Full

Bram W.
van Kooij
Cystoid macular edema

in uveitis
More than meets the eye
Bram Wouter van Kooij
Cystoid macular edema in uveitis
Utrecht University, Faculty of Medicine, The Netherlands
ISBN: 90-39341796
Cover: Twin design, Culemborg, The Netherlands
Printed by: Torendruk, Nijkerk, The Netherlands
Copyright © 2006 by B.W. van Kooij. No part of this thesis may be reproduced or
transmitted in any form or by any means, electronic or mechanic, including photocopy,
recording, or any information storage and retrieval system, without permission of the
copyright owner.
2
Cystoid macular edema in uveitis
Cystoid macula oedeem in uveitis

Daar zit meer achter
(met een samenvatting in het Nederlands)
Proefschrift
ter verkrijging van de graad van doctor

aan de Universiteit Utrecht op gezag van de
Rector Magnificus, Prof. dr. W. H. Gispen,
ingevolge het besluit van het College voor Promoties
in het openbaar te verdedigen op
dinsdag 28 maart 2006 des middags te 4.15 uur
door
Bram Wouter van Kooij

geboren op 22 mei 1975 te Amersfoort
3
Promotoren: Prof. dr. A. Rothova, Universiteit Utrecht
Prof. dr. W.F. Treffers, Universiteit Utrecht
Co-promotor: Dr. R. Fijnheer, Universiteit Utrecht
This thesis was supported by the Dr. F.P. Fisher Foundation
Additional financial support for the printing of this thesis by Landelijke Stichting voor
Blinden en Slechtzienden, Rotterdamse Vereniging Blindenbelangen, Visio, Alcon
Nederland BV, Allergan BV, Carl Zeiss BV, DORC Nederland BV, Eli Lilly Nederland BV,
Ipsen Farmaceutica BV, Laméris Ootech BV, MSD BV, Novartis Pharma BV, Pfizer BV,
Thea Pharma NV and Tramedico BV is gratefully acknowledged.
4
Voor mijn ouders
Voor Hanneke
5
Contents
Chapter 1 Introduction. 9
Chapter 2 The impact of macular edema on visual acuity 19

in patients with uveitis.
(Ophthalmology, in press)
Chapter 3 IgG concentration in the aqueous humor of 31

patients with uveitis and cystoid macular edema.
(submitted for publication)
Chapter 4 Distinct cytokine and chemokine profiles in the 41

aqueous of patients with uveitis and cystoid
macular edema
Chapter 5 Prognosticators of cystoid macular edema in 49

uveitis patients.
Chapter 6 Trace microalbuminuria in inflammatory cystoid 61

macular edema.
(American Journal of Ophthalmology 2004;138:1010–1015)
Chapter 7 A randomized, masked, cross-over trial of 75

lisinopril for inflammatory macular edema.
(American Journal of Ophthalmology, in press)
Chapter 8 The effect of non-steroidal anti-inflammatory 89

drugs on inflammatory cystoid macular edema.
(American Journal of Ophthalmology 2005;140:563–564)
Chapter 9 The pro’s and contra’s of intravitreal 99

triamcinolone injections for uveitis and
inflammatory macular edema.
(Ocular Immunology and Inflammation, in press)
6
Chapter 10 Rapid progression of diabetic retinopathy in 123
eyes with posterior uveitis.
Chapter 11 Summary and conclusions 131

Samenvatting en conclusies
Dankwoord 149
Curriculum Vitae 153
List of publications 154
7
8
Chapter 1
Chapter 1
Introduction
9
Introduction
Introduction
Cystoid macular edema (CME) is an accumulation of fluid in the most
central part of the retina and represents an important, vision
threatening disorder occurring in the course of various ocular diseases.
CME may complicate a multitude of underlying conditions such as
uveitis, diabetic retinopathy, retinal vein occlusions, retinitis
pigmentosa and may also develop following intraocular surgery.1-4 In
uveitis, (as well as in diabetes mellitus and retinal vascular occlusions)
CME forms a major cause of visual impairment and blindness, explicitly
in 41% and 29%, respectively. Approximately 31 to 47% of all patients
with uveitis develop CME in at least one eye.3,5,6
Pathogenesis of CME
The exact pathogenesis of CME in patients with uveitis is not clear. The
cause of pathologic fluid accumulation is attributed to a disruption of
the retinal vessels in the macular area i.e. inner blood retinal barrier
(BRB) and/or a dysfunction of the retinal pigment epithelium (RPE) i.e.
outer BRB. The inner BRB is formed by tight junctions between the
endothelial cells of the retinal vessels; the outer BRB is located at the
level of the RPE. The RPE functions as a permeability barrier between
the choroid and the retina by blocking the migration of small molecules
from the choriocapillaris into the retina and by actively transporting
ions and fluid from the retina into the choroid. The carbonic anhydrase
system (which can be enhanced by acetazolamide) is responsible for
this active transport and is localized at the apical and basal membranes
of the RPE. Intraocular inflammation may lead to the disruption of both
the BRBs via the release of local inflammatory mediators. In addition,
diverse systemic diseases, which are frequently associated with
intraocular inflammation might induce a systemic release of diverse
(inflammatory) mediators influencing both BRBs.
Vascular endothelial growth factor (VEGF) is a major known factor
in angiogenesis and vasopermeability in the eye. VEGF increases
vascular permeability and promotes BRB breakdown in diabetic
retinopathy. The VEGF concentration was elevated in the vitreous fluid
of patients with CME of diabetic origin compared to diabetic patients
without CME and non-diabetic patients.7-9 In addition, VEGF
concentration was elevated in the aqueous humor of 26 patients with
10
Chapter 1
diabetic and 9 with inflammatory CME compared to 28 patients without

diabetes and 11 patients without inflammatory CME.10,11 It was also
reported that a high VEGF level predicted a risk for the postoperative
development of CME in non-proliferative diabetic retinopathy patients.12
Transforming growth factor β (TGFβ) is, in contrast to VEGF, an
inhibitor of angiogenesis. TGFβ has been reported to be increased in
experimental autoimmune uveoretinitis together with VEGF and may
prevent effects of VEGF as its endothelial growth activity.13 Interleukin-
6 (IL-6) is a pro-inflammatory cytokine which may indirectly cause an
increase of vascular permeability and angiogenesis by inducing the
expression of VEGF. IL-6 was reported to be elevated in the vitreous
fluid and aqueous humor of 52 patients with diabetic CME.8,11 VEGF and
IL-6 were not only significantly correlated with macular edema, but
also showed a strong correlation with each other.8,11 Angiotensin II (A-
II) is a key factor in cardiovascular homeostasis and regulates the
growth of vascular smooth muscle cells and stimulates the induction of
various vascular growth factors. A-II and VEGF levels correlated with
the severity of diabetic CME, suggesting that A-II contributes to the
expansion of diabetic CME in collaboration with VEGF.7 Tumor necrosis
factor α (TNFα) and interleukin-1β (IL-1β) are potent pro-inflammatory
cytokines that appear to promote retinal vascular leakage by increasing
permeability of the endothelial cell tight junctions and by increasing
vesicular transport across the retinal vein endothelium in experimental
autoimmune uveoretinitis (EAU).13,14 VEGF, TNFα and IL-1β also induce
adhesion of leukocytes to the endothelium by increasing the production
of intercellular adhesion molecule-1 (ICAM-1).14,15 BRB breakdown
induced by VEGF appears to be partly leukocyte dependent, because
inhibition of ICAM-1 prevents BRB breakdown in rats. Soluble ICAM-1
was reported to be elevated in the vitreous fluid of patients with
diabetic CME.15
Until now inflammatory CME was thought to be a result of the
disturbance of the retinal vessels and RPE by the inflammatory
reaction. However, this does not explain the apparent arbitrariness of
uveitis patients getting CME. Although the association of CME with
uveitis of long standing duration was reported, it is not feasible to
predict which patients with uveitis will develop CME.
11
Introduction
One additional factor contributing to the development of CME in

uveitis might be associated vascular damage from other origins. The
support for this hypothesis is outlined in the underlying paragraph.
The condition of the retinal vessels, especially of endothelium,
plays an important role in the development of CME. Cardiovascular
diseases and subsequent retinal vascular changes are involved in the
pathogenesis of CME in patients with diabetes mellitus and retinal vein
occlusions and also after intraocular surgery.16 CME is strongly
associated with hypertension and the presence of microalbuminuria,
particularly in patients with diabetes mellitus.17-25 Microalbuminuria
reflects the increased permeability of damaged renal vascular walls.
The role of microalbuminuria as a marker of early vascular disease was
examined not only in diabetes mellitus, but also in non-diabetes
mellitus patients.26-31 In the non-diabetic population, microalbuminuria
was an independent indicator for cardiovascular risk factors and
cardiovascular morbidity. Moreover, in uveitis, elevated levels of
cellular fibronectin, which is excreted by the endothelium and is
elevated in systemic vascular disorders, were found in the intraocular
fluid.32-34
This led us to the hypothesis that cardiovascular diseases and/or
its risk factors might play a role in the development of CME in a disease
seemingly unrelated to systemic vascular pathology such as uveitis.
The cardiovascular diseases together with an increased permeability of
the retinal and choroidal vessels might form an underlying
predisposition with already damaged vessels and increased
permeability. The additional damage by inflammation might cause an
early onset of CME even after a short duration of uveitis.
Treatment of CME
The treatment of inflammatory CME is complicated, usually prolonged
and often characterized by a delicate balance between the beneficial
effects on CME and the adverse systemic and ocular effects. In the
past, CME was treated only when visual acuity was already
compromised. Despite frequently aggressive treatments, the
progression of inflammatory CME with accompanying visual loss was
common. This formed the reason for changed treatment policy of
inflammatory CME, for which early treatment regimens are now
12
Chapter 1
recommended, even for patients with (still) undamaged visual acuity.35

The treatment of CME is focused on the specific underlying ocular
condition in combination with acetazolamide and/or corticosteroids.
Furthermore, NSAIDs and a variety of immunosuppressive drugs are
used, all of which might be associated with significant adverse
effects.35 The effect of somatostatin analogues on CME has not yet
been systematically investigated and their role in the management of
inflammatory CME is still unclear.36 In the past years, the intravitreous
administration of triamcinolone has become popular for the treatment
of CME of various origins and has had a beneficial, but temporary effect
on CME. Despite the various possibilities of managing inflammatory
CME, many patients are resistant or intolerant to the treatments and
new therapeutic options, preferably with few side effects, are needed.
In conclusion
CME is a major complication of uveitis and its pathogenesis is yet
unclear. The retinal and choroidal vasculature plays an important role
in the pathogenesis of CME. The changes of ocular vasculature due to
systemic vascular disease were implicated in the development of CME;
however, the exact mechanism of this relation remains to be
elucidated. The recently used treatment options are frequently
insufficient because of the resistance of CME or intolerance to the
medication. For the effective management of inflammatory CME, it is
important to know the impact of CME on visual acuity, the prognosis of
CME for different anatomic locations and entities, and the patients at
risk for CME.
Aim of the Thesis

The aim of this thesis is to improve our knowledge in the pathogenesis
of inflammatory cystoid macular edema, specifically to elucidate the
role –if any- of vascular diseases in the development of inflammatory
CME. Furthermore we assess treatment options on their efficacy and
subsequently attempt to find new therapeutic alternatives for
inflammatory CME.
13
Introduction
Outline of the Thesis

To understand the impact of CME on visual acuity in patients with
uveitis and to determine potentially high risk groups for developing it,
we analyze the general and ophthalmologic data of 529 uveitis patients
in a cross-sectional study in chapter 2.
In chapter 3 we evaluate the quality of the BRB in uveitis and link
its disturbances to the development of CME. We determine the
breakdown of the BRB by measuring the leakage of immune globulin G
(IgG) through the BRB and relate its levels to clinical features of
uveitis, especially to the development of CME.
In chapter 4 we investigate if specific cytokines and chemokines
are involved in the pathogenesis of uveitis and inflammatory CME. We
determine the concentration of various mediators in the aqueous
humor of uveitis patients and relate them to clinical features such as
intraocular infection, activity of inflammation and CME.
Our hypothesis that the presence of generalized vascular damage
enhances the development of CME in uveitis is examined in chapters 5,
6 and 7. In chapter 5 we perform a retrospective study to investigate
the role of age, cardiovascular diseases and their risk factors on the
development of CME in uveitis. We do this by analyzing a questionnaire
on cardiovascular diseases and their risk factors and ophthalmologic
data in 54 patients with uveitis without CME and 43 patients with both
uveitis and CME. In chapter 6, we further explore our hypothesis by
examining the possible role of systemic microvascular damage on the
development of CME. In a matched case-control study we determine
and analyze urinary albumin excretion in 24 uveitis patients with CME
and 24 uveitis patients without CME in relation to ophthalmologic data.
In chapter 7, we attempt to analyze the effect of lisinopril, an
angiotensin converting enzyme inhibitor, on inflammatory CME and
visual acuity as well as on microalbuminuria. Lisinopril has a beneficial
effect on microvascular leakage (microalbuminuria) and has a blood
pressure lowering effect in patients with diabetes mellitus and systemic
hypertension. In this study we make use of a randomized, double-
blind, cross-over trial in 40 patients with inflammatory CME.
In chapters 8 and 9 we evaluate the efficacy of different treatment
regimens on inflammatory macular edema. In chapter 8 we examine
the effect of NSAIDs on inflammatory CME in 66 patients (38 patients
14
Chapter 1
treated with rofecoxib and 28 patients treated with naproxen) in a

prospective pilot study. In chapter 9 we discuss the efficacy of
intravitreal triamcinolone acetonide, a new administration possibility for
corticosteroids, in uveitis and inflammatory CME in a review of
literature and also include our experience with this technique of
corticosteroid application.
In Chapter 10 we report on 2 diabetic patients with asymmetric,
rapid progression of diabetic retinopathy after suffering from infectious
posterior uveitis.
The English and Dutch summary and conclusions form chapter 11.
15
Introduction
References
1. Bresnick GH. Diabetic macular edema. A review. Ophthalmology 1986;93:989-
997.
2. Klein R, Klein BE, Moss SE, Davis MD, DeMets DL. The Wisconsin epidemiologic
study of diabetic retinopathy. IV. Diabetic macular edema. Ophthalmology
1984;91:1464-1474.
3. Rothova A, Suttorp-van Schulten MS, Treffers WF, Kijlstra A. Causes and
frequency of blindness in patients with intraocular inflammatory disease. Br J
Ophthalmol 1996;80:332-336.
4. Weinberg D, Seddon J. Venous occlusive disease of the retina. In Albert D,
Jacobiec F, editors. Principles and practice of ophthalmology. Philadelphia: WB
Saunders, 1994:735-746.
5. Suttorp-Schulten MS, Rothova A. The possible impact of uveitis in blindness: a
literature survey. Br J Ophthalmol 1996;80:844-848.
6. Durrani OM, Tehrani NN, Marr JE, Moradi P, Stavrou P, Murray PI. Degree,
duration, and causes of visual loss in uveitis. Br J Ophthalmol 2004;88:1159-
1162.
7. Funatsu H, Yamashita H, Ikeda T, Nakanishi Y, Kitano S, Hori S. Angiotensin II
and vascular endothelial growth factor in the vitreous fluid of patients with
diabetic macular edema and other retinal disorders. Am J Ophthalmol
2002;133:537-543.
8. Funatsu H, Yamashita H, Ikeda T, Mimura T, Eguchi S, Hori S. Vitreous levels of
interleukin-6 and vascular endothelial growth factor are related to diabetic
macular edema. Ophthalmology 2003;110:1690-1696.
9. Funatsu H, Yamashita H, Ikeda T, Mimura T, Shimizu E, Hori S. Relation of
diabetic macular edema to cytokines and posterior vitreous detachment. Am J
Ophthalmol 2003;135:321-327.
10. Fine HF, Baffi J, Reed GF, Csaky KG, Nussenblatt RB. Aqueous humor and
plasma vascular endothelial growth factor in uveitis-associated cystoid macular
edema. Am J Ophthalmol 2001;132:794-796.
11. Funatsu H, Yamashita H, Noma H, Mimura T, Yamashita T, Hori S. Increased
levels of vascular endothelial growth factor and interleukin-6 in the aqueous
humor of diabetics with macular edema. Am J Ophthalmol 2002;133:70-77.
12. Funatsu H, Yamashita H, Noma H, Shimizu E, Mimura T, Hori S. Prediction of
macular edema exacerbation after phacoemulsification in patients with
nonproliferative diabetic retinopathy. J Cataract Refract Surg 2002;28:1355.
13. Vinores SA, Chan CC, Vinores MA, Matteson DM, Chen YS, Klein DA, Shi A, Ozaki
H, Campochiaro PA. Increased vascular endothelial growth factor (VEGF) and
transforming growth factor beta (TGFbeta) in experimental autoimmune
uveoretinitis: upregulation of VEGF without neovascularization. J Neuroimmunol
1998;89:43-50.
14. Luna JD, Chan CC, Derevjanik NL, Mahlow J, Chiu C, Peng B, Tobe T,
Campochiaro PA, Vinores SA. Blood-retinal barrier (BRB) breakdown in
experimental autoimmune uveoretinitis: comparison with vascular endothelial
growth factor, tumor necrosis factor alpha, and interleukin-1beta-mediated
breakdown. J Neurosci Res 1997;49:268-280.
16
Chapter 1
15. Funatsu H, Yamashita H, Sakata K, Noma H, Mimura T, Suzuki M, Eguchi S, Hori

S. Vitreous levels of vascular endothelial growth factor and intercellular adhesion
molecule 1 are related to diabetic macular edema. Ophthalmology
2005;112:806-816.
16. Tranos PG, Wickremasinghe SS, Stangos NT, Topouzis F, Tsinopoulos I, Pavesio
CE. Macular edema. Surv Ophthalmol 2004;49:470-490.
17. Jain R, Stevens JD, Bunce CV, Garrett C, Hykin PG. Ischaemic heart disease
may predispose to pseudophakic cystoid macular oedema. Eye 2001;15:34-38.
18. Jalkh AE, Trempe CL. Macular edema in branch retinal vein occlusion: types and
treatment. Ophthalmic Surg 1989;20:26-32.
19. Lopes de Faria JM, Jalkh AE, Trempe CL, McMeel JW. Diabetic macular edema:
risk factors and concomitants. Acta Ophthalmol Scand 1999;77:170-175.
20. Wallow IH, Danis RP, Bindley C, Neider M. Cystoid macular degeneration in
experimental branch retinal vein occlusion. Ophthalmology 1988;95:1371-1379.
21. Gilbert RE, Tsalamandris C, Allen TJ, Colville D, Jerums G. Early nephropathy
predicts vision-threatening retinal disease in patients with type I diabetes
mellitus. J Am Soc Nephrol 1998;9:85-89.
22. Knudsen ST, Bek T, Poulsen PL, Hove MN, Rehling M, Mogensen CE. Macular
edema reflects generalized vascular hyperpermeability in type 2 diabetic
patients with retinopathy. Diabetes Care 2002;25:2328-2334.
23. Dodson PM, Gibson JM. Long-term follow-up of and underlying medical
conditions in patients with diabetic exudative maculopathy. Eye 1991;5 ( Pt
6):699-703.
24. Klein R, Klein BE, Moss SE, Cruickshanks KJ. The Wisconsin Epidemiologic Study
of Diabetic Retinopathy. XV. The long-term incidence of macular edema.
Ophthalmology 1995;102:7-16.
25. Roy MS, Klein R. Macular edema and retinal hard exudates in African Americans
with type 1 diabetes: the New Jersey 725. Arch Ophthalmol 2001;119:251-259.
26. Hillege HL, Janssen WM, Bak AA, Diercks GF, Grobbee DE, Crijns HJ, van Gilst
WH, de Zeeuw D, de Jong PE. Microalbuminuria is common, also in a
nondiabetic, nonhypertensive population, and an independent indicator of
cardiovascular risk factors and cardiovascular morbidity. J Intern Med
2001;249:519-526.
27. Hillege HL, Fidler V, Diercks GF, van Gilst WH, de Zeeuw D, van Veldhuisen DJ,
Gans RO, Janssen WM, Grobbee DE, de Jong PE. Urinary albumin excretion
predicts cardiovascular and noncardiovascular mortality in general population.
Circulation 2002;106:1777-1782.
28. Pedrinelli R, Dell'Omo G, Penno G, Mariani M. Non-diabetic microalbuminuria,
endothelial dysfunction and cardiovascular disease. Vasc Med 2001;6:257-264.
29. Pedrinelli R, Dell'Omo G, Di B, V, Pontremoli R, Mariani M. Microalbuminuria, an
integrated marker of cardiovascular risk in essential hypertension. J Hum
Hypertens 2002;16:79-89.
30. Roest M, Banga JD, Janssen WM, Grobbee DE, Sixma JJ, de Jong PE, de Zeeuw
D, Der Schouw YT. Excessive urinary albumin levels are associated with future
cardiovascular mortality in postmenopausal women. Circulation 2001;103:3057-
3061.
17
Introduction
31. Romundstad S, Holmen J, Hallan H, Kvenild K, Kruger O, Midthjell K.

Microalbuminuria, cardiovascular disease and risk factors in a
nondiabetic/nonhypertensive population. The Nord-Trondelag Health Study
(HUNT, 1995-97), Norway. J Intern Med 2002;252:164-172.
32. Kanters SD, Banga JD, Algra A, Frijns RC, Beutler JJ, Fijnheer R. Plasma levels
of cellular fibronectin in diabetes. Diabetes Care 2001;24:323-327.
33. Lockwood CJ, Peters JH. Increased plasma levels of ED1+ cellular fibronectin
precede the clinical signs of preeclampsia. Am J Obstet Gynecol 1990;162:358-
362.
34. Peters JH, Maunder RJ, Woolf AD, Cochrane CG, Ginsberg MH. Elevated plasma
levels of ED1+ ("cellular") fibronectin in patients with vascular injury. J Lab Clin
Med 1989;113:586-597.
35. Rothova A. Medical treatment of cystoid macular edema. Ocul Immunol Inflamm
2002;10:239-246.
36. Kuijpers RW, Baarsma S, van Hagen PM. Treatment of cystoid macular edema
with octreotide. N Engl J Med 1998;338:624-626.
18
Chapter 1
Chapter 2
The impact of macular edema

on visual acuity in uveitis
Charlotte Lardenoye1* MD, PhD, Bram van Kooij2* MD, Aniki Rothova2
MD, PhD.
*
These authors contributed equally to the work.
1
St. Elisabeth Hospital, Dept of Ophthalmology, P.O. Box 90151, 5000
LC Tilburg, The Netherlands.
2
F.C.Donders Institute, Dept of Ophthalmology, University Medical
Center Utrecht, P.O. Box 85 500, 3508 GA Utrecht, The Netherlands.
(Ophthalmology, in press)
19
The impact of CME on visual acuity in uveitis
Abstract
Objective
To investigate the impact of cystoid macular edema (CME) on visual
acuity in patients with uveitis.
Design
Cross-sectional study
Methods
Participants: The data of 529 patients (842 eyes) with uveitis were
analyzed. Main outcome measures: We recorded gender and age of the
patients, anatomic site and diagnosis of uveitis, associations with
systemic diseases, onset and duration of uveitis, presence of CME as
well as optimal visual acuity and the causes of decrease in visual
acuity.
Results
CME was noted in 175 (33%) of all uveitis patients of whom 77 (44%)
had visual acuity equal to or less than 20/60 in at least one eye. The
mean visual acuity for eyes with CME was significantly worse than for
eyes without CME (0.25 versus 0.4; P = .003). Of all uveitis patients,
185 (35%) developed visual acuity equal to or less than 20/60 in at
least one eye, in 77 (42%) patients caused by CME. Poor visual acuity
in patients with CME was associated with the advanced age of the
patients, chronic inflammation and various specific uveitis entities. The
development of visually impaired or blind eyes in patients with pan-
and intermediate uveitis was caused in the majority of cases by CME
(59% and 85%, respectively).
Conclusion
CME was a major cause of visual loss in patients with uveitis. The
unsatisfactory visual acuity in patients with uveitis underlines the need
for improved management of this complication.
20
Chapter 2
Introduction
Cystoid macular edema (CME) complicates the course of many ocular
disorders such as uveitis, diabetic retinopathy, retinal vein occlusion,
retinitis pigmentosa, radiation retinopathy, and may also complicate
intraocular surgery. CME regularly leads to a reduction in visual acuity
and has been suggested to be a major cause of decreased visual acuity
in patients with uveitis.1-4 Few data, however, are available on the
impact of CME on visual acuity for different uveitis entities.1-4 The aim
of this study was to evaluate the impact of CME on visual acuity in a
large uveitis population, specified for various uveitis entities according
to their anatomic location, etiology and associations with systemic
diseases.
Material and methods

In this cross-sectional study we included 529 patients with uveitis,
chosen from more than 2000 patients with uveitis from our institute,
that combines secondary and tertiary ophthalmologic care. The random
choice was generated with Statistical Package for the Social Sciences
(SPSS 12.0).
All patients underwent a standard screening protocol, depending
upon the anatomic classification of the inflammation, which included
erythrocyte sedimentation rate, red and white blood cell counts,
glucose levels, HLA-B27 typing, angiotensin converting enzyme and
lysozyme levels, syphilis serology, chest X ray and, for children under
16, antinuclear antibody levels. Selected patients (depending on
history, character and activity of their ocular disease as well as the
outcome of the laboratory and radiographic screening procedures)
underwent special tests and diagnostic procedures (tailored approach).
Systemic diseases were diagnosed according to current diagnostic
criteria; the diagnosis of sarcoidosis was always confirmed by
histological examination. The following data were recorded: sex and
age of the patients, anatomic site, clinical diagnosis, etiologic agent of
uveitis, associations with systemic diseases, onset and duration of the
uveitis, presence of CME in one or both eyes, visual acuity and whether
CME was the main cause of decreased visual acuity. Legal blindness
was defined as a best corrected visual acuity equal to or less than
20/200 for the better eye, and visual impairment as a best corrected
21
visual acuity equal to or less than 20/60 but better than 20/200.5 The
term ‘legally blind and/ or visually impaired eye’ was used to describe
visual loss in individual affected eyes. For patients with more than one
possible cause of visual loss, the first complication which caused the
decrease in vision was considered responsible.
CME was defined by both clinical and angiographic criteria.6 For
statistical analysis Snellen visual acuity values were converted into
LogMar visual acuity (logarithm of the minimal angle of resolution).
Student’s-t test and the Chi-square test were used for statistical
analysis. We also used repeated-measurement analysis of variance to
assess relationships between recorded data and final visual outcome.
P< .05 was considered significant.
Results
The mean age of patients was 46 years (range 6-89) and the male-to-
female ratio was 1 to 1.13 (Table 1). The majority of patients
(388/529; 73%) were aged 21 to 60 years. We found no significant
difference in age between patients with and without CME (47 years
versus 45 years; P = .30) or gender (male 45 years, female 46 years;
P = .30). The mean duration of follow-up was 5.4 years (range 6
months-46 years) with no difference between those with CME (5.6
years) and those without (5.3 years; P = .60) (Table 1).
CME in at least one eye was observed in 175 out of 529 patients
(33%), of whom 77/175 (44%) developed a visually impaired or blind
eye (VA≤20/60). Our study group included 5 patients with diabetes
mellitus, 2 (40%) of them had CME. The mean visual acuity for eyes
with CME was significantly worse than for eyes without CME (20/80
versus 20/50; P = .003). Vision equal to or less than 20/60 in at least
one eye was more commonly observed in patients with CME (77/175,
44%) than in patients without CME (108/354, 31%; P = .002).
Of all uveitis patients,185/529 (35%) had vision equal to or less
than 20/60 in at least one eye, 77/185 (42%) of which was caused by
CME. Forty-eight/529 (9%) patients were either legally blind or visually
impaired (legal blindness in 20/529, 4% and visual impairment in
28/529, 5% of patients). Twenty-five/48 (52%) of either legal
blindness or visual impairment was caused by CME (respectively in
6/20, 30% of patients and in 19/28, 68% of patients).
22
Chapter 2
The relationships between visual loss and CME for different

anatomic sites of uveitis are given in Table 2. Visual impairment or
blindness was most commonly seen in panuveitis (63/99, 64%; P < .01
compared to other anatomical types), and was caused by CME in 37/63
(59%) of patients. Only 20/72 (28%) of patients with intermediate
uveitis suffered from visual impairment or blindness, but the proportion
of CME causing the visual loss was higher than in other anatomic
locations (17/20, 85%; P < .04 for all locations). Patients with
intermediate uveitis and panuveitis had the highest frequency of CME
(respectively, 43/72, 60% and 65/99, 66%) and a similar frequency of
CME leading to visual loss (respectively 17/43, 40% and 34/65, 52%; P
= .19).
Table 1: General characteristic of patients with uveitis
Total Non-CME* CME*
Patients 529 354 (67%) 175 (33%)

Number of affected eyes 842 564 (67%) 278 (33%)
Male to female ratio 1 : 1.13 1:1.05 1:1.33
Mean age in years (range) 46 (6-89) 45 (6-89) 47 (9-87)
Mean follow-up in years (range) 5.4 (0.5-46) 5.3 5.6
Mean visual acuity 20/60 20/50# 20/80#
*
#
p= 0.003
Table 2: Anatomical types of uveitis, visual loss and cystoid macular edema (CME)
Patients with visual acuity (VA)

of ≤ 20/60 in at least one eye
Proportion of
Patients CME leading to
Location Total with CME Total with CME VA of ≤ 20/60
Anterior 213 23 (11%) 49 (23%) 7 (14%) 7/23 (30%)

Intermediate 72 43 (60%) 20 (28%) 17 (85%) 17/43 (40%)
Posterior 122 41 (34%) 45 (37%) 13 (29%) 13/41 (32%)
Pan 99 65 (66%) 63 (64%) 37 (59%) 34/65 (52%)
Scleritis 23 3 (13%) 8 (35%) 3 (38%) 3/3 (100%)
Total 529 175 (33%) 185(35%) 77 (42%) 77/175 (44%)
23
Table 3: Specific uveitis entities, visual loss and cystoid macular edema (CME)
Patients with visual acuity (VA)

of ≤ 20/60 in at least one eye
Proportion of
Cause of uveitis or association Patients CME leading to
with systemic disease Total with CME Total with CME VA of ≤ 20/60
HLA-B27-associated 58 7 (12%) 6 (10%) 2 (33%) 2/7 (29%)

Toxoplasmosis 56 12 (21%) 21 (38%) 4 (19%) 4/12 (33%)
Sarcoidosis 29 17 (59%) 9 (31%) 6 (67%) 6/17 (35%)
Fuchs’ heterochromic iridocyclitis 22 3 (14%) 11 (50%) 2 (18%) 2/3 (67%)
Trauma or intraocular surgery 21 6 (29%) 10 (48%) 4 (40%) 4/6 (67%)
Viral infections of anterior segment* 18 2 (11%) 5 (28%) 1 (20%) 1/2 (50%)
Juvenile rheumatoid arthritis 10 6 (60%) 5 (50%) 2 (40%) 2/6 (33%)
Behçet disease 8 5 (63%) 5 (63%) 2 (40%) 2/5 (40%)
Systemic autoimmune disease 8 1 (13%) 3 (38%) 0 (0%) 0/1 (0%)
Intraocular bacterial infection# 7 3 (43%) 2 (29%) 1 (50%) 1/3 (33%)
Birdshot retinochoroidopathy 6 6 (100%) 5 (83%) 5 (100%) 5/6 (83%)
Acute retinal necrosis 5 5 (100%) 5 (100%) 3 (60%) 3/5 (60%)
Intraocular lymphoma 5 1 (20%) 3 (60%) 1 (33%) 1/1 (100%)
Miscellaneous+ 39 14 (36%) 23 (59%) 7 (30%) 7/14 (50%)
Undetermined 237 87 (37%) 72 (30%) 37 (51%) 37/87 (43%)
Total 529 175 (33%) 185 (35%) 77 (42%) 77/175(44%)
*
Patients with posterior segment involvement as in acute retinal necrosis were excluded.
#
All intraocular bacterial infections were proven by analysis of intraocular fluids.
+
The group miscellaneous included patients with multiple sclerosis, serpiginous chorioretinopathy,
acute posterior multifocal placoid epitheliopathy, intraocular tumors other than lymphoma, diabetes
mellitus associated uveitis, Vogt-Koyanagi-Harada disease, systemic fungal infection, parasitic
infections other than toxoplasmosis, systemic syphilis or gonorrhea infections, and Eales’ disease.
Specific uveitis entities with their visual acuities and CME development
are indicated in Table 3. Visual loss (with and without CME) was mainly
seen in acute retinal necrosis (5/5, 100%) followed by birdshot
chorioretinopathy (5/6, 83%), Behçet’s disease (5/8, 63%), intraocular
lymphoma (3/5, 60%), juvenile rheumatoid arthritis (5/10, 50%) and
Fuchs’ heterochromic iridocyclitis (11/22, 50%). The uveitis entities
complicated by CME in more than 50% of patients included sarcoidosis
(17/29, 59%), juvenile rheumatoid arthritis (6/10, 60%), Behçet’s
disease (5/8, 63%), birdshot chorioretinopathy (6/6,100%) and acute
retinal necrosis (5/5, 100%). Visual loss caused by CME was mainly
seen in birdshot chorioretinopathy, sarcoidosis and acute retinal
necrosis (Table 3). Visual loss was least frequently observed in HLA-
B27 associated uveitis (6/58, 10%; Table 3). CME was least frequently
24
Chapter 2
seen in viral infections of the anterior segment (11%; Table 3) and

HLA-B27 associated uveitis (12%; Table 3).
Using multiple regression analysis, we found that poor visual
acuity was associated with the presence of CME, the duration of the
uveitis and the age of the patients (independent associations; P < .001
for each). The duration of the uveitis and increasing age of the patients
were related (P = .03). The association of poor visual acuity in
inflammatory CME with increasing age remained significant when
adjusted to the duration of uveitis. Uveitis with a duration of more than
1 year was associated with CME (P = .004).
CME developed in 32% of the cases in patients younger than 50,
43% of whom suffered significant visual loss. Thirty-six % of the
patients older than 50 exhibited CME, 67% of whom suffered visual
impairment or blindness (P = .002). The gender of the patients did not
influence the visual outcome (P = .85). Among the patients in the
working age group (18-65), 120/423 (28%) had a visually impaired or
blind eye, which was caused by CME in 53/120 (44%) of the patients.
Discussion
In our cross-sectional study, 35% of all uveitis patients developed at
least one visually impaired or legally blind eye, which was caused by
CME in 42% of the cases. This finding implicates that CME is an
important cause of visual loss in patients with uveitis. The chance of
developing visual impairment in the uveitis eye is much higher if CME is
present (P < .003). In present series, CME caused visual loss in 44% of
patients between 18 and 65 years old and is therefore an important
factor of visual loss in the working age group of the uveitis population.
In our previous study, we observed the development of visually
impaired or blind eyes in 35% of patients with uveitis, which is in
accordance with our present findings.1 Durrani et al performed a
retrospective study in 315 patients with uveitis with a mean duration of
follow-up of 37 months. They had documented a visual loss (<6/18) at
some point during follow-up in 70% of patients with uveitis. In their
series, visual loss was attributable to CME in 47% of the cases (27%
caused by CME solely, 20% caused by CME in combination with
cataract), which is very similar to our finding of 42%.3 In patients with
chronic severe uveitis, legal blindness developed in 3% of patients and
25
18% of patients retained CME despite the treatment.4 The differences

in results are probably caused by a difference in the study designs. We
have shown that the impact of CME on visual acuity differed for diverse
anatomic types of uveitis. Panuveitis was associated with poor visual
outcome compared to other anatomical types of uveitis (P < .01),
which is in accordance with other studies.1,3,4 In addition, CME was
more severe in patients with panuveitis, as 52% of panuveitis patients
with CME suffered from visual loss compared to 40% for intermediate
uveitis and 30% for anterior and posterior uveitis. Although the visual
loss in intermediate uveitis (28%) was less frequent than in posterior
and panuveitis (Table 2), 85% of the visual loss was caused by CME,
which is more frequent than in other anatomic uveitis types (P < .04).
The visual acuity differed for specific uveitis entities. HLA B27-
associated uveitis had a rather favorable visual outcome: only 10% of
HLA B27-positive patients developed a visually impaired or legally blind
eye and none had bilateral legal blindness or visual impairment. These
observations are in accordance with the results of previous studies on
HLA B27-associated uveitis.1,7,8 CME played a minor role in the visual
loss of HLA-B27-associated uveitis, as only 12% of patients with a
visually impaired eye developed CME. The similar percentage of CME in
HLA B27 positive patients was reported previously.9,10 The visual
acuity in sarcoid-associated uveitis was similar to already reported
findings.1,11 CME developed in 59% of patients with sarcoid-associated
uveitis, which is similar to a study performed by Dana et al (58%).12
The patients with Behçet's disease had a better visual outcome than
usually reported, which may be due to ethnic and geographic
differences as well as to the small number of patients included in our
study.13,14 We found CME in 63% of patients with ocular Behçet’s
disease. This frequency is similar to the previously reported 45-
63%.13,15 The patients with birdshot chorioretinopathy had poor visual
outcome, which is in concordance with the literature.16-18 The
prominent role of CME in the visual loss occurring in birdshot
chorioretinopathy was also previously reported.17-19 The surprisingly
high frequency of visual impairment in Fuchs’ heterochromic
iridocyclitis may be explained by the presence of (not yet removed)
cataracts. CME was not a frequent feature in Fuchs’ heterochromic
iridocyclitis and toxoplasmosis (respectively 14% and 21%).
26
Chapter 2
The poor visual acuity in inflammatory CME was clearly associated

with increasing age (P < .002). Hypothetical explanations for this
finding might include decreased function and/or degeneration of retinal
cells during life, a process which may be enhanced by an intra-ocular
inflammation.20 The eventual compromised condition of the vascular
wall in aged persons, associated with increasing leakage of fluid
through the vessels, might also play a role.
The drawbacks of our study included the absence of evaluation for
severity (except of duration) and/or activity of uveitis, although an
association between these characteristics and the development of CME
undoubtedly exists. The reasons included the retrospective analysis of
the data and the absence of an appropriate score for the severity of
uveitis. The large number of the patients included might however
compensate for this drawback. In addition, the analysis of preceding
intraocular surgeries was not performed, which might also exacerbate
or induce the CME in patients with uveitis. The evaluation of the
changes of CME following surgery would require systematically
performed FA’s or OCTs and these were not available in our cross-
sectional study.
In conclusion, we established that CME is a major complication
causing visual decrease in uveitis, and is especially severe among
elderly patients and those with chronic disease. Our findings emphasize
the need for improved management of inflammatory CME.21
27
References
Ophthalmol 1996;80:332-336.
1162.
4. Bodaghi B, Cassoux N, Wechsler B, Hannouche D, Fardeau C, Papo T, Huong DL,
Piette JC, LeHoang P. Chronic severe uveitis: etiology and visual outcome in 927
patients from a single center. Medicine (Baltimore) 2001;80:263-270.
5. Kraut JA, McCabe CP. The problem of low vision. Definition and common
problems. In Albert DM, Jacobiec FA, editors. Principles and practice of
ophthalmology. Philadelphia: Saunders,W.B., 1994:3664-3666.
6. Clinical features of uveitis. In Kanski JJ, editor. Clinical Ophthalmology. London:
Butterworth-Heinemann, 1994:152-154.
7. Power WJ, Rodriguez A, Pedroza-Seres M, Foster CS. Outcomes in anterior
uveitis associated with the HLA-B27 haplotype. Ophthalmology 1998;105:1646-
1651.
8. Rodriguez A, Akova YA, Pedroza-Seres M, Foster CS. Posterior segment ocular
manifestations in patients with HLA-B27-associated uveitis. Ophthalmology
1994;101:1267-1274.
9. Uy HS, Christen WG, Foster CS. HLA-B27-associated uveitis and cystoid macular
edema. Ocul Immunol Inflamm 2001;9:177-183.
10. Monnet D, Breban M, Hudry C, Dougados M, Brezin AP. Ophthalmic findings and
frequency of extraocular manifestations in patients with HLA-B27 uveitis: a
study of 175 cases. Ophthalmology 2004;111:802-809.
11. Lobo A, Barton K, Minassian D, du Bois RM, Lightman S. Visual loss in sarcoid-
related uveitis. Clin Experiment Ophthalmol 2003;31:310-316.
12. Dana MR, Merayo-Lloves J, Schaumberg DA, Foster CS. Prognosticators for
visual outcome in sarcoid uveitis. Ophthalmology 1996;103:1846-1853.
13. Tugal-Tutkun I, Onal S, tan-Yaycioglu R, Huseyin AH, Urgancioglu M. Uveitis in
Behcet disease: an analysis of 880 patients. Am J Ophthalmol 2004;138:373-
380.
14. Sakamoto M, Akazawa K, Nishioka Y, Sanui H, Inomata H, Nose Y. Prognostic
factors of vision in patients with Behcet disease. Ophthalmology 1995;102:317-
321.
15. Gedik S, Akova Y, Yilmaz G, Bozbeyoglu S. Indocyanine green and fundus
fluorescein angiographic findings in patients with active ocular Behcet's disease.
Ocul Immunol Inflamm 2005;13:51-58.
16. Kiss S, Ahmed M, Letko E, Foster CS. Long-term follow-up of patients with
birdshot retinochoroidopathy treated with corticosteroid-sparing systemic
immunomodulatory therapy. Ophthalmology 2005;112:1066-1071.
17. Rothova A, Berendschot TT, Probst K, van KB, Baarsma GS. Birdshot
chorioretinopathy: long-term manifestations and visual prognosis.
28
Chapter 2
18. Thorne JE, Jabs DA, Peters GB, Hair D, Dunn JP, Kempen JH. Birdshot
retinochoroidopathy: ocular complications and visual impairment. Am J
Ophthalmol 2005;140:45-51.
19. Gasch AT, Smith JA, Whitcup SM. Birdshot retinochoroidopathy. Br J Ophthalmol
1999;83:241-249.
20. Weiter JJ, Roh S, Pruett RC. Aging in the retina and choroid. In Albert DM,
Jacobiec FA, editors. Principles and practice of ophthalmology. Basic Sciences.
Philadelphia: WB Saunders, 1994:718-722.
2002;10:239-246.
29
30
Chapter 2
Chapter 3
Immunoglobulin G
concentration in the aqueous
humor of patients with uveitis
and cystoid macular edema.
Bram van Kooij1 MD, Annemarie. Weersink2 MD, PhD, David van de
Vijver2 PhD, Jolanda de Groot – Mijnes1,2 PhD, Aniki Rothova1 MD, PhD.
1
F.C. Donders Institute of Ophthalmology and 2Department of Virology,
Eijkman-Winkler Center, University Medical Center, Utrecht, The
Netherlands.
31
IgG concentration in the aqueous of patients with uveitis and CME
Abstract
Purpose
To determine the relationship between intraocular immunoglobulin G
(IgG) levels in aqueous humor and various clinical features, in
particularly cystoid macular edema (CME) in patients with uveitis.
Methods
IgG concentrations were determined in 112 aqueous humor samples
from uveitis patients. Clinical data, including the specific diagnosis,
duration of uveitis, presence and duration of CME, and all systemic
medications at the time of sampling were documented. Fluorescein
angiography and/or fundoscopic examination were monitored to
evaluate the condition of the retina and the choroid. Controls consisted
of 20 aqueous humor samples from patients with age-related cataract
without uveitis or other ocular diseases. The relationship between the
intraocular IgG levels and the clinical data was statistically analyzed.
Results
Intraocular IgG concentrations were elevated in patients with uveitis
compared to control patients (0.07 mg/mL versus 0.01 mg/mL; P <
.001). Uveitis patients with CME had higher intraocular IgG levels
compared to those without CME (0.16 mg/mL versus 0.05 mg/mL; P <
.001). Inflammatory CME and retinal ischemia were independent
predictors of high intraocular IgG levels (P = .017 and P = .001,
respectively). An association was found between CME and choroidal
leakage (P = .043) but not between choroidal leakage and IgG
concentrations, suggesting that the outer blood-retinal barrier (BRB) is
selectively permeable for fluid, but not for large molecules like IgG
Conclusions
High intraocular IgG levels were found in patients with uveitis,
particularly in those with CME and/or retinal ischemia. Our study
revealed a possible selective permeability of the outer BRB for fluid, but
not large proteins.
32
Chapter 3
Introduction
Cystoid macular edema (CME) is a major complication of uveitis and
negatively influences the visual outcome of patients with uveitis. CME
is the result of an increased permeability of the blood-retinal barrier
(BRB). Immunoglobulin G (IgG) is a protein with a molecular weight of
150 000, which is present in peripheral blood and lymph fluid and
makes up about 75% of the total serum immunoglobulins. An intact
BRB prevents the leakage of IgG into ocular fluids, including the
aqueous humor, and IgG is not normally produced in the quiescent
eye.1 Therefore, under these conditions the IgG concentration in the AH
is very low (<0.04 mg/mL).2,3 However, when the BRB is damaged, as
in cases of uveitis and/or CME, IgG might leak into the AH. We
hypothesized that in uveitic eyes the presence of CME is associated
with elevated intraocular IgG levels, due to more severe leakage
through the BRB. Here, we determined the levels of IgG in the aqueous
humor of patients with uveitis and relate the results to clinical features
of uveitis, especially to CME.
Material and methods

Patients
From October 2001 to August 2003 139 aqueous humor samples from
123 patients with an active uveitis were collected for diagnostic
purposes at the department of Ophthalmology of the University Medical
Center Utrecht, The Netherlands, a secondary and tertiary
ophthalmologic care center. All clinical data at the time of aspiration
were registered, including specific diagnosis and duration of uveitis,
presence and duration of CME, and systemic medications. Patients with
a laboratory-diagnosed intraocular infection were considered to have
infectious uveitis. CME, papillary leakage, retinal vasculitis, choroidal
leakage, ischemia and (active) retinal lesions were evaluated on
fluorescein angiograms (FAs). FAs performed between 6 months before
and 6 months after the aqueous tap were considered suitable for
examination, provided no change in medication had occurred and
uveitis activity remained stable. When an FA was not available we
evaluated clinical data based on slit lamp and fundoscopic examinations
at the time of the aqueous sampling. Eleven patients were excluded
from the final analysis because of the lack of clinical data. When a
33
patient had more than one aqueous humor tap, 1 sample was
randomly chosen for analysis. One hundred twelve samples and
patients were available for analysis. Controls consisted of 20 aqueous
samples from patients with age-related cataract without uveitis or
other associated ocular disease. This study was performed according to
the tenets of the Declaration of Helsinki.
Sample collection and processing

The ocular fluid samples taken at the uveitis clinic of the UMCU were
stored at –80°C in sterile screw-cap tubes within 5 hours of collection
prior to processing for laboratory analysis.
Determination of intraocular IgG production concentrations

Total intraocular IgG concentrations were determined by an in-house
ELISA as described previously (de Groot-Mijnes et al. forthcoming).
Briefly, flat-bottom microtiter plate wells (Nunc) were coated by
incubation with 100 µL of PBS containing 1 µg/mL goat F(ab’)2 anti-
human IgG (Southern Biotechnology Associates) at 4°C for 16 hours.
The wells were washed four times with 400 µL of PBS-Tw20.
Subsequently, serial two-fold, followed by four serial four-fold dilutions
(1/500 to 1/1024000) of the aqueous humor samples were prepared in
phosphate buffered saline (PBS) supplemented with 0.05% Tween® 20
(PBS-Tw20) and 2% Protifar plus (Nutricia) and 100 µl was added to
the pre-coated wells. To be able to calculate the IgG concentration,
seven serial two-fold dilutions (1/50000 to 1/3200000) of a
nephelometer standardized human serum sample (Peliclass human IgG
subclass kit, Sanquin) were included. Incubation was for 1 hour at
37ºC. The wells were washed four times with 400 µL of PBS-Tw20.
Then, 100 µL of horseradish peroxidase-conjugated goat anti-human
IgG (Southern Biotechnology Associates) was added to each well at a
dilution of 1/5,000 to 1/15,000, as had been determined in advance for
every new batch. After a 1-hour incubation at 37 °C, the wells were
washed three times with 400 µL of PBS-Tw20 and twice with 350 uL
demineralized water. Finally, 100 µL of TMB Super Slow 1-C HRP
Microwell Substrate (BioFX) was added to each well. After 10 minutes
the reaction was stopped by adding 100 µL of 0.5M of sulphuric acid
and the absorption was measured at 450-nm.
34
Chapter 3
Data analysis was performed using SPSS (version 12.0). The log
transformed IgG concentration was compared between all groups with
means of ANOVA with a post hoc Bonferroni test. A multivariate
analysis was performed using a linear regression analysis submitting all
variables that showed statistical significance (P < .05) The chi-square,
or if applicable the Fisher’s exact test, was used to compare categorical
variables. Results are presented as the geometric mean which is the
antilog of the mean of the log transformed IgG concentrations.
Results
We reviewed 112 aqueous humor samples for total IgG concentration.
General data of the patients are given in Table 1. The average age was
45 years (range 11 to 82 years). Fifty-seven (51%) patients were
male. Forty (36%) patients had uveitis anterior, 3 (3%) had
intermediate uveitis, 39 (35%) had uveitis posterior, 28 (25%)
panuveitis and 2 (2%) had scleritis. Of these 112 patients, 34 (30%)
suffered from CME at the time of the aqueous humor tap.
Table 1: General characteristics of uveitis patients
Number of patients 112

Average age in years (range) 45 (11 – 82)
Male-to-female ratio 57 : 55
Number of patients with duration of uveitis (%):

Less than 3 months 44 (39%)
Between 3 and 12 months 25 (22%)
More than 12 months 43 (38%)
Anatomic location of uveitis (%):

Anterior 40 (36%)
Intermediate 3 (3%)
Posterior 39 (35%)
Pan 28 (25%)
Scleritis 2 (2%)
Patients with cystoid macular edema (CME) (%) 34 (30%)

CME duration < 3 months (%) 18 (53%)
CME duration 3 to 12 months (%) 10 (29%)
CME > 12 months (%) 6 (18%)
35
Table 2: Geometric mean IgG concentrations and correlations of general and

ophthalmological features
Geometric mean IgG (mg/mL) Multivariate

Univariate linear regression
analysis analysis
Present (n) Absent (n) P-value* P-value*
General data:
Male gender 0.06 (57) 0.08 (55) .20
Age (years): n.a.† n.a. † .50
Diabetes 0.09 (6) 0.07 (106) .65
Ophthalmological data:
Cystoid macular edema 0.16 (34) 0.05 (78) <.001 .017
Papillary leakage 0.11 (52) 0.04 (57) <.001 .61
Vasculitis 0.14 (26) 0.05 (84) .004 .39
Choroidal leakage 0.13 (24) 0.07 (35) .09
Retinal spots 0.06 (19) 0.06 (80) 1.00
Retinal ischemia 0.42 (10) 0.05 (101) <.001 .001
Infectious uveitis 0.09 (26) 0.06 (86) .33
Location of uveitis:
Anterior 0.05 (40) 0.07 (72) .20
Intermediate 0.06 (3) 0.07 (109) .98
Posterior 0.06 (39) 0.07 (73) .73
Panuveitis 0.10 (28) 0.06 (84) .14
Scleritis 0.21 (2) 0.07 (110) .29
Uveitis duration:
< 3 months 0.06 (44) 0.07 (68) .64
3 to 12 months 0.07 (25) 0.06 (87) .76
> 12 months 0.07 (43) 0.06 (69) .84
Cystoid macular edema duration:

< 3 months 0.15 (18) 0.16 (16) .15
3 to 12 months 0.13 (10) 0.17 (24) .33
> 12 months 0.23 (6) 0.14 (28) .22
*
ANOVA test
†
not applicable
The geometric mean IgG concentrations in aqueous humor and their

associations is given in Table 2. The geometric mean IgG concentration
in the non-uveitis control samples (0.01 mg/mL) was significantly lower
than in uveitis eyes (0.07 mg/mL; P < .001) and in all subgroups (P ≤
.04), except for patients with intermediate uveitis (3 patients; P =
.23). In the eyes of patients with CME, papillary leakage, vasculitis or
retinal ischemia, the IgG concentration was significantly higher than in
patients lacking these ophthalmological features (all P-value were
≤.004). No association was found between a higher IgG concentration
36
Chapter 3
Table 3: Correlations of cystoid macular edema (CME) with general and ophthalmological
features.
P-value (2-sided)* Linear regression
General data:
Female gender .04 .27
Age (years):
<30, <40 n.s. †
>50 .01 .023
Diabetes .11
Ophthalmological data:
Papillary leakage <.001 .17
Vasculitis .05 .23
Choroidal leakage .003 .043
Retinal spots .78
Retinal ischemia .06
Infectious uveitis .47
Anterior <.001‡ .38
Intermediate .17
Posterior .39
Panuveitis .02 .13
Scleritis .54
Uveitis duration:
< 3 months .40
3 to 12 months .05 .11
> 12 months .39
* chi-square or Fisher exact test.

‡
Uveitis anterior was inversely correlated with CME.
†
not significant.
and the duration of uveitis or CME. Linear regression showed that CME
(P = .017) and retinal ischemia (P = .001) were independent predictors
of a higher IgG concentration.
The association between CME and other ophthalmological features
is given in Table 3. CME was correlated with papillary leakage (P <
.01), vasculitis (P = .05), choroidal leakage (P < .01), female gender
(P = .04), age above 50 years (P = .01), panuveitis (P = .02) and a
uveitis duration of more than 3 months (P = .05). Uveitis anterior was
inversely correlated with CME. Linear regression showed that choroidal
leakage (P = .043) and age above 50 (P = .023) were independent
predictors of CME.
37
Discussion
We found a strong association between elevated IgG concentrations
and CME as well as retinal ischemia (P = .017 and P = .001,
respectively) in the aqueous humor of patients with uveitis. The IgG
concentration was higher for patients with uveitis compared to the
controls (P ≤ .04) irrespective of the anatomic type, except for patients
with intermediate uveitis. The latter might be explained by the small
sample size of patients with intermediate uveitis (n = 3). In addition,
we found an association between CME and choroidal leakage (P = .043)
and CME and age above 50 years (P = .023).
The healthy BRB is impermeable for plasma proteins. The blood-
aqueous barrier (BAB) is partly permeable for plasma proteins.
Therefore small amounts of IgG can be found in the aqueous of healthy
eyes.4 IgG levels in the aqueous humor of uveitis patients were
previously investigated by Murray et al.5 The authors found a mean IgG
concentration of 0.03 mg/mL in control patients with age-related
cataract, which is similar to our findings (mean 0.02 mg/ml).5
The elevated IgG concentrations in the aqueous humor of patients
with retinal ischemia and CME can be logically argued. Ischemia causes
damage to the retinal vasculature, resulting in a free flow of proteins
from the peripheral blood into the interior of the eye. The elevated IgG
concentrations in patients with inflammatory CME can be expected
since CME is the result of the breakdown of the outer and inner BRB.
One would expect that proteins smaller than IgG could also be found in
the aqueous humor of patients with uveitis with or without CME. Bloch-
Michel et al. found elevated aqueous humor levels of IgG and albumin
in patients with herpetic uveitis compared to controls.2 The elevated
IgG concentrations in the aqueous humor of patients with uveitis can
be explained by the disruption of the BRB and/or BAB and/or
intraocular IgG production. Since the two most important factors
associated with high IgG levels in patients with inflammatory CME
(CME itself and retinal ischemia) are located in the retina, the
disruption of the blood-retina barrier is the most likely cause of
elevated IgG concentrations. Additional studies on leakage of plasma
proteins into the interior of the eye of patients with CME will further
improve our understanding of the mechanism of BRB breakdown.
38
Chapter 3
Previously, others noted an association between the prolonged

duration of uveitis and the development of CME.6,7 Surprisingly, we did
not find a correlation between the duration of the uveitis and CME.
However, we did find an association between advanced age and CME.
This could be explained by a decreased function and/ or degeneration
of retinal cells during life, a process which may be enhanced by an
intra-ocular inflammation.8 Age-related deterioration of the vascular
wall, which is associated with increasing leakage of fluid through the
vessels, might also play a role.8
We found a correlation between CME and choroidal leakage but
not between choroidal leakage and the IgG concentration. Possibly, the
damaged retinal pigment epithelium, which is a part of the outer BRB,
might be selectively permeable for fluid originating from the choroidal
layers, but not for large molecules like IgG. Thus, choroidal leakage
may add to the edema, but not to the IgG concentrations.
Our study shows that patients with uveitis have elevated levels of
intraocular IgG, particularly those with CME, retinal ischemia or both.
These observations indicate that a damaged BRB leads to both an
accumulation of fluid in the macula and increased IgG concentration.
The finding of associations between high intraocular IgG levels and
CME and/ or retinal ischemia in patients with uveitis improves our
understanding of the pathogenesis of CME. The clinical relevance and
utility require further investigation.
39
References
1. Dernouchamps JP. The proteins of the aqueous humour. Doc Ophthalmol
1982;53:193-248.
2. Bloch-Michel E, Lambin P, Debbia M, Tounsi Y, Trichet C, Offret H. Local
production of IgG and IgG subclasses in the aqueous humor of patients with
Fuchs heterochromic cyclitis, herpetic uveitis and toxoplasmic chorioretinitis. Int
Ophthalmol 1997;21:187-194.
3. Stur M, Grabner G, Dorda W, Zehetbauer G. The effect of timolol on the
concentrations of albumin and IgG in the aqueous humor of the human eye. Am
J Ophthalmol 1983;96:726-729.
4. Freddo TF. Shifting the paradigm of the blood-aqueous barrier. Exp Eye Res
2001;73:581-592.
5. Murray PI, Hoekzema R, Luyendijk L, Konings S, Kijlstra A. Analysis of aqueous
humor immunoglobulin G in uveitis by enzyme-linked immunosorbent assay,
isoelectric focusing, and immunoblotting. Invest Ophthalmol Vis Sci
1990;31:2129-2135.
6. Dana MR, Merayo-Lloves J, Schaumberg DA, Foster CS. Prognosticators for visual
outcome in sarcoid uveitis. Ophthalmology 1996;103:1846-1853.
7. Lardenoye CW, van Schooneveld MJ, Treffers WF, Rothova A. Grid laser
photocoagulation for macular oedema in uveitis or the Irvine-Gass syndrome. Br
J Ophthalmol 1998;82:1013-1016.
40
Chapter 3
Chapter 4
Distinct cytokine and

chemokine profiles in the
aqueous of patients with uveitis
and cystoid macular edema
Bram van Kooij1 MD, Aniki Rothova1 MD, PhD, Ger Rijkers2, PhD,
Jolanda de Groot – Mijnes1,3 PhD.
1
F.C. Donders Institute of Ophthalmology, 2Department of Pediatric
Immunology, Wilhelmina Children's Hospital and, 3Department of
Virology, Eijkman-Winkler Center, University Medical Center Utrecht,
Utrecht, The Netherlands.
41
Mediator profiles in the aqueous of patients with uveitis and CME
Abstract
Purpose
To investigate the global cytokine and chemokine expression pattern in
the aqueous humor (AH) of uveitis patients and relate them to clinical
features.
Methods
In a cross-sectional study, we measured in 31 AH samples from uveitis
patients, the concentration of mediators by a multiplex immunoassay.
Eleven control samples were included.
Results
Uveitis samples had higher levels of interleukin-6, interleukin-8, soluble
intercellular adhesion molecule, soluble vascular cell adhesion molecule
(sVCAM), and interferon-inducible protein-10 (IP-10) than non-uveitis
controls. Active uveitis samples had higher levels of interleukin-6,
interleukin-10, interferon-γ, sVCAM, RANTES, and IP-10 than quiescent
uveitis samples. Infectious uveitis was associated with higher levels of
interleukin-10 than non-infectious uveitis (P<.03 for all subgroups). No
significant differences were found between CME and non-CME samples.
Conclusions
Elevated levels of specific mediators were found in active and in
infectious uveitis, but not in CME. Mediator profiles might lead to a
better understanding of the pathogenesis of uveitis
42
Chapter 4
Cytokines, chemokines and soluble adhesion molecules modulate

immune reactions, they influence the communication between various
cell types and can alter the properties of the vascular endothelium. To
investigate the role of cytokines, chemokines and adhesion molecules
in uveitis, thirty one diagnostic aqueous humor (AH) samples from
patients with uveitis, which were stored at –80°C in sterile screw-cap
tubes within 5 hours of collection, were analyzed for 16 mediators
(Table) in a multiplex immunoassay as described previously.1 Controls
consisted of 11 AH samples from patients with age-related cataract
without uveitis or other associated ocular disease. This study was
performed according to the tenets of the Declaration of Helsinki.
The findings were related to several clinical features, such as
intraocular infection, activity of inflammation and cystoid macular
edema (CME). Only patients with a laboratory confirmed intraocular
infection were considered to have infectious uveitis. The activity of
uveitis at the time of sampling was registered and the presence of CME
was evaluated on fluorescein angiograms. Concentrations above or
below the detection limit were given as the highest or lowest detectable
value. Data analysis was performed using SPSS (version 12.0). Results
are presented as the geometric mean and the median (Table). The
mediator levels were compared between all groups with the Mann
Whitney U test. A P-value < .05 was considered to be significant.
Patients with uveitis had higher levels of interleukin-6 (IL; P
<.001), IL-8 (P=.002), soluble intercellular adhesion molecule (sICAM;
P=.002), soluble vascular cell adhesion molecule (sVCAM; P=.003),
and interferon-inducible protein-10 (IP-10; P<.001) compared to non-
uveitic controls (Table). During active uveitis, higher levels of IL-6
(P=.020), IL -10 (P=.028), interferon-γ (IFN-γ; P=0.030), sVCAM
(P=.009), RANTES (P=.016), and IP-10 (P<.001) were found than
during quiescent uveitis. Patients with infectious uveitis had higher
levels of IL-10 (P=.034) than patients with non-infectious uveitis. No
significant differences were found between uveitis patients with or
without CME
43
Table: Mediator concentrations in the aqueous humor of patients with uveitis and controls
44
active quiescent infectious non infectious
Uveitis Controls P- CME non CME P- disease disease P- uveitis uveitis P-
n=31 n=11 value n=12* n=17* value n=17 n=14 value n=11 n=20 value
Mean age
(years) 50 66 57 43 44 56 52 49
Geometric Mean Concentration, Median, (pg/ml)

(range)
IL1β 2.6, <2.5 2.5, <2.5 .394 2.5, <2.5 2.6, <2.5 .401 2.8, <2.5 2.5, <2.5 .192 2.7, <2.5 2.6, <2.5 .698
(<2.5-6.7) (<2.5) (<2.5) (<2.5-6.7) (<2.5-6.7) (<2.5) (<2.5-5.2) (<2.5-6.7)
IL-2 6.9 <5.6 5.6, <5.6 .217 6.0, <5.6 7.0, <5.6 .706 8.1, <5.6 5.6, <5.6 .056 6.7, <5.6 7.0, <5.6 .671
(<5.6-41) (<5.6) (<5.6-12) (<5.6-41) (<5.6-41) (<5.6) (<5.6-34) (<5.6-40)
IL-4 2.6, <2.3 2.3, <2.3 .291 2.3, <2.3 2.5, <2.3 .227 2.8, 2.3 2.3, <2.3 .104 2.9, <2.3 2.5, <2.3 1.00
(<2.3-21.3) (<2.3) (<2.3) (<2.3-3.9) (<2.3-21) (<2.3) (<2.3-21) (<2.3-3.9)
IL-6 637, 254 38, 47 < .001 338, 221 693, 240 .756 1413, 1100 242, 138 .020 676, 779 615, 246 .708
(39->13821) (<7.0-240) (70->13822) (39.47->13821.81) (78->13822) (39->13822) (39->13822) (62->13822)
IL8 43, 21 12, <9.8 .002 21, 17 42, 20 .498 73, 41 23, 11 .076 75, 41 32, 19 .488
(<9.8->4544) (<9.8-79) (<9.8-94) (<9.8-1693) (<9.8->4544) (<9.8-362) (<9.8->4544) (<9.8-1693)
IL-10 17, <8.1 8.1, <8.1 .066 9.9, <8.1 14, <8.1 .211 27, <8.1 9.0, <8.1 .028 42, <8.1 10, <8.1 .034
(<8.1->4975) (<8.1) (<8.1-82) (<8.1-503) (<8.1->4975) (<8.1-32) (<8.1->4975) (<8.1-39)
IL-12p70 6.6, <6.4 6.4, <6.4 .551 6.4, <6.4 6.4, <6.4 1.00 6.8, <6.4 6.4, <6.4 .364 7.1, <6.4 6.4, <6.4 .178
(<6.4-20) (<6.4) (<6.4) (<6.4) (<6.4-20) (<6.4) (<6.4-20) (<6.4)
IL-13 2.6, <2.1 2.2, <2.1 .666 2.3, <2.1 2.6, <2.1 .706 3.0, <2.1 2.1, <2.1 .056 2.5, <2.1 2.6, <2.1 .671
(<2.1-16) (<2.1-2.4) (<2.1-5.2) (<2.1-16) (<2.1-16) (<2.1) (<2.1-14) (<2.1-16)
IL-18 8.1, <5.0 8.0, <5.0 .756 6.3, <5.0 8.7, <5.0 .296 9.3, <5.0 6.9, <5.0 .626 7.0, <5.0 8.8, <5.0 .184
(<5.0-74) (<5.0-23) (<5.0-23) (<5.0-36) (<5.0-79) (<5.0-23) (<5.0-78) (<5.0-36)
Table: Mediator concentrations in the aqueous humor of patients with uveitis and controls (continued)
active quiescent infectious non infectious
Uveitis Controls P- CME non CME P- disease disease P- uveitis uveitis P-
n=31 n=11 value n=12* n=17* value n=17 n=14 value n=11 n=20 value
Geometric Mean Concentration, Median, (pg/ml)

(range)
IFNγ 5.0, <4.1 4.1, <4.1 .162 4.1, <4.1 5.2, <4.1 .076 6.0, <4.1 4.1, <4.1 .030 5.4, <4.1 4.9, <4.1 .821
(<4.1-52) (<4.1) (<4.1) (<4.1-25) (<4.1-52) (<4.1) (<4.1-52) (<4.1-25)
TNFα 2.4, <2.1 2.1, <2.1 .217 2.2, <2.1 2.3, <2.1 .738 2.7, <2.1 2.1, <2.1 0.56 2.5, <2.1 2.4, <2.1 .723
(<2.1-15) (<2.1) (<2.1-4.0) (<2.1-7.2) (<2.1-15) (<2.1) (<2.1-15) (<2.1-7.2)
sICAM 440, 401 106, 64 .002 373, 290 428, 401 .507 625, 448 287, 277 .104 379, 401 478, 396 .869
(34-3806) (30-665) (136-1566) (34-3806) (98-3806) (34-1133) (34-3045) (98-3806)
sVCAM 813, 746 269, 304 .003 598, 629 942, 1157 .199 1285, 1157 466, 420 .009 802, 892 819, 712 .967
(156->6531) (60-697) (189-2472) (156->6531) (249->6531) (156-1553) (156-4312) (189->6531)
RANTES 95, 85 73, 73 .674 55, 38 120, 135 .067 148, 154 55, 33 .016 103, 85 90, 67 .659
(<27-1016) (<27-248) (<27-343) (<27-1016) (<27-1016) (<27-331) (<27-745) (<27-1016)
Eotaxin 5.1, <4.6 4.6, <4.6 .217 4.8, <4.6 5.2, <4.6 .769 5.5, <4.6 4.6, <4.6 .056 5.9, <4.6 4.7, <4.6 .065
(<4.6-28) (<4.6) (<4.6-6.6) (<4.6-28) (<4.6-28) (<4.6) (<4.6-28) (<4.6-5.7)
IP-10 836, 911 83, 134 < .001 585, 708 978, 911 .268 1955, 1834 298, 202 <.001 1487, 2693 609, 708 .076
(40->7384) (<5.0-448) (40-5183) (62->7384) (324->7384) (40->7384) (62->7384) (40->7384)
CME = Cystoid macular edema, IL = interleukin, INF = interferon, TNF = tumor necrosis factor, sICAM = soluble intercellular adhesion molecule, sVCAM = soluble vascular cell
adhesion molecule, IP-10 = interferon-inducible protein 10
* CME could not be evaluated in two patients
45
Chapter 4
Our findings confirm and extend previous findings of elevated levels of

IL-6, IL-8, and IFN-γ in patients with uveitis.2,3 Elevated IL-10 levels in
intraocular infections were also noted by others4, but in contrast to our
study, these levels were compared to non-uveitis controls. Our data
show that IL-10 may be an indicator for an active and/or infectious
uveitis. Within these subgroups there is variation in the IL-10 levels,
which warrants more in depth analysis of the microbiological causes of
the infection.
Elevated IL-6 and sICAM levels previously were reported in
patients with CME associated with diabetic retinopathy and branch
retinal vein occlusion.5,6 In contrast, in our patients with uveitis
associated CME, we found no differences in the levels of these
mediators, nor in the other cytokines and chemokines analyzed. We did
not observe clear Th1 (IFN-γ, IL-2 and TNFα) or Th2 cytokine profiles
(IL-4) in patients with uveitis. In contrast, we found a strong
correlation of active uveitis with sVCAM and IP-10, both of which are
important mediators for leukocyte recruitment.7
The differences with previous findings might in part be explained
by the complex biological function of the mediators, transient
production during the course of the inflammatory process, with specific
uveitis entities and medications at the time of sampling.4
Clearly, future research of cytokines and other mediators involved
in specific types of uveitis is needed to elucidate the role of these
proteins in the inflammation process. As different mediators possess
biological overlapping functions and often influence the production of
each other, the ability to simultaneously monitor changes in multiple
cytokines would hold both a greater discriminatory as well as
instructive power than single cytokine analysis. Knowledge of these
local cytokine and adhesion molecule biosignatures, which can
nowadays easily be obtained even in small samples using the multiplex
immunoassay, might lead to a better understanding of the
pathogenesis of uveitis and CME, and help to design new therapies.
46
Chapter 4
References
1. de Jager W, Prakken BJ, Bijlsma JW, Kuis W, Rijkers GT. Improved multiplex
immunoassay performance in human plasma and synovial fluid following removal
of interfering heterophilic antibodies. J Immunol Methods 2005;300:124-135.
2. Curnow SJ, Falciani F, Durrani OM, Cheung CM, Ross EJ, Wloka K, Rauz S, Wallace
GR, Salmon M, Murray PI. Multiplex bead immunoassay analysis of aqueous humor
reveals distinct cytokine profiles in uveitis. Invest Ophthalmol Vis Sci
2005;46:4251-4259.
3. Wallace GR, John CS, Wloka K, Salmon M, Murray PI. The role of chemokines and
their receptors in ocular disease. Prog Retin Eye Res 2004;23:435-448.
4. Ongkosuwito JV, Feron EJ, van Doornik CE, Van der Lelij A, Hoyng CB, La Heij EC,
Kijlstra A. Analysis of immunoregulatory cytokines in ocular fluid samples from
patients with uveitis. Invest Ophthalmol Vis Sci 1998;39:2659-2665.
5. Funatsu H, Yamashita H, Sakata K, Noma H, Mimura T, Suzuki M, Eguchi S, Hori S.
Vitreous levels of vascular endothelial growth factor and intercellular adhesion
molecule 1 are related to diabetic macular edema. Ophthalmology 2005;112:806-
816.
6. Noma H, Minamoto A, Funatsu H, Tsukamoto H, Nakano K, Yamashita H, Mishima
HK. Intravitreal levels of vascular endothelial growth factor and interleukin-6 are
correlated with macular edema in branch retinal vein occlusion. Graefes Arch Clin
Exp Ophthalmol 2005.
7. Sorensen TL. Targeting the chemokine receptor CXCR3 and its ligand CXCL10 in
the central nervous system: potential therapy for inflammatory demyelinating
disease? Curr Neurovasc Res 2004;1:183-190.
47
48
Chapter 4
Chapter 5
Prognosticators of cystoid
macular edema in uveitis
patients
Kiki Probst1* MD, Bram van Kooij1* MD, Rob Fijnheer2 MD, PhD, Mark
Roest2 PhD, Wolter de Loos3 MD PhD †, Aniki Rothova1 MD, PhD.
*
These authors contributed equally to the work.
1
F.C. Donders Institute of Ophthalmology, 2Department of Hematology
and 3Department of Internal Medicine, University Medical Center,
Utrecht, The Netherlands.
49
Prognosticators of CME in uveitis patients
Abstract
Purpose
To determine ophthalmologic and systemic factors contributing to the
development of cystoid macular edema (CME) in patients with uveitis.
Methods
Retrospective cross-sectional study in which 97 consecutive patients
with uveitis filled in an extensive questionnaire for the presence of
cardiovascular diseases and its risk factors. An analysis of the
ophthalmologic and questionnaire data was conducted.
Results
CME was present in 44% (43/97) of patients. Its development was
strongly associated with increasing age (P = .001) and age at onset of
uveitis (P < .001). For patients older than 50 years, the risk of
developing CME was 3.8-fold larger than for younger patients. The
most frequent anatomic location of uveitis associated with CME was
panuveitis (49%). Papillary leakage on fluorescein angiography was
associated with CME (P < .001), independently of other risk factors.
After adjustment for age, multivariate logistic regression showed no
association between cardiovascular disease and its risk factors and the
development of CME.
Conclusions
Age, independent of duration of uveitis, was a major risk factor for the
development of CME in uveitis. A positive correlation between CME and
papillary leakage on angiography was noted.
50
Chapter 5
Introduction
The development of cystoid macular edema (CME) has a decisive effect
on the visual outcome of many ocular disorders including uveitis,
diabetic retinopathy and retinal vein occlusions.1 Despite aggressive
treatments, progression of CME with accompanying visual loss is
common, and photoreceptor dysfunction and/or loss often cannot be
prevented.2,3 The unsatisfactory visual prognosis in inflammatory CME
has led to a recent recommendation of early treatment and underlines
the need for improved management of this complication of uveitis.3 The
pathogenesis of inflammatory CME is largely unknown, as are the
factors which determine either visual recovery or irreversible loss of
vision in an individual patient with CME, clarified. Several studies
pointed out the increased frequency of systemic cardiovascular
disorders (CVD) in CME due to diabetic retinopathy, central retinal vein
occlusion and post intra-ocular surgery.4-8 Recent reports showed that
systemic microvascular damage, which is a risk factor for developing
CVD, was associated with inflammatory and diabetic CME.9,10 Herein
we evaluate the impact of age, presence of CVD, CVD risk factors and
ophthalmologic data in uveitis patients with and without CME.
Methods
This study included 100 consecutive patients with uveitis who consulted
our uveitis clinic. All patients answered an extensive questionnaire,
developed for screening of cardiovascular diseases and used in the
department of vascular medicine.11 Three uveitis patients were left out
of the analysis, because of incomplete data sets.
The variables collected were the general medical history,
medication, age, age at onset of uveitis, duration of uveitis and CME,
family history of cardiovascular disorders, presence of CVD and its risk
factors. The diagnosis of cardiovascular disease included a history of
myocardial infarction, angina pectoris, coronary arterial bypass graft,
aortic aneurysm, ischemic stroke, transient ischemic attacks, carotid
artery occlusion and intermittent claudication. Risk factors for CVD in
this study were hypertension, hypercholesterolemia, smoking, pack-
years and higher BMI.11 Pack-years were calculated by multiplying the
duration of smoking by the number of cigarettes smoked per day,
51
divided by 20.11 Patients with diabetes mellitus were excluded from this
study; therefore this risk factor was not included in our calculations.
In addition, we conducted a retrospective analysis of the
ophthalmologic data of all patients with uveitis, with the emphasis on
specific diagnosis and location of uveitis, presence, onset and severity
of CME, the presence of vasculitis, papillary leakage, retinal occlusions
or multiple retinal hemorrhages. CME was defined by both clinical
and/or angiographic criteria.12,13 The fluorescein angiograms (n=83)
were evaluated by an experienced, masked ophthalmologist.
For statistical analysis (Chi-square, Kruskall-Wallis, Mann-Whitney
U tests for non-parametric analysis, univariate and multivariate logistic
regression) SPSS for Windows 11.0 statistical package (SPSS Inc.
Chicago, USA) was used.
Results
Ophthalmologic and cardiovascular data of our uveitis patients are
given in Table 1. The mean age at participation in our study was 47.5
years (range 12.3 – 82.5; median 45.3) and the mean follow-up (date
of onset of uveitis until the date of completing the questionnaire) 8.4
years (range 0.0 – 40.9, median 6.6). The male-to-female ratio was 1 :
1.9 (34/63).
Overall analysis of our group of 97 uveitis patients showed CME to
be present in 43/97 (44%), vasculitis in 41/85 (48%), papillary
leakage in 41/77 (53%) and retinal occlusions in 10/78 (13%; Table
1). In patients with CME, panuveitis was the most frequent anatomic
location of uveitis (49%, P = .001). Uveitis patients with CME also
exhibited papillary leakage on fluorescein angiography (25/38 vs.
16/39, P < .001) more often. The presence of CME was strongly
associated with a higher age of our patients (54.7 vs. 41.8 years, P <
.001). Uveitis patients with CME were older at the onset of uveitis
(47.1 vs. 32.9, P < .001) and, in univariate analysis, showed higher
BMI (25.0 versus 23.6, P = .045) and more pack-years (14.2 versus
5.1 years, P = .002; Table 1). The duration of uveitis was not
associated with the development of CME (P = 0.40). The prevalence of
CVD among patients with or without CME did not differ significantly
(7/43; 16% versus 3/54; 6%; P = .10).
52
Chapter 5
Table 1: General and ophthalmological characteristics of uveitis patients, with and

without cystoid macular edema (CME).
Total CME+ CME- p-value
Number of patients 97 43 (44%) 54 (56%)

Male / female 1 : 1.9 1 : 1.5 1 : 2.2 n.s.
Mean age (years) 47.5 54.7 41.8 < .001
Mean age at onset of uveitis (years) 39.2 47.1 32.9 < .001
Mean duration of uveitis (years) 8.4 7.6 9.0 n.s.
Uni-/bilateral disease 22/74 7/36 15/38 n.s.
Anterior 7(7%) 0 (0%) 7 (13%) .010
intermediate 14 (15%) 8 (19%) 6 (11%) n.s.
posterior 43 (46%) 14 (33%) 29 (54%) .020
panuveitis 30 (32%) 21 (49%) 9 (17%) .001
Retinal changes:
Vasculitis 41/85 (48%) 24/41 (59%) 17/44 (39%) n.s.
Papillary leakage 41/77 (53%) 25/38 (66%) 16/39 (41%) < .001
Retinal occlusions 10/78 (13%) 7/38 (18%) 3/40 (8%) n.s.
Retinal hemorrhages 19/80 (24%) 14/39 (36%) 5/41 (12%) n.s.
Cardiovascular data:
Cardiovascular disease 10 (10%) 7 (16%) 3 (6%) n.s.
Hypertension 13 (13%) 8 (19%) 5 (9%) n.s.
Hypercholesterolemia 12 (12%) 6 (14%) 6 (11%) n.s.
Current smoking 28 (29%) 12 (43%) 16 (30%) n.s.
Mean pack- years 9.2 14.2 5.1 .002
Mean body mass index 24.2 25.0 23.6 .045
* Univariate analysis of CME+ versus CME- uveitis patients
The prevalence of CVD in our overall uveitis group was 10/97 (10%)
and was associated with advancing age (average 67.3 years in patients
with CVD vs. 45.2 years in non-CVD patients, P < .001); higher age at
onset of uveitis (58.4 vs. 37.0 years; P = .001); systemic hypertension
(10/87, 11% in non-CVD patients; 3/10, 30% in CVD patients; P =
.04) and smoking (21/87, 24% in non-CVD patients; 7/10, 70% in CVD
patients; P = .003). CVD was not associated with gender (p = 0.1);
duration (8.3 years in non-CVD patients versus 9.0 in CVD patients; P
= .80) or location of uveitis, nor with clinical symptoms CME (37/87,
43% in non-CVD patients; 7/10, 70% in CVD patients; P = .10), retinal
vasculitis (30/70, 43% in non-CVD patients; 6/10, 60% in CVD
patients; P = .60), papillary leakage (36/68, 53% in non-CVD patients;
53
5/9, 56% in CVD patients; P = .90), retinal occlusions (6/67, 9% in

non-CVD patients; 2/9, 22% in CVD patients; P = .30) or hemorrhages
(13/70, 19% in non-CVD patients; 4/10, 40% in CVD patients; P =
.20).
Multivariate logistic regression showed that age at onset and
papillary leakage were the most important independent predictors of
CME (Table 2). There was a trend indicating retinal occlusions as an
independent risk factor for the development of CME (odds ratio 9.9;
95% CI. 0.7-141.7; Table 2), but due to the lack of statistical power
this finding did not reach significance. The relationship between BMI,
pack-years, other CVD factors, location of uveitis (panuveitis) and CME
diminished after adjustment for age.
Table 2: Multivariate logistic regression analysis to obtain independent predictors of

cystoid macular edema*
Significance Odds Ratio ( 95% C.I.)
Age at onset (years) <.01 1.09 (1.0 - 1.1 )

Papillary leakage .01 6.06 (1.4 - 25.6)
Retinal occlusions .09 9.9 (0.7 - 141.7)
C.I., confidence interval.
* By stepwise backward binary logistic regression.
Uveitis patients older than 50 years (n = 41), had a 3.8-fold higher risk
of developing CME than patients younger than 50 (n = 56; 95% CI 1.6-
9.0) and patients above 70 (n = 12) had even a 4.5-fold higher risk
(95% CI 1.2-19.6). In patients above 50 who also exhibited papillary
leakage, the prevalence of CME was 81% (13/16). Patients younger
than 50 with concomitant papillary leakage, showed CME in 48%
(12/25), whereas in younger patients without papillary leakage, the
prevalence of CME was only 11% (2/19, P = .008). CME in patients
above 50 with panuveitis was noted in 11/13 (85%) of patients in
contrast to 14/28 (50%) in patients without panuveitis (P = .03).
54
Chapter 5
Discussion
We observed a strong association between advancing age and
development of CME in uveitis patients. The risk of developing CME in
patients older than 50 years was 3.8-fold higher compared to those
younger than 50 years. This association was independent of the
duration of CME. In addition, papillary leakage was an independent
predictor of increased CME risk. CME in uveitis was also associated with
increased BMI, prolonged heavy smoking and location of uveitis
(panuveitis), but these associations diminished after adjustment for
age.
Determinants of early development of CME have not been
systematically investigated in the past. Incidental reports showed
several characteristics which preceded the development of CME in
uveitis, including decreased contrast sensitivity, abnormal color vision
and photoreceptor dysfunction and/or loss.2,14 The prevalence of CME
has further been associated with specific uveitis entities including pars
planitis, Behçet’s syndrome, birdshot chorioretinopathy, and peripheral
multifocal choroiditis.15-17 In this present study, the most important
prognosticator for the development of inflammatory CME was the
advancing age of the patient, which seemed to be independent of the
duration of uveitis.
An association between the prolonged duration of uveitis and the
development of CME has been noted previously by others;18,19
however, the effect of the absolute age of the uveitis patients on CME
was not investigated. Surprisingly, no association between the duration
of the uveitis and CME was noted in our study. This contradiction with
previous reports can be explained by the large number of patients with
chronic uveitis included in our series (the mean duration of uveitis in
our CME group was 7.6 and in our non-CME group 9.0 years). It is
probable that the duration of uveitis might influence the development
of CME, especially during the first months or years after the onset of
the intra-ocular inflammation. The influence of duration of the
inflammatory process on the development of CME probably decreases
in cases which had already become chronic (as in the majority of our
patients).
In the literature, the negative influence of age on the visual
outcome of CME has been described in retinal occlusions and diabetic
55
retinopathy, and was attributed to insufficiency of the vascular

endothelium with a subsequent increase of leakage of fluids, or inability
to reabsorb fluid in older patients.20-23 The explanation for this
phenomenon might lie in age-induced changes of retinal and choroidal
vessels as well as RPE abnormalities. Loss of endothelial cells and
vascular pericytes, RPE cell damage and decrease of integrity of the
blood-retinal barrier in the elderly were observed previously.24 In
normal aging, the rigidity of the small and medium-sized retinal vessels
increases, while the vascular density in the choroid and the
microcirculatory blood flow diminishes. With the increased size of the
foveal avascular zone and thinning of the choroid in the macular area,
a frail equilibrium develops in the elderly eye.24 Combined with events
induced by uveitis, the poor vascular resistance of the macula might
quickly lead to the development of manifest CME. In diabetic CME the
correlation with increasing age was attributed to a longer duration of
hyperglycemic damage to the microvasculature, with subsequent loss
of the blood-retina barrier, rather than to age-induced changes itself.
22,25
In our patients, we did not observe a significant correlation

between CVD and/or its risk factors except for age (and diabetes
mellitus, which was excluded in this series) and the development of
CME. An objective method to measure generalized vascular disease
would require microalbuminuria and/or intima-media thickness
measurements, which was beyond the scope of this study. In a recent
study, however, we reported that trace microalbuminuria was noted in
patients with uveitis and CME in contrast to uveitis patients without
CME. This indicates that systemic microvascular leakage is associated
with the development of CME in uveitis patients.10 Therefore, we
cannot entirely exclude the association between CVD and/or its risk
factors and inflammatory CME, also because our study included only a
limited number of patients with CVD. To analyze the actual impact of
CVD and its risk on the development and prognosis of CME, a large age
and gender-matched study of preferably elderly uveitis patients would
be required.
The use of a questionnaire to assess the presence of CVD is a
frequently employed and fairly reliable tool, but remains a subjective
method, relying on the patients’ co-operation and memory.26 We used
fluorescein angiography to analyze the inflammatory CME. In a
56
Chapter 5
condition of chronic CME, one may be looking for subtle changes that
can be picked up only on optical coherence tomography (OCT), which is
a more sensitive technique for evaluating macular thickness.
Unfortunately, OCT was not available for our study.
Unexpectedly, a firm association between papillary leakage and
CME was observed. Although the association of CME and optic disc
leakage has been reported following cataract extraction (Irvine-Gass
syndrome), it has - to our knowledge- not been reported for phakic
patients with uveitis. The pathogenesis of Irvine-Gass syndrome is
attributed to mechanical forces in the vitreous and attachments of the
vitreous around the macula and optic disc, as well as to an
inflammatory component). 27,28 In phakic uveitis patients, the
pathogenesis of concomitant optic disc leakage with CME is unknown,
but mechanical traction or adhesions of the vitreous might also play a
role, considering the improvement of inflammatory CME after pars
plana victrectomy.3,29 It is not probable, however, that mechanical
forces are the only explanation for concomitant CME and optic disc
leakage. An analysis of the natural history of the development of CME
would be necessary to clarify whether or when inflammatory CME is
preceded or accompanied by leakage of the optic disc.
Our study points out that elderly patients or patients with onset of
uveitis at advanced age and patients with optic disk leakage carry an
increased risk of developing CME. These observations open new
potentials for the management of patients with uveitis and might
improve their visual prognosis.
57
References
1. Tessler H, Lam S. Cystoid macular edema. In Pepose JS, Holland GN, Wilhelmus
KR, editors. Ocular infection and Immunity. St. Louis: Mosby, 1996:553-555.
2. Lardenoye CW, Probst K, DeLint PJ, Rothova A. Photoreceptor function in eyes
with macular edema. Invest Ophthalmol Vis Sci 2000;41:4048-4053.
2002;10:239-246.
4. Jain R, Stevens JD, Bunce CV, et al. Ischaemic heart disease may predispose to
pseudophakic cystoid macular oedema. Eye 2001;15:34-38.
8. Gilbert RE, Tsalamandris C, Allen TJ, et al. Early nephropathy predicts vision-
threatening retinal disease in patients with type I diabetes mellitus. J Am Soc
Nephrol 1998;9:85-89.
9. Knudsen ST, Bek T, Poulsen PL, et al. Macular edema reflects generalized
vascular hyperpermeability in type 2 diabetic patients with retinopathy. Diabetes
Care 2002;25:2328-2334.
10. van Kooij B, Fijnheer R, Roest M, Rothova A. Trace microalbuminuria in
inflammatory cystoid macular edema. Am J Ophthalmol 2004;138:1010-1015.
11. Simons PC, Algra A, Bots ML, et al. Common carotid intima-media thickness and
arterial stiffness: indicators of cardiovascular risk in high-risk patients. The
SMART Study (Second Manifestations of ARTerial disease). Circulation
1999;100:951-957.
12. Photocoagulation for diabetic macular edema. Early Treatment Diabetic
Retinopathy Study report number 1. Early Treatment Diabetic Retinopathy Study
research group. Arch Ophthalmol 1985;103:1796-1806.
13. 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. Ophthalmology
1987;94:761-774.
14. Ibanez HE, Lesher MP, Singerman LJ, et al. Prospective evaluation of the effect
of pseudophakic cystoid macula edema on contrast sensitivity. Arch Ophthalmol
1993;111:1635-1639.
15. Freeman G, Matos K, Pavesio CE. Cystoid macular oedema in uveitis: an
unsolved problem. Eye 2001;15:12-17.
16. Lardenoye CW, Van der Lelij A, de Loos WS, et al. Peripheral multifocal
chorioretinitis: a distinct clinical entity? Ophthalmology 1997;104:1820-1826.
17. Rothova A, Berendschot TT, Probst K, et al. Birdshot chorioretinopathy: long-
term manifestations and visual prognosis. Ophthalmology 2004;111:954-959.
18. Dana MR, Merayo-Lloves J, Schaumberg DA, Foster CS. Prognosticators for
visual outcome in sarcoid uveitis. Ophthalmology 1996;103:1846-1853.
58
Chapter 5
19. Lardenoye CW, van Schooneveld MJ, Treffers FW, Rothova A. Grid laser
J Ophthalmol 1998;82:1013-1016.
997.
21. Glacet-Bernard A, Coscas G, Chabanel A, et al. Prognostic factors for retinal vein
occlusion: prospective study of 175 cases. Ophthalmology 1996;103:551-560.
22. Klein R, Klein BE, Moss SE, et al. The Wisconsin epidemiologic study of diabetic
retinopathy. IV. Diabetic macular edema. Ophthalmology 1984;91:1464-1474.
23. Moss SE, Klein R, Klein BE. The 14-year incidence of visual loss in a diabetic
population. Ophthalmology 1998;105:998-1003.
25. Vitale S, Maguire MG, Murphy RP, et al. Clinically significant macular edema in
type I diabetes. Incidence and risk factors. Ophthalmology 1995;102:1170-
1176.
26. Klungel OH, de Boer A, Paes AH, et al. Cardiovascular diseases and risk factors
in a population-based study in The Netherlands: agreement between
questionnaire information and medical records. Neth J Med 1999;55:177-183.
27. Blair NP, Kim SH. Cystoid macular edema after ocular surgery. In Albert DM,
Jacobiec FA, editors. Principles and Practice of Ophthalmology. Basic Sciences.
Philadelphia: WB Saunders company, 1994:903-904.
28. Sebag J, Balazs EA. Pathogenesis of cystoid macular edema: an anatomic
consideration of vitreoretinal adhesions. Surv Ophthalmol 1984;28 Suppl:493-
498.
29. Bovey EH, Herbort CP. Vitrectomy in the management of uveitis. Ocul Immunol
Inflamm 2000;8:285-291.
59
60
Chapter 5
Chapter 6
Trace microalbuminuria in
inflammatory cystoid macular
edema.
Bram van Kooij1 MD, Rob Fijnheer2 MD, PhD, Mark Roest2 PhD, Aniki
Rothova1 MD, PhD.
1
F.C. Donders Institute of Ophthalmology and 2Department of
Hematology, University Medical Center, Utrecht, The Netherlands.
(American Journal of Ophthalmology, 2004 Dec;138(6):1010-1015)
61
Trace microalbuminuria in inflammatory CME
Abstract
Purpose
To assess the role of cardiovascular morbidity, its risk factors, and
microalbuminuria in the development of inflammatory cystoid macular
edema (CME).
Design
A matched case-control study.
Methods
Study population: We included 24 consecutive patients with uveitis and
CME. Twenty four uveitis patients without CME, matched for age and
duration of uveitis served as controls. Intervention and observation
procedures: Patients and controls were interviewed for the presence of
cardiovascular risk factors and cardiovascular morbidity. All
medications were registered. Morning urinary albumin concentration
was measured, as well as blood pressure, C-reactive protein, and
creatinine in blood. Patients suffering from diabetes mellitus were
excluded from this study. Main Outcome Measures: The presence of
cardiovascular morbidity and its risk factors and microalbuminuria in
uveitis patients with and without CME.
Results
We found a positive association between trace- and/or
microalbuminuria and inflammatory CME (P = .001; odds ratio 13.0, 95
% CI 2.5 to 68.1 and P = .015; odds ratio 5.9, 95 % CI 1.6 to 22.6),
but no relation between CME and cardiovascular morbidity or its risk
factors. No additional association between trace- and/or any
microalbuminuria and general characteristics of patients, specific
factors related to general disease as a cause of ocular inflammation,
location of uveitis, duration of uveitis, and medication was found.
Conclusion
The presence of trace- and/or microalbuminuria in inflammatory CME
might indicate the presence of early systemic vascular disease and
carry the risk of developing CME. This finding brings new insight into
the pathogenesis of CME and could open up new avenues for the
treatment of CME.
62
Chapter 6
Introduction
Cystoid macular edema (CME) is a major cause of visual impairment
(41%) and blindness in uveitis (29%).1,2 Additionally, a large survey on
uveitis indicated that 26% of all patients with uveitis develop CME.1
The treatment of CME is usually initiated when visual acuity drops
below 20/40, and consists of causal treatment of underlying disorder
and symptomatic treatment of CME. The pathogenesis of CME in uveitis
patients is not clear and the groups of patients with uveitis at risk of
developing severe CME have not yet been identified.
Cardiovascular diseases and retinal vascular changes are involved
in the pathogenesis of CME in patients with diabetes mellitus and
retinal vein occlusions, however the underlying mechanism is not yet
clarified.3-7 In particular, the CME in patients with diabetes mellitus is
strongly associated with hypertension and the presence of
microalbuminuria. 5,7-11 Microalbuminuria reflects the increased
permeability of damaged vascular walls. The role of microalbuminuria
as a marker of early vascular disease was recently examined not only
in diabetes mellitus, but also in non-diabetes mellitus patients.12-17 In
the nondiabetic population, microalbuminuria was an independent
indicator for cardiovascular risk factors and cardiovascular morbidity.
Microalbuminuria in patients with uveitis could point out the patients
with early vascular disease and might be associated with the
development of CME.
In this report, we analyze the possible association between
microalbuminuria, cardiovascular diseases, and inflammatory CME in
patients with uveitis.
Patients and Methods

This matched case-control study included 24 uveitis patients with CME
and 24 uveitis patients without CME, matched for age and duration of
uveitis. Consecutive patients with uveitis and CME were selected from
the uveitis clinic of the ophthalmologic department of the University
Hospital of Utrecht. For each patient, a control patient was selected
from the same clinic population with a similar age (± 5 years, mean ±
1.8 years) and duration of uveitis (subdivided in 0 ≤ 2, 2 ≤ 5, 5 ≤ 10,
and 10 > years) but without CME or previous CME. CME was diagnosed
using clinical criteria and fluorescein angiography in all cases.18 CME
63
was graded by a masked ophthalmologist on late phase fluorescein

angiography. Patients with no leakage were graded 0, those with
leakage less than 25% as grade 1, leakage between 25% and 66% was
graded as 2, and leakage of more than 66% was graded as 3.19
Patients were assessed for inflammation activity and cataract
extraction status at time of inclusion.
Patients and controls were interviewed for the presence of
cardiovascular disease using a questionnaire, including information on
weight, all medications, smoking habits, presence of systemic
hypertension, diabetes mellitus, hypercholesterolemia, cerebral
vascular accidents, transient ischemic attacks, angina pectoris,
myocardial infarction, abdominal aortic aneurysm, thrombosis, and
lung emboli. In addition to the patients with systemic hypertension
diagnosed in the past and already treated, the patients with 3
independent measurements with an interval of at least 1 week and a
diastolic blood pressure of 95 mm Hg or higher were also diagnosed as
having hypertension. Cardiovascular morbidity was defined as the
presence of or history of cerebral vascular accidents, transient ischemic
attacks, angina pectoris, myocardial infarction, and abdominal aortic
aneurysm.20 Cardiovascular risk factors were defined as the presence
or a history of hypertension, hypercholesterolemia, and current
smoking.20 Venous thrombo-embolic disease was scored separately.
Patients suffering from diabetes mellitus were excluded from this
study; so were patients with a non fasting blood glucose level of 7.0
mmol/l at time of uveitis screening.
Data analysis was performed using the Statistical Package for the
Social Sciences (SPSS 11.0). Differences between the patients and
controls were compared with the Fisher's exact test and a multivariate
logistic regression analysis. A P-value < .05 was considered to be
significant.
Blood samples were taken from 46 patients to determine C-
reactive protein (CRP) and creatinine. Two patients refused blood
sampling.
Morning urinary albumin excretion was determined in all by a
commercial immunoturbidimetry assay with a sensitivity of 5.5 mg/l
and inter- and intra-assay coefficients of variation of 5.2% and 13.2%
respectively (Dade Behring Diagnostics, Liederbach, Germany).
Morning urinary albumin excretion ≤ 5 mg/l was defined as no
64
Chapter 6
albuminuria, of 5 ≤ 20 mg/l was defined as trace microalbuminuria and

morning urinary albumin excretion of 20 ≤ 200 mg/l was defined as
microalbuminuria. 13,21
Creatinine clearance was calculated using the serum creatinine
levels as a basis with the Cockcroft-Gault equation.22,23
Results
General data from the CME patients and non-CME controls with uveitis
are given in Table 1. The average age was 45 years for the CME
patients and 45 years for the controls. The male-to-female ratio in the
CME group was 3:5, in the non-CME group 1:2. No differences in
gender, inflammation activity, and cataract extraction status (cataract
extraction 0 ≤ 2 years before inclusion) were found in both groups (P =
1.0; P = .77; P = .42).
Cardiovascular morbidity was present in 3 patients (6%) (2/24,
8% with CME, 1/24, 4% in non-CME group, P = .60). Cardiovascular
risk factors were present in 26 patients (54%) (14/24, 58% with CME,
12/24, 50% in non-CME group, P = .77) Hypertension was present in 5
patients (10%) (3/24, 12% with CME, 2/24, 8% in non-CME group, P =
1.0). Additionally, 19 patients (40%) were current smokers (11/24,
46% with CME, 8/24, 33% in non-CME group, P = .56).
Table 1. General characteristics of uveitis patients with and without cystoid macular
edema (CME).
CME Non-CME
n = 24 n = 24
Average age (range) 45 (11-74) 45 (12-69)

Male-to-female ratio 3:5 1:2
Location:
Anterior (%) 1 (4.2) 7 (29.2)
Intermediate (%) 6 (25.0) 2 (8.3)
Posterior (%) 5 (20.8) 7 (29.2)
Pan (%) 12 (50.0) 8 (33.3)
Association with systemic disease (%) 7 (29.2)* 8 (33.3)+
Recent systemic medication for uveitis and/or CME (%)++ 18 (75.0) 9 (37.5)
Medication for cardiovascular diseases or risk factors (%) 4 (16.7) 2 (8.3)
*
2 patients with sarcoidosis, 1 patient with HLA-B27+ ankylosing spondylitis, 1 patient with juvenile
idiopathic arthritis, 1 patient with Behçet's disease, 1 patient with granuloma annulare, 1 patient with
borreliosis; + 1 patient with sarcoidosis, 2 patients with HLA-B27+ ankylosing spondylitis, 2 patients
with juvenile idiopathic arthritis, 2 patients with Behçet's disease, 1 patient with multiple sclerosis; ++
Includes acetazolamide, corticosteroids, non-steroidal anti-inflammatory drugs, cytostatics.
65
Twenty-seven patients (56%) had systemic medication for uveitis

and/or CME during the sampling (18/24, 75% with CME, 9/24, 38% in
non-CME group P = .019). Six patients (13%) had medication for
cardiovascular disease or cardiovascular risk factors.
Trace microalbuminuria was found in 54% of the patients (13/24)
with CME in contrast to 8% of the patients (2/24) in non-CME group (P
= .001; odds ratio 13.0, 95 % CI 2.5 to 68.1); and any
microalbuminuria (5 ≤ 200 mg/l) was found in 13/24 versus 4/24
patients (P = .015; odds ratio 5.9, 95 % CI 1.6 to 22.6) (Table 2 and
Table 3). Microalbuminuria (≥ 20 mg/l) was found in 2 patients (8% in
non-CME group, not significant); one patient had IgG paraproteinemia
with low creatinine clearance of 59 ml/min, the other was treated for
10 years with cyclosporin A; his creatinine clearance was 74 ml/min.
Renal function, measured using creatinine clearance was determined in
46 patients and was below 80 ml/min in 12 patients (6 with CME and 6
in non-CME group controls, P = 1.0).
Table 2: General and laboratory characteristics of uveitis patients with and without
microalbuminuria
No micro- Trace- and any

albuminuria microalbuminuria P-value
≤ 5 mg/l (> 5 mg/l)
n=31 n=17
CME positive 11 (35.5) 13 (76.5) .015

Gender (male) 8 (25.8) 9 (52.9) ns
Age >50 years 10 (32.2) 9 (52.9) ns
Associated systemic disease 11 (35.8) 4 (23.5) ns
High blood pressure 2 (6.5) 3 (17.6) ns
Cardiovascular disease 1 (3.2)* 2 (11.8)+ ns
Current smoking 11 (35.5) 8 (47.1) ns
C-reactive protein above 7 mg/L++ 9 (29.0) 3 (17.6) ns
Systemic medication
Acetazolamide 4 (12.9) 5 (29.4) ns
Corticosteroids 5 (16.1) 6 (35.3) ns
Non-steroidal anti-inflammatory drugs 8 (25.8) 9 (52.9) ns
Cytostatics 7 (22.6) 5 (29.4) ns
Duration of uveitis ns
>5 years 26 (83.9) 13 (76.5) ns
>10 years 14 (45.2) 5 (29.4) ns
CME = cystoid macular edema, ns = not significant

*
Myocardial infarction; + Myocardial infarction and abdominal aneurysm aorta; ++
Two patients
were not tested.
66
Chapter 6
Trace- and/or microalbuminuria was not related to the presence of

cardiovascular risk factors (P = .37) and morbidity (P = .23),
hypertension (P = 1.0), smoking (P = .54), current and previous
treatment with cyclosporin A (P = 1.0), current medication with
corticosteroids (P = .16), acetazolamide (P = .25), cytostatics (P =
.73), NSAID's (P = .11) or to elevated CRP (p = .49), creatinine
clearance <80 (n = 46, P = .18), age > 50 years (P = .22), gender (P
= .11), specific cause of ocular inflammation, location of uveitis,
duration of uveitis, interval between the onset of uveitis and
development of CME (n = 20, P = .16), gradation of CME and presence
of systemic diseases (P = .52) (Table 2).
The correlation between trace- and/or microalbuminuria and CME
was not only found in elderly patients but also in younger patients.
Adjustment of gender, age, location of uveitis, activity of uveitis,
specific diagnosis, cardiovascular risk factors and morbidity, and
cataract extraction status did not change the correlation between
trace- and any microalbuminuria and CME.
Table 3: General and ocular characteristics of uveitis patients with and without cystoid
macular edema (CME)
CME Non-CME P-value

n = 24 n = 24
Trace microalbuminuria 13 (54.2) 2 (8.3) .001

Any microalbuminuria 13 (54.2) 4 (16.6) .015
Gender (male) 9 (37.5) 8 (33.3) ns
Age >50 years 9 (37.5) 10 (41.7) ns
Associated systemic disease 7 (29.2) 8 (33.3) ns
High blood pressure 3 (12.5) 2 (8.3) ns
Cardiovascular disease 2 (8.3)* 1 (4.2)+ ns
Current smoking 11(45.8) 8 (33.3) ns
C-reactive protein above 7 mg/L ++ 7 (29.2) 5 (22.7) ns
Systemic medication 18 9 .019
Acetazolamide 9 (36.0) 0 (0.0) .002
Corticosteroids 7 (29.2) 4 (16.6) ns
Non-steroidal anti-inflammatory drugs 13 (54.2) 4 (16.6) .013
Cytostatics 8 (33.3) 4 (16.6) ns
Duration of uveitis ns
>5 years 17 (70.8) 22 (91.7) ns
>10 years 8 (33.3) 11 (45.8) ns
ns = not significant;
* myocardial infarction; + myocardial infarction and abdominal aneurysm aorta; ++
two patients
without CME were not tested.
67
Discussion
We report on a positive correlation between trace- and/or
microalbuminuria and inflammatory CME (respectively, P = .001; odds
ratio 13.0, 95 % CI 2.5 to 68.1 and P = .015; odds ratio 5.9, 95 % CI
1.6 to 22.6).
The association between microalbuminuria and cardiovascular
morbidity, mortality and risk factors, and all-cause death was
previously reported for both, diabetic and nondiabetic patients, as well
as the associations with diabetes mellitus, nephropathy, endothelial
dysfunction, inflammation, increasing age, and malignancies. 12-17,24-29
Additionally retinal micro vascular abnormalities were associated with
long term cardiovascular mortality.30
Previous studies on microalbuminuria used different excretion
values for (trace) microalbuminuria and are therefore difficult to
compare. In the general population, 76% of individuals were reported
to have urinary albumin excretion of 0 ≤ 10 mg/l and urinary albumin
excretion above 20 mg/l was present in 7% to 15%.12,29 In our series
these percentages were 88% (42/48) and 4% (2/48), respectively.
Urinary albumin excretion in the range 5 ≤ 10 mg/l in the healthy
population was to our knowledge not yet reported. In our series, 23%
of all uveitis patients (11/48) (10/24 CME patients versus 1/24 non-
CME patients, P = .004) had urinary albumin excretion of 5 ≤ 10 mg/l.
The majority of discrepancies of trace microalbuminuria between CME
and non-CME patients was observed in the excretion range of 5 ≤ 10
mg/l.
The prevalence of systemic hypertension and myocardial infarction
in urinary albumin excretion of 0 ≤ 10 mg/l found in our series was
similar to that reported earlier.13 However, in our series, the significant
associations between cardiovascular morbidity and cardiovascular risk
factors with trace- and/or microalbuminuria were not noted, probably
because of the limited number of patients included in this study.12
Furthermore, large recent studies showed that any degree of
microalbuminuria (including trace microalbuminuria) was an
independent indicator of cardiovascular morbidity.13,16,29 The risk of
cardiovascular morbidity was associated with increasing albumin
excretion, which might indicate that our patients with trace- and/or
microalbuminuria might develop cardiovascular diseases in the future.
68
Chapter 6
We included a limited, heterogeneous population of consecutive

patients with uveitis, with different anatomic locations and specific
diagnoses. The sample size is limited and the borderline P-values
observed do not exclude significant associations. We performed a
multivariate logistic regression analysis controlling for these factors,
which showed no influence on our results of the correlation between
CME and trace- and/or microalbuminuria. To explore these possible
associations and confounders a larger sample size is needed.
In diabetic patients, the presence of microalbuminuria and CVD
indicated a greater risk of developing CME and severe
retinopathy,5,7,8,31-33 but the relation between trace microalbuminuria
and CME and/ or diabetic retinopathy is not yet known.
The treatment of inflammatory CME includes various combinations
of acetazolamide, systemic or periocular corticosteroids, NSAID’s and
various immunosuppressive drugs, all of which might be associated
with significant adverse effects. Despite this therapeutic arsenal the
current treatment is often not effective. Since urinary albumin
excretion is considered to be a marker of damage to the vascular
endothelium and since this damage seems to be present at very low
levels of microalbuminuria, it is plausible that the finding of trace-
and/or microalbuminuria in patients with uveitis and CME might
indicate the presence of (early) systemic vascular disease. This
condition, in combination with the pathological changes in the retinal
and choroidal vessels induced by uveitis could lead to the increased
leakage of fluid in the macula and might explain the correlation
between (trace) microalbuminuria and CME. The beneficial effect of
early angiotensin-converting enzyme inhibitor (ACE inhibitor) treatment
in diabetic patients with microalbuminuria is widely recognized. ACE
inhibitors are prescribed routinely for these patients and reduce
albuminuria, progression of nephropathy, and hypertension.34-37
Several reports suggested initiating treatment with ACE inhibitors in
diabetic patients even in the absence of microalbuminuria and
hypertension.38-41 The majority of studies reported a beneficial effect of
ACE inhibitors in slowing the progression of diabetic retinopathy.34,39,42-
44
The effect of ACE inhibitors on diabetic CME has to our knowledge
not yet been systematically studied. Since the use of ACE inhibitors has
a low incidence of side effects and diminishes renal vascular
69
permeability, we could hypothesize that this type of treatment might

be of beneficial influence on inflammatory CME.
In conclusion, we found a positive association between
inflammatory macular edema and trace microalbuminuria. The
presence of trace microalbuminuria in CME might indicate a presence of
early systemic vascular disease and carry a risk of increased
permeability of retinal and choroidal vessels and thereby early
development of CME. This finding could bring new insights into the
pathogenesis of CME and could open up new avenues for the treatment
of CME.
70
Chapter 6
References
Ophthalmol 1996;80:332-336.
3. Jain R, Stevens JD, Bunce CV, Garrett C, Hykin PG. Ischaemic heart disease may
predispose to pseudophakic cystoid macular oedema. Eye 2001;15:34-38.
7. Gilbert RE, Tsalamandris C, Allen TJ, Colville D, Jerums G. Early nephropathy
predicts vision-threatening retinal disease in patients with type I diabetes
mellitus. J Am Soc Nephrol 1998;9:85-89.
9. Dodson PM, Gibson JM. Long-term follow-up of and underlying medical
conditions in patients with diabetic exudative maculopathy. Eye 1991;5 ( Pt
6):699-703.
11. Roy MS, Klein R. Macular edema and retinal hard exudates in African Americans
with type 1 diabetes: the New Jersey 725. Arch Ophthalmol 2001;119:251-259.
12. Hillege HL, Janssen WM, Bak AA, Diercks GF, Grobbee DE, Crijns HJ, van Gilst
WH, de Zeeuw D, de Jong PE. Microalbuminuria is common, also in a
nondiabetic, nonhypertensive population, and an independent indicator of
cardiovascular risk factors and cardiovascular morbidity. J Intern Med
2001;249:519-526.
13. Hillege HL, Fidler V, Diercks GF, van Gilst WH, de Zeeuw D, van Veldhuisen DJ,
Gans RO, Janssen WM, Grobbee DE, de Jong PE. Urinary albumin excretion
predicts cardiovascular and noncardiovascular mortality in general population.
Circulation 2002;106:1777-1782.
14. Pedrinelli R, Dell'Omo G, Penno G, Mariani M. Non-diabetic microalbuminuria,
endothelial dysfunction and cardiovascular disease. Vasc Med 2001;6:257-264.
15. Pedrinelli R, Dell'Omo G, Di B, V, Pontremoli R, Mariani M. Microalbuminuria, an
integrated marker of cardiovascular risk in essential hypertension. J Hum
Hypertens 2002;16:79-89.
16. Roest M, Banga JD, Janssen WM, Grobbee DE, Sixma JJ, de Jong PE, de Zeeuw
D, Der Schouw YT. Excessive urinary albumin levels are associated with future
cardiovascular mortality in postmenopausal women. Circulation 2001;103:3057-
3061.
71
17. Romundstad S, Holmen J, Hallan H, Kvenild K, Kruger O, Midthjell K.

Microalbuminuria, cardiovascular disease and risk factors in a
nondiabetic/nonhypertensive population. The Nord-Trondelag Health Study
(HUNT, 1995-97), Norway. J Intern Med 2002;252:164-172.
18. Cystoid macular oedema. In Kanski JJ, editor. Clinical Ophthalmology. London:
J Ophthalmol 1998;82:1013-1016.
20. Criqui MH. Epidemiology of cardiovascular disease. In Goldman L, Bennett JC,
editors. Cecil textbook of medicine. Philadelphia: W.B.Saunders Company,
2000:167-170.
21. Bangstad HJ, Try K, Dahl-Jorgensen K, Hanssen KF. New semiquantitative
dipstick test for microalbuminuria. Diabetes Care 1991;14:1094-1097.
22. Cockcroft DW, Gault MH. Prediction of creatinine clearance from serum
creatinine. Nephron 1976;16:31-41.
23. Verhave JC, Balje-Volkers CP, Hillege HL, de Zeeuw D, de Jong PE. The reliability
of different formulae to predict creatinine clearance. J Intern Med
2003;253:563-573.
24. Zappasodi P, Pascutto C, Bosoni T, Corso A. Urinary proteins in multiple
myeloma: strong correlation with the indices of tumor burden. Haematologica
2001;86:878.
25. Pedersen LM, Sorensen PG. Increased urinary albumin excretion rate in breast
cancer patients. Acta Oncol 2000;39:145-149.
26. Pedersen LM, Sorensen PG. Clinical significance of urinary albumin excretion in
patients with non-Hodgkin's lymphoma. Br J Haematol 1999;107:889-891.
27. Pedersen LM, Milman N. Microalbuminuria in patients with lung cancer. Eur J
Cancer 1998;34:76-80.
28. Redon J, Rovira E, Miralles A, Julve R, Pascual JM. Factors related to the
occurrence of microalbuminuria during antihypertensive treatment in essential
hypertension. Hypertension 2002;39:794-798.
29. Gerstein HC, Mann JF, Yi Q, Zinman B, Dinneen SF, Hoogwerf B, Halle JP, Young
J, Rashkow A, Joyce C, Nawaz S, Yusuf S. Albuminuria and risk of cardiovascular
events, death, and heart failure in diabetic and nondiabetic individuals. JAMA
2001;286:421-426.
30. Wong TY, Klein R, Nieto FJ, Klein BE, Sharrett AR, Meuer SM, Hubbard LD,
Tielsch JM. Retinal microvascular abnormalities and 10-year cardiovascular
mortality: a population-based case-control study. Ophthalmology
2003;110:933-940.
31. Wirta O, Pasternack A, Mustonen J, Laippala P, Lahde Y. Retinopathy is
independently related to microalbuminuria in type 2 diabetes mellitus. Clin
Nephrol 1999;51:329-334.
32. Cruickshanks KJ, Ritter LL, Klein R, Moss SE. The association of
microalbuminuria with diabetic retinopathy. The Wisconsin Epidemiologic Study
of Diabetic Retinopathy. Ophthalmology 1993;100:862-867.
of Diabetic Retinopathy: XVII. The 14-year incidence and progression of diabetic
72
Chapter 6
retinopathy and associated risk factors in type 1 diabetes. Ophthalmology

1998;105:1801-1815.
34. Ravid M, Savin H, Jutrin I, Bental T, Katz B, Lishner M. Long-term stabilizing
effect of angiotensin-converting enzyme inhibition on plasma creatinine and on
proteinuria in normotensive type II diabetic patients. Ann Intern Med
1993;118:577-581.
35. Mogensen CE, Neldam S, Tikkanen I, Oren S, Viskoper R, Watts RW, Cooper ME.
Randomised controlled trial of dual blockade of renin-angiotensin system in
patients with hypertension, microalbuminuria, and non-insulin dependent
diabetes: the candesartan and lisinopril microalbuminuria (CALM) study. BMJ
2000;321:1440-1444.
36. Nielsen FS, Rossing P, Gall MA, Skott P, Smidt UM, Parving HH. Long-term effect
of lisinopril and atenolol on kidney function in hypertensive NIDDM subjects with
diabetic nephropathy. Diabetes 1997;46:1182-1188.
37. Poulsen PL, Ebbehoj E, Mogensen CE. Lisinopril reduces albuminuria during
exercise in low grade microalbuminuric type 1 diabetic patients: a double blind
randomized study. J Intern Med 2001;249:433-440.
38. Solomon CG. Reducing cardiovascular risk in type 2 diabetes. N Engl J Med
2003;348:457-459.
39. Schrier RW, Estacio RO, Esler A, Mehler P. Effects of aggressive blood pressure
control in normotensive type 2 diabetic patients on albuminuria, retinopathy and
strokes. Kidney Int 2002;61:1086-1097.
40. Yusuf S, Sleight P, Pogue J, Bosch J, Davies R, Dagenais G. Effects of an
angiotensin-converting-enzyme inhibitor, ramipril, on cardiovascular events in
high-risk patients. The Heart Outcomes Prevention Evaluation Study
Investigators. N Engl J Med 2000;342:145-153.
41. Randomised placebo-controlled trial of lisinopril in normotensive patients with
insulin-dependent diabetes and normoalbuminuria or microalbuminuria. The
EUCLID Study Group. Lancet 1997;349:1787-1792.
42. Chase HP, Garg SK, Harris S, Hoops S, Jackson WE, Holmes DL. Angiotensin-
converting enzyme inhibitor treatment for young normotensive diabetic
subjects: a two-year trial. Ann Ophthalmol 1993;25:284-289.
43. Jackson WE, Holmes DL, Garg SK, Harris S, Chase HP. Angiotensin-converting
enzyme inhibitor therapy and diabetic retinopathy. Ann Ophthalmol 1992;24:99-
103.
44. Chaturvedi N, Sjolie AK, Stephenson JM, Abrahamian H, Keipes M, Castellarin A,
Rogulja-Pepeonik Z, Fuller JH. Effect of lisinopril on progression of retinopathy in
normotensive people with type 1 diabetes. The EUCLID Study Group. EURODIAB
Controlled Trial of Lisinopril in Insulin-Dependent Diabetes Mellitus. Lancet
1998;351:28-31.
73
74
Chapter 6
Chapter 7
A randomized, masked, cross-

over trial of lisinopril for
inflammatory macular edema.
Bram van Kooij1 MD, Rob Fijnheer2 MD, PhD, Joke de Boer1 MD, PhD,
Ninette ten Dam – van Loon1 MD, Imke Bartelink3 PhD, Mark Roest2
PhD, Aniki Rothova1 MD, PhD.
1
F.C. Donders Institute of Ophthalmology, 2Department of Hematology
and 3Department of Pharmacy, University Medical Center, Utrecht, The
Netherlands.
75
A randomized, masked, cross-over trial of lisinopril for inflammatory CME
Abstract
Purpose
To analyze the effect of ACE inhibitor lisinopril on inflammatory cystoid
macular edema and visual acuity.
Design
Randomized, double-blind, placebo-controlled cross-over trial.
Methods
Setting:Out-patient clinic of the Department of Ophthalmology at the
University Medical Center of Utrecht. Patients:Forty patients with
inflammatory cystoid macular edema were included. Intervention: Each
patient received lisinopril (10 mg daily) or placebo for three months.
After 2 months of a lisinopril/placebo free wash-out period the groups
received the reverse study medication for 3 months. Fluorescein
angiography was performed to evaluate the retina. Main outcome
measures: Cystoid macular edema, best corrected visual acuity and
contrast sensitivity.
Results
Lisinopril had no effect on cystoid macular edema, visual acuity,
papillary leakage, retinal vasculitis and choroidal leakage. In a
subgroup analysis we observed a decrease in blood pressure (lisinopril,
14/36 patients; placebo, 5/36 patients; P = .02) and a decrease in
morning urinary albumin excretion (lisinopril, 23/35 patients; placebo
10/34 patients, P = .003) was observed.
Conclusions
Although lisinopril had no effect on inflammatory cystoid macular
edema and visual acuity, we found a positive effect on the vascular
system.
76
Chapter 7
Introduction
Trace microalbuminuria (morning urinary albumin concentration 5≤20
mg/L) was reported to be present more often in patients with
inflammatory cystoid macular edema (CME) than in patients with
uveitis without CME.1 This association indicates the presence of (early)
systemic vascular disease in patients with inflammatory CME, which
might be a predisposing factor for the (early) development of CME. This
theory is supported by the finding that systemic vascular diseases are
involved in the pathogenesis of CME in patients with diabetes mellitus
and retinal vein occlusions.2 In particular, the CME in patients with
diabetes mellitus is strongly associated with hypertension and the
presence of microalbuminuria.3-5
CME, an accumulation of fluid in the macula, is a vision-
threatening complication common to various ocular disorders.6-9 CME is
a major cause of visual impairment (41%) and blindness (29%) in
uveitis; of all patients with uveitis approximately 31 to 47% develop
CME.8,10,11 Early treatment was recommended for this complication of
uveitis.12 The treatment includes, besides treatment of the underlying
ocular condition, various combinations of acetazolamide, corticosteroids
in various administrations, NSAIDs and various other
immunosuppressive drugs, all of which might be associated with
significant adverse effects.12 Despite this therapeutic arsenal many
patients are resistant to or intolerant of these treatments. Therefore it
is very important to find new therapeutic options, preferably with few
side-effects.
Since vascular permeability diminishes and endothelial function
improves with the use of ACE inhibitors,13-16 and ACE inhibitors have a
low incidence of side-effects, we hypothesized that this type of
treatment might be beneficial for inflammatory CME. In this study we
attempt to analyze the effect of lisinopril on the inflammatory CME and
visual acuity in patients with uveitis.
Patients and methods

In this study were included adult patients with inactive or chronic low
grade activity of uveitis (with or without treatment), and persistent
CME which was documented on fluorescein angiography and which
caused a decrease in visual acuity and did not respond to the current
77
standard treatment. The study protocol was approved by the medical

ethical committee from the University Medical Center of Utrecht and
adhered to the tenets of the Declaration of Helsinki. All patients signed
an informed consent before enrollment in the study. Exclusion criteria
included age below 18 years, a history of hypersensitive reaction to
ACE inhibitors, use of ACE inhibitors, nephropathy, pregnancy or
lactation. Anatomic classification and specific diagnoses of uveitis were
based on the current criteria.17
Study design
Before the study, we set the criteria for success, one of which was that
the medication should be considered to be successful when at least
35% of patients showed improvement. Using a two-sided statistical
test with an alpha error of 0.05 and a beta error of 0.1, we calculated
that 40 patients should be included in a parallel setting. Moreover, the
cross-over design allows this study to have additional statistical value
where each case is serving as its own control.
Randomization and Study Protocol

In this randomized, double-blind, cross-over trial, each patient was
randomized prior to the study in either the lisinopril group or the
placebo group. Lisinopril and placebo were placed in identical capsules
by the Pharmacy Department, which also planned the randomization.
Both patients and investigators were masked to the randomization.
Lisinopril was given in a dose of 10 mg daily for 3 months. After 2
months of a lisinopril or placebo free wash-out period the groups
received the reverse study medication for 3 months (Fig. 1).
Patients received information about the study both orally and in
writing. After signing an informed consent patients received a monthly
clinical examination that included: best corrected visual acuity on the
Snellen chart, contrast sensitivity on the Pelli Robson chart, eye
pressure measurement, slit lamp examination, fundoscopy and blood
pressure measurement. Adverse events were monitored monthly via a
standardized list and were scored and defined according to the World
Health Organization criteria.18 Fluorescein angiography and laboratory
tests (creatinine, c-reactive protein and morning urinary albumin
78
Chapter 7
Figure 1: Flow of patients with inflammatory cystoid macular edema (CME) in the cross-
over trial. Study period: December 2003 – December 2004
40 Randomized
21 assigned to receive 19 assigned to receive

lisinopril for 3 months placebo for 3 months
1 non-study related
complication 1 pregnant
1 adverse effects
Wash-out 2 months Wash-out 2 months
1 deterioration of
uveitis and CME
17 received lisinopril 19 received placebo

for 3 months for 3 months
2 deterioration of
uveitis and CME
36 included for analysis 36 included for analysis
excretion) were performed before the study and repeated after the first
and second medication/placebo period (after 3 and 8 months). In
addition optical coherence tomography (Zeiss, Germany) to measure
central macular thickness was performed in some patients depending
on the availability of the instrument. A change in macular thickness
exceeding 12% was reported to be a significant change.19 Furthermore
patients were checked for a history of cardiovascular diseases or their
risk factors using a questionnaire which included myocardial infarction,
angina pectoris, abdominal aortic aneurysm, cerebral vascular
79
accidents, transient ischemic attacks, intermittent claudication,

presence of systemic hypertension, diabetes mellitus,
hypercholesterolemia, thrombosis, lung emboli, smoking and drinking
habits. Weight and height were measured and all medications were
registered. Patients continued their prior treatment for uveitis during
the study. The criteria for the withdrawal of patients from the study
were: visual acuity decrease of more than 2 Snellen lines caused by the
progression of CME, increase in uveitis activity requiring a change of
anti-inflammatory medication, unexpected ophthalmologic
complications requiring a change in medication or ocular surgery and
pregnancy. If adverse events became intolerable, the patients were
also taken off the study and were informed about their
medication/placebo via an independent ophthalmologist.
Urinary albumin measurements

Morning urinary albumin excretion was determined with Quikread U-
ALB, (Orion Diagnostica, Espoo, Finland). The inter-assay precision of
this assay is 6.6%; the intra-assay precision is 4.7% (n = 10, for an
urine sample with 10mg/L albumin). The detection limit is < 5mg/L.
Morning urinary albumin excretion ≤ 5mg/L was defined as no
albuminuria; that of 5≤20 mg/L was defined as trace microalbuminuria
and morning urinary albumin excretion of 20≤200 mg/L was defined as
microalbuminuria.
Study endpoints
Primary endpoints of the study were the amount of CME measured by
fluorescein angiography, best corrected visual acuity and contrast
sensitivity. The amount of CME on late-phase fluorescein angiograms
(10 minutes after injection of the dye) were graded by two
independent masked observers, who estimated the area of the macula
affected by edema. Patients with no leakage on the late-phase
fluorescein angiogram were graded 0, those with leakage less than
25% of the macular area as grade 1, leakage between 25% and 66%
was graded as 2 and leakage of more than 66% was graded as 3.20
Decrease of CME was defined as a decrease of at least one grade.
Improvement of visual acuity was defined as an increase of at least two
Snellen lines. Improvement in contrast sensitivity was defined as an
increase of 1 line with 3 symbols.
80
Chapter 7
Secondary endpoints of the study were papillary leakage, retinal

vasculitis and choroidal leakage measured on fluorescein angiography;
morning urinary albumin excretion, blood pressure and adverse-
effects; c-reactive protein and creatinine.
Sub-analysis was planned for different age groups (≤ 40 years, >
40 years; ≤ 50 years, > 50 years; ≤ 60 years, > 60 years), different
duration of CME (≤ 2 years, ≤ 5 years, > 5 years) and uveitis (≤ 2
years, ≤ 5 years, > 5 years).
Statistical analysis
Data analysis was performed using the Statistical Package for the
Social Sciences (SPSS 12.0). Differences between the medication and
the placebo group were compared with the chi-square test or the
Fisher’s exact test. A P-value of < .05 was considered to be significant.
Results
Forty patients were included in this study from December 2003 to
December 2004. General data of the patients are given in Table 1. The
average age was 54.6 years (SD 12.3, range 29 to 74 years). The
male-to-female ratio was 2:3. The mean duration of uveitis was 9
years (SD 6.6, range 1 to 25 years). The mean duration of CME was
6.0 years (SD 5.0, range 0.5 to 20 years). Two (5%) patients had
anterior uveitis, 10 (25%) intermediate uveitis, 13 (32.5%) posterior
uveitis and 15 (37.5%) panuveitis. The specific ophthalmologic
diagnoses are given in Table 1. Thirty-two (80%) patients were on
systemic medication for uveitis during the trial.
Trace microalbuminuria (urinary albumin excretion of 5≤20 mg/L)
was observed in 30/40 (75%) of patients. Microalbuminuria (urinary
albumin excretion of >20 mg/L) was observed in 6/40 (15%) of
patients. Twenty-three/40 (58%) patients had either cardiovascular
disease or risk factors (Table 1). From the above-mentioned
cardiovascular diseases and risk factors, only hypertension had a
borderline association with urinary albumin excretion of > 10mg/L (P =
.08) and urinary albumin excretion of > 20 mg/L ( P= .09). There was
no association between trace and/or any microalbuminuria (5≤20
mg/L) and cardiovascular diseases or its risk factors. There was no
correlation between any microalbuminuria and c-reactive protein and
creatinine.
81
Table 1: General and ocular characteristics of patients with inflammatory CME*
Number of patients 40
Mean age in years (range) 55 (29 – 74)
Male-to-female ratio 2:3
Mean duration of uveitis in years (range) 9 (1 - 25)
Mean duration of CME in years (range) 6 (0.5 - 20)
Location:
Anterior (%) 2 (5%)
Intermediate (%) 10 (25%)
Posterior (%) 13 (33%)
Pan (%) 15 (38%)
Association with systemic disease (%) 5 (13%)
Recent systemic medication for uveitis and/or CME (%) 32 (80%)
Microalbuminuria
<5 mg/L (%) 4 (10%)
5 ≤ 10 mg/L (%) 20 (50%)
10 ≤ 20 mg/L (%) 10 (25%)
> 20 mg/L (%) 6 (15%)
Myocardial infarction and angina pectoris (%) 1 (3%)
Hypertension (%) 8 (20%)
Diabetes mellitus (%) 2 (5%)
Hypercholesterolemia (%) 6 (15%)
Current smokers (%) 12 (30%)
Medication for cardiovascular diseases or risk factors (%) 9 (23%)
*
Thirty-six patients completed the lisinopril branch and 36 patients the

placebo branch of the trial and were available for analysis (Fig. 1). The
study had 6 drop-outs; two patients dropped out in the first part of the
study (lisinopril branch; one patient developed a corneal edema, which
made fundoscopy and fluorescein angiography impossible and the other
patient experienced serious dizziness and refused to complete the
trial). In the wash-out period one additional patient (after completing
the placebo part) suffered from increased inflammation activity and
CME and required additional medication before she could start the
lisinopril branch. One patient became pregnant at the end of the first
part of the study (placebo medication); fluorescein angiogram at 4
months was not made and she was excluded from further study. Two
patients dropped out in the second period during the placebo branch
because of increasing inflammation activity and CME which required
additional medication. Anti-inflammatory medication was not changed
during the trial in any of the other patients.
82
Chapter 7
Table 2: Results of lisinopril and placebo treatment in inflammatory CME*
Intent to treat (n=40)

Lisinopril Placebo P-value
Improvement in: No % No %
Visual acuity 3/40 (8%) 4/40 (10%) 1.0

Contrast sensitivity 10/40 (25%) 13/40 (33%) .46
Cystoid macular edema 10/40 (25%) 9/40 (23%) .79
Papillary leakage 7/40 (18%) 3/40 (8%) .31
Retinal vasculitis 0/40 (0%) 2/40 (5%) .49
Choroidal leakage 5/40 (13%) 5/40 (13%) 1.0
Patients’ subjective improvement 6/40 (15%) 3/40 (8%) .48
Presence of:
Adverse effects 13/36 (36%) 3/36 (8%) .009
Decreased blood pressure 14/36 (39%) 5/36 (14%) .016
Decreased microalbuminuria 23/35 (66%) 10/34 (29%) .003
*
For the patients in the lisinopril group (intent to treat), we found no

effect on visual acuity (improvement in 3/40, 8% treated with
lisinopril; 4/40, 10% treated with placebo; P = 1.0), CME
(improvement in 10/40, 25% treated with lisinopril; 9/40, 23% treated
with placebo; P = .79), contrast sensitivity (improvement in 10/40,
25% treated with lisinopril; 13/40, 33% treated with placebo; P = .46),
papillary leakage (improvement in 7/40, 18% treated with lisinopril;
3/40, 8% treated with placebo; P = .31), retinal vasculitis
(improvement in 0/40, 0% treated with lisinopril; 2/40, 5% treated
with placebo; P = .49), choroidal leakage (improvement in 5/40, 13%
treated with lisinopril; 5/40, 13% treated with placebo; P = 1.0) or on
subjective grading of visual acuity by patients (improvement in 6/40,
15% treated with lisinopril; 3/40, 8% treated with placebo; P = .48)
(Table 2). An “as treated” analysis was also performed and showed
similar results. Sub-analysis for different age groups, different
durations of CME, different durations of uveitis, and different anatomic
locations did not show any differences between the placebo and
lisinopril medication. A sub-analysis showed no improved CME in the
placebo branch after first being treated with lisinopril (P = .63).
We found a significant effect of lisinopril on blood pressure (P =
.016) and trace- and microalbuminuria (P = .004 and P = .003)
83
compared to the placebo (Table 2). No correlation was found between

decreased morning urinary albumin excretion (in the lisinopril group)
and visual acuity, CME, contrast sensitivity, papillary leakage, retinal
vasculitis, choroidal leakage and on the subjective grading of visual
acuity by patients.
Adverse effects were reported in 13 patients during the study and
consisted of: cough (n=3), dizziness (n=5), fatigue (n=7) and
headache (n=1) in patients treated with lisinopril and fatigue (n=2)
and headache (n=1) in patients treated with the placebo (for all side
effects P=0.009; Table 2). Out of these, only one required the
withdrawal of lisinopril medication.
In 17 (12 lisinopril, 5 placebo) eyes optical coherence tomography
was performed and could be analyzed for central retinal thickness. In
the lisinopril group, 2/12 eyes showed decreased retinal thickness,
2/12 eyes showed increased retinal thickness and 8/12 eyes remained
unchanged. In the placebo group 1/5 eyes showed increase in retinal
thickness and the 4 remaining eyes remained unchanged.
Discussion
Three months of the administration of 10 mg lisinopril had no effect on
chronic CME, visual acuity, papillary leakage, retinal vasculitis and
choroidal leakage in patients with uveitis. In patients with lisinopril
medication, we observed a decrease in blood pressure (P = .02) and a
decrease in morning urinary albumin excretion (P = .003).
ACE-inhibitors and angiotensin II receptor antagonists have
clinically equal effects.21 Because of their well known beneficial effect
on the vascular permeability, they could make a favorable contribution
to the treatment of pathological retinal and choroidal vascular leakage.
Several studies have investigated the effect of ACE inhibitors and
angiotensin II receptor antagonists on the progression of diabetic
retinopathy, diabetic CME and visual acuity, but results were
conflicting.22-25 Knudsen et al. performed a randomized double-masked
study with 24 patients to analyze the effect of a 4-month treatment
with losartan (an angiotensin II receptor antagonist) on diabetic
maculopathy in type 2 diabetic patients. They found an increased
central retinal thickness in the losartan group) and no effect on the
visual acuity.25 The authors could not give an explanation for this
84
Chapter 7
unexpected finding, but a relatively small number of patients and a

short follow-up period might have played a role. These results are in
concordance with our negative findings. Although we found no effect of
lisinopril on CME and visual acuity, we consider it of value to perform
further studies on the effect of ACE-inhibitors or other vessel-protecting
drugs on CME of limited duration. A prolonged use of these treatments
on CME might also be of interest.
ACE inhibitors were reported by the EUCLID study group and by
Kvetny et al. to lower the urinary albumin excretion even in the trace
microalbuminuria range in patients with insulin-dependent diabetes
mellitus.26,27 The EUCLID study group also found a lower blood pressure
in normotensive patients in the treatment group than in the placebo
group. We showed that lisinopril significantly decreased the morning
urinary albumin excretion in the trace microalbuminuria range and
blood pressure in non-diabetic patients, which might indicate a
beneficial effect of lisinopril on subclinical vascular pathology.
Inflammatory and diabetic macular edema are associated with
generalized microvascular permeability. The drug with a beneficial
effect on the vascular endothelium and on lowering urine albumin
excretion, even in the trace microalbuminuria range, had however no
effect on longstanding macular edema (in our study series the average
CME duration was 6 years). The flaw in our study consists of the
inclusion of patients with a variable duration of CME and treatment with
lisinopril of only 3 months which might have influenced the negative
results obtained. The results might be different for patients with early
CME because in patients with a longer duration of CME the retinal
vessels could well already have been subjected to irreversible changes
on which lisinopril had no effect or needed a longer period to be
effective. It is also possible that the lisinopril was effective on the inner
blood retinal barrier, formed by the endothelial tight junctions of the
retinal vessels but had no effect on the retinal pigment epithelium,
responsible for the clearing of the interstitial fluid. Studies on the effect
of lisinopril on early CME and over a prolonged period are needed.
Although the frequency of side effects was substantial in the
lisinopril group (P = .009; 33%, Table 2) the influence of the side
effects on the continuation of the therapy was small, and only one
patient required the withdrawal of medication. Our study included 10%
dropouts and all possible corrections for these dropouts did not
85
significantly change the results. We used a cross-over setting to

increase the patients recruitment in each group. Despite this our
sample size might have been too small which might have influenced
the results. Theoretically an improvement in the lisinopril branch could
perpetuate even long after cessation of the drug, however a sub-
analysis showed no improved CME in the placebo branch after first
being treated with lisinopril (P = .63).
We used fluorescein angiography to analyze the primary endpoint
for leakage. Fluorescein angiograms were evaluated independently by
two masked observers to exclude subjectivity as much as possible. In a
condition of chronic cystoid macular edema of a mean duration of 6
yrs, one may be looking for subtle changes that can be picked up only
on optical coherence tomography (OCT), which is a more sensitive
technique to evaluate macular thickness. Unfortunately we could use
this technology only in 17 eyes (12 on lisinopril and 5 on placebo)
because OCT was not available to us at the start of the trial.
In conclusion, we found no effect of 3-months course of lisinopril
on inflammatory CME and visual acuity, although a positive effect on
the vascular system was observed. The treatment was associated with
mild side effects. The future studies on the efficacy of various vessel-
protecting drugs are needed for this vision-threatening macular
disorder.
86
Chapter 7
References
1. van Kooij B, Fijnheer R, Roest M, Rothova A. Trace microalbuminuria in
inflammatory cystoid macular edema. Am J Ophthalmol 2004;138:1010-1015.
2. Tranos PG, Wickremasinghe SS, Stangos NT, Topouzis F, Tsinopoulos I, Pavesio
CE. Macular edema. Surv Ophthalmol 2004;49:470-490.
997.
7. Klein R, Klein BE, Moss SE, Davis MD, DeMets DL. The Wisconsin epidemiologic
study of diabetic retinopathy. IV. Diabetic macular edema. Ophthalmology
1984;91:1464-1474.
Ophthalmol 1996;80:332-336.
9. Weinberg D, Seddon J. Venous occlusive disease of the retina. In Albert D,
Jacobiec F, editors. Principles and practice of ophthalmology. Philadelphia: WB
Saunders, 1994:735-746.
1162.
2002;10:239-246.
13. Mogensen CE, Neldam S, Tikkanen I, Oren S, Viskoper R, Watts RW, Cooper ME.
Randomised controlled trial of dual blockade of renin-angiotensin system in
patients with hypertension, microalbuminuria, and non-insulin dependent
diabetes: the candesartan and lisinopril microalbuminuria (CALM) study. BMJ
2000;321:1440-1444.
14. Nielsen FS, Rossing P, Gall MA, Skott P, Smidt UM, Parving HH. Long-term effect
of lisinopril and atenolol on kidney function in hypertensive NIDDM subjects with
diabetic nephropathy. Diabetes 1997;46:1182-1188.
15. Poulsen PL, Ebbehoj E, Mogensen CE. Lisinopril reduces albuminuria during
exercise in low grade microalbuminuric type 1 diabetic patients: a double blind
randomized study. J Intern Med 2001;249:433-440.
16. Ravid M, Savin H, Jutrin I, Bental T, Katz B, Lishner M. Long-term stabilizing
effect of angiotensin-converting enzyme inhibition on plasma creatinine and on
proteinuria in normotensive type II diabetic patients. Ann Intern Med
1993;118:577-581.
87
17. Bloch-Michel E, Nussenblatt RB. International Uveitis Study Group

recommendations for the evaluation of intraocular inflammatory disease. Am J
Ophthalmol 1987;103:234-235.
18. Edwards IR, Aronson JK. Adverse drug reactions: definitions, diagnosis, and
management. Lancet 2000;356:1255-1259.
19. Massin P, Audren F, Haouchine B, Erginay A, Bergmann JF, Benosman R, Caulin
C, Gaudric A. Intravitreal triamcinolone acetonide for diabetic diffuse macular
edema: preliminary results of a prospective controlled trial. Ophthalmology
2004;111:218-224.
J Ophthalmol 1998;82:1013-1016.
21. Barnett AH, Bain SC, Bouter P, Karlberg B, Madsbad S, Jervell J, Mustonen J.
Angiotensin-receptor blockade versus converting-enzyme inhibition in type 2
diabetes and nephropathy. N Engl J Med 2004;351:1952-1961.
22. Comer GM, Ciulla TA. Pharmacotherapy for diabetic retinopathy. Curr Opin
Ophthalmol 2004;15:508-518.
23. Funatsu H, Yamashita H, Shimizu E, Mimura T, Nakamura S, Hori S. Quantitative
measurement of retinal thickness in patients with diabetic macular edema is
useful for evaluation of therapeutic agents. Diabetes Res Clin Pract
2004;66:219-227.
24. Pradhan R, Fong D, March C, Jack R, Rezapour G, Norris K, Davidson MB.
Angiotensin-converting enzyme inhibition for the treatment of moderate to
severe diabetic retinopathy in normotensive Type 2 diabetic patients. A pilot
study. J Diabetes Complications 2002;16:377-381.
25. Knudsen ST, Bek T, Poulsen PL, Hove MN, Rehling M, Mogensen CE. Effects of
losartan on diabetic maculopathy in type 2 diabetic patients: a randomized,
double-masked study. J Intern Med 2003;254:147-158.
26. Randomised placebo-controlled trial of lisinopril in normotensive patients with
insulin-dependent diabetes and normoalbuminuria or microalbuminuria. The
EUCLID Study Group. Lancet 1997;349:1787-1792.
27. Kvetny J, Gregersen G, Pedersen RS. Randomized placebo-controlled trial of
perindopril in normotensive, normoalbuminuric patients with type 1 diabetes
mellitus. QJM 2001;94:89-94.
88
Chapter 7
Chapter 8
The effect of non-steroidal anti-

inflammatory drugs on
edema.
Bram van Kooij1 MD, Joke de Boer1 MD, PhD, Ninette ten Dam-van
Loon1 MD, Rob Fijnheer2 MD, PhD, Aniki Rothova1 MD, PhD.
1
F.C. Donders Institute of Ophthalmology and 2Department of
Hematology, University Medical Center, Utrecht, The Netherlands.
(Published as a Brief Report in American Journal of Ophthalmology

2005 Sep;140(3):563-564)
89
The effect of NSAIDs on inflammatory CME
Abstract
Purpose
To assess the role of systemic non-steroidal anti-inflammatory drugs
(NSAIDs) in the treatment of inflammatory cystoid macular edema
(iCME).
Design
Pilot study.
Methods
We included 66 patients with iCME and treated them with naproxen (2
x 250 mg daily; n = 28) or rofecoxib (1 x 25 mg daily; n = 38). After 4
months of therapy, visual acuity and inflammation activity were
measured. We evaluated the grade of CME, retinal vasculitis, and
papillary leakage with fluorescein angiography.
Results
No clear effect of NSAIDs on iCME, visual acuity, and intraocular
inflammation was observed. A beneficial effect was noted in less than
8% of the affected eyes. Drop-out after 4 months of treatment was
52%, because of adverse effects or inefficacy of the treatment.
Improvement of visual acuity was slightly better in patients who were
treated with naproxen compared to rofecoxib (p ≤ .05).
Conclusion
Systemic NSAIDs have a limited role (if any) in the treatment of iCME.
90
Chapter 8
Introduction
Cystoid macular edema (CME) is a major cause of visual impairment
and blindness in uveitis.1-3 Approximately 31% to 47% of patients with
uveitis develop CME1,3 and the efficacy of treatment is frequently
insufficient and associated with significant side effects.
The symptomatic treatment of inflammatory CME includes various
combinations of acetazolamide, systemic or periocular corticosteroids,
non-steroidal anti-inflammatory drugs (NSAIDs) and other
immunosuppressive drugs, all of which might be associated with
significant adverse effects. The role of systemic administration of
NSAIDs was considered controversial, but one report claimed that -for
inflammatory CME- oral NSAIDs used for at least several months
matched the efficacy of periocular corticosteroid injections.4
Classic NSAIDs (for example, naproxen) are inhibitors of
cyclooxygenase (COX) 1 and 2. Inhibition of COX is responsible for the
therapeutic effect and the side effects of NSAIDs. Selective COX-2
inhibitors have fewer side effects on the gastrointestinal tract than
classic NSAIDs.
We performed a pilot study to assess the possible effect of
naproxen and rofecoxib on inflammatory CME.
Patients and Methods

We included 66 consecutive patients with inflammatory CME and
treated them randomly with naproxen (2 x 250 mg per daily) or
rofecoxib (1 x 25 mg daily) for at least 4 months. Patients with
naproxen received an additional gastric acid-related disease medication
in contrast to rofecoxib patients. An analysis before and 4 months after
the commencement of therapy included the measurement of visual
acuity (using Snellen charts) and inflammation activity.5 The grade of
CME,6 activity of retinal vasculitis, and papillary leakage were evaluated
by fluorescein angiography that was performed at the commencement
of treatment and between 4 and 12 months thereafter. Diagnosis of
CME was based on both clinical criteria and fluorescein angiography.7
CME was graded by a masked observer on late-phase fluorescein
angiograms. The therapeutic effect was considered successful in
patients with an improved visual acuity of at least 2 Snellen lines or
improvement on fluorescein angiography of at least 1 grade. Treatment
91
was considered to have failed in patients with a loss of visual acuity of

at least 2 lines or with an increase of CME, inflammation activity or
both to such an extent that they required additional medication.
For analysis of visual acuity the LogMar visual acuity (logarithm of
the minimal angle of resolution) was used. Differences between the two
groups were compared using the chi square test. A P-value < .05 was
considered to be significant.
Results
General data of the patients are given in the Table. Sixty-six patients
were treated with either naproxen (28 patients) or rofecoxib (38
patients). Forty patients (61%) (57% naproxen; 63% rofecoxib; P =
.62) were females. Mean age at the commencement of treatment was
50 years (range 8-83; average 47 years for naproxen, range 8-74;
average 52 years for rofecoxib, range 13-83; n.s.) CME was bilateral in
30 of the patients which resulted in 96 affected eyes.
Forty patients were on systemic medications for uveitis, CME or
both of whom 10 patients (4 with naproxen, 6 with rofecoxib; P = .87)
were on multiple systemic medications and thirty patients (12 with
naproxen, 18 with rofecoxib; P = .72) were on systemic monotherapy.
The therapy includes corticosteroids, methotrexate, cyclosporine and
acetazolamide that was given either in a combination or as single
drugs. The remaining patients used topical therapy or received
periocular corticosteroid injections previously.
Thirty-four of 66 patients (52 %) (13/28, 46 % with naproxen and
21/38, 55% with rofecoxib; P = .48) did not finish 4 months of NSAID
use, either because of side effects of medication or because of the
worsening of their visual acuity, ocular inflammation or both.
Seventeen of 66 patients (26 %) (5/28, 18% with naproxen; 12/38,
32% with rofecoxib, P = .21) stopped their medication because of
adverse effects. Naproxen patients reported breathing difficulties (n =
1), blurred vision (n = 1), nausea (n = 1), insomnia (n= 1), hair loss
(n = 1); and rofecoxib patients had breathing difficulties (n = 2),
gastrointestinal symptoms (n = 6), nausea (n= 2), limb edema (n= 2),
and paraesthesias (n = 1).
92
Table. Characteristics of patients with inflammatory cystoid macular edema treated with naproxen or rofecoxib.
Finished treatment of On treatment at time of 2nd

Intention to treat with: 4 months with: fluorescein angiography with:
P- P- P-
Variable Naproxen Rofecoxib value Naproxen Rofecoxib value Naproxen Rofecoxib value
Patients (n) 28 38 15 17 13 13
Male/female (n) 12/16 14/24 n.s. 6/9 9/8 n.s. 5/8 7/6 n.s.
Average age in years (range) 47 (8-74) 52 (13-83) n.s. 43 (8-56) 51 (13-80) n.s. 46 (31-56) 50 (26-77) n.s
Mean follow-up in months (range) n/a n/a 4 4 8 (5-12) 6 (4-8) n.s
Additional medication
Multiple systemic1 4 (14%) 6 (16%) n.s. 2 (13%)3 (18%)n.s.
Mono systemic1 14 (50%) 20 (53%) n.s. 6 (40%)8 (47%)n.s.
Topical or periocular injection only 10 (36%) 12 (31%) n.s. 7 (47%)6 (35%)n.s.
Treatment failures (n) 13 (46%) 21 (55%) n.s.

Adverse effects of medications2 5 (18%) 12 (32%) n.s. n/a n/a n/a n/a
Remainder3 8 (29%) 9 (24%) n.s. n/a n/a n/a n/a
Total number of affected eyes (n)4 40 56 20 23 17 17

Improved (n)5
Visual acuity 3 (8%) 0 (0%) .04 3 (15% ) 0 (0%) .05 2 (12%) 0 (0%) n.s.
Cystoid macular edema 3 (8%) 2 (4%) n.s. n/a n/a 3 (18%) 2 (13%)6 n.s.
Retinal vasculitis 0 (0%) 1 (2%) n.s. n/a n/a 0 (0%) 1 (6%) n.s.
Papillitis 0 (0%) 2 (4%) n.s. n/a n/a 0 (0%) 2 (12% ) n.s.
Inflammation activity 0 (0%) 3 (5%) n.s. 0 (0%) 3 (13%) n.s. 3 (18%) 2 (12% ) n.s.
n/a = not applicable; n.s = not significant.

1
Corticosteroids, methotrexate, cyclosporine and acetazolamide that were given either in a combination or as single drugs.
2
Patients treated with naproxen reported breathing difficulties (n = 1), blurred vision (n = 1), nausea (n = 1), insomnia (n = 1), hair loss (n = 1); patients with rofecoxib reported
breathing difficulties (n = 2), gastrointestinal symptoms (n = 6), nausea (n = 2), limb edema (n = 2), and paresthesias (n = 1).
3
Includes loss of visual acuity of at least 2 Snellen lines and/or increase of cystoid macular edema and/or inflammatory activity that required additional medication.
4
Cystoid macular edema was bilateral in 30 of the patients which resulted in 96 affected eyes.
5
For definitions of improvement and grading, see text and appropriate references.
6
Because of technical problems in 15 of 17 eyes, cystoid macular edema could be graded.
93
Chapter 8
The results of the analysis after 4 months are shown in the Table. From
all affected eyes 3 of 96 (3%) showed improved visual activity (3/40,
8% with naproxen compared to none with rofecoxib; P = .037).
Inflammation activity decreased in 3 of 96 eyes (3%) (none with
naproxen; 3/56, 5% with rofecoxib; P = .14). Of those who finished 4
months of treatment in 3 of 43 eyes (7%) visual acuity improved
(3/20, 15% with naproxen compared to none with rofecoxib; P =
.054). Inflammation activity decreased in 3 of 43 eyes (7%) (none with
naproxen; 3/23, 13% with rofecoxib; P = .094).
At the time of the second fluorescein angiography a decrease in
CME was noted in 5 of 96 eyes (5%) (3/40, 8% with naproxen; 2/54,
4% with rofecoxib; P = .42). Visual activity improved in 2 of 96 eyes
(2%) (2/40, 5% with naproxen; none with rofecoxib; P = .091).
Inflammatory activity decreased in 5 of 96 (3/40, 8% with naproxen;
2/56, 4% with rofecoxib; P = .39). Retinal vasculitis and papillitis did
not improve with the use of naproxen, whereas rofecoxib improved
vasculitis and papillitis in respectively 1 (2%; P = .40) and 2 (4%; P =
.23) eyes (Table). From the eyes that were still under treatment at the
time of the second fluorescein angiography, a decrease in CME was
noted in 5 of 34 eyes (15%) (3/17, 18% with naproxen; 2/15, 13%
with rofecoxib; P = .74). Visual acuity improved in 2 of 34 eyes (6%)
(2/17, 12% with naproxen; none with rofecoxib; P = .15). Retinal
vasculitis and papillitis improved in respectively 1 (6%; P = .31) and 2
(12%; P = .15) eyes treated with rofecoxib, whereas eyes treated with
naproxen did not improve. Inflammation activity decreased in 5 of 34
eyes (15%) (3/17, 18% with naproxen; 2, 12% with rofecoxib; P =
.63; Table).
Discussion
The effect of systemic NSAIDs on inflammatory CME, visual acuity, and
intraocular inflammation was observed in less than 8% of affected eyes
(Table). Improvement of visual acuity was slightly better in patients
who were treated with naproxen (P ≤ .05). In 34 of 66 patients (52%),
the treatment failed, in 17 patients (26%) because of adverse effects
of NSAIDs.
The use of topical NSAIDs for pseudophakic CME is widely
accepted,8,9 but the role of systemic NSAIDs on inflammatory CME is
94
Chapter 8
unclear. A previous study of Rojas et associates4 reported that NSAIDs

had a positive effect on inflammatory CME, similar to that of periocular
corticosteroids injections. Shortcomings of their study included a
limited number of patients, no evaluation of fluorescein angiograms,
and no specified NSAID medications and dosages. The authors stated
that the best visual acuity was observed after 6 months of oral
medication with NSAIDs, though the exact data are not specified. The
discrepancy with our results is difficult to explain but might be caused
in part by the small number of patients included in the previous study,
methods and differences in additional medications. The drawbacks of
our study included a non-controlled design, extended time interval for
second fluorescein angiography, the use of prophylactic stomach
protection in patients who received only naproxen, and a high drop-out
because of side effects or worsening of the disease. The high drop-out
rate (26%) that was caused by adverse effects of the medication points
out the potential complications of this class of drug.
Classic NSAIDs (for example, naproxen) are inhibitors of COX 1
and 2. The inhibition of COX 1 is responsible for inhibiting platelet
aggregation and vasodilatation and COX 2 for pain killing and the anti-
inflammatory effects. The side effects on the gastro-intestinal tract are
caused by the inhibition of COX 1. Platelet aggregation and
vasoconstriction are caused by COX 2.10,11 It is not known which
specific effect might be responsible for the decrease of CME, but it is
assumed to be the anti-inflammatory effect, which might in theory be
achieved by the inhibition of only COX 2.
Vascular endothelial growth factor (VEGF) is elevated in the
aqueous humor and vitreous body of patients with CME12-15 and COX-2
has been shown to modulate the expression of the VEGF ligand and its
receptors, also in the retina.16-18 However, in an experimental rat
model of VEGF-induced retinal vascular leakage, COX inhibitors showed
no effect on VEGF-mediated blood-retinal barrier breakdown.19 Our
results are in accordance with this experimental study.
Rofecoxib is a selective COX 2 inhibitor. Large studies showed that
rofecoxib had, in contrast to traditional NSAIDs, few side effects on the
gastrointestinal tract.10,11 In our study 6 of 38 patients stopped
rofecoxib medication because of gastrointestinally related side effects.
This cannot be explained by additional systemic medications (P = .92).
In contrast, all patients with naproxen received prophylactics against
95
gastric acid-related disease and none of them had to stop the

medication because of gastrointestinal side effects. During the
preparation of this manuscript Merck & Co announced a worldwide
withdrawal of rofecoxib because of an increased relative risk of
cardiovascular events. So far none of our patients have had
cardiovascular events.
Our study of NSAIDs for inflammatory CME was characterized by a
drop-out of at least 52% because of adverse effects, inefficacy or both
and we conclude that systemic NSAIDs have a limited role (if any) in
the treatment of inflammatory CME.
96
Chapter 8
References
Ophthalmol 1996;80:332-336.
1162.
4. Rojas B, Zafirakis P, Christen W, Markomichelakis NN, Foster CS. Medical
treatment of macular edema in patients with uveitis. Doc Ophthalmol
1999;97:399-407.
5. Whitcup SM. Examination of the patient with uveitis. In Nussenblatt RB, Whitcup
SM, editors. Uveitis: Fundamentals and Clinical Practice. Philadelphia: Mosby,
2004:54-65.
J Ophthalmol 1998;82:1013-1016.
7. Cystoid macular oedema. In Kanski JJ, editor. Clinical Ophthalmology. London:
8. Rossetti L, Autelitano A. Cystoid macular edema following cataract surgery. Curr
Opin Ophthalmol 2000;11:65-72.
9. Miyake K, Ibaraki N. Prostaglandins and cystoid macular edema. Surv
Ophthalmol 2002;47 Suppl 1:S203-S218.
10. FitzGerald GA, Patrono C. The coxibs, selective inhibitors of cyclooxygenase-2. N
Engl J Med 2001;345:433-442.
11. Peura DA. Gastrointestinal safety and tolerability of nonselective nonsteroidal
anti-inflammatory agents and cyclooxygenase-2-selective inhibitors. Cleve Clin J
Med 2002;69 Suppl 1:SI31-SI39.
12. Fine HF, Baffi J, Reed GF, Csaky KG, Nussenblatt RB. Aqueous humor and
plasma vascular endothelial growth factor in uveitis-associated cystoid macular
edema. Am J Ophthalmol 2001;132:794-796.
13. Funatsu H, Yamashita H, Noma H, Mimura T, Yamashita T, Hori S. Increased
levels of vascular endothelial growth factor and interleukin-6 in the aqueous
humor of diabetics with macular edema. Am J Ophthalmol 2002;133:70-77.
14. Funatsu H, Yamashita H, Ikeda T, Nakanishi Y, Kitano S, Hori S. Angiotensin II
and vascular endothelial growth factor in the vitreous fluid of patients with
diabetic macular edema and other retinal disorders. Am J Ophthalmol
2002;133:537-543.
15. Funatsu H, Yamashita H, Ikeda T, Mimura T, Eguchi S, Hori S. Vitreous levels of
interleukin-6 and vascular endothelial growth factor are related to diabetic
macular edema. Ophthalmology 2003;110:1690-1696.
16. Ayalasomayajula SP, Kompella UB. Celecoxib, a selective cyclooxygenase-2
inhibitor, inhibits retinal vascular endothelial growth factor expression and
vascular leakage in a streptozotocin-induced diabetic rat model. Eur J Pharmacol
2003;458:283-289.
97
17. Liu XH, Kirschenbaum A, Yao S, Stearns ME, Holland JF, Claffey K, Levine AC.
Upregulation of vascular endothelial growth factor by cobalt chloride-simulated
hypoxia is mediated by persistent induction of cyclooxygenase-2 in a metastatic
human prostate cancer cell line. Clin Exp Metastasis 1999;17:687-694.
18. Skold M, Cullheim S, Hammarberg H, Piehl F, Suneson A, Lake S, Sjogren A,
Walum E, Risling M. Induction of VEGF and VEGF receptors in the spinal cord
after mechanical spinal injury and prostaglandin administration. Eur J Neurosci
2000;12:3675-3686.
19. Castro MR, Lutz D, Edelman JL. Effect of COX inhibitors on VEGF-induced retinal
vascular leakage and experimental corneal and choroidal neovascularization. Exp
Eye Res 2004;79:275-285.
98
Chapter 8
Chapter 9
The pro’s and contra’s of

intravitreal triamcinolone
injections for uveitis and
edema.
Bram van Kooij1 MD, Philip de Vries2 PhD, Aniki Rothova1 MD, PhD.
F.C. Donders Institute of Ophthalmology1 and Department of

Pharmacy2, University Medical Center Utrecht, The Netherlands.
(Ocular Immunology and Inflammation, in press)
99
The pro’s and contra’s of IVTA injections for uveitis and CME
Abstract
Intravitreal triamcinolone acetonide (IVTA) injections are gaining in
popularity and are nowadays regularly performed for various ocular
diseases. We performed a literature review on the use, efficacy and
complications of IVTA application in non-infectious uveitis and
inflammatory cystoid macular edema. IVTA applications brought a
quick improvement in vision in the majority of cases. The drawbacks
included the temporary duration of the effect with the need for
repeated injections which re-exposed patients to the risk of
complications. The risk of bacterial endophthalmitis was 0.5% and was
influenced also by the specific IVTA preparation. To diminish the risk of
endophthalmitis, we have chosen for ready-for-use IVTA-injections
prepared by our pharmacy department in which 90 % of the toxic
additives was removed and the dispensed dose of triamcinolone
acetonide was validated. Elevated IOP was seen in 30-43% of eyes and
cataract developed in 29% of eyes of patients, who were usually of
advanced age. In conclusion, the rapid effect of IVTA might be of value
in severe presentations of non-infectious uveitis and CME and might
shorten the time interval needed for the improvement
100
Chapter 9
Introduction
Intravitreal triamcinolone acetonide (IVTA) injections had their first
experimental application in the beginning of the eighties,1-3 and were
reported to be effective in experimental animal models of subretinal
neovascularisation, blood-retinal barrier breakdown and proliferative
vitreoretinopathy (PVR).3-6 Since the first use in human eyes at the end
of the nineties, IVTA injections are gaining in popularity and are
nowadays regularly performed for various ocular diseases including age
related macular degeneration,7-9 uveitis10;11 and cystoid macular edema
(CME) from diverse causes such as diabetes mellitus,12-15 following
intraocular surgery,16;17 uveitis18-20 and retinal vein occlusions.21-23 The
beneficial effect of IVTA injections was also reported in PVR and is
therefore also employed during the pars plana vitrectomy (PPV).24;25
The intravitreous administration of small amounts of triamcinolone
acetonide (2-25mg) provides high intraocular levels of corticosteroids
but with limited or no systemic side effects compared to other ways of
administration.26
Uveitis is an often chronic and potentially blinding disease. Of all
causes of blindness in the United States, approximately 10 to 15 % is a
result of uveitis.27 Thirty-five % of uveitis patients exhibited blindness
or visual impairment.28 Non-infectious uveitis is usually treated with
steroidal or non-steroidal immunosuppressive agents, however lack of
efficacy and adverse effects of medications form frequent problems.
CME is one of the most common sight-threatening complications in
uveitis and manifests itself not only during the active phase of uveitis
but also over prolonged periods even when inflammation has
subsided.29-31 Inflammatory CME is often responsive to corticosteroids
given either systemically or as periocular injections. Other treatment
options include acetazolamide, non-steroidal anti-inflammatory drugs
(NSAIDs), immunomodulatory agents, laser grid photocoagulation, and
PPV. Despite the diverse therapeutic possibilities, corticosteroids still
represent a major option for the treatment of inflammatory CME.
However, high dosages and a need for repeated administrations
constitute a problem because of their systemic and ocular adverse
effects. In the pursuit for additional treatments for uveitis and
inflammatory CME, IVTA application might represent a new therapeutic
option.
101
In this review, we address the efficacy of intravitreal triamcinolone

acetonide in uveitis and inflammatory CME.
Uveitis
So far (Pubmed search March 2005, used terms uveitis and
triamcinolone), 3 studies have been published on IVTA administration
in non-infectious uveitis (2 case series10;11 and 1 case report32). In
these three studies 14 patients (14 eyes) were included (7 with
Behçet's disease, 4 with idiopathic vasculitis, 1 with idiopathic
panuveitis, 1 with pars planitis, 1 with idiopathic uveitis; Table 1). IVTA
injection has also been reported as an additional therapeutic tool in two
patient with sympathetic ophthalmia.33;34
The indications for IVTA included exacerbation of uveitis despite
systemic multiple drug treatment in all patients as well as adverse
effects (3 patients) and non-compliance (2 patients) to systemic
corticosteroids and /or other immunosuppressive agents (Table 1).
Mean visual acuity before treatment with IVTA injection was 0.05
(range: hand movements to 0.2). Best mean visual acuity after
treatment (1 week post treatment in 10 eyes and 4-6 weeks in 4 eyes;
Table 2) was 0.6 (range: 0.2 to 1.0; Table 1). In 1 eye visual acuity
returned to the pre-treatment level within 4 months. In 6 patients
systemic treatment could be reduced (Table 2). However, in at least
9/14 (64%) of eyes, IVTA had only a temporary effect. Eight/14 (57%)
eyes required repeated IVTA injections for exacerbations.
Application of IVTA solely for the postsurgical or post-traumatic
endophthalmitis had no beneficial effect on visual acuity compared to
treatment with intravitreal antibiotics alone or to a combination of
intravitreal antibiotics and IVTA application.35;36
The use of IVTA injections in non-infectious uveitis is limited,
probably because of the temporary effect. However, it could become a
supportive tool in the management of non-infectious uveitis as the
additional medication in acute exacerbations or as a part of initial
treatment to quickly accomplish the improvement before the less
quickly working drugs reach the desirable effect.
102
Table 1: General characteristics of patients treated with intravitreal triamcinolone acetonide injection for non-infectious uveitis
Publication location of uveitis, diagnosis duration previous indication initial best

(N) age or associated disease of uveitis treatment of IVTA VA VA
Benitez vasculitis, not specified not specified systemic + periocular CS, CyA 0.05 0.6
del Castillo vasculitis, not specified not specified systemic + periocular CS, CyA 0.1 0.7
Sanchez, vasculitis, not specified not specified systemic + periocular CS, CyA exacerbation 0.05 0.9
2001 (n=10) vasculitis, not specified not specified systemic + periocular CS, CyA, MTX of uveitis or 0.05 0.8
ref. no: 7 panuveitis, not specified not specified systemic + periocular CS, CyA, Aza inefficacy of HM 0.8
mean: 27 Behçet's disease not specified systemic + periocular CS, CyA standard 0.05 0.6
range: 8-34 Behçet's disease not specified systemic + periocular CS, CyA treatment HM 0.7
Behçet's disease not specified systemic + periocular CS, CyA 0.1 0.8
Behçet's disease not specified CyA 0.05 0.6
pars planitis, not specified not specified systemic + periocular CS 0.1 0.6
Degenring, 17 not specified 5 years systemic + periocular CS, CyA exacerbation 0.1 0.5
2003 (n=1) adverse effects
ref. no: 27
Kramer, 13 incomplete Behçet's disease 5 years CS, CyA, MTX exacerbation, HM 1.0
2004 (n=3) non-compliance
ref. no: 8 adverse effects
18 incomplete Behçet's disease 3-5 years CS, CyA, MTX exacerbation, 0.05 0.2
non-compliance
50 incomplete Behçet's disease 3-5 years CS, CyA, MTX exacerbation, 0.2 0.5
adverse effects
Mean VA: 0.05 0.6
IVTA = intravitreal triamcinolone acetonide, VA = visual acuity, CS = corticosteroids, CyA = cyclosporine A, MTX = methotrexate,
Aza = azathioprine, HM = hand movements; adverse effects=intolerance for standard treatment, na = not applicable
103
Chapter 9
104
Table 2: Outcome of patients treated with intravitreal triamcinolone acetonide for non-infectious uveitis
duration of reduction patients with

Number maximum effect effect after repeated of systemic elevated IOP
Publication: of eyes on VA within follow-up injection injections treatment (treatment)
Benitez 10 1 week 3 to18 months ≥ 3 months 5 eyes 6 patients 2 (topical)

del Castillo
Sanchez
Degenring 1 5 weeks 4 months <4 months no no 1 (topical)
Kramer 3 4 to 6 weeks not specified 2 to 6 months 3 eyes no 1 (topical)
VA = visual acuity, IOP = intra-ocular pressure

Table 3: General characteristics of patients treated with intravitreal triamcinolone acetonide injection for inflammatory CME
duration
Publication location of uveitis, diagnosis of CME indication initial best
(N) age or associated disease (months) previous treatment of IVTA VA VA
Young, 48 intermediate, unknown 6 systemic + periocular CS, CyA refractory CME 0.33 0.66
2001 (n=6) 57 posterior, unknown 6 systemic + periocular CS refractory CME 0.33 0.66
ref. no: 16 25 intermediate, unknown 9 periocular CS refractory CME 0.33 1.0
23 intermediate, unknown 12 systemic + periocular CS, CyA, Myco refractory CME 0.25 0.66
28 intermediate, unknown 48 systemic + periocular CS, CyA refractory CME 0.25 0.5
38 intermediate, unknown 7 systemic + periocular CS, CyA refractory CME <0.1 0.66
Antcliff, 59 retinal vasculitis, unknown 14 systemic + periocular CS, CyA refractory CME 0.30 0.5
2001 (n=6) 33 retinal vasculitis, unknown 132 systemic + periocular CS, CyA, Aza refractory CME 0.25 0.33
ref. no: 15 adverse effects
44 retinal vasculitis, unknown 24 systemic + periocular CS, CyA refractory CME 0.17 0.4
56 HLA B27+ AAU 96 systemic + periocular CS, CyA refractory CME 0.15 0.2
54 HLA B27+AAU 72 systemic + periocular CS, CyA refractory CME 0.01 0.02
34 intermediate, MS 60 systemic + periocular CS, CyA refractory CME 0.02 0.25
Karacorlu, mean: 37 Behçet's disease 2 to 4 acetazolamide refractory CME mean: 0.22 0.33
2004 (n=10) range: 28-47
ref. no: 17
Martidis, 60 Birdshot chorioretinopathy n.s. topical NSAID’s refractory CME 0.33 0.8
2001 (n=2) 38 Birdshot chorioretinopathy n.s. periocular CS recurrence of CME 0.10 0.66
ref. no: 32
Mean VA: 0.2 0.4
Sorensen, mean: 48 posterior, unknown n.s. systemic CS and/or CyA refractory CME not specified
2005 (n=17) range: 20-70
ref. no: 33
CME = cystoid macular edema, IVTA = intravitreal triamcinolone acetonide, VA = visual acuity, AAU acute anterior uveitis, CS = corticosteroids, CyA = cyclosporine A,
Aza = azatriopine, Myco = mycophenylate, NSAID’s = non-steroidal anti-inflammatory drugs, n.s = not specified
105
Chapter 9
106
Table 4: Outcome of patients treated with intravitreal triamcinolone acetonide for inflammatory CME
maximum duration of reduction patients with

number of effect on maximum effect effect after repeated of systemic elevated IOP,
Publication patients VA within on OCT within follow-up injection injections treatment (treatment)
Young 6 1-6 months 1 to 3 months 12 months 3 to > 6 months 0 no 5 (4 topical, 1 TE)
1 2
Antcliff 6 1-3 months 6 to 7 days 3-16 months 1.5 to 3 months 3 2 patients 1 (TE)
Karacorlu 10 1-6 months 1 to 3 months 6 months 3 to > 6 months 5 no 3 (topical)
Martidis 2 2-3 months 2 to 3 months 6 months > 6 months 0 no 0
Sorensen 17 2-3 months 1 week to 1 month 6 months 3 to 4 months 0 no not specified
1 2
VA = visual acuity, OCT = optical coherence tomography , IOP = intra-ocular pressure, TE = trabeculectomy, no effect in 1 patient, repeated
injections after IVTA were periocular injections
Chapter 9
Inflammatory CME
So far, 5 studies have been published (Pubmed search March 2005) on
the use of IVTA injections for inflammatory CME (5 case series). These
studies included 41 patients (41 eyes; 10 with Behçet's disease, 6 with
intermediate uveitis, 3 with idiopathic vasculitis, 2 with Birdshot
choroidoretinopathy, 2 with HLA-B27 associated uveitis, 18 with
idiopathic uveitis posterior; Table 3).18-20;37;38 The indications for IVTA
injections included resistance to standard treatment for CME (40
patients, in 1 patient combined with adverse effects of systemic
treatment) and recurrence of CME (1 patient). Mean pre-treatment
duration of CME was 23.5 months (range 2-132 months). Mean pre-
treatment visual acuity was 0.2 (range: 0.01 to 0.5). Mean maximum
visual acuity after treatment (1 to 6 months; Table 4) was 0.4 (range:
0.02 to 1.0). CME diminished or disappeared on optical coherence
tomography (systematically done before and after IVTA injection in 35
patients) usually within 6 days to 3 months. Systemic treatment could
be reduced in 2 patients. Duration of the effect was usually between
1.5 months and 6 months. In 8 eyes repeated injections were
required.18,20 In total, 31/41 eyes (76%) had recurrence of CME within
6 months (Table 4).
IVTA administration seems to be an effective supplementary tool
in the management of refractory CME when standard treatment failed
to control CME, however the effect is often temporary. The effect of
IVTA injection on long term prognosis of CME will be very difficult to
evaluate since it also depends on the chronicity and activity of uveitis
and other medications used.
Duration of effect
Logically, the duration of the effect depends on the duration of the
intraocular availability of triamcinolone acetonide. From experimental
studies in the rabbit eye we know that the clearance of triamcinolone
acetonide from the vitreous body is most rapid in eyes after vitrectomy
and lens extraction (6.5 days), followed by vitrectomized eyes only
(16.8 days), ultimately followed by not-operated eyes (41 days).2
Mason et al. demonstrated detectable intravitreal triamcinolone
acetonide concentrations in four human eyes (phakic and pseudophakic
eyes with intact capsular bag) after 1.25 to 2.75 months after a single
107
injection (4 mg in 0.1 ml).39 These eyes underwent therapeutic

vitrectomy at various post injection intervals (range 1.25 to 5 months).
The mean estimated half-life of an intravitreal injection of 4 mg was
between 15.4 and 18.6 days in the non-vitrectomized eyes.40;41 In the
vitrectomized eyes half live was reported to be 3.2 days.41 Beer et al.
calculated that triamcinolone acetonide could be detectable for
approximately 3 months in aqueous humor of non-vitrectomized
eyes.41 The maximum duration of the effect (using 4 mg triamcinolone
acetonide in 0.1 ml) was 4 to 5 months, and was similar for eyes with
diabetic and inflammatory CME.40 Maximum effect on visual acuity was
found between 1 and 6 months after IVTA.
In conclusion, triamcinolone acetonide (4 mg in 0.1 ml) is
detectable for approximately 3 months after a single intravitreal
injection which corresponds with observations of clinical efficacy of
approximately 2 to 6 months. Jonas et al. found that intravitreal
injection of approximately 20-25 mg was effective in the order of 7 to 8
months and was even detectable in aqueous humor up to 1.5 years
after the injection.13;42
Adverse effects
The adverse effects of IVTA injections discussed in this section are
based on studies of patients with a variety of ocular disorders, not
uveitis solely.
Endophthalmitis
Bacterial endophthalmitis is a most serious complication of IVTA
injections and had an estimated prevalence of 0.5 % (10/2154).43;44
In the studies dealing with endophthalmitis after IVTA injections,
the terms endophthalmitis, non-infectious endophthalmitis, sterile
endophthalmitis and pseudoendophthalmitis have sometimes been
used inconsistently and the estimated prevalence of all types of
endophthalmitis was reported to be 1.1 % (24/2154).43;44
In a study performing the injection with so called “washed and
filtered” triamcinolone acetonide 0 of 520 injections were complicated
by endophthalmitis.44 In other studies with non-washed triamcinolone
acetonide direct from the vial (Kenacort-A 40 or Kenalog 40), the
108
Chapter 9
percentage of bacterial endophthalmitis was 0.6% (10/1634) and 1.5%

for all types of endophthalmitis (24/1634).43
The term sterile endophthalmitis has been suggested to be used
for cases with a true inflammation as a result of the IVTA and
represents probably a toxic reaction to triamcinolone acetonide, the
preservative or the solvent and not an infectious process. The term
pseudoendophthalmitis has been suggested to be used for cases were
the injected triamcinolone acetonide itself creates a masquerade
picture that looks like an inflammation but consists actually of
triamcinolone acetonide crystals migrating from the vitreous into the
anterior chamber, masquerading as inflammatory cells, sometimes
forming a pseudohypopyon (Fig.1).43;45 Pseudoendophthalmitis was
documented in three large studies and was more common in patients
with a history of cataract extraction and/or pars plana vitrectomy than
in patients with no history of intraocular surgery (19 versus 2 eyes).45-
47
All 19 previously described non-culture positive cases presented

within 3 days after the injection with the following symptoms (eyes):
acute loss of vision (13), pain (4), anterior chamber cells (and/ or
crystals) (14), vitritis (and/ or crystals) (14) and (pseudo)hypopyon
(18) and all resolved within 2 weeks.45-47 All 10 culture positive
endophthalmitis eyes presented in an average of 8.5 days (range 1-15
days) with anterior chamber cells (10), pain (7), decreased vision (8),
hypopyon (9), red eye (6), vitritis (10).The management included
vitreous biopsy and intraocular antibiotics in 8 eyes and vitrectomy in 2
eyes. Visual acuity did not recover to pre-treatment levels in 6/10
treated eyes.46;48 Distinction between genuine and pseudo-
endophthalmitis is essential since endophthalmitis is an extremely
serious sight threatening complication. The time interval between the
IVTA and onset of symptoms might form an indication as all cases of
non-infectious endophthalmitis were manifest within 2 days in contrast
to the later onset of infectious endophthalmitis which manifested
usually between 10 and 15 days after the injection. However, the
distinction based on these intervals is not categorical as 3 eyes
developed infectious endophthalmitis within 2 days after the IVTA
injection. The diagnosis of pseudo-endophthalmitis seems reasonable
when typically, within 2 days after IVTA administration, triamcinolone
acetonide crystals in the vitreous body and/or anterior chamber are
109
recognized. Otherwise, the inflamed eyes should be closely followed

and carefully managed as not to miss infectious endophthalmitis. The
otherwise dramatic signs of infectious endophthalmitis as pain and
redness might be reduced due to the effect of injected steroids.
Possible risk factors comprise the use of multidose vials, presence
of diabetes mellitus, filtering bleb and blepharitis.48 The optimal
prevention of this complication is suggested by various studies and
includes the application of the injection under sterile conditions,
administration of topical antibiotics before and after injection in high
risk cases and removal of preservatives and vehicle components.44;49;50
Elevated Intraocular Pressure

After injection of 4 mg triamcinolone acetonide, elevated intraocular
pressure (IOP ≥ 5 mmHg above baseline IOP or ≥ 24 mmHg)
developed in 30% to 43% of injected eyes (102/277 of eyes, 37%).51-
53
After injection of 20-25 mg triamcinolone acetonide, 41% (112/272)
of patients developed an IOP higher than 21 mm Hg.54 In comparison,
after posterior subtenon's injection of 40 mg triamcinolone acetonide
30 to 36% developed elevated IOP (≥ 8 mmHg above baseline IOP or
≥ 21 mmHg). 55,56 Elevated IOP is usually measured 1 week to 3
months after injection but in some cases elevated IOP was noted within
1 week and in one patient later than 9 months after injection,
indicating that it is imperative to monitor IOP for an extended period of
time.51;53;54;57 The elevated pressure is usually controlled with
temporary topical treatment; however cases of filtering surgery (in
3/305, 1% of injected eyes) and vitrectomy to remove the
triamcinolone acetonide remnants have been reported. 51;54;58;59
Repeated injections did not influence the incidence of elevated IOP,
however specific data of steroid responders and non-responders were
not given.53;60;61
Higher risk of getting elevated IOP (RR = 2.1, P < .01) had eyes
with baseline IOP above 15 mmHg.53 No correlation was found between
IOP elevation and age, gender, ocular disease for which the IVTA
injection was given, diabetes mellitus, hypertension, smoking and
visual acuity. Glaucoma represents a logical contraindication to IVTA
application; however Jonas et al. and Smithen et al. reported that in 26
patients with preexistent glaucoma who received IVTA, the IOP rise
was not higher then in patients without glaucoma. This surprising result
110
Chapter 9
might be in part explained by the small number of patients included. In

addition, the treatment of glaucoma patients and their IOP before and
after IVTA was not specified.53;54 Jonas et al reported also a correlation
between IOP level and younger age, although the groups compared
were of advanced age (71 years versus 77 years).54
Occasional excessive increase of IOP directly after the
administration of injection was reported and can be associated with
damage to retinal ganglion cells and might even provoke retinal
vascular occlusion.62 Some ophthalmologists therefore take
prophylactic measures, like an aqueous humor tap, pressure lowering
agents or lowering IOP before the injection by oculopressure.8;60;63
Cataract
Corticosteroids have a cataractogenic effect. Cataract surgery was
performed in 29% eyes treated with triamcinolone acetonide versus
5% in placebo group (P = .003) after 12 to 34 months of follow up.51
Another study reported a significantly increased degree of cataract in
all layers of the lens after a mean follow-up of 7 months (range 3 to 20
months). The increase in the degree of cataract correlated significantly
with the duration of follow-up and the number of injections with IVTA.14
The patients in these studies had an average age of more than 70
years. In younger patients, the data about the development of cataract
following IVTA are not available. Cataract progression after subtenon's
injections of triamcinolone acetonide was reported to occur in 18%
(mean follow-up, 15 months) to 36% (mean follow-up, 23 months) of
injected eyes.55;56
Jonas et al speculated that the steroids injected in the eye may
affect the internal structures of the eye and might cause more and
unusual types of surgical complications after cataract surgery. They
evaluated 22 lens extractions after IVTA injection and did not find more
or a changed profile of complications.64
An interesting report hypothesized that the development of
posterior subcapsular cataract following IVTA is induced by elevated
IOP. Fifty-one % of IOP responders had progression of cataract versus
3 % of the non-responders (P < .0001, log-rank test).65 This
association suggests a similar mechanism responsible for the
development of corticosteroid-induced cataract and corticosteroid-
induced high IOP.
111
Sporadic complications
An unintended sub-retinal application of triamcinolone acetonide has
been reported, resulting in sub-retinal hemorrhage and atrophy.66
Other possible complications might include corneal erosion, central
serous retinopathy, allergic reaction to triamcinolone acetonide or the
vehicle, conjunctival ulceration, retinal detachment and vitreous
hemorrhage.
Experiences in the UMC Utrecht

In our institution, we have chosen an intraocular administration of so
called “washed” triamcinolone acetonide preparation by van de Poll and
de Blois (personal communication, department of pharmacy of the
Maxima Medical Center and Catharina Hospital, The Netherlands) and
distinct from that Jonas et al. used (Table 5). The preparation of
“filtered” triamcinolone used by Jonas requires many steps with the
potential risk of microbial contamination (Table 5). Our decision was
dictated by the fact that the intraocular use of triamcinolone acetonide
suspension directly from the vial was associated with more
complications.
Preparation of the ”washed” IVTA

The methods of preparation and analysis of IVTA were adapted from
the methods used by van de Poll and de Blois (personal
communication). In short, IVTA injections were prepared aseptically
under laminar flow conditions in the pharmacy department of our
hospital. The available preparation, Kenacort-A 40 (Bristol-Meyers
Squibb) contains not only 40 mg of TA per ml but also the preservative
benzylalcohol (10mg/ml), the wetting agent Polysorbate 80 (E433) and
0.63% of the suspending agent carmellose (E466). As these additives
are potentially harmful to the eye they should be removed. This was
done by centrifuging the vial with Kenacort-A 40 at 3000 rpm. The
clear solution (approx. 0.9 ml) was aseptically removed and replaced
with saline solution. The resulting suspension retained a concentration
of 40 mg/ml. Then 0.1 ml of this suspension containing 0.4 mg of TA
was filled into a 1 ml syringe. The syringes were made to order a few
hours before administration. Each 1 ml vial of Kenacort-A 40 yielded
approx. 7 syringes.
112
Chapter 9
Table 5: Technique of triamcinolone acetonide preparation.
Washed and filtered (Jonas)
• 0.625 ml triamcinolone acetonide (40 mg/1 ml) was extracted directly from the Kenacort vial in
a syringe.
• The extracted volume was diluted in Ringer’s solution and pressed through a Millipore membrane
filter (pore size 5 µm), placed on top of the syringe. Triamcinolone crystals remained in the
syringe.
• This procedure was repeated 3 times
• At the end 0.2 ml of solution was left in the syringe.
Washed (van de Poll and de Blois)
• The vial with triamcinolone acetonide (40 mg/1 ml) was centrifuged at 3000 rpm.
• The clear solution (approx. 0.9 ml) was aseptically removed and replaced with saline solution.
• 0.1 ml of the suspension was filled into a 1 ml syringe.
Directly from the vial
• 0.1 ml triamcinolone acetonide (40 mg/1 ml) was extracted direct from the Kenacort/ Kenalog
vial in a syringe.
Garcia-Arumi et al. described a similar technique with the exception

that they did not perform the procedure aseptically and under laminar
flow conditions, introducing the potential risk of microbial
contamination.67 They also compared different techniques for
purification of TA including the technique of Jonas et al. and overnight
sedimentation and concluded that all techniques effectively reduced the
concentration of benzylalcohol. However the final concentration of TA
was much lower in the technique by Jonas et al., probably because of
the entrapment of TA in the filter.67 They recommended the
centrifugation technique as the best way of administering the drug.
To validate our method of preparation we analyzed all the
syringes that were not dispensed. The concentration of TA and of
benzylalcohol in the IVTA was measured using a HPLC method that was
newly developed. Over a period of 7 months we prepared a total of 162
syringes of which 76 were dispensed and 86 were analyzed. The
average TA concentration found was 3.8 mg/0.1 ml (sd. 0.6 mg/0,1
ml). The average concentration of benzylalcohol found was 0.089
mg/0.1 ml (sd. 0.016 mg/0,1 ml) in contrast to 1.0 mg/0.1 ml in the
original suspension. Garcia-Arumi et al. found an average concentration
of 0.097mg / 0.1 ml which is very similar to our finding.67 Not only
113
benzylalcohol was removed for more than 90 % but also the potential
toxic agents Polysorbate 80 (E433) and carmellose (E466).
Technique of IVTA injections

Our technique was performed according to the recommendations for
intravitreal injections published earlier and was given in a supine
position in a room for minor surgical procedures in our out-patients
department.49;50 In short, 10% of topical povidone-iodine solution and
topical anesthetics were applied before the IVTA. After placement of a
sterile eyelid speculum, intravitreal injection with “washed”
triamcinolone (4 mg/0.1ml) was performed with a 27-gauge needle in
the lower quadrant, 3.5 mm (pseudophakic patients) or 4.0 mm
(phakic patients) from limbus. Patients with IOP of >10 mmHg received
250 mg acetazolamide orally prior to the injection. Patients were
contacted one day after the IVTA application by phone and examined
within 1 week after the injection.
Results
The results of our IVTA treatment for inflammatory CME are given in
Table 6. We treated 19 eyes (17 patients) of whom 10 had an increase
of vision 0.5 to 6 months after the IVTA injection, however in 6 of
these 10 eyes a relapse in visual acuity loss occurred. The other 7 eyes
showed no improvement in visual acuity, however all of these had
chronic inflammatory CME for many years (Table 6).In 3/19 eyes (1
patient aphakic; 2 patients pseudophakic) triamcinolone acetonide
crystals were visible 1 or 2 days after IVTA and all of whom
spontaneously recovered within several days (Fig 1). Elevated IOP was
noted in 5/19 (26%) eyes and was topically treated in all patients.
Three out of the 19 patients had immediately following injection
subjective symptoms of total loss of vision and on fundoscopy pulsation
of the veins was visible; all these patients had elevated intraocular
pressure (all above 50 mmHg). In these patients aqueous tap was
directly performed and in all vision restored within seconds and IOP
normalized. This led us to change our policy and perform the aqueous
tap before the IVTA in patients with compromised retinal vasculature.
114
Table 6: Own experiences: Patients treated with intravitreal triamcinolone acetonide injections for refractory inflammatory CME
Duration months
of CME initial best to reach VA end of
No. Age (years) Location, etiology of uveitis Previous treatment VA VA best VA follow-up
Follow-up 6 months 6 months

1 82 10 panuveitis, unknown acetazolamide, periocular CS, NSAID's 0.4 0.6 2 0.3
2 57 2 panuveitis, unknown acetazolamide, NSAID's, laser grid 0.16 0.16 - 0.16
3 41 10 posterior with occlusive vasculitis, unknown systemic CS, MTX, CyA, Aza, colchicine 0.25 0.6 3 0.25
4 43 4 panuveitis, Behçet's disease systemic CS, MTX, CyA, Aza, colchicine 0.6 1.0 0.5 1.0
5 62 5 panuveitis, prior bartonella infection acetazolamide, periocular CS 0.6 1.0 6 1.0
6 59 10 panuveitis, unknown acetazolamide, periocular CS, MTX, NSAID's 0.7 0.7 - 0.6
7 73 10 panuveitis, sarcoidosis periocular CS, laser grid 0.1 0.1 - 0.1
8 66 5 panuveitis, unknown acetazolamide, periocular CS, laser grid 0.05 0.05 - 0.05
9 55 4 anterior, HLA-B27+ acetazolamide, systemic + periocular CS, MTX 0.05 0.05 - 0.05
10 41 15 posterior, unknown systemic CS, acetazolamide, MTX FC FC - FC
11 39 4 anterior, unknown systemic CS, grid laser 0.1 0.1 - 0.1
12 65 0.5 posterior, sarcoidosis acetazolamide, periocular CS 0.16 0.25 0.5 0.2
13 43 1.5 posterior, Crohn’s disease systemic + periocular CS, MTX, CyA, infliximab 0.1 0.16 0.5 0.16
14 55 7 posterior, unknown acetazolamide, periocular CS 0.3 0.3 - 0.3
Mean VA 0.16 0.2 0.16
Follow-up 3 months 3 months

15 59 0.2 posterior, birdshot MTX, systemic CS 0.4 0.7 1 0.6
16 59 0.2 posterior, birdshot MTX, systemic CS 0.5 0.7 2 0.7
17 41 10 posterior with occlusive vasculitis, unknown systemic CS, MTX, CyA, Aza, colchicine 0.6 0.6 - 0.6
18 60 1.5 panuveitis, unknown systemic CS 0.7 1.0 3 1.0
19 62 0.3 posterior, unknown periocular CS 0.2 0.3 1 0.2
Mean VA 0.4 0.6 0.5
CME = cystoid macular edema, VA = visual acuity, CS = corticosteroids, CyA = cyclosporine A, Aza = azatriopine, NSAID’s = non-steroidal anti-inflammatory drugs, FC = finger
counting
115
Chapter 9
Figure 1: Pseudo hypopyon formed by triamcinolone acetonide crystals in the anterior

chamber of patient 6 (Table 6). The pseudo hypopyon was noted one day after IVTA and
disappeared within 1 week
Conclusions
IVTA application brought in the majority of cases with non-infectious
uveitis and inflammatory CME a quick improvement in vision, even in
patients with otherwise intractable and prolonged CME. The IVTA might
therefore shorten the time interval needed to achieve improvement or
remission in those patients otherwise treated with slowly acting
medications. The role of IVTA might especially be of value in the acute
presentations and/or severe recurrences of intraocular inflammation.
The disadvantages included the limited duration of the effect and the
need for repeated injections which re-expose patients to the risk of
endophthalmitis, elevated IOP and cataract formation. The high relapse
rate we found in this review could be due to the design of the discussed
studies. There was no protocol in how to manage the concomitant
systemic immunosuppression after IVTA, which could influence the
results. A randomized clinical trial instead of case series would give
more information about the duration of the effect. The risk of bacterial
endophthalmitis was 0.5% of all injected eyes and might be further
reduced by performing the procedure under strict sterile circumstances.
IOP should be monitored carefully, because of the risk of elevated IOP
in 30 to 43% of eyes. Although the slow-release devices for intraocular
corticosteroid administration might solve the temporary effect of IVTA,
116
Chapter 9
the enhanced risk of glaucoma might become a serious disadvantage.

Further studies, preferably randomized clinical trials with higher
number of patients of various uveitis entities and longer follow-up are
needed to determine the exact place of IVTA in the treatment of
(chronic) uveitis and inflammatory CME, and to recognize which uveitis
entities might especially benefit from IVTA applications.
117
Reference List
1. McCuen BW, Bessler M, Tano Y, Chandler D, Machemer R. The lack of toxicity of
intravitreally administered triamcinolone acetonide. Am.J.Ophthalmol.
1981;91:785-8.
2. Schindler RH, Chandler D, Thresher R, Machemer R. The clearance of intravitreal
triamcinolone acetonide. Am.J.Ophthalmol. 1982;93:415-7.
3. Tano Y, Chandler D, Machemer R. Treatment of intraocular proliferation with
intravitreal injection of triamcinolone acetonide. Am.J.Ophthalmol. 1980;90:810-
6.
4. Wilson CA, Berkowitz BA, Sato Y, Ando N, Handa JT, de Juan E Jr. Treatment
with intravitreal steroid reduces blood-retinal barrier breakdown due to retinal
photocoagulation. Arch.Ophthalmol. 1992;110:1155-9.
5. Ishibashi T, Miki K, Sorgente N, Patterson R, Ryan SJ. Effects of intravitreal
administration of steroids on experimental subretinal neovascularization in the
subhuman primate. Arch.Ophthalmol. 1985;103:708-11.
6. Antoszyk AN, Gottlieb JL, Machemer R, Hatchell DL. The effects of intravitreal
triamcinolone acetonide on experimental pre-retinal neovascularization. Graefes
Arch.Clin.Exp.Ophthalmol. 1993;231:34-40.
7. Challa JK, Gillies MC, Penfold PL, Gyory JF, Hunyor AB, Billson FA. Exudative
macular degeneration and intravitreal triamcinolone: 18 month follow up.
Aust.N.Z.J.Ophthalmol. 1998;26:277-81.
8. Gillies MC, Simpson JM, Luo W, Penfold P, Hunyor AB, Chua W et al. A
randomized clinical trial of a single dose of intravitreal triamcinolone acetonide
for neovascular age-related macular degeneration: one-year results.
Arch.Ophthalmol. 2003;121:667-73.
9. Jonas JB, Degenring RF, Kreissig I, Friedemann T, Akkoyun I. Exudative age-
related macular degeneration treated by intravitreal triamcinolone acetonide. A
prospective comparative nonrandomized study. Eye 2004.
10. Benitez Del Castillo Sanchez JM, Garcia SJ. [Intravitreal injection of
triamcinolone acetonide in non infectious uveitis]. Arch.Soc.Esp.Oftalmol.
2001;76:661-4.
11. Kramer M, Ehrlich R, Snir M, Friling R, Mukamel M, Weinberger D et al.
Intravitreal injections of triamcinolone acetonide for severe vitritis in patients
with incomplete Behcet's disease. Am.J.Ophthalmol. 2004;138:666-7.
12. Sutter FK, Simpson JM, Gillies MC. Intravitreal triamcinolone for diabetic macular
edema that persists after laser treatment: three-month efficacy and safety
results of a prospective, randomized, double-masked, placebo-controlled clinical
trial. Ophthalmology 2004;111:2044-9.
13. Jonas JB, Degenring RF, Kamppeter BA, Kreissig I, Akkoyun I. Duration of the
effect of intravitreal triamcinolone acetonide as treatment for diffuse diabetic
macular edema. Am.J.Ophthalmol. 2004;138:158-60.
14. Jonas JB, Kreissig I, Sofker A, Degenring RF. Intravitreal injection of
triamcinolone for diffuse diabetic macular edema. Arch.Ophthalmol.
2003;121:57-61.
15. Massin P, Audren F, Haouchine B, Erginay A, Bergmann JF, Benosman R et al.
Intravitreal triamcinolone acetonide for diabetic diffuse macular edema:
118
Chapter 9
preliminary results of a prospective controlled trial. Ophthalmology

2004;111:218-24.
16. Karacorlu M, Ozdemir H, Karacorlu S. Intravitreal triamcinolone acetonide for the
treatment of chronic pseudophakic cystoid macular oedema. Acta
Ophthalmol.Scand. 2003;81:648-52.
17. Conway MD, Canakis C, Livir-Rallatos C, Peyman GA. Intravitreal triamcinolone
acetonide for refractory chronic pseudophakic cystoid macular edema. J.Cataract
Refract.Surg. 2003;29:27-33.
18. Antcliff RJ, Spalton DJ, Stanford MR, Graham EM, ffytche TJ, Marshall J.
Intravitreal triamcinolone for uveitic cystoid macular edema: an optical
coherence tomography study. Ophthalmology 2001;108:765-72.
19. Young S, Larkin G, Branley M, Lightman S. Safety and efficacy of intravitreal
triamcinolone for cystoid macular oedema in uveitis.
Clin.Experiment.Ophthalmol. 2001;29:2-6.
20. Karacorlu M, Mudun B, Ozdemir H, Karacorlu SA, Burumcek E. Intravitreal
triamcinolone acetonide for the treatment of cystoid macular edema secondary
to Behcet disease. Am.J.Ophthalmol. 2004;138:289-91.
21. Park CH, Jaffe GJ, Fekrat S. Intravitreal triamