. Du ring vitreoretinal su rgery, the eye m ay be filled w ith air or
gas, wh ich has the p otential to decenter the lens from its original location. Both ReSTOR
and ReZoom are mu ltifocal len ses. These lenses m ay induce aberration s that can interfere
w ith intraoperative visu alization of the retina. Th is can be esp ecially bothersome to the
vitreoretinal su rgeon du ring mem brane peeling. In add ition, one m ay also need to re-
focu s the m icroscope when operating from the posterior p ole to the periphery.
What Type of IOL Should Be Considered in an Eye With Vitreoretinal Disease? 147
Summary
The selection of an appropriate IOL is important in eyes w ith vitreoretinal pathology.
For eyes that m ay requ ire vitreoretinal su rgery, silicone lenses of any type shou ld be
avoided.
References
1. Holekamp NM, Shui YB, Beebe DC. Vitrectomy surgery increases oxygen exposure to the lens: a possible
mechanism for nuclear cataract formation. Am J O phthalm ol. 2005;139 (2):302-310.
2. Khawly JA, Lambert RJ, Jaffe GJ. Intraocular lens changes after short- and long-term exposure to intraocular
silicone oil: an in vivo study. O phthalm ology. 1998;105(7):1227-1233.
3. Sparrow JR, Miller AS, Zhou J. Blue light-absorbing intraocular lens and retinal pigment epithelium protection
in vitro. J Cataract Refract Surg. 2004;30 (4):873-878.
4. Marshall J, Cionni RJ, Davison J, et al. Clinical results of the blue-light filtering AcrySof Natural foldable acrylic
intraocular lens. J Cataract Refract Surg. 2005;31(12):2319-2323.
5. Rezai KA, Gasyna E, Seagle BL, Norris JR Jr, Rezaei KA. AcrySof natural filter decreases blue light-induced
apoptosis in human retinal pigment epithelium. Graefes Arch Clin Exp O phthalm ol. 2008;246 (5):671-676.
6. Falkner-Radler CI, Benesch T, Binder S. Blue light-filter intraocular lenses in vitrectomy combined with cataract
surgery: results of a randomized controlled clinical trial. Am J O phthalm ol. 2008;145(3):499-503.
7. Tewari A, Shah GK. Presbyopia-correcting intraocular lenses: what retinal surgeons should know. Retina.
2008;28 (4):535-537.
37 QUESTION
WHEN A POSTERIOR CAPSULAR

TEAR O CCURS DURING CATARACT
SURGERY, WHAT NEXT?
Timothy W. Olsen, MD
A p osterior capsu lar tear w ith vitreou s in the anterior cham ber is a potential su rgical
complication of cataract extraction, even in exp erienced hand s. Th is intraop erative com-
plication can contribute to fu rther challenges, such as corneal decomp en sation, pupillary
irregu larity, intraocu lar lens movement, cystoid m acu lar edem a (Figu re 37-1), retinal
break and detach ment, or endophthalm itis. Therefore, proper m anagem ent of the vitre-
ou s is imperative in order to m in im ize such risks and is the key to a successfu l su rgical
outcome.1-3
The first step is to recogn ize that a capsu lar ruptu re has occu rred. Th is is not always
easy to detect, especially in subtle cases. Sudden deep en ing of the anterior cham ber,
increased or decreased len s mobility, pupillary irregu larities, or a change in the occlu sion
of the phacoemu lsification tip m ay all suggest that the p osterior capsu le has ruptu red.
Visu alization of the capsu lar op en ing is facilitated by a good red reflex and proper focu s-
ing of the su rgical m icroscope.
Once a ruptu re has been con firm ed, the goal is to prevent exten sion of the tear and
to remove any vitreou s from the pupillary space and anterior cham ber. The follow ing
principles shou ld be kept in m ind :
• Avoid rapid changes in intraocu lar pressu re.
• Use viscoelastic agents.
• Avoid pu lling on the vitreou s.
• Use the gu illotine cutters as opposed to scissors.
• Visu alize the vitreou s.
• Remove vitreou s from wou nd s.
• Avoid “fish ing” for posteriorly d islocated lens fragments.
149
150 Question 37
Fig u re 37-1. Cysto id m a cula r e de m a

fo llo win g ca ta ra ct surge ry. No te th e
pe ta llo id pa tte rn o f hype rfluo re sce n ce
in th e re circula tio n ph a se o f th e a n gio -
gra m a s we ll a s disc hype rfluo re sce n ce ,
co n siste n t w ith Irvin e -Ga ss syn dro m e .
Stop the in fu sion of flu id th rough the handpiece im m ed iately in order to m in im ize
the size of the cap su lar tear. Inject a d ispersive viscoelastic agent 4 prior to the removal of
the phacoemu lsification handpiece; in the setting of a sm all ruptu re, th is m aneuver m ay
rep osit the vitreou s back into the posterior segm ent and tamponade the anterior hyaloid
face. If any crystalline lens fragments are still present, the viscoelastic agent can separate
the len s m aterial from the ruptu re site and prevent lens m aterial from falling th rough the
capsu lar open ing into the vitreou s cavity. When the anterior cham ber has been stabilized,
the size and extent of the capsu lar deh iscence m ay be assessed, and the intraocu lar pres-
su re shou ld then be m aintained at lower pressu res. Occasionally, conversion to an extra-
capsu lar procedu re m ay be necessary to safely remove len s fragments.
Any prolapsed vitreou s shou ld be cut rather than pu lled to m in im ize traction on the
retina. Gu illotine-style vitreou s cutters remove very sm all segments of vitreou s, thu s per-
m itting safer removal. Faster cut rates translate into less traction on the vitreou s du ring
the aspiration cycle of the cutter and in tu rn, less traction on the retina. Lower aspiration
is safer than h igher aspiration.
A separate in fu sion u sing an irrigation handpiece or anterior cham ber m aintainer is
preferred. An in fu sion sleeve is suboptim al becau se it creates flow away from the cutting
tip. Avoid sharp need le inju ry to the cap su le or other anterior cham ber structu res.
Visu alization of the vitreou s is important and can be ach ieved by u sing tangential illu-
m ination, such as from a fiberoptic light sou rce or by u sing preservative-free triamcino-
lone. Injection of up to 0.10 m L in a slow and deliberate m an ner allow s the triamcinolone
to ad here to vitreou s strand s, en hancing visu alization.5
If the posterior capsu le is violated, vitreou s can then extend from the p osterior seg-
m ent toward the site of lowest pressu re, such as any op en wou nd where vitreou s is
com mon ly fou nd. A vitreou s cutter entering from a separate entry site into the anterior
cham ber allow s the su rgeon to cut the vitreou s more tangentially and where the vitreou s
strand enters the anterior segment. A good analogy is to cut seaweed closer to the base
of the plant rather than trying to trim the tips as they sw ing in the cu rrents. Introducing
the gu illotine cutter th rough the cap su lar rent u sing a h igh cut rate and low aspiration
setting is reasonable on ly if the cutting tip can be read ily visu alized. Avoid aggressive
When a Posterior Capsular Tear Occurs During Cataract Surgery, What Next? 151
m an ipu lation and cutting into the posterior segment, especially if visu alization is lim ited.
Also, avoid “fish ing” in the posterior cham ber for len s fragments.
When you feel that all of the vitreou s has been removed from the anterior segment,
check again for residu al vitreou s. Some clues that vitreou s is still present include the
presence of pupillary irregu larities such as the classic “peaked pupil” and/ or len s decen-
tration.
Some anterior segment su rgeons are com fortable introducing in stru ments th rough the
pars plana. Th is is also a reasonable approach, but one shou ld be prepared to exam ine the
peripheral retina w ith scleral depression intraoperatively follow ing closu re of the eye for
retinal breaks that m ay occu r from these m aneuvers.
An intraocu lar lens can then be placed either in the su lcu s or in the anterior cham ber
dep end ing on the amou nt of capsu lar support rem ain ing.
If any len s fragm ents rem ain in the vitreou s cavity, referral to a vitreoretinal su rgeon
for fu rther evaluation is warranted.
Financial disclosure: Research support from N ational Institutes of Health, N ational Institute on A ging,
N ational Eye Institute, Emtech Biotechnology, Research to Prevent Blindness, Georgia Research A lliance,
and N eurotech.

Jou ng Yon Kim, MD
Assistant Professor, Director, Comprehen sive Ophthalmology
Emory Un iversity, Atlanta, GA
References
1. Girard LJ, Nieves R, Hawkins RS. Ultrasonic fragmentation for vitrectomy and associated surgical procedures.
Trans Sect O phthalm ol Am Acad O phthalm ol O tolaryngol. 1976;81(3 pt 1):432-450.
2. Kanski JJ, Ramsay JH. Vitrectomy techniques in the management of complications in cataract surgery. Trans
O phthalm ol Soc U K. 1980;100 (pt 1):216-218.
3. O sher RH, Cionni RJ. The torn posterior capsule: its intraoperative behavior, surgical management, and long-
term consequences. J Cataract Refract Surg. 1990;16 (4):490 -494.
4. Blumenthal M, Assia E, Neuman D. Lens anatomical principles and their technical implications in cataract
surgery. Part II: the lens nucleus. J Cataract Refract Surg. 1991;17(2):211-217.
5. Wu MC, Bhandari A. Managing the broken capsule. Curr O pin O phthalm ol. 2008;19 (1):36-40.
38 QUESTION
WHEN SHOULD A PATIENT WITH

DROPPED CATARACT MATERIAL OR
DISLOCATED INTRAOCULAR LENS BE
REFERRED TO A RETINAL SPECIALIST?
Steven D. Schwartz, MD
When was the last time you d ropp ed len s m aterial? An intraocu lar len s (IOL)? Most
likely, it is not a com mon occu rrence. With any luck, you are read ing th is chapter ju st in
case. But, if you have recently encou ntered such a circu m stance, I hope these thoughts
help you m anage you r patient successfu lly.
Encou ntering an in frequent and potentially sight-th reaten ing complication in the
operating room evokes inten se emotions in the su rgeon, ranging from “fight (for you r
patient!)” to “flight (from the scene!).” With th is in m ind, I u n iform ly suggest to residents
and accomplished su rgeons alike, m aintain you r poise and control, stay w ith in you r skill
set, and remem ber that a second su rgery for d ropped lens m aterial or an IOL is far better
than creating even more complication s.
Retained Lens Material

When referred early, eyes w ith d islocated lens m aterial typically do very well w ith
vitreoretinal intervention—regard less of the size or typ e of the lens fragment left in the
eye and regard less of whether an IOL is placed prim arily. Attention shou ld be d irected
toward anterior vitrectomy and m eticu lou s wou nd closu re (sutu res) as opposed to heroic
attempts to “rescue” fragments of lens from the vitreou s cavity. Thu s, I generally suggest
again st potentially risky su rgical m aneuvers such as chasing len s fragments into the
153
154 Question 38
posterior segment w ith need les, viscoelastics, phacoemu lsification probes, or other instru-
ments. Exceptions can always be fou nd, but control and precision typically keep eyes out
of irreversible trou ble.
The literatu re suggests that patients w ith retained len s m aterial can, on occasion, be
safely observed du ring the postoperative period. However, in my opin ion, the vast m ajor-
ity of these patients shou ld be referred to a retinal su rgeon on a relatively u rgent basis.
Time to referral can impact a nu m ber of clin ically sign ificant find ings, includ ing severity
of uveitis, intraocu lar pressu re, and corneal clarity. Sim ilarly, the sooner the patient is
referred, the sooner the retinal su rgeon can reassu re the patient and fam ily.
Retained lens m aterial cases range from the m in im al (cortical rem nants w ith in the
capsu lar bag) to the severe (entire lens d islocated into the p osterior segm ent). Vision loss
in these patients can be cau sed by a nu m ber of cond itions, includ ing but not lim ited to
in flam m ation (uveitis, cystoid m acu lar edem a, second ary glaucom a), vitreou s traction
(retinal detach ment, giant retinal tear, cystoid m acu lar edem a), and su rgical trau m a
(hyphem a, vitreou s hemorrhage, iris trau m a, ciliary body detach ment and the fu ll spec-
tru m of retinal detach ment). Becau se all of these problem s are p otentially devastating,
and largely preventable or treatable when intervention is early, I routinely recom mend
early referral and intervention. The “early referral” practice pattern is supported by two
lines of reason ing. First, evidence suggests that modern vitreoretinal su rgery is far more
precise and controlled tod ay than ever before. Second, early evalu ation and m anagement
by a retinal specialist valid ates the decision to m anage med ically.
Contrast th is approach w ith a delayed or non referral scenario such as when an eye
is treated med ically u nder the prim ary su rgeon’s care for weeks or month s follow ing a
complicated cataract su rgery w ith retained len s m aterial. The intention is to ach ieve an
acceptable outcome wh ile avoid ing the potential risk of a second intervention. Pressu re
and in flam m ation are interm ittently controlled, and the lens fragments d isappear slowly.
Vision can be comprom ised, care is intensive, exp ectations are typically not m et, and
emotions range from concerned to u n happy. Worse still, if and when the patient comes to
a second su rgery, he or she m ay face an increased risk of complication s. Most agree that
op erating on an in flamed eye is probably less controlled than op erating on a qu ieter eye
in the im m ed iate post-cataract su rgery period.
In other word s, if you d rop the len s, I recom mend you essentially “stop, sutu re, and
refer.” Anterior vitrectomy is ind icated when vitreou s gel is prolapsed into the anterior
segm ent. Care shou ld be taken to ensu re that no vitreou s is allowed to rem ain in the
wou nd or to interfere w ith the iris sph incter. Intraocu lar len s placement in the ciliary
su lcu s is acceptable, in my opin ion, on ly if an adequ ate anterior vitrectomy has been p er-
formed and if adequ ate capsu lar support is present. Anterior cham ber IOL placem ent can
exacerbate corneal, intraocu lar pressu re, and iris issues and shou ld be u ndertaken w ith
caution or be avoided du ring the in itial procedu re. Virtually every vitreoretinal su rgeon
w ill place a sutu re in the cataract incision at the begin n ing of vitreou s su rgery u n less one
has been placed prim arily, so I suggest you sutu re the incision. Referral shou ld be im me-
d iate. In my opin ion, it is better for the patient, the eye, and the referring su rgeon.
When Should a Patient With Dropped Cataract Material Be Referred? 155
Dislocated Intraocular Lens

Cases of d islocated IOLs also fall into a sp ectru m of severity ranging from m inor
decentration (len s sublu xation) to complete d islocation into the posterior segment. Once
the vitreou s body is d istu rbed, it is best approached via the pars plana w ith appropri-
ate illu m ination and in fu sion. Thu s, I generally advise against potentially risky su rgical
m aneuvers such as attempted IOL retrieval w ith need les, viscoelastics, forceps or anterior
vitrectomy in stru ments. Rather, if an IOL is d islocated intraoperatively, I recom m end
anterior vitrectomy to remove any vitreou s gel from the anterior segment, carefu l wou nd
closu re (sutu re), and relatively im med iate referral. I do not recom mend a second IOL be
placed prior to retrieval of the d islocated IOL.
As is true w ith all complex patients, the best way forward is an honest path w ith the
patient’s best interest in m ind. Succu m bing to the emotional tendency to subject the
patient to heroic and imprecise rescue m aneuvers m ay not be in the patient’s best interest.
Stop, sutu re, and refer; the eye, the patient, and the referring su rgeon shou ld do well.
39 QUESTION
HOW DO I MANAGE A
SUPRACHOROIDAL HEMORRHAGE?
Odette M. Houghton, MD
A suprachoroid al hemorrhage is a potentially devastating complication of intraocu lar

su rgery, particu larly anterior segment su rgical procedu res. Suprachoroid al hemorrhages
can occu r intraoperatively or in the early postoperative p eriod.
The key to preserving an eye w ith an intraoperative suprachoroid al hemorrhage is
early recogn ition and prompt action prior to the expu lsion of intraocu lar contents. In
order to exped itiou sly m anage an intraoperative suprachoroid al hemorrhage, prepare
you rself for the event prior to every intraocu lar su rgical procedu re. Acknowledge that
the p ossibility of a suprachoroid al hemorrhage does exist, even though it is u ncom mon.
Craft a m anagement strategy in advance and have all of the necessary tools im med iately
available.
Early sign s of a suprachoroid al hemorrhage include eye pain; increased posterior pres-
su re; loss of the red reflex; en largement of the pupil; shallow ing of the anterior cham ber
w ith forward d isplacement of the iris, len s, or lens implant; and vitreou s prolapse. If a
suprachoroid al hemorrhage is su spected, im m ed iate closu re of the wou nd shou ld elevate
the intraocu lar pressu re su fficiently to tamponade the suprachoroid al bleed ing. Ideally,
close the wou nd w ith 8-0 sutu res. If there is no time to place sutu res, approxim ate the
wou nd edges w ith d igital pressu re or toothed forceps. Wou nd closu re is much faster w ith
preplaced sutu res and well-constructed incisions. If you have a penetrating keratoplasty
incision, close the globe w ith a temporary keratoprosthesis.
The risk of vitreou s incarceration m ay be reduced w ith m aintenance or reform ation
of the anterior cham ber w ith saline or air, but a h igh ly cohesive viscoelastic substance is
u sually requ ired. Tissue incarceration is u su ally best m anaged du ring a second ary pro-
cedu re. However, if you are convinced the hemorrhage is lim ited and there is no fu rther
bleed ing, then reopen ing, clean ing, and closing sm all sections of the wou nd in sequence
m ay be considered 1 but are not generally preferred.
157
158 Question 39
For extensive suprachoroid al hemorrhage, external d rainage via sclerotom ies m ay be

necessary to allow approxim ation of the anterior segment wou nd edges and thu s close
the globe. However, prim ary d rainage is hazardou s and shou ld be avoided if possible.
Drainage cou ld reverse the tamponade effect of the second arily elevated intraocu lar
pressu re and resu lt in add itional suprachoroid al bleed ing. Conversely, d rainage m ay be
u nsuccessfu l due to the suprachoroid al hemorrhage clotting before an em ergency scle-
rotomy can be p erformed. Follow ing external d rainage, if performed, the sclerotom ies
shou ld be sutu red closed, and the anterior cham ber shou ld be reformed, if p ossible.
Most lim ited intraoperative and delayed postoperative choroid al hemorrhages resolve
w ith in 6 weeks w ithout sequelae or treatm ent. Du ring the postoperative p eriod, topical
and oral pred n isone (1mg/ kg/ d ay) is recom mended to control in flam m ation and reduce
the likelihood of intraocu lar tissue ad hesion. Elevated intraocu lar pressu re requ ires
aggressive topical and sometim es oral therapy. Eye pain can be m anaged w ith topical
cycloplegia and oral analgesia.
Not all patients benefit from a second ary procedu re. Ind ications for su rgical m anage-
m ent and choroid al d rainage include central retinal apposition that does not resolve
w ith in 1 to 2 weeks, shallow/ flat anterior cham ber w ith second ary intraocu lar pressu re
elevation, hypotony, persistent flat anterior cham ber, or intractable eye pain. Make su re
that there is no wou nd leak as the cau se of hypotony or shallow/ flat anterior cham ber.
Drainage m ay need to be com bined w ith a vitrectomy if there is rhegm atogenou s or
tractional retinal detach ment, incarceration of intraocu lar tissues, retained len s m aterial,
vitreou s hemorrhage, or extensive su bretinal hemorrhage.
A suprachoroid al hemorrhage is easier to d rain after 7 to 14 d ays becau se the clotted
blood has a greater chance of being liquefied. Echograph ic characteristics of liquefaction
include reduced reflectivity on A-scan and a more echolucent pattern on B-scan.2
Many su rgeons believe that the optim al d rainage site is the h ighest point of choroi-
d al elevation, as determ ined by echography or ophthalmoscopy. Sclerotom ies shou ld be
rad ial and created w ith a cut-dow n incision. Isolation of the rectu s mu scles u sing silk ties
or traction sutu res w ill enable good exposu re of the sclera (Figu re 39-1). Alternatively, an
anterior sclerotomy such as created for pars plana vitreou s su rgery m ay be u sed. Usu ally
more than one sclerotomy is requ ired in opp osite qu ad rants. Drainage can be promoted
by m an ipu lating the sclerotomy w ith forceps or cotton-tip applicators. Alternatively, 6-0
traction sutu res can be placed on the wou nd edge.
Th roughout the d rainage procedu re, the intraocu lar pressu re mu st be m aintained w ith
an intraocu lar in fu sion. In order to reduce the risk of iatrogen ic d am age to the anteri-
orly d isplaced retina, the in fu sion can be introduced th rough the lim bu s via a 23-gauge
need le, a sm all gauge vitrectomy in fu sion can nu la, or an anterior cham ber m aintainer.
Establish the in fu sion prior to placement of the sclerotom ies to avoid hypotony. Precise
intraocu lar pressu re is d ifficu lt to m aintain u sing a syringe or gravity in fu sion. A con-
stant intraocu lar in fu sion pressu re (preset between 20 m m Hg and 30 m m Hg) is easier to
m aintain w ith a continuou s pressu rized in fu sion system.3
The stand ard intraocu lar in fu sion com mon ly u sed is balanced saline solution. If there
is adequ ate com mu n ication between the anterior and posterior cham bers, viscoelastic
or air can be utilized. When vitreoretinal su rgery is anticipated, the choice of substitute
is lim ited to clear liqu id s. Perfluorocarbon liqu id is particu larly u sefu l as a prim ary
How Do I Manage a Suprachoroidal Hemorrhage? 159
Fig u re 3 9 -1. An in tra o pe r-

a tive ph o to o bta in e d fro m
a vide o o f a ch o ro id a l
dra in a ge be in g pe rfo rm e d
u sin g a ra dia l cut-do w n
in cisio n . Go o d e xpo sure o f
th e scle ra is a ch ie ve d usin g
6 -0 silk tra ctio n tie s wh ich
a re lo o pe d a ro un d th e re c-
tus m uscle s. Th e dra in a ge
is fa cilita te d by m a n ipula t-
in g th e scle ro to m y w ith
two co tto n -tip a pplica to rs.
Ph o to co urte sy o f Dr. De a n
Elio tt.
su bstitute or as an adju nct in th is instance. Perfluorocarbon liqu id d isplaces vitreou s, lens

fragments, and suprachoroid al and su bretinal hemorrhage anteriorly. When u sing per-
fluorocarbon liqu id s, placing the sclerotom ies anterior to the equ ator m ay assist d rainage
of the suprachoroid al hemorrhage.4
Once norm al anatom ic con figu ration of the pars plana has been re-established w ith
d rainage, a conventional 3-port pars plana arrangem ent can be performed. If there is
vitreou s or retinal incarceration before or du ring d rainage, anterior vitrectomy th rough
the lim bu s or pars plana shou ld be considered prior to fu rther d rainage. Once d rainage
is complete, the retina m ay need to be delam inated from anterior structu res. A relaxing
retinotomy or scleral buckle m ay be requ ired for residu al vitreoretinal traction. A nonex-
pansile long-acting gas such as 14% C 3F8 can be u sed for internal tamp onade. You m ay
prefer to u se silicone oil in eyes in the follow ing situ ations: large relaxing retinotom ies,
h igh risk of proliferative vitreoretinopathy, hypotony, or recu rrent suprachoroid al hemor-
rhage. If you u se silicone oil, be aware that insu fficient silicone oil tamp onade m ay follow
complete reabsorption of the suprachoroid al hemorrhage.
Any residu al choroid al hemorrhage rem ain ing after d rainage shou ld completely
resolve over a few weeks to month s. Eyes shou ld be mon itored for postoperative compli-
cation s such as hyp otony, retinal detach ment, proliferative vitreoretinopathy, p ersistent
pain, and even phth isis.
References
1. Kuhn F, Morris R, Mester V. Choroidal detachment and expulsive choroidal hemorrhage. O phthalm ol Clin
N orth Am. 2001;14(4):639-650.
2. Chu TG, Green RL. Suprachoroidal hemorrhage. Surv O phthalm ol. 1999;43(6):471-486.
3. Lakhanpal V, Schocket SS, Elman MJ, Nirankari VS. A new modified vitreoretinal surgical approach in the
management of massive suprachoroidal hemorrhage. O phthalm ology. 1989;96 (6):793-800.
4. Desai UR, Peyman GA, Chen CJ, et al. Use of perfluoroperhydrophenanthrene in the management of supra-
choroidal hemorrhages. O phthalm ology. 1992;99 (10 ):1542-1547.
40 QUESTION
WHEN SHOULD I SUSPECT

ENDOPHTHALMITIS IN MY POSTOPERATIVE
CATARACT PATIENT AND WHAT ARE THE
TREATMENT O PTIONS?
Bernard H. Doft, MD
Postcataract su rgery endophthalm itis is one of the most d readed complication s of cata-
ract su rgery. The incidence m ay be rising in correspondence w ith increased u se of clear
corneal su rgery, and it cou ld be as h igh as 2 to 3 per 1000 operations. Every ophthalmolo-
gist recogn izes the typical case of a patient presenting a few d ays after su rgery w ith a
pain fu l red eye, cou nt fingers (CF) vision, a hypopyon, and no view posteriorly. However,
there are much more su btle presentation s, and early detection can resu lt in better out-
comes. Making an early d iagnosis can sometim es be d ifficu lt. When shou ld you su spect
endophthalm itis? Once you su spect it, what do you do? To add ress these questions, let u s
look at a few patients presenting at variou s time points after cataract su rgery.
Patient One comes in on the first postoperative d ay. He has a bit of d iscom fort. Vision
is 20/ 80. At slit lamp, you see a defin ite hypopyon. When you look beh ind the intraocu-
lar len s (IOL), the vitreou s is pretty clear. Is th is endophthalm itis? It can be hard to tell.
If endophthalm itis is presenting th is early, such as on the first postoperative d ay, it is
probably a pretty viru lent organ ism. Yet the vitreou s in th is patient is clear. Does that
m ake sen se? What do you do? The answer is watch the patient for a few hou rs and have
another look. Ok, so it doesn’t get worse in those few hou rs. What now? Well, it still
cou ld be endophthalm itis, but more likely you need to th in k about toxic anterior segment
synd rome (TASS). Watch a bit longer, and look again in a few more hou rs. If the vitreou s
rem ains qu iet, then it is likely TASS. Th is synd rome is not in fectiou s endophthalm itis,
and m anagem ent is entirely d ifferent than for endophthalm itis.
161
162 Question 40
Patient Two comes in on postoperative d ay 1 ju st like Patient One. You wait a few hou rs
and look again. The hypopyon is bigger, fibrin in the anterior cham ber (AC) is worse, and
the vitreou s is becom ing cloudy. The patient becam e worse in those few hou rs. Now you
have you r d iagnosis. Endophthalm itis that presents th is early is a very viru lent organ ism,
cau sing the rapid increase in in flam m ation. Emergency treatment is necessary.
Let u s look at Patient Th ree. It is 5 to 6 d ays postoperatively. The patient comes in for a
routine exam . She has no complaints, but states, “My vision isn’t perfect, doc.” On exam,
vision is 20/ 50-, the cornea is clear, the IOL is in good p osition, and w ith a carefu l exam,
you see defin ite cells in the AC but no hypopyon. It is a bit more in flam m ation than
you are u sed to seeing, but not much, and there are also som e cells in the vitreou s. “My
su rgery went well,” you say. “Why shou ld there be a cellu lar reaction in the AC?” Keep
asking you rself that question. Don’t let the absence of pain or the absence of a hypopyon
fool you. Perform gon ioscopy. There are tim es that one can see a sm all layering of cells in
the in ferior angle w ith a gon ioscopy lens when one can not really appreciate it at the slit
lamp alone. Remem ber, you don’t have to have a hypopyon to d iagnose endophthalm itis.
Th is is another patient to watch carefu lly. See the patient several hou rs later. If late in the
d ay th ings don’t look worse, look again very soon, perhaps the next morn ing. However,
if th ings are getting worse when you see the patient later that d ay, there is a good chance
it’s endophthalm itis.
Ok, now let u s look at Patient Fou r. It is 3 to 5 d ays after su rgery. The patient says, “My
eye aches.” Of cou rse, you r staff had the patient com e right over. On exam, visual acu ity
is 20/ 800, and the eye is red. The AC show s a 1 m m layered hyp opyon, cells, and fibrin
strand s (Figu re 40-1). The vitreou s is cloudy. With ind irect ophthalmoscopy, you can ju st
barely m ake out the retina. No mystery here! Th is is classic postoperative endophthalm i-
tis. Now, here’s a variant. Patient Five presents ju st like Patient Fou r except he is d iabetic.
Managem ent m ight d iffer, so read on. One more patient is Patient Six. He has the exact
same presentation as Patient Fou r except that vision on presentation is light perception
(LP) on ly. The d iagnosis is obviou s in Patient Fou r, Patient Five, and Patient Six and the
presentation is textbook.
Here is ou r last patient, Patient Seven. It is about 6 or more weeks postsu rgery or even
1 year or longer. The patient is in for a routine exam. There are on ly m in im al complaints
about slightly blu rred vision. Acu ity is 20/ 40. The eye looks wh ite. There is no hypopyon,
but there is a m ild AC reaction w ith a few keratic precipitates (Figu res 40-2 and 40-3). You
see some creamy opacification of the posterior capsu le. The vitreou s has som e in flam m a-
tion that m ight look like a sm all string of pearls. But, it is m ild. The eye appears to have
m ild uveitis. Most likely, th is is ch ron ic indolent endophthalm itis. Management is qu ite
d ifferent than for patients who present w ith more acute d isease.
Well, you have m ade you r d iagnosis, or at least you are su spiciou s enough to consider
it. Here are a few points. If you are a cataract su rgeon, and if you even th in k of endo-
phthalm itis, assu me that is what it is u ntil you prove it otherw ise. It is much safer for
you r patient if you th in k of it that way. I have often had cataract su rgeons call m e and say,
“I saw th is patient yesterd ay morn ing, and the eye looked in flamed, so I increased the
frequency of topical antibiotics. Now, it is much worse.” Endophthalm itis is never treated
alone w ith postoperative topical d rugs. If you are th in king in fection, refer the patient to
a retinal specialist. Using increased frequency of topical antibiotics w ill ju st cau se you to
lose valu able time. If the eye is in fected, it w ill on ly get worse, and you w ill have delayed
When Should I Suspect Endophthalmitis in My Postoperative Cataract Patient? 163
Fig u re 4 0 -1. Cla ssic po sto pe ra tive e n do -

ph th a lm itis with hypo pyo n .
Fig u re 4 0 -2 . Ke ra tic pre cipita te s in a n e ye

with ch ro n ic in do le n t e n do ph th a lm itis.
Fig u re 4 0 -3 . Cre a m y wh ite o pa cifica tio n s

o f th e ca psule in ch ro n ic in do le n t e n do -
ph th a lm itis.
164 Question 40
appropriate treatm ent. Rem em ber, when in dou bt, send it out! (That is, send it to you r
local friend ly retina sp ecialist for another opin ion.)
What Are the Treatment Options?

Ok, the d iagnosis is m ade. Let’s go patient by patient. First, Patient One. The patient
presented on the first postoperative d ay w ith a hypopyon, but clear vitreou s, and d id not
progress to have vitreou s involvem ent w ith observation. Th is patient w ith TASS does not
have endophthalm itis. There is no space to deal w ith m anagem ent of TASS here, but it
involves treating w ith steroid s, controlling pressu re, corneal in flam m ation, etc. Look it up
if you are not fam iliar w ith th is synd rom e.
Patient Two, Patient Th ree, and Patient Fou r are all m anaged the same. Each is a patient
w ith acute postcataract su rgery endophthalm itis presenting w ith vision of cou nting
fingers or better. Stand ard of care treatment involves obtain ing cu ltu res of the AC and
vitreou s, injection of intravitreal antibiotics, along w ith p ostop erative topical antibiot-
ics. Depend ing on the patient’s d rug allergies, intravitreal antibiotics include intravitreal
vancomycin 1 mg in 0.1 cc and either am ikacin 0.4 mg in 0.1 cc or ceftazid ime 2 mg in
0.1 cc. Vitrectomy is u su ally not requ ired when patients present w ith CF or better vision.
Unless you are out in the boon ies and have no im med iate access to a retinal specialist,
refer th is patient out im med iately. Don’t wait u ntil the next d ay. If you r retinal specialist
is not available, find another one! The patient can probably be m anaged as an outpatient,
and if the patient is not a d iabetic, he or she w ill have a better than 60% chance of end ing
up 20/ 40 or better.
Patient Five is a bit d ifferent as he is d iabetic. He m ay have a better outcome if a vitrec-
tomy is perform ed, as opposed to ju st a tap and injection of antibiotics alone.
In Patient Six, the presenting vision is LP on ly. Th is patient requ ires a vitrectomy, and
he need s it im med iately. Even w ith th is severe presentation, outcomes can be excellent
w ith one-th ird of eyes ach ieving a visual acu ity of 20/ 40 or better.
Ou r last patient, Patient Seven, most likely has ch ron ic indolent endophthalm itis. Th is
is not an em ergency. Depend ing on presentation, if you are not su re of the d iagnosis, you
m ight want to treat th is patient w ith topical steroid s and watch. Som etimes th ings w ill
get better, but then relap se. When it does, it is tim e to refer th is patient out for defin itive
treatm ent. The organ ism s are often sequestered in the lens capsu le. Management options
vary, but often the best outcomes w ill occu r w ith vitrectomy, intravitreal antibiotics, and
lens exchange or removal.
Suggested Reading
Results of the Endophthalmitis Vitrectomy Study: a randomized trial of immediate vitrectomy and of intravenous
antibiotics for the treatment of postoperative bacterial endophthalmitis. Endophthalmitis Vitrectomy Study
Group. Arch O phthalm ol. 1995;113(12):1479-1496.
Doft BH, Wisniewski SR, Kelsey SF, Fitzgerald SG, Endophthalmitis Vitrectomy Study Group. Diabetes and postop-
erative endophthalmitis in the endophthalmitis vitrectomy study. Arch O phthalm ol. 2001;119 (5):650-656.
Mamalis N, Edelhauser HF, Dawson DG, Chew J, LeBoyer RM, Werner L. Toxic anterior segment syndrome. J
Cataract Refract Surg. 2006;32(2):324-333.
41 QUESTION
HOW DO I FOLLOW A PATIENT WITH

A PRESUMED CHOROIDAL NEVUS?
Prithvi Mruthyunjaya, MD
Choroid al melanocytic proliferation s follow a range of clin ical m an ifestation s that

include ben ign, prem alignant, and m alignant cond itions. As the vast m ajority of these
lesion s are defined by their clin ical featu res and less com mon ly by a con firm atory biopsy,
there can be some controversy and con fu sion over how to characterize a particu lar
lesion.
A choroid al nevu s is a com mon find ing in the adu lt Wh ite popu lation and although
on ly 1 in 9000 w ill likely progress into a m alignant uveal melanom a (UM), it is important
for the clin ician to follow and re-evalu ate these lesion s.1 The clin ical challenge is to define
parameters for a particu lar ben ign or pre-m alignant lesion that m ake it more or less likely
to progress into a m alignancy.
Ben ign cond itions includ e choroid al melanocytosis, or choroid al freckle, wh ich is
typically a flat pigmented lesion less than 2 d isc d iameters in d iameter that rem ains sta-
tionary over time. A ben ign choroid al nevu s is typically a m in im ally elevated, variably
pigmented lesion that has a base between 1 to 10 m m and an apical height typically less
than 1.5 m m .2 Ben ign nevi often have a smooth su rface w ith d ru sen and m in im al retinal
pigment epithelial (RPE) d isruption (Figu re 41-1). The more concern ing spectru m is what
I refer to as an indeterm inate choroid al lesion (ICL). These are the atypical or “su spi-
ciou s” choroid al nevi, but not defin itively a UM, that m ay exh ibit one or more featu res
that h istorically m ay either pred ict futu re grow th or m ay be fou nd more com mon ly w ith
an active UM. It is important to exclude cond itions that com mon ly m im ic or m asquerade
as a choroid al nevu s or UM, includ ing choroid al vascu lar lesions, m etastatic tu mors to
the choroid, or su bretinal hemorrhage from non m acu lar choroid al neovascu larization
(ectopic d isciform lesion).
Melanocytic choroid al lesion s are d ivided into 3 broad categories in ou r ocu lar oncol-
ogy clin ic: 1) ben ign (low su spicion for m alignancy), 2) indeterm inate (med iu m to h igh
165
166 Question 41
Fig u re 41-1. Ch o ro ida l n e vus w ith

o n ly ove rlyin g druse n e stim a te d to
h a ve a lo w risk fo r gro w th o r m a lig-
n a n t tra n sfo rm a tio n .
su spicion for m alignancy), and 3) m alignant. Th is is based on com bin ing patient h istory
w ith the clin ical exam ination, ancillary im aging stud ies, and stand ard ized u ltrasonogra-
phy. We sp ecifically ask patients if they are having any visu al symptom s that m ay suggest
flu id exud ation or lesion grow th (eg, photop sias or visu al field changes) and aggressively
seek any previou s docu mentation of the lesion, includ ing evidence of change or grow th
of the lesion from old photographs or u ltrasou nd s. Key clin ical exam ination featu res
include pigmentation pattern (predom inantly melanotic, amelanotic, or m ixed pigmenta-
tion), an estim ate of the size and d istance to the fovea and optic nerve (in m illimeters or
d isc d iameters), the presence of su bretinal flu id (eg, overlying or adjacent to the lesion),
lip ofu scin accu mu lation (orange pigment), d ru sen, or associated RPE alterations near
the tu mor ind icative of previou s su bretinal flu id (“h igh water m ark”) (Figu re 41-2). All
lesions are photographed and optical coherence tomography im ages (OCT) of the lesion
are obtained. A stand ard 10 MH z B-scan u ltrasou nd probe is u sed to m easu re the lesion
base and height. Th is baseline in form ation is vital in formu lating an accu rate d iagnosis
and for futu re follow-up.
Ophthalm ic im aging is invalu able in the d iagnosis and mon itoring of these lesions.
There is no su bstitute for good fu ndu s photographs that captu re all m argins of the tu mor
and can provide a reference view of the m acu la and optic nerve. Use of a comp osite or
montage im age featu re fou nd on m any d igital fu ndu s cameras or com m ercially avail-
able w ide-angle fu ndu s cameras provides a lesion “footprint” that m ay be compared on
follow-up. We have fou nd that OCT scan n ing over the lesion su rface provides excellent
u ltrastructu ral in form ation. The OCT can help d istingu ish between intraretinal cystic
flu id accu mu lation, typically a sign of ch ron icity, and active neu rosen sory retinal separa-
tion from su bretinal flu id, typically a sign of an active lesion (Figu re 41-3). Early stud ies
have demonstrated an association between active su bretinal flu id on OCT and su bsequent
grow th of ICLs, suggesting that such find ings mu st be mon itored very closely.5
Retrospective stud ies have identified several featu res at presentation that m ay help pre-
d ict grow th of a sm all ICL that is su bsequently d iagnosed as a UM. These featu res include
tu mor th ickness ≥ 2 m m, proxim ity to the optic nerve head or foveola, the presence of
How Do I Follow a Patient With a Presumed Choroidal Nevus? 167
Fig u r e 41-2 . In de te rm in a te ch o ro ida l

le sio n w ith m o de ra te risk o f gro w th a n d
pro gre ssio n in to a u ve a l m e la n o m a .
Am e la n o tic n evus w ith m ild RPE a lte ra -
tio n a n d druse n .
Fig u re 41-3 . Op tica l co h e re n ce to m o gra ph y ove r th e le sio n de m o n stra te s n e uro -

se n so ry re tin a l se pa ra tio n co n siste n t w ith subre tin a l fluid, a sig n th a t h a s be e n
a sso cia te d w ith subse que n t gro w th .
lipofu scin (orange) pigm ent overlying the tu mor, su bretinal flu id, or visu al symptom s.3,4
Risk stratification suggests that patients w ith ICLs associated w ith more than one of these
featu res be cou nseled to the sign ificant chance of grow th over a 5-year p eriod.4 Such
lesions requ ire very close follow-up or, preferably, an evaluation by an ocu lar oncologist.
Carefu l follow-up is imp ortant w ith any lesion. The presence of even the most ben ign-
appearing lesion shou ld be conveyed to the patient and h is or her prim ary ophthalmolo-
gist to ensu re a routine yearly follow-up. I mon itor med iu m to h igh su spicion ICLs at
least every 4 months for 1 year and, if completely stable, m ay su bsequently extend the
interval to every 6 months. In the case of a h igh ly su spiciou s ICL or a sm all melanom a
that the patient elects to not have treated, re-evalu ations are conducted no longer than 2
to 3 month s apart and sooner if new visu al symptom s are reported.
Any evidence of grow th or the development of new featu res that arise du ring the
cou rse of routine follow-up, such as expansion of lesion borders, new su bretinal flu id,
or change in u ltrasou nd characteristics, shou ld prompt fu rther evalu ation by an ocu lar
oncology sp ecialist who m ay then consider continued close observation (2- to 3-month
intervals), intraocu lar biopsy, or defin itive treatm ent.
168 Question 41
References
1. Singh AD, Kalyani P, Topham, A. Estimating the risk of malignant transformation of a choroidal nevus.
O phthalm ology. 2005;112(10 ):1784-1989.
2. Augsburger JJ, Corrêa ZM, Trichopoulos N, Shaikh A. Size overlap between benign melanocytic choroidal nevi
and choroidal malignant melanoma. Invest O phthalm ol Vis Sci. 2008;49 (7):2823-2828.
3. Singh AD, Mokashi AA, Bena JF, Jacques R, Rundle PA, Rennie IG. Small choroidal melanocytic lesions:
Features predictive of growth. O phthalm ology. 2006;113(6):1032-1039.
4. Shields CL, Cater J, Shields JA, Singh AD, Santos MC, Carvalho C. Combination of clinical factors predictive
of growth of small choroidal melanocytic tumors. Arch O phthalm ol. 2000 ;118 (3):360-364.
5. Espinoza G, Rosenblatt, Harbour JW. O ptical coherence tomography in the evaluation of retinal changes
associated with suspicious choroidal melanocytic tumors. Am J O phthalm ol. 2004;137(1):90-95.
42 QUESTION
HOW DO I DISTINGUISH O NE
PIGMENTED LESION FROM ANOTHER?
Amy C. Schefler, MD
The d ifferential d iagnosis of a pigmented choroidal lesion includes choroid al nevu s,

choroid al melanom a, congenital hypertrophy of the retinal pigment epitheliu m (CH RPE),
melanocytom a, and extram acu lar d isciform lesion. Su spiciou s choroid al nevi accou nt for at
least 50% of the referrals that are sent to ou r center to ru le out uveal melanom a. Given that a
m issed uveal melanom a d iagnosis can be life threatening, the general ophthalmologist and
retinal specialist shou ld be well versed in d istingu ishing these lesions from each other.
Choroidal Nevi and Melanoma

Pigmented choroid al nevi are extremely com mon in the Wh ite popu lation, and it is
estim ated that 10% to 15% of adu lts have at least one such lesion. Uveal melanom a, on the
other hand, is a rare d isease w ith on ly 1200 cases per year in the Un ited States. Thu s, the
task of d istingu ish ing the rare m alignant lesion from m any ben ign lesion s is challeng-
ing. The most important d iagnostic tools in th is endeavor are fu ndu s exam ination w ith
docu mentary photography and ophthalm ic u ltrasou nd.
There are 6 featu res (in order of decreasing importance) outlined below that aid in
d istingu ish ing a choroid al nevu s from a melanom a. These criteria shou ld also help in
identifying nevi that present a h igher risk of m alignant tran sform ation and necessitate
more frequent follow-up 1:
• Size: Size is the most important risk factor/ determ in ing factor in classifying uveal
m elanom as. The Collaborative Ocu lar Melanom a Study (COMS) defined the cut-
off between sm all m elanom as (wh ich shou ld be observed) and med iu m tu mors
(wh ich shou ld be treated) as 2.5 m m in apical height.2 In the hand s of an excel-
lent echographer, the apical height for a choroid al m ass shou ld be reproducible
169
170 Question 42
to w ith in 0.2 m m . Basal d imen sions of choroid al tu mors, especially those w ith
shallow-angled borders, are more d ifficu lt to measu re and to reproduce but are cer-
tain ly considered h igh risk if greater than 16 m m . In general, we follow the COMS
gu idelines and define lesions greater than 2.5 m m in height as m elanom as.
• Ultrasound characteristics: On B-scan u ltrasonography, choroid al melanom as are
more likely to be vascu lar, whereas nevi are more com mon ly avascu lar. On A-scan,
melanom as tend to have low to m ed iu m internal reflectivity versu s the h igh inter-
nal reflectivity that often characterizes nevi.
• Orange pigment: When choroid al nevi develop orange pigment on their su rface, the
chance of m alignant tran sform ation increases.
• Presence of serous fluid: Subretinal flu id associated w ith a choroid al nevu s can be
ind icative of choroid al neovascu larization, but th is is also a h igh-risk featu re for
m alignant transform ation and a sign ificant nu m ber of m elanom as have associated
serou s retinal detach m ents.
• Presence of drusen: The presence of d ru sen is a strong ind icator of ch ron icity and
ben ign ity. Lesions w ith d ru sen are rarely melanom as or u ndergo m alignant trans-
form ation.
• Location: Ju xtapapillary location renders a nevu s more likely to be/ becom e m elano-
m as than those that are not near the optic nerve.
There is sign ificant controversy on the issue of the defin ition of a sm all melanom a.
Sm all lesions (by COMS criteria) m ay be reasonably observed w ithout an increased rate
of metastasis or death, reserving treatm ent on ly for docu mented grow th of the lesion or
su spiciou s changes.3 However, sm aller lesion s that have already demonstrated obviou s
grow th on reliable serial photography or echography and/ or have one or more other h igh-
risk featu res such as orange pigment need to be treated.
Weigh ing all of the risk factors associated w ith a particu lar lesion takes clin ical expe-
rience and judgm ent. As a general ru le, we feel that a patient w ith h igh-risk featu res
for size or 2 other risk factors shou ld be referred for assessm ent by an ocu lar oncologist
(Figu re 42-1). Sm all, flat peripheral lesions (Figu re 42-2) w ith no orange pigment and w ith
associated d ru sen can be observed and docu mented w ith serial photography by a general
ophthalmologist or retina sp ecialist. When in dou bt, an ophthalmologist who has lim ited
exp erience w ith tu mors shou ld always refer a patient to an ocu lar oncologist for an evalu-
ation.
Other Pigmented Choroidal Lesions

Other pigmented lesions that are important to include in the d ifferential d iagnosis
of choroid al melanom a are CH RPE, melanocytom a, and extram acu lar d isciform lesion s
associated w ith age-related m acu lar degeneration. Its sharply dem arcated borders and
intralesional lacu nae can d istingu ish CH RPE. They are generally flat and change slowly,
if at all, over tim e. They can occu r w ith in the m acu la, affecting central vision. When
occu rring as mu ltiple lesions and con figu red in an oval shap e, these lesions raise the su s-
picion for the d iagnosis of Gard ner’s synd rome, and such patients shou ld be advised to
u ndergo a colonoscopy to ru le out cancerou s polyps.
How Do I Distinguish One Pigmented Lesion From Another? 171
Fig u re 4 2 -1. (A) A ch o ro ida l n e vus is sh o wn

A with h igh -risk ch a ra cte ristics, in cludin g a pica l
th ickn e ss, o ra n ge pigm e n t, a n d subre tin a l fluid.
Th e 2 m a n a ge m e n t o ptio n s fo r th is pa tie n t a re
clo se se ria l o bse rva tio n a n d pla que bra ch y-
th e ra py. (B) Stra tu s OCT de m o n stra te s th e
subre tin a l fluid a sso cia te d with th is h igh -risk
le sio n .
A B
Fig u re 4 2 -2 . (A) Ch o ro ida l n e vus is pre se n te d with lo w-risk ch a ra cte ristics, in cludin g n o a pica l
th ickn e ss, pre se n ce o f druse n , n o subre tin a l fluid, a n d la ck o f o ra n ge pigm e n t. (B) B-sca n ultra -
so un d o f th e fla t le sio n in Figure A is sh o w n .
Melanocytom as are typically jet black rather than the slate gray/ brow n hue of m elano-
m as and most com mon ly occu r in a peripapillary location (Figu re 42-3). They are much
more com mon in persons of African descent than Wh ites. They most com mon ly demon-
strate h igh internal reflectivity and avascu larity on u ltrasou nd. They can rarely have asso-
ciated choroid al neovascu larization or vascu lar occlu sive events and can u ndergo sm all
changes in size over time. There have been several rare reports of these lesions tran sform-
ing into choroid al melanom a and as such, these lesions shou ld be followed serially.
172 Question 42
Fig u re 4 2-3 . Th is is a m e la n o cyto m a

o f th e o ptic disc with typica l je t bla ck
pigm e n t in a da rkly pigm e n te d in di-
vidua l.
An extram acu lar d isciform lesion, also referred to as peripheral exud ative hemorrhagic
chorioretinopathy, can produce d iagnostic con fu sion when associated w ith subretinal
and/ or vitreou s hemorrhage. These can typically be d istingu ished from choroid al m ela-
nom a by their anterior location and presence of su bretinal blood in variou s stages of reso-
lution. Wh ile uveal melanom as can u ncom mon ly be associated w ith subretinal hemor-
rhage, the degree of hemorrhage is not generally as profou nd as is its association w ith an
extram acu lar d isciform lesion. One helpfu l d iagnostic clue is the presence of peripheral
d ru sen in association w ith the d isciform lesion.

Timothy G. Mu rray, MD, MBA, FACS
Professor, Ophthalmology, Director, Ocu lar Oncology Service
Departm ent of Ophth almology, Bascom Palm er Eye In stitute, Miam i, FL
References
1. Factors predictive of growth and treatment of small choroidal melanoma: CO MS Report No. 5. The
Collaborative O cular Melanoma Study Group. Arch O phthalm ol. 1997;115(12):1537-1544.
2. Diener-West M, Earle JD, Fine SL, et al. The CO MS randomized trial of iodine 125 brachytherapy for choroidal
melanoma, II: characteristics of patients enrolled and not enrolled. CO MS Report No. 17. Arch O phthalm ol.
2001;119 (7):951-965.
3. Sobrin L, Schiffman JC, Markoe AM, Murray TG. O utcomes of iodine 125 plaque radiotherapy after initial
observation of suspected small choroidal melanomas: a pilot study. O phthalm ology. 2005;112(10 ):1777-
1783.
43 QUESTION
HOW DO I KNOW IF THERE IS A

RETINAL DETACHMENT OR TUMOR WHEN
I AM DOING ULTRASONOGRAPHY?
Cathy DiBernardo, CDOS, ROUB

and James T. Handa, MD
With u ltrasonography of the norm al eye, the vitreou s cavity is echolucent (devoid of
any signals) and the p osterior layers of the eye (retina, choroid, and sclera) are ind istin-
gu ishable from one another. It is on ly when signals are present w ith in the vitreou s cavity
that you need to p erform fu rther evalu ation to d ifferentiate one echo from another.
The follow ing are key word s that are u sed regu larly to describe u ltrasou nd character-
istics:
• The gain setting is the decibel level. Use h igh or m aximu m gain to evalu ate the vit-
reou s. Decrease the gain settings as needed to improve resolution of the echoes.
• Perpendicularity is ach ieved when the sou nd beam is aligned to target ocu lar struc-
tu res, resu lting in m axim ally h igh signals. An oblique sou nd beam align ment rela-
tive to the ocu lar structu res resu lts in d im in ished or false echo in form ation.
• A fter movement is an ind ication of the kinetic prop erties of an echo. You shou ld
look at how the echoes move du ring “dynam ic” echography.
• Reflectivity refers to the brightness of a signal produced by an echo on B-scan and
the amplitude of the echo spikes on A-scan.
• Structure refers to the d istribution of echoes created by the internal h istologic
organ ization of the tissue. On A-scan, th is is evaluated by the d istribution of the
echo spikes w ith in the tissue, and on B-scan, it is evaluated by variations in echo
den sity w ith in the tissue being evalu ated.
173
174 Question 43
Retinal Detachment
The retina is one of the h ighest reflective su rfaces in the eye, and when it is partially
or totally detached, it has d istinct echograph ic characteristics that set it apart from other
structu res. The structu re that most com mon ly mu st be d istingu ished from the retina is
the posterior hyaloid layer of a posterior vitreou s detach ment. The posterior hyaloid is
typically less reflective and th in ner than the retina (Figu re 43-1A). Generally, when the
retina is detached, B-scan u ltrasonography w ill demonstrate a mem brane that is h igh ly
reflective, th ick, and often folded (Figu re 43-1B). If the retinal detach m ent involves the
p eripapillary region, then when you im age the optic nerve, you w ill see a mem brane (the
detached retina) insert at the optic nerve, wh ile the mem brane of a total p osterior vitre-
ou s detach ment (PVD) w ill appear above the nerve w ith in the vitreou s cavity. When the
p osterior hyaloid rem ains ad herent to the posterior pole, it is u sually near the edge of
the optic d isc as opposed to the center of the d isc as occu rs w ith a retinal detach ment.
Du ring dynam ic exam ination, you w ill typically see some movement of a detached retina.
Th is movem ent can range from a jiggling motion when the detach ment is acute (w ithout
m em branes) to stiff movement when it is ch ron ic (w ith mem branes). In contrast, the
vitreou s is typically more mobile than the retina. You can determ ine the exact extent of
a retinal detach ment w ith a complete evaluation of all qu ad rants of the eye. Tran sverse
(cross-sectional view s) and longitud inal (rad ial sections) scans shou ld be employed du r-
ing the screen ing process to aid in the d iagnosis and determ ine the extent of the retinal
detach m ent. Be organ ized. You shou ld move the u ltrasou nd probe in a system atic fash ion
from the lim bu s (sou nd aimed posteriorly) to the forn ix (sou nd aimed to the periphery).
You m ay m iss retinal tears and traction mem branes w ithout an organ ized approach. If
stand ard ized A-scan is available to you, the spike created from a retinal detach m ent is
u sually m axim ally h igh (as h igh as the in itial signal on the im age).
Intraocular Tumors
Clinically, intraocu lar tu mors can m im ic one another. If the patient history does not add
inform ation to aid in the d iagnosis, u ltrasou nd is u sefu l to d ifferentiate one lesion from
another and to provide precise measu rements and details about nearby ocu lar structu res.
Tu mors can be any shap e or size. You can u se B-scan u ltrasonography for docu m ent-
ing the top ograph ic featu res such as location, shap e, extent, and gross estim ation of the
height. You can obtain th is in form ation by u sing transverse and longitud inal scans to
bu ild a th ree-d imen sional (3-D) im age and localize the lesion. As you interpret the d if-
ferent scans, create a composite, 3-D im age of the lesion in you r m ind. Once the location
has been determ ined, docu ment the shap e and measu re the m axim al height and basal
d iameter. Since m ass lesions are typically solid, you shou ld not see after movement from
the lesion su rface. However, you m ay see movem ent w ith in the tu mor resu lting from its
intrinsic circu lation.
The structu re of the lesion creates echo characteristics that reveal its internal h istologic
organ ization that can be u sed to categorize it. On stand ard ized A-scan, th is is evalu ated
by the d ifferences in amplitude of the echo spikes w ith in the lesion, and on B-scan, it is
evaluated by variations in echo density w ith in the tu mor. When the tu mor h istology is
How Do I Know if There Is a Retinal Detachment or Tumor? 175
Fig u re 43 -1. (A) Po ste rio r vitre o us de ta ch m e n t. To p : Tra n sve rse sca n sh o win g a po s-
te rio r vitre o us de ta ch m e n t (a rrow ). Ce n te r: Lo n gitudin a l sca n sh o win g n o in se rtio n
a t th e po ste rio r po le o r o ptic n e rve h e a d (ON). Bo tto m : Sta n da rdize d A-sca n sh o win g
o n ly a m in im a l spike pro duce d by th e po ste rio r hya lo id surfa ce . (B) Rh e gm a to ge n o us
re tin a l de ta ch m e n t. To p: Tra n sve rse sca n sh o w in g a de n se , fo lde d, h ighly re fle ctive
m e m bra n e (a rro w ). Ce n te r: Lo n gitudin a l sca n sh o w in g th e de n se m e m bra n e a dh e r-
e n t to th e o ptic n e rve h e a d (a rro w a n d ON). Bo tto m : Sta n da rdize d A-sca n sh o w in g a
m a xim a lly h igh , th ick spike pro duce d by th e re tin a l de ta ch m e n t.
homogeneou s, the echo pattern rem ains con sistent on A-scan and a u n iform appearance
w ith in the tu mor on B-scan. The reflectivity refers to the amplitude of the spikes w ith in
the body of the tu mor that is seen on A-scan and the signal brightness on B-scan. The
more organ ized the tu mor w ith sm all cells, the lower the amplitude of the echo spikes
and the lower the brightness becau se more of the sou nd waves pass th rough the tu mor
instead of reflecting back to the probe at d ifferent interfaces. The structu re (d istribution
of cells) and the reflectivity (size of cells) can be d ifficu lt to determ ine w ith B-scan alone.
Add ing the A-scan vector (sup erimposed A-scan im age over the B-scan im age) is u su ally
not helpfu l becau se there is no stand ard setting for these scans. Stand ard ized A-scan, on
the other hand, has been u sed successfu lly to d ifferentiate one tu mor from another.
176 Question 43
The most com mon echograph ic featu res of som e of the most com mon intraocu lar
tu mors are described below:
• Choroidal melanoma: B-scan u ltrasonography demonstrates the follow ing find ings:
solid, collar button or dome-shaped, regu larly structu red, and intrin sic vascu lar-
ity (Figu re 43-2). Low reflectivity is seen on A-scan u ltrasonography. If you see a
collar-button shap e, it is a certain ind ication that the tu mor has grow n th rough
Bruch‘s mem brane. Becau se the tu mor is typically well organ ized (ie, the spind le
cells form a pseudosyncitiu m), you w ill see a regu lar structu re on A- and B-scan
and med iu m-low amplitude echoes (reflectivity) on A-scan. We consider the ampli-
tude as “m ed iu m” in height w hen it is half of the height of the spike corresp ond ing
to the neu rosen sory retina. Du ring dynam ic u ltrasou nd, you can see blood flow
movement w ith in the tu mor, ind icative of its intrin sic vascu larity.
• Choroidal metastasis: B-scan u ltrasonography demon strates the follow ing find ings:
solid, dome-shaped, often w ith a central area of excavation, irregu larly structu red,
and m in im al or no vascu larity. The central excavation is often due to necrosis w ith-
in the tu mor since it does not have its ow n vascu lar structu re as does a melanom a.
A choroid al metastasis is irregu larly structu red and its content is heterogeneou s
due to the d isorgan ized cellu lar grow th pattern, in flam m ation, and necrosis.
• Choroidal nevus: B-scan u ltrasonography demon strates the follow ing find ings:
solid, flat or slightly dom e-shaped w ith m in im al elevation, regu larly structu red,
and no vascu larity (Figu re 43-3). H igh reflectivity is seen on A-scan u ltrasonogra-
phy. The spike height helps to d istingu ish th is from a melanom a, and the regu lar
structu re is not typical for a metastasis.
• Choroidal hemangioma: B-scan u ltrasonography demonstrates the follow ing find-
ings: solid, dome-shaped, regu larly structu red, and no vascu larity. H igh reflectiv-
ity is seen on A-scan u ltrasonography. On u ltrasou nd, th is lesion is d ifficu lt to
d ifferentiate from a choroid al nevu s becau se the tu mor height is u su ally m in im al
so that the internal structu re is d ifficu lt to characterize. It is important to integrate
you r clin ical observation s to d istingu ish between the lesions. A hem angiom a w ill
app ear orange-red w ith ind istinct borders such that the lesion almost “blend s” in
w ith the su rrou nd ing choroid al backgrou nd, wh ile a nevu s is brow n or amela-
notic.
• M elanocytoma: B-scan u ltrasonography demon strates the follow ing find ings: solid,
dome-shaped over the optic d isc, regu larly structu red, and no vascu larity. H igh
reflectivity is seen on A-scan u ltrasonography. Th is lesion is d ifferent from a mela-
nom a becau se it has h igh reflectivity wh ile a m elanom a has low reflectivity on
A-scan.
How Do I Know if There Is a Retinal Detachment or Tumor? 177
Fig u re 43 -2 . Ch o ro ida l m e la n o m a . To p le ft: Tra n sve rse sca n sh o w in g a sligh tly pe a ke d m a ss

le sio n (a rrow ). Th e cro ssh a irs sh o w th e la te ra l dim e n sio n o f th e tum o r. To p righ t: Lo n gitudin a l
sca n sh o win g th e le sio n a t th e po ste rio r po le (a rro w ) clo se to th e o p tic n e rve h e a d (ON). Bo tto m
le ft: Axia l sca n sh o win g th e le sio n a t th e po ste rio r po le clo se to th e o ptic n e rve h e a d (ON).
Bo tto m righ t: Sta n da rdize d A sca n sh o w in g th e lo w re fle ctivity th a t is a co n siste n t fin din g in
ch o ro ida l m e la n o m a (a rro w ).
178 Question 43
Fig u re 43 -3 . Ch o ro ida l n evus. To p le ft: Tra n sve rse sca n sh o win g a m in im a lly e leva te d, do m e
sh a pe d le sio n (a rro w ). To p righ t: Lo n gitudin a l sca n sh owin g th e tum o r (a rro w ) a n d its proxim -
ity to th e o p tic n e rve (ON). Bo tto m le ft: Sta n da rdize d A sca n a t tissue se n sitivity sh o win g th e
m o de ra te ly h igh re fle ctivity (a rro w ). Bo tto m righ t: De cre a se th e ga in fo r m e a sure m e n t o f le sio n
th ickn e ss (a rrow ).
Suggested Reading
Blumenkranz MS, Byrne SF. Standardized echography (ultrasonography) for the detection and characterization of
retinal detachment. O phthalm ology. 1982;89 (7):821-831.
Byrne SF, Green RL. Ultrasound of the Eye and O rbit. 2nd ed. St. Louis, MO : Mosby; 2002.
DiBernardo CW, Greenberg, EF. O phthalm ic Ultrasound: A Diagnostic Atlas. 2nd ed. New York, NY: Thieme;
2006.
44 QUESTION
AFTER A CLOSED GLOBE INJURY,

WHAT VITREORETINAL FINDINGS SHOULD
I EXPECT AND WHEN DO I REFER?
Esther M. Bowie, MD
Accord ing to the Ocu lar Trau m a Classification System, an op en globe inju ry is defined
as a fu ll-th ickness defect of the cornea and/ or sclera, and open globes are d ivided into
ruptu res or lacerations depend ing on the mechan ism of inju ry (ruptu res are cau sed by
blu nt objects, wh ile laceration s are cau sed by sharp ones). Laceration s are fu rther su b-
d ivided into penetrating inju ry, intraocu lar foreign body inju ry, and perforating inju ry.1
A closed globe inju ry is defined as an intact or partial th ickness wou nd of the cornea
and/ or sclera. After a closed globe inju ry, there are several vitreoretinal find ings that
m ay be present:
• Commotio retinae: Com motio retinae is a com mon find ing that m an ifests as gray
or wh ite outer retinal opacification w ith or w ithout scattered intraretinal hemor-
rhages (Figu re 44-1). Visu al acu ity can be reduced to 20/ 200. Patients need weekly
follow-up w ith retinal detach m ent (RD) warn ings. Retinal transparency and visu-
al acu ity u sually norm alize spontaneou sly w ith in 4 d ays to 4 weeks. Incomplete
visu al recovery is u su ally due to second ary retinal pigm ent epitheliu m (RPE)
d am age or m acu lar hole development. For m acu lar hole form ation, con sider refer-
ral for su rgical repair.
• Cotton-wool spots: Cotton-wool spots app ear as focal areas of nerve fiber layer
wh iten ing less than 0.25 d isc d iameter in size. They are cau sed by obstruction of a
retinal arteriole. Cotton-wool spots u su ally resolve w ithout sign ificant sequelae.
• Choroidal rupture: A choroid al ruptu re is u su ally seen as a yellow-wh ite cu rvilinear,
crescent-shap ed lesion concentric to the optic nerve head and deep to the retina. It
m ay be obscu red by su bretinal hemorrhage. Choroid al ruptu re m ay occu r in the
179
180 Question 44
Fig u re 4 4 -1. Co m m o tio

re tin a e : re tin a wh ite n in g
a n d in tra re tin a l h e m o r-
rh a ge s fo llo win g clo se d
glo be in jury.
fovea w ith an im m ed iate decrease in vision. If the ruptu re is not d irectly u nder
the fovea, vision m ay not be affected, or spontaneou s visual recovery m ay occu r
after reabsorption of the hemorrhage. Patients w ith good visu al acu ity need to be
followed regu larly to detect the development of choroid al neovascu larization. An
Am sler grid for home testing is helpfu l to detect early symptom s associated w ith
choroid al neovascu larization.
• Retinitis sclopetaria: Sclopetaria classically occu rs after nonp enetrating, h igh-veloc-
ity inju ry. Typically, an object strikes the globe tangentially and lodges in the orbit.
Acutely, there is an area of absent retina and choroid w ith su rrou nd ing retinal
opacification, sub- and intraretinal or vitreou s hemorrhage, and exposed sclera.
Patients shou ld be followed closely wh ile the hemorrhage reabsorbs becau se retinal
breaks and/ or detach m ent m ay develop.
• Vitreous hemorrhage: Dam age to vessels of the ciliary body, retina, or choroid m ay
resu lt in a visu ally sign ificant vitreou s hemorrhage. Most sm all hemorrhages
resolve w ithout treatment. A meticu lou s exam ination of the entire retina to ru le-
out a retinal break or detach ment is essential. When dense vitreou s hemorrhage is
present, u ltrasonography is helpfu l to detect a retinal break/ detach ment, p osterior
vitreou s detach ment (PVD), or choroid al detach ment, wh ich m ay otherw ise be
obscu red. The absence of a retinal break on u ltrasonography does not ru le out
its presence. If no retinal break is detected, then observation w ith serial fu ndu s
and echograph ic exam is essential. Bed rest w ith the patient’s head elevated at 30
degrees m ay facilitate clearing of vitreou s blood from the visu al axis.
If a retinal break and/ or detach m ent is detected w ith exam ination or echography, or
if a trau m atic vitreou s hemorrhage does not clear rapid ly, referral for evaluation by a
retinal specialist is recom m ended. When a retinal break is present, laser retinopexy or
cryotherapy is perform ed to prevent RD. Su rgery is ind icated for patients w ith a RD and
a non-clearing vitreou s hemorrhage.
Preretinal hemorrhage in the su bhyaloid or su binternal lim iting mem brane space is
seen as a boat-shap ed collection of blood from gravitational pooling.
Su bretinal/ su bpigment epithelial hemorrhages are often due to trau m atic choroid al
ruptu re. Subretinal pigment epithelial hemorrhage is more brow n than red. A break in
Bruch’s m em brane can later resu lt in choroid al neovascu larization.
After a Closed Globe Injury, What Vitreoretinal Findings Should I Expect? 181
Fig u re 4 4 -2 . Re tin a l dia lysis: 32

ye a r o ld with re tina l dia lysis a n d
de ta ch m e n t fo llo win g tra um a .
Retinal Break and Detachment

A closed globe inju ry has the greatest impact in the in ferotemp oral qu ad rant, where
the orbital rim least protects the eye. Symptom s of retinal breaks/ detach ments include
decreased visual acu ity, photop sia, floaters, and visu al field defects. Exam ination m ay
show a decrease in intraocu lar pressu re, “tobacco du st,” or hemorrhage in the vitreou s.
Patients often need su rgical intervention.
Trau m atic retinal breaks m ay be single or mu ltiple and u sually occu r at the posterior
edge of the vitreou s base. A retinal d ialysis (Figu re 44-2) is com mon after trau m a and
m ay be accompan ied by vitreou s base avu lsion. The presence of vitreou s base avu lsion is
pathognomon ic for a closed globe inju ry.
RD occu rs in 35% of closed globe inju ries, mostly in you ng m ales u nder age 40. Retinal
breaks occu rring at the time of inju ry do not necessarily lead to detach m ent. Twelve p er-
cent to 30% of trau m atic RDs are d iagnosed at presentation, 30% to 40% at 1 month, and
80% at 2 month s. RDs shou ld be referred for repair.2
Traumatic Optic Neuropathy

Trau m atic optic neu ropathy m ay resu lt from trau m a to the globe and/ or orbital rim .
Patients present w ith sign ificantly decreased visual acu ity and a relative afferent pupil-
lary defect. Optic d isc pallor develops approxim ately 6 weeks later. Intravenou s steroid
u se du ring the acute phase of the inju ry is controversial. If trau m atic optic neu ropathy
is associated w ith an expand ing retrobu lbar hemorrhage, then orbital decompression is
necessary.
182 Question 44
Optic Nerve Avulsion

Optic nerve avu lsion is a rare inju ry after severe trau m a to the globe. Patients present
w ith sign ificant visu al loss and an afferent pupillary defect. Ophthalmoscopy m ay show
depression of the optic d isc; abnorm al or abruptly end ing retinal vessels; and/ or hemor-
rhage in the vitreou s, su bhyaloid, intraretinal, or su bretinal space. Computed tomography
can be u sefu l in d iagnosing optic nerve avu lsion.

Berd ine M. Bu rger, MD
Ophthalmology Resid ent, Storm Eye In stitute
Med ical Un iversity of South Carolina, Charleston, SC
References
1. Kuhn F, Morris R, Witherspoon, CD, Heimann K, Jeffers JB, Treister G. A standardized classification of ocular
trauma. O phthalm ology. 1996;103(2):240 -243.
2. Regillo CD, Brown GC, Flynn HW Jr. Vitreoretinal Disease: The Essentials. New York, NY: Thieme; 1999.
45 QUESTION
WHAT SHOULD I TELL PATIENTS

WITH RETINITIS PIGMENTOSA ABOUT
THEIR PROGNOSIS AND SHOULD I
PRESCRIBE VITAMIN SUPPLEMENTS?
Raymond Iezzi, MD, MS
Retin itis pigmentosa (RP) is a heterogeneou s group of genetic d iseases that is associ-
ated w ith a sim ilar clin ical phenotyp e. Cu rrently, over 45 genetic loci and their associ-
ated genes have been identified in non-synd rom ic RP.1 Although genotyp e-phenotyp e
correlations are occasionally observed in RP, from the genetic stud ies it is clear that one
genotype does not necessarily equate to one phenotyp e. In add ition, w ith in one fam ily,
the phenotypic presentation of a genotype is a product of an ind ividu al’s genom e and
environ mental factors. Mu ltiple m em bers of a fam ily m ay carry the sam e genotype but
their phenotyp e and d isease cou rse m ay vary con siderably.
Given the genetic heterogeneity and ind ividu al variation of RP, it can be very d if-
ficu lt or impossible to accu rately pred ict prognosis. Cu rrently, genotyping of RP can
give in sight into prognosis, as som e mutations carry a particu larly worse prognosis. For
example, patients w ith X-lin ked RP (5% to 15% of cases) are more likely to develop the
most sign ificant vision loss. However, I tell patients that although RP is a progressive
d isease, the chances they w ill retain some fu nctional vision for life is good.
Psychophysical and objective testing m ay give lim ited in form ation about prognosis.
Patients who exh ibit an extingu ished scotopic electroretinogram (ERG) w ith m in im al to
moderate loss of the photopic ERG are at greatest risk of developing visu al acu ity loss,
as it is well established that loss of h igh-resolution cone vision is preceded by loss of rod
fu nction. Cone ERG amplitudes (measu red at a 30 H z flicker rate) typically demonstrate
an exponential decline over the cou rse of a patient’s d isease. Thu s, complete loss of cone
183
184 Question 45
ERG amplitude can be pred icted accord ing to a patient’s baseline 30-H z average, filtered
cone flicker ERG amplitude, u sing the m athem atical formu la for exponential decay
derived by Berson.2
From my experience, I feel that one helpfu l pred ictor of visual prognosis, w ithout
h igh ly specialized ERG testing, is the natu ral h istory of affected fam ily mem bers. With
the caveat that there is phenotypic heterogeneity even w ith in a single fam ily, exam ina-
tion of fam ily mem bers and their clin ical cou rse can be helpfu l in gain ing in sight into a
patient’s prognosis. It is also important to note that some patients can retain good central
acu ity even in the presence of an extingu ished ERG.
Faced w ith the d iagnosis of RP and the prognosis of a slow decline in visual fu nction,
most patients w ith RP take an active role in their d isease. It is important for you to also
take an active role and provide patients w ith qu ality resou rces and in form ation about
treatments. The Fou nd ation for Fighting Blind ness (w w w.blind ness.org) and the Research
to Prevent Blind ness (w w w.rpbu sa.org) Web resou rces for patients are excellent. Optim ize
you r patient’s visu al fu nction th rough appropriate referrals to a low-vision specialist
and screen for treatable complication s of RP such as m acu lar edem a and cataracts. Many
patients are interested in seeking any treatment that m ay help preserve vision. In a recent
su rvey, 96% of RP patients reported the u se of complementary and alternative m ed icine,
w ith 75% u sing nutritional or vitam in therapy.3
Ophthalmologists are caught between their RP patients who are seeking and u ndergo-
ing treatments w ithout rigorou s safety and efficacy d ata and an evidence-based m ed i-
cal com mu n ity waiting for safety and efficacy d ata before m aking recom m end ation s.
It seem s that the evidence-based com mu n ity is outpaced by patient dem and s. When
d iscu ssing recom mend ation s of nutritional supplem ents or vitam ins, en su re that first
you do no harm . Second, recogn ize that ou r knowledge about the efficacy and safety of
nutritional supplements and vitam ins is lim ited.
In a random ized trial, RP patients taking 15,000 IU of vitam in A (retinyl palm itate)
were show n to have a slower decline in cone ERG fu nction compared w ith controls.4 The
study is controversial becau se psychophysical resu lts d id not d iffer between study and
control su bjects. Although there were no seriou s adverse events in th is trial, the safety
profile for patients taking vitam in A, p otentially for decades, is u n know n. Vitam in A
is contraind icated in pregnant wom en becau se it is a know n teratogen and in smokers
becau se of a know n increased risk of lu ng cancer. Vitam in A has also been associated w ith
hepatotoxicity and osteopen ia. I d iscu ss the risks and benefits of vitam in A supplementa-
tion w ith patients and support its u se for those interested. For patients who elect to take
h igh-dose vitam in A, their liver fu nction and bone density shou ld be mon itored.
Fish oil, sp ecifically docosahexaenoic acid (DH A), has theoretical benefits in RP and
has show n prom ise in early trials. The m ain risk in taking nutritional supplements is
the poor regu lation of the indu stry. There are seriou s concerns about low-quality fish oil
contain ing polych lorinated biphenyls (PCBs), d ioxin s, and mercu ry; however, modern
molecu lar d istillation method s of fish oil pu rification remove these contam inants. An
excellent su rvey of the m anu factu ring stand ard s and pu rification quality of com mercially
available fish oils is available on line from the Environ mental Defen se Fu nd.5 I recom-
m end the u se of 2 to 3 g/ d ay of h igh-qu ality fish oil for all RP patients, as omega-3 fatty
acid s are converted by lipoxygenase en zymes into resolvin s. Resolvin s are potent endoge-
nou s anti-in flam m atory molecu les produced by the body to induce the resolution phase of
What Should I Tell Patients With Retinitis Pigmentosa About Their Prognosis? 185
in flam m ation. Given that the clin ical and h istological appearance of RP in both hu m an s
and an im al models demonstrates a sign ificant neu roin flam m atory component, RP is not
a m isnomer. Consequently, it seem s reasonable that omega-3 fatty acid s m ay have a role
in the treatm ent of RP. I tell patients that fish oil can interfere w ith some med ication s and
can prolong bleed ing tim es. At h igher doses, it has also been associated w ith the risk of
hemorrhagic stroke.
Other nutritional supplem ents that m ay be of benefit include lutein and zeaxanth in.
These supplements have demonstrated no toxicity or side effect concerns in both hu m an
and an im al stud ies. Little is know n about the effects of lutein and zeaxanth in in RP;
however, I do not d iscou rage interested patients from taking them due to their favorable
system ic safety profile.
Fu nd amentally, antioxid ants are anti-in flam m atory. We have demonstrated in the
laboratory that the Age-Related Eye Disease Study (AREDS) formu lation, for example,
was capable of suppressing retinal neu roinflam m ation in a rat model of RP. The AREDS
formu la has not been clin ically tested in patients w ith RP, however. It is also important to
remem ber that some antioxid ants can becom e pro-oxid ants after m etabolic degrad ation.
For example, the u se of vitam in E in RP patients accelerates ERG decline.4 Consequently,
since there are no prospective clin ical trials of the AREDS formu lation in patients w ith
RP, I do not recom mend th is for my patients due to its vitam in E content. It is important to
consider the fact that nutritional supplements m ay exh ibit some u ntoward effects depend-
ing upon the d isease. (For example, in a mou se model of Stargardt’s d isease, vitam in A has
been show n to increase d ru sen accu mu lation since the d isease is associated w ith the toxic
bu ildup of A2E, a metabolic byproduct of vitam in A m etabolism. Consequently, I d iscou r-
age my Stargardt’s patients from taking vitam in supplements that contain vitam in A.)
RP is a progressive d isease, but patients shou ld be reassu red that except for a sm all
p ercentage, most patients w ill retain som e vision for life. You shou ld d iscu ss the risks and
benefits of treatment w ith nutritional and vitam in therapy, keeping in m ind the lack of
h igh-qu ality efficacy d ata. I recom mend supporting the u se of vitam in A in appropriate
patients and recom m end u sing fish oil, lutein, and zeaxanth in in my RP patients who are
h igh ly motivated to take nutritional supplements.

Jam in S. Brow n, MD
Vitreoretinal Fellow, Kresge Eye In stititute, Departm ent of Ophthalmology
Wayne State Un iversity School of Med icine, Detroit, MI
References
1. Retinal Information Network. Available at: http://www.sph.uth.tmc.edu /Retnet/. Accessed June 15, 2009.
2. Berson EL. Long-term visual prognoses in patients with retinitis pigmentosa: the Ludwig von Sallmann lecture.
Exp Eye Res. 2007;85(1):7-14.
3. Kiser AK, Dagnelie G. Reported effects of non-traditional treatments and complementary and alternative
medicine by retinitis pigmentosa patients. Clin Exp O ptom. 2008;91(2):166-176.
4. Berson EL, Rosner B, Sandberg MA, et al. A randomized trial of vitamin A and vitamin E supplementation for
retinitis pigmentosa. Arch O phthalm ol. 1993;111(6):761-772.
5. Environmental Defense Fund. Fish oil supplements: is the brand you’re taking safe? Available at: http://www.
edf.org/page.cfm?tagID =16536. Accessed June 15, 2009.
46 QUESTION
WHICH PATIENTS SHOULD I REFER

FOR ELECTROPHYSIOLOGY TESTING AND
WHAT MIGHT THE TEST SHOW?
Amani A. Fawzi, MD
There are a plethora of electrophysiologic tests for the retina. I w ill review them by
category and h igh light wh ich cond ition s are best m atched w ith each exam ination.
Electro-oculography
Generally sp eaking, electro-ocu lography (EOG) evaluates the retina pigm ent epithe-
liu m (RPE) w ith input from the photoreceptors. I find the EOG u sefu l in the setting of
vitelliform m acu lar lesions to establish the d iagnosis of Best’s vitelliform dystrophy (nor-
m al electroretinogram [ERG], along w ith a severely reduced EOG) and to d ifferentiate it
from the pattern dystroph ies, especially from adu lt vitelliform dystrophy (m ild to moder-
ate EOG reduction). I always obtain an EOG in conju nction w ith an ERG in cases of su s-
p ected generalized photoreceptor dysfu nction as well as su spected toxic retinopath ies.
Flash Visual Evoked Potential

A bright flash visual evoked potential (VEP) signal is d riven by the m acu la and optic
nerve. It is an evoked potential m easu red over the occipital cortex that does not requ ire
patient coop eration. The VEP is a global qu alitative resp onse that can not qu antify m ild to
moderate levels of visu al dysfu nction. It m ay be u sefu l in cases of total hysterical blind-
ness or in a ped iatric practice to screen nonverbal in fants su spected of total blind ness.
187
188 Question 46
When u sed in eyes w ith opaque m ed ia, it is imp ortant to note that, by itself, the flash VEP
can not pred ict visu al acu ity. Comparing the tracing to that of the norm al fellow eye m ay
shed some light on the extent of optic nerve/ m acu lar comprom ise.
Pattern Visual Evoked Potential

In an adu lt practice, th is test is u su ally obtained in patients su sp ected of nonorgan ic
visu al loss. Un like the flash VEP, the pattern VEP relies heavily on patient cooperation
(actu ally “looking at the screen” du ring the test), attention to the screen pattern, and
corrected refraction. Professional m alingerers are able to defocu s, ignore the screen, and
“blin k excessively” to yield an abnorm al test resu lt. Carefu l observation du ring the test
m ay improve patient compliance. That being said, a well-formed pattern-reversal VEP is
suggested to be incompatible w ith acu ity of ~20/ 150 or worse.1 Variou s tech n iques for elic-
iting the VEP have been u sed in in fants to assess visual acu ity, fu sion, and stereop sis.2,3
Full-Field Electroretinography
As retina specialists, fu ll-field electroretinography (ERG) is the test w ith wh ich we are
most fam iliar. It is a m ass response that assesses the fu ll host of rod and cone photorecep -
tors as well as the post-tran sduction pathways in the in ner retina. I order a fu ll-field ERG
when I su sp ect or want to ru le out generalized retinal dysfu nction. When I have a patient
w ith a m acu lar lesion, I order a fu ll-field ERG to d ifferentiate m acu lar or pattern dystro-
phy (norm al to m ild ly abnorm al cone ERG) from a cone or cone-rod dystrophy (severely
abnorm al to flat cone ERG). In patients w ith retin itis pigmentosa (RP), the ERG can d if-
ferentiate a cone-rod pattern (when cone ERG is equal to or more severely affected rod
ERG) from a rod-cone pattern (rod s affected more severely than cones). The rod ERG is
u sually severely attenu ated at an early age, w ith the exception of som e form s of autosom al
dom inant RP, where adu lts m ay retain som e rod fu nction. Th is find ing helps d ifferentiate
RP from pseudo-RP and ru bella retinopathy, where the retinal find ings m ay be severe in
the face of a norm al to m ild ly affected ERG. The role of ERG testing in patients w ith RP is
important in m any ways: to gu ide genetic screen ing, to narrow the d ifferential d iagnosis,
and to offer som e prognostic gu id ance.
Multifocal Electroretinography
Th is is a new and continu ally evolving tech nology that is capable of assessing focal
electrophysiologic responses in the posterior retina. The record ing involves specialized
stimu li and m athem atical modeling of the resp onse, lead ing to a top ograph ical m ap of
tracings across the posterior fu ndu s. More recently, the mu ltifocal ERG has fou nd clin ical
u se in evalu ating plaquen il retinopathy. It is also u sefu l as a research tool in a variety of
retinal dystroph ies, glaucom a, and d iabetic retinopathy.
Which Patients Should I Refer for Electrophysiology Testing? 189
Specific Disease Entities Where

Electrophysiology May Be Diagnostic
ELECTRONEGATIVE ELECTRORETINOGRAPHY
Th is waveform is typically described u nder test cond ition s of m aximu m d ark-ad apted
bright flash stimu lation and con sists of a large a-wave and a severely depressed b-wave
that m ay on ly reach the baseline. The d ifferential d iagnosis includes congen ital stationary
n ight blind ness (both autosom al recessive and X-lin ked), juven ile X-lin ked retinosch isis,
ischem ic retinal cond ition s, paraneoplastic melanom a (melanom a-associated retinopathy
[MAR]), qu in ine retinal toxicity, occasionally in bird shot chorioretinopathy, and in some
early/ transitional form s of RP (eg, cone-rod dystroph ies in the setting of GUCY2D and
CRX R41W mutations).
CANCER-ASSOCIATED RETINOPATHY AND MELANOMA-ASSOCIATED

RETINOPATHY
These cond ition s are both characterized by m in im al pigmentary changes to norm al
fu ndu s find ings in the face of severe ERG dysfu nction. Cancer-associated retinopathy
(CAR) is associated w ith severe generalized ERG attenu ation, wh ile MAR is associated
w ith the electronegative waveform.
CONGENITAL STATIONARY NIGHT BLINDNESS (X-LINKED OR

AUTOSOMAL RECESSIVE)
A com bination of an electronegative ERG in the m aximu m d ark-ad apted respon se,
along w ith a non record able rod ERG, m ild ly abnorm al to norm al cone response, fu ll
visu al field, and norm al fu ndu s (or myopic fu ndu s) wou ld clinch the d iagnosis.
ENHANCED S-CONE SYNDROME (GOLDMANN-FAVRE SYNDROME,

CLUMPED PIGMENTARY DEGENERATION)
When th is entity is su sp ected based on the clin ical app earance (clu mp ed or nu m mu lar
pigmentary changes, esp ecially along the arcades, in association w ith m acu lar sch isis in
a patient w ith long-stand ing n ight blind ness), then the clin ician m ay consider ordering
sp ecialized spectral ERG (blue on am ber). If th is special test is not available, a fu ll-field
ERG shou ld be obtained. When the rod ERG is barely record able or non record able and
the cone respon se is very sim ilar to the m axim al d ark-ad apted resp onse, then th is entity
shou ld be con sidered. Once the d iagnosis is m ade, genetic screen ing for N R2E3 mutations
m ay u ltim ately clinch the d iagnosis.
190 Question 46
Miscellaneous
BIRDSHOT CHORIORETINOPATHY
Some physicians u se variations in the ERG amplitudes (30-H z flicker and scotopic
respon ses) as an objective tool to mon itor d isease progression and the response to im mu-
nosuppressant therapy.
PLAQUENIL AND O THER DRUG TOXICITIES

The role of the fu ll-field ERG and the EOG is controversial in the detection of early
changes of hyd roxych loroqu ine retinopathy. Clin ical exam ination, a red Am sler grid,
and red target visu al field testing have been advocated as in itial screen ing tests. Recently,
fu ndu s autofluorescence and mu ltifocal ERG testing have emerged as u sefu l adju ncts.
References
1. Heckenlively JR, Arden GB, eds. Principles and Practice of Clinical Electrophysiology of Vision. 2nd ed. Boston,
MA: The MIT Press; 2006.
2. Halliday AM, McDonald WI. Visual evoked potentials. In: Stalberg E, Young RR, eds. N eurology I: Clinical
N europhysiology. London: Butterworths; 1981.
3. Birch E, Petrig B. FPL and VEP measures of fusion, stereopsis and stereoacuity in normal infants. Vision Res.
1996;36 (9):1321-1327.
47 QUESTION
HOW DO I DIFFERENTIATE ALL OF

THOSE WHITE DOT SYNDROMES?
Sunil K. Srivastava, MD
The wh ite dot synd rom es represent a collection of in flam m atory eye d iseases char-
acterized by sp ots in the retina/ choroid. There are several d iseases that are lu mp ed in
th is category of ocu lar in flam m ation and m any have abbreviation s that can m ake the
d iagnosis con fu sing. In order to treat these patients, it is important not to get lost in the
abbreviations, but in stead to system ically evalu ate the patient.
The Big Picture

The wh ite dot synd romes are like all in flam m atory d iseases in the eye in that most of
them w ill requ ire some anti-in flam m atory regimen, but som e are in fectiou s in origin. It is
important to not m iss potential in fectiou s cau ses that can worsen w ith im mu nosuppres-
sive therapy. It is equ ally important to ensu re that prompt im mu nosuppressive therapy
is in stituted for those non in fectiou s in flam m atory d iseases that can lead to sign ificant
vision loss.
The History
After realizing that a patient has a wh ite dot synd rom e, I step back and ensu re that I
have reviewed h is or her h istory and review of system s thorough ly. A complete review of
system s is needed becau se positive find ings in the review can help w ith the d ifferential
d iagnosis. Those w ith a h istory of a recent viral illness fit w ith d iseases such as mu ltiple
evanescent wh ite dot synd rome (MEWDS) and acute posterior mu ltifocal placoid pig-
m entary epitheliopathy (APMPPE).
191
192 Question 47
I also want to clarify the patient’s symptom s and ask about photopsias, nyctalopia,
dysch rom atopsia, floaters, and metamorphopsia. These symptom s are more com mon in
some synd romes (ie, nyctalopia and dysch rom atopsia in bird shot choroidopathy). Is the
patient myopic? Some wh ite dot synd rom es occu r more often in myopic patients (pu nctate
in ner choroidopathy [PIC]).
The Examination
A complete eye exam ination is obviou sly needed. Many of the wh ite dot synd rom es
have m in im al in flam m ation, thu s the presence of in flam m atory cells and/ or signs of
previou s ocu lar in flam m ation can help lim it the d ifferential d iagnosis.
When exam in ing the retina, I look at the size, color, and characteristics of the wh ite
dots. Are they sm all (less than the caliber of the central retinal artery), large (bigger than
the optic nerve), or in between? Is there one lesion or are there mu ltiple? Where are they
located? Are the borders sharp? Are the dots atroph ic, wh ite, or yellow? Any associ-
ated hemorrhage, pigmentary change, vascu lar sheath ing? Finally, are there any changes
arou nd the nerve (ie, peripapillary atrophy, tilted d isc)?
I obtain im aging for most patients, u sually a fluorescein angiogram and sometimes an
indocyan ine green angiogram. Both allow me to determ ine areas of activity. Active areas
w ill u su ally hyp erfluoresce late and have ind istinct borders, wh ile inactive areas u su ally
hypofluoresce or stain and have sharp borders. I sometim es also w ill u se optical coher-
ence tomography to con firm the presence of cystoid m acu lar edem a or su bretinal flu id.
Developing a Differential
It is now time to formu late a d ifferential and a treatment plan. The specifics of each
particu lar wh ite dot synd rome are beyond the scope of th is chapter, but I w ill provide an
algorith m to allow one to develop a d ifferential.
As mentioned above, there can be a nu m ber of in fectiou s cau ses that can present as a
“wh ite dot” synd rome. Syph ilis can m im ic m any wh ite dot synd romes (Figu re 47-1) and
thu s shou ld be tested for w ith a seru m rapid plasm a reagin (RPR) and fluorescent trep o-
nem al antibody absorption (FTA-Abs). Other in fectiou s cau ses that shou ld be con sidered
are tu bercu losis and toxoplasmosis. The latter typically has a single wh ite choroid al/ reti-
nal lesion w ith sign ificant in flam m ation. Tu bercu losis typically presents w ith mu ltiple
deep choroid al lesions w ith sign ificant in flam m ation. A PPD and toxoplasmosis titers are
not needed in each patient but shou ld be ordered in su spiciou s cases.
The next step is to d ifferentiate the size and nu m ber of spots/ dots and determ ine the
amou nt of in flam m ation. If the spots are sm all to m ed iu m (50 to 125 µm), my d ifferential
includes MEWDS, PIC, acute m acu lar neu roretinopathy, mu ltifocal choroid itis (MFC),
ocu lar h istoplasmosis synd rome (OH S), sarcoidosis, and in fectiou s cau ses. If there is
m in im al in flam m ation, MEWDS, PIC, acute m acu lar neu roretinopathy, and OH S are
more likely. With increased in flam m ation, MFC, sarcoidosis, and in fectiou s cau ses are
more likely.
How Do I Differentiate All of Those White Dot Syndromes? 193
A B
Fig u re 47-1. Th is figure illustra te s a pa tie n t w ith syph ilis (A) wh e re th e pre se n ta tio n o f a de e p
re tin a l ye llo wish pla que ca n lo o k sim ila r to a ca se o f a cute po ste rio r m ultifo ca l pla co id pigm e n -
ta ry e pith e lio pa thy (APMPPE) (B).
If the patient has larger spots (greater than 125 µm) or plaques, my d ifferential includes
APMPPE, serpiginou s choroidopathy, sarcoidosis, bird shot choroidopathy, and in fectiou s
cau ses. If there is m in im al in flam m ation, APMPPE and serpiginou s choroidopathy are
more likely. With sign ificant in flam m ation, sarcoidosis, bird shot choroidopathy, and
in fectiou s etiologies are more likely.
So how do you tell the d ifference between all of these d iseases even after lim iting you r
d ifferential? The an swer is that you can not always do so, but there are som e sp ecific pat-
terns that seem to fit some of these d iseases.
For example, MEWDS (Figu re 47-2) tend s to be u n ilateral, w ith sm all pu nctate dots
in the m acu lar area and a characteristic pattern on angiography. Th is occu rs in you ng
wom en preceded by a viral illness and resolves on its ow n. PIC (Figu re 47-3) also tend s to
occu r in women, typically myop es w ith no sign s of in flam m ation. The spots are u su ally
more atroph ic and can be pigm ented. MFC can be sim ilar in appearance to PIC, but has
in flam m ation and can have subretinal fibrosis in the later stages.
APMPPE (see Figu re 47-1B) u su ally presents w ith a h istory of a preced ing viral illness
w ith m in im al in flam m ation. The spots are yellow, large, and plaque-like. On angiography,
the lesions block early and hyp erfluoresce late. Serpiginou s choroidopathy (Figu re 47-4)
can look sim ilar but typically has a continuou s large lesion that is helical w ith a yellow-
ish lead ing edge. On angiography, the lesion s typically hypofluoresce early and hyper-
fluoresce late. Bird shot choroidoretinopathy (Figu re 47-5) has larger atroph ic-appearing
spots, more prom inent nasally. Th is d isease is u su ally bilateral but can be asym metric
and has sign ificant in flam m ation and cystoid m acu lar edem a w ith vascu lar leakage on
angiography.
194 Question 47
Fig u re 47-2 . Fun dus ph o to gra ph o f a

pa tie n t with MEWDS. No te th e sm a ll
wh ite do ts in th e m a cula r re gio n .
Fig u re 47-3 . Th is fun dus ph o to gra ph illus-

tra te s a ca se o f PIC. No te th e ye llo w ish
a tro ph ic a ppe a ra n ce o f th e spo ts a n d th e
a re a s o f pig m e n ta tio n .
Fig u re 47-4 . Th is pa tie n t h a s se rpigi-

n o us ch o ro ido pa th y. Th e a ctive e dge s
o f th is le sio n h a ve a ye llo wish a ppe a r-
a n ce , wh ile th e in a ctive a re a s a re m o re
pigm e n te d.
How Do I Differentiate All of Those White Dot Syndromes? 195
A B
Fig u re 47-5 . Fun dus ph o to gra ph (A) o f a pa tie n t w ith birdsh o t ch o ro ido re tin o pa thy. No te th e
sub tle ye llo w ish subre tin a l a re a s. La te ph a se fra m e o f a fluo re sce in a n gio gra m (B) illustra te s
sign ifica n t va scula r a n d m a cula r le a ka ge .
The Next Steps

Usually, I order an RPR/ FTA-Abs, a chest x-ray, angiotensin converting en zym e, and
often a PPD. I w ill add other tests depend ing on my review of system s and exam find-
ings.
I base my decision to treat on my clin ical su spicion. Som e wh ite dot synd romes are
self-lim ited (MEWDS and APMPPE). I continue to follow these patients closely to mon itor
improvement and begin oral pred n isone if they do not improve.
Other d iseases such as PIC, MFC, bird shot, and serpiginou s choroidopathy requ ire
im mu nosuppression. I begin oral pred n isone and dose it based on d isease severity.
Bird shot and serpiginou s choroidopathy u sually requ ire long-term im mu nosuppression,
so I begin a steroid-sparing agent soon after starting pred n isone. Both PIC and MFC can
also requ ire steroid-sparing agents.
As these d iseases can look sim ilar, I mon itor the patient’s resp onse closely. If he or she
does not respond, I take a step back and re-exam ine my d iagnosis and con sider others. I
keep in my m ind in fectiou s m asquerades and in some older patients, I also con sider p os-
sible neoplastic d iseases.
Summary
Although the wh ite dot d iseases can be con fu sing, a system ic approach to the evalu-
ation of the patient lead s to a lim ited d ifferential and p ossible d iagnosis. It is important
to consider in fectiou s etiologies and to treat patients w ith im mu nosuppression when
needed. Remem ber to mon itor the patient carefu lly, collaborate w ith their intern ist, and
re-assess the d iagnosis if the patient does not respond as exp ected.
196 Question 47
References
1. Nussenblatt RB, Whitcup SM. Uveitis: Fundam entals and Clinical Practice. 3rd ed. Philadelphia, PA: Mosby;
2003.
2. Foster CS, Vitale AT, Schmitt BD, et al. Diagnosis and Treatm ent of Uveitis. Philadelphia, PA: Saunders;
2002.
3. Jabs DA, Rosenbaum JT, Foster CS, et al. Guidelines for the use of immunosuppressive drugs in patients with
ocular inflammatory disorders: recommendations of an expert panel. Am J O phthalm ol. 2000 ;130 (4):492-
513.
4. Kedhar SR, Thorne JE, Wittenberg S, Dunn JP, Jabs DA. Multifocal choroiditis with panuveitis and punctate
inner choroidopathy: comparison of clinical characteristics at presentation. Retina. 2007;27(9):1174-1179.
5. Thorne JE, Jabs DA, Peters GB, Hair D, Dunn JP, Kempen JH. Birdshot retinochoroidopathy: ocular complica-
tions and visual impairment. Am J O phthalm ol. 2005;140 (1):45-51.
48 QUESTION
HOW DO I WORK-UP A PATIENT WITH

INTERMEDIATE UVEITIS OR PANUVEITIS?
Thomas A. Albini, MD
When approach ing a patient w ith interm ed iate uveitis or panuveitis, it is imp ortant to
keep a healthy su spicion for all possible etiologies and obtain appropriate ancillary test-
ing to work th rough the d ifferential. Keep in m ind that m any autoim mu ne synd romes
m an ifest them selves more fu lly over time and rep eat exam ination s on a tim ely basis can
finally clinch the d iagnosis long after the in itial presentation. Begin w ith a h istory includ-
ing presenting symptom s; recent travel; exposu re to in fectiou s d iseases and their vectors
such as kittens, puppies, ticks, etc; risk factors for hu m an im mu nodeficiency viru s (H IV)
and/ or syph ilis in fection; rheu m atologic, neu rologic, and other med ical cond ition s; and a
fu ll review of symptom s, includ ing exam ination of skin rashes and joints. Fam ily h istory
of autoim mu ne d isorders m ay reveal a pred isp osition to system ic autoim mu ne cond i-
tions. Review of eth n ic backgrou nd m ay reveal racial pred ilections for d iseases such as
Behçet’s d isease, mu ltiple sclerosis (MS), or Vogt-Koyanagi-H arad a (VKH ) synd rome.
Ophthalm ic exam ination is essential in determ in ing the location (anterior, posterior,
intermed iate, or panuveitis)1 and qu ality (granu lom atou s versu s non-granu lom atou s) of
intraocu lar in flam m ation. Follow ing th is in itial h istory and system ic and ophthalm ic
evalu ation s, a d ifferential d iagnosis is established and a d irected ancillary evalu ation is
in itiated. Frequently obtained evalu ation s include the follow ing:
• PPD (Mantou x test) to evalu ate for TB. Consider interferon-γ release assay in
patients w ith positive PPD or in im mu nocomprom ised patients.
• Blood pressu re: Mon itor if hypertensive retinopathy is present, if there is cu rrent
or plan ned cyclosporine u se, or if ocu lar ischem ia is su sp ected.
• Urinalysis to evalu ate for evidence of neph ropathy as in system ic lupu s erythem a-
tosu s (SLE) or tu bu lointerstitial neph ritis-associated uveitis.
• Chest x-ray to evalu ate for h ilar lymphadenopathy, cavitary granu lom atou s d is-
ease con sistent w ith tubercu losis, or pu lmonary m alignancy.
197
198 Question 48
Blood Sample
GENERAL
• Fu ll blood cou nt w ith d ifferential: Evaluate for leu kem ia, in fection (left sh ift),
extent of im mu nosuppression, and/ or eosinoph ilia.
• Glucose: Cu rrent or plan ned system ic steroid u se.
• Liver fu nction tests: Cu rrent or plan ned antimetabolite u se.
• Creatin ine/ BUN : Evalu ate renal fu nction prior to ad m in istration of med ication s
metabolized by the kid neys (eg, acyclovir, cyclosporine). Evaluate for neph ritis as
seen in SLE.
AUTOIMMUNE PHENOMENA
• Angioten sin-converting en zyme (ACE) is elevated in sarcoid or other processes
involving m acrophage activation; levels are typically h igher in ch ild ren and can be
elevated in mu ltiple autoim mu ne d iseases.2
• Antinuclear antibody (ANA) is associated w ith SLE.
• Antineutroph il cytoplasm ic antibod ies (ANCA): p -ANCA suggests m icroscopic
polyangiitis, wh ile c-ANCA suggests Wegener’s granu lom atosis. ELISA for anti-
myeloperoxid ase is more sp ecific than p -ANCA and shou ld be p erform ed in p -
ANCA p ositive sera, wh ile anti-proteinase 3 is con firm atory in c-ANCA positive
sera.3
• Dou ble-stranded deoxyribonucleic acid (d sDNA) is suggestive of SLE.
• Rheu m atoid factor (RF) is p ositive in rheu m atoid arth ritis, negative in sp ondylo-
arth ropath ies.
• Eryth rocyte sed imentation rate (ESR) is elevated in system ic in flam m atory pro-
cesses.
• C-reactive protein: Acute phase reactant also elevated in system ic in flam m atory
d isorders.
• Complement assays: Complem ent can be depleted in system ic vascu litis or elevat-
ed as an acute phase reactant.
• Hu m an leu kocyte antigen (H LA) B27 can be u sed to evaluate for RF-negative spon-
dyloarth ropath ies.
• H LA typing can be u sed to evalu ate for im mu nogenetic association s w ith su s-
pected cond itions (Table 48-1).4
INFECTIOUS SEROLOGY
• MH A-TP or FTA-Abs (antibod ies to Treponema pallidum [ie, syph ilis])
• RPR or VDRL (ind icators of active syph ilis)
• Toxoplasmosis IgG and IgM
How Do I Work-Up a Patient With Intermediate Uveitis or Panuveitis? 199
Table 48-1
Human Leukocyte Antigens in Uveitis

Disease/Syndrome HLA Uveitis
Acute anterior u veitis B27 Anterior, unilateral, sudden onset
Behçet ‘s disease B51, B52 Anterior, acute, posterior acute, or chronic
Birdshot chorioretinopathy A29 Chronic bilateral posterior u veitis
Idiopathic intermediate DR15 Chronic/recurrent, bilateral intermediate
u veitis/pars planitis u veitis
Ju venile rheumatoid (or DPB1*0202 Chronic bilateral anterior uveitis
chronic) arthritis
Vog t-Koyanag i-Harada DR1, DR4; Bilateral panu veitis of acute onset followed
syndrome DRB1*0405 in Asians by chronic anterior u veitis
Adapted from Levinson RD. Immunog enetics of ocular inflammatory disease. Tissue Antigens.
2007;69(2):105 -112.
• Toxocara ELISA
• H IV
• Hu m an T-lymphotroph ic viru s (H TLV)
• Bartonella henselae and qu intana
• Lyme d isease panel
• Anti-viral antibod ies to evaluate for Herp es simplex, Varicella zoster, Epstein-Barr
viru s, or Cytomegaloviru s exposu re.
Additional Imaging Studies

• CT scan of chest: Consider spiral CT or CT/ PET scan for h ighest sensitivity for
sarcoidosis. Evalu ate for old granu lom as, h ilar lymphadenopathy, and lu ng m alig-
nancy.
• MRI of brain: To evaluate for wh ite m atter plaques in MS, ring-en hancing lesion s
w ith su rrou nd ing edem a in central nervou s system (CNS) toxoplasmosis or CN S
lymphom a, or com mu n icating hyd rocephalu s, en larged ventricles and/ or paren-
chym al calcification s and granu lom as in Cryptococcus.
• Lu m bar pu nctu re: To evaluate for cerebral spinal flu id (CSF) pleocytosis or pig-
ment laden m acrophages in VKH, CSF VDRL in neu rosyph ilis, or cytology in
su spected CN S lymphom a.
200 Question 48
H EMATOLOGIC DISORDERS
• Plasm a viscosity and seru m and u rine protein electrophoresis: For evalu ation of
Walden strom’s m acroglobu linem ia and other hypergam m aglobu linem ias, wh ich
can m an ifest as hyp erviscosity retinopathy or serou s retinal detach m ents.
• Lupu s anticoagu lant and anticard iolipin antibod ies: Cau ses of autoim mu ne hyper-
coagu lable states.
• Factor V Leiden gene mutation, homocysteine, protein C and S deficiency, anti-
th rom bin III deficiency, and proth rom bin gene mutation if a hypercoagu lable state
is su sp ected.
ELECTROPHYSIOLOGY
• Fu ll-field ERG and/ or mu ltifocal ERG to evaluate for deficiency in retinal fu nction
in ch ron ic d iseases such as bird shot chorioretinopathy.
BIOPSY
• Renal, pu lmonary, or skin: As ind icated to better evalu ate for system ic processes.
• Conju nctival biop sy can be helpfu l in evalu ating for sarcoid if conju nctival nodu les
are present in su sp ected cases.
• Anterior cham ber tap : For viral p olymerase chain reaction (PCR); Toxoplasmosis
PCR; Toxocara ELISA.
• Diagnostic vitrectomy: Cytology, flow cytometry; gene rearrangem ent stud ies,
cytokine profile (interleu kin-10/ 6 ratio); cu ltu res and stains of aspirated vitreou s;
vitreou s PCR and ELISA.
• Chorioretinal biop sy m ay be u ndertaken in select cases where occu lt in fection or
lymphom a is su spected and extensive evalu ation has not rendered a d iagnosis.
References
1. Jabs DA, Nussenblatt RB, Rosenbaum JT, Standardization of Uveitis Nomenclature (SUN) Working Group.
Standardization of uveitis nomenclature for reporting clinical data. Results of the First International Workshop.
Am J O phthalm ol. 2005;140 (3):509-516.
2. McCallum RM, Bylund DJ. Vasculitis. In: McPherson RA, Pincus MR, eds. Henry’s Clinical Diagnosis and
Managem ent by Laboratory Methods. 21st ed. Philadelphia, PA: Saunders; 2007.
3. Abraham Jr NZ, Carty RP, DuFour DR, Pincus MR. Clinical enzymology. In: McPherson RA, Pincus MR, eds.
Henry’s Clinical Diagnosis and Managem ent by Laboratory Methods. 21st ed. Philadelphia, PA: Saunders;
2007.
4. Levinson RD. Immunogenetics of ocular inflammatory disease. Tissue Antigens. 2007;69 (2):105-112.
49 QUESTION
HOW DO I DIFFERENTIATE THE

INFECTIOUS POSTERIOR UVEITIDES AND
WHAT ARE THE CURRENT TREATMENTS?
JP Dunn, MD
Posterior uveitis m ay be in flam m atory, neoplastic, trau m atic, or in fectiou s in origin.

Not on ly does in fectiou s posterior uveitis requ ire specific therapy dep end ing on the
cau sative agent, but a delay in m aking the correct d iagnosis is more likely to affect
visu al outcom es adversely than a delay in treating in flam m atory or neoplastic uveitides.
Fu rthermore, inappropriate therapy, such as the u se of system ic or intravitreal cortico-
steroid s, m ay accelerate the cou rse of in fectiou s uveitis and render it less responsive to
antim icrobial therapy. Therefore, you shou ld always th in k in fection first when evalu ating
a patient w ith posterior uveitis. A h igh index of su spicion, the u se of pattern recogn ition,
and an awareness of the available d iagnostic tools are essential.
Let u s begin w ith a list of in fectiou s p osterior uveitides (Table 49-1). Overall, posterior
uveitis accou nts for about 15% of all uveitides; however, in fectiou s etiologies accou nt
for a h igher p ercentage of posterior uveitis than for anterior or interm ed iate uveitis.1
For simplification, I have included in th is d iscu ssion cond itions that are not tech n ically
uveitis (eg, retin itis, endophthalm itis) but that requ ire sim ilar d iagnostic and therap eutic
approaches and are, therefore, logically considered in the sam e context. A proper h is-
tory and review of system s w ill often su bstantially narrow the d ifferential d iagnosis.
Obviou sly, a previou s h istory of ocu lar in fection is sign ificant; toxoplasm ic retin itis, for
example, is the most com mon in fectiou s cau se of posterior uveitis and has a h igh recu r-
rence rate. A recent on set of sh ingles shou ld increase you r su spicion of acute retinal
necrosis. Ind igent and foreign-born patients m ay be at increased risk of ocu lar tu bercu-
losis. Im mu nosuppressed patients (eg, those w ith H IV/ AIDS or receiving chemotherapy)
are much more likely than otherw ise healthy patients to develop ocu lar opportu n istic
201
202 Question 49
Table 49-1
Causes of Infectious Uveitis

Viral
• Cytomegalovirus (CMV) retinitis
• Necrotizing herpetic retinopathy
• Acute retinal necrosis
• Progressive outer retinal necrosis
• West Nile virus retinopathy
• Rubella panuveitis
• Subacute sclerosing panencephalitis
Bacterial
• Lyme borreliosis
• Syphilitic uveitis
• Tuberculous choroiditis
• Cat-scratch neuroretinitis
• Metastatic bacterial endophthalmitis (many species)
Fungal
• Candida albicans
• Ocular histoplasmosis syndrome
• Coccidiomycosis
• Cryptococcal choroiditis
• Aspergillosis
Protozoal/Helminthic
• Toxoplasmic retinitis
• Toxocariasis
• Diffuse unilateral subacute neuroretinitis (DUSN)
• Pneumocystis carinii choroidopathy
in fection s, such as cytom egaloviru s (CMV) retin itis. Injection d rug u sers, hospitalized
patients w ith indwelling catheters, or outpatients who have u ndergone invasive d iagnos-
tic procedu res, such as colonoscopy or even teeth clean ing, m ay develop metastatic bacte-
rial or fu ngal endophthalm itis. Recent travel m ay prompt consideration of a particu lar
organ ism, such as cryptococcosis in som eone who has been to Californ ia’s Central Valley.
Conversely, geography helps to ru le out other d iseases. The patient w ith panuveitis who
has never been outside of the San Anton io, Texas area does not need to be worked up for
ocu lar Lyme borreliosis. Finally, consider the age of the patient. Some in fectiou s uveitides
occu r almost exclu sively in the you ng (eg, toxocariasis), wh ile others can occu r at any
age but m ay present d ifferently depend ing on the sou rce of tran sm ission (eg, congen ital
versu s acqu ired syph ilis).
Many types of uveitis, includ ing in fectiou s ones, are d iagnosed clin ically. There
are some helpfu l generalizations regard ing pattern recogn ition. In fectiou s uveitis is
more likely to be focal. Isolated retinal vascu litis is more com mon ly non in fectiou s
How Do I Differentiate the Infectious Posterior Uveitides? 203
(eg, sarcoidosis). Some in fectiou s uveitides are u sually u n ilateral (eg, acute retinal
necrosis), whereas others (eg, syph ilis, Lyme d isease) are bilateral, although som etimes
asym metric. The development of necrotizing retin itis at the edge of a chorioretinal scar
strongly suggests toxoplasmosis. A m ixed pattern of hemorrhage and necrosis w ith a
granu lar-app earing border suggests CMV retin itis.
Laboratory tests can support a clin ical d iagnosis but rarely serve as the sole means
of d iagnosis. For example, m any adu lts are serologically p ositive for such pathogens as
Toxoplasma gondii, CMV, or Borrelia burgdorferi w ithout having d isease. Therefore, the pre-
d ictive negative value of a test is u sually more helpfu l w ith these tests than the pred ictive
positive value. In a patient w ith bilateral, mu ltifocal choroid itis, for example, a p ositive
serologic toxoplasm a IgG (and IgM) by ELISA is probably not relevant, but a negative test
virtually ru les out the d isease. One exception to th is generalization is syph ilis, wh ich
(becau se it can cau se almost any m an ifestation of uveitis) shou ld always be con sidered in
the d ifferential d iagnosis. For th is reason, I test all uveitis patients w ith both an RPR (or
VDRL) and an FTA-Abs (or MH A-TP) becau se the former can be falsely negative in up to
30% of patients w ith syph ilitic uveitis.
In cases in wh ich the d iagnosis is not m ade clin ically, or the respon se to treatment is
poor, invasive testing is ind icated. The polymerase chain reaction (PCR) test is now w idely
available and demonstrates good sensitivity and specificity for Herp es simplex types 1
and 2, Varicella zoster, and CMV; it is not as u sefu l for toxoplasmosis. Either aqueou s 2 or
vitreou s 3 can be u sed ; the former is easier to obtain, but the latter m ay have h igher yield.
The same specimen can be u sed for bacterial and fu ngal cu ltu res and for cytology. Rarely,
endoretinal or chorioretinal biopsy is necessary.4 In som e instances, the com bination of a
characteristic clin ical appearance in conju nction w ith key system ic featu res w ill d irect the
appropriate therapy (eg, a focal choroid al nodu le in a patient w ith a positive PPD skin test
and an abnorm al chest x-ray merits an aggressive anti-tubercu lou s regimen).
With all d iagnostic tests, it is critical that you speak w ith the pathologist before obtain-
ing any sp ecimen so that you know exactly how the specimen shou ld be processed.
Noth ing is more fru strating to a patient than to u ndergo an invasive d iagnostic procedu re,
on ly to have an in su fficient quantity, the specimen placed in the w rong med iu m, or left at
the w rong temp eratu re so that the correct test can not be performed. Referral to a uveitis
sp ecialist is ind icated if prop er hand ling and processing can not be gu aranteed.
Treatm ent of in fectiou s posterior uveitis is obviou sly best d irected again st a sp ecific
pathogen, but in a patient w ith a rapid ly progressive cou rse, atypical featu res, or non-
d iagnostic laboratory tests, broad-sp ectru m therapy m ay be necessary. In such cases, you
shou ld u se one d rug to cover mu ltiple pathogens if p ossible (eg, foscarnet is effective
again st the viru ses cau sing both acute retinal necrosis and CMV retin itis) and treat for
the cond ition that is likely to progress most qu ickly (eg, if you can not decide if a wh ite
necrotizing m id-p eripheral lesion represents toxoplasmosis or cand id iasis 5, be su re to
u se antifu ngal therapy becau se the cand id iasis is the greater im med iate th reat to vision).
Both system ic and intravitreal agents m ay be u sed. Referral to a uveitis specialist is often
ind icated for treatment d irection.
204 Question 49
References
1. McCannel CA, Holland GN, Helm CJ, Cornell PJ, Winston JV, Rimmer TG. Causes of uveitis in the general
practice of ophthalmology. UCLA Community-Based Uveitis Study Group. Am J O phthalm ol. 1996;121(1):35-
46.
2. Westeneng AC, Rothova A, de Boer JH, de Groot-Mijnes JD. Infectious uveitis in immunocompromised
patients and the diagnostic value of polymerase chain reaction and the Goldmann-Witmer coefficient in aque-
ous analysis. Am J O phthalm ol. 2007;144 (5):781-785.
3. Margolis R. Diagnostic vitrectomy for the diagnosis and management of posterior uveitis of unknown etiology.
Curr O pin O phthalm ol. 2008;19 (3):218-224.
4. Rutzen AR, O rtega-Larrocea G, Dugel PU, et al. Clinicopathologic study of retinal and choroidal biopsies in
intraocular inflammation. Am J O phthalm ol. 1995;119 (5):597-611.
5. Shah CP, McKey J, Spirn MJ, Maguire J. O cular candidiasis: a review. Br J O phthalm ol. 2008;92(4):466-468.

Curbside Consultation in Retina

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Curbside Consultation in Retina

Uploaded by

Copyright:

Available Formats

Curbside

Pu blished by: SLACK Incorp orated

Library of Congress Cataloging-in-Pu blication Data

Question 13: How Do I Work-Up and Manage a Patient With a

D ean Eliott, MD is Director of the Doheny Retina In stitute and Professor of

Karen M . Gehrs, M D (Question 7) Raymond Iezzi, M D, M S (Question 45)

Jennifer I. Lim, M D (Question 20) Franco M . Recchia, M D (Question 3)

Catherine B. M eyerle, M D (Question 23) Kourous A . Rezaei, M D (Question 36)

William F. M ieler, M D (Question 27) SriniVas R. Sadda, M D (Question 18)

Steven D. Schwartz, M D (Question 38) William E. Smiddy, M D (Question 16)

Sharon Fekrat, M D, FACS

WHAT ARE SOME TIPS TO

What I Tell the Patient

WHAT RETINAL FINDINGS WOULD

A complete d ilated fu ndu s exam ination shou ld be perform ed prior to cataract or

Retinal Detachment Prevention

Epiretinal Membrane and Macular Hole

Acknowledgment of Significant Collaboration

WHAT SHOULD I TELL A MYOPE

Acknowledgment of Significant Collaboration

Suber S. Huang, MD, MBA

Risk Factors for Retinal Detachment

Prophylactic Treatment of Lattice Degeneration

WHAT RETINAL BREAKS

Ron Afshari Adelman, MD, MPH

Which Retinal Breaks Are Risky?

Treatment Options for Different Types of Breaks

Acknowledgment of Significant Collaboration

WHAT LESIONS IN THE FELLOW EYE OF

HOW SOON SHOULD A PATIENT

HOW LONG SHOULD I WAIT TO

DOES CATARACT SURGERY IN

WHAT SHOULD I DO WHEN I SEE A

Fig u re 10 -1. Wh ite ca lcific e m bo li fro m a

Fig u re 10 -2 . (A) Le ft fun dus o f a 33-ye a r-o ld

Recommended Systemic Evaluation of Patients

WHAT SHOULD I DO WHEN I

Jorge G. Arroyo, MD, MPH

Fig u re 11-1. Th e CWS, a lso ca lle d

Fig u re 11-2 . Ea rly ligh t m icro sco -

Etiologies of Cotton-Wool Spots

Immune-mediated collagen vascular disease/immune complex disease

Fig u re 11-3 . Seve re re tin a l disto rtio n

Fig u re 11- 4 . Tra um a tic la ce ra tio n

WHAT SHOULD I DO WHEN I SEE

Intraretinal hemorrhages associated w ith su b- and preretinal hemorrhage can be

Acknowledgment of Significant Collaboration

HOW DO I WORK-UP AND

vitrectomy. More recently, pharm acologic approaches to clearing vitreou s hemorrhages

Financial disclosure: Research support from A llergan & N ovartis.

Acknowledgment of Significant Collaboration

HOW CAN O PTICAL COHERENCE

Fig u re 14 -1. In tra re tin a l cystic ch a n ge s co n siste n t with cysto id m a cula r e de m a .

Fig u re 14 -2 . Epire tin a l m e m bra n e ca usin g th icke n in g o f th e m a cula , lo ss o f n o rm a l

Fig u re 14 -3 . Po ste rio r hya lo id in se rtin g in to th e m a cula re sultin g in bum py a ppe a r-

WITH THE RAPIDLY ADVANCING

Acknowledgment of Significant Collaboration

HOW IMPORTANT IS THE PROMPT

Are All Macular Holes Treated With Vitrectomy?

When Should Cataract Surgery Be Considered

WHEN SHOULD I REFER A PATIENT

Fig u re 17-1. Th is pa tie n t pre se n te d

Fluorescein angiography is a valu able d iagnostic tool in detecting epiretinal mem-

When Do I Refer a Patient With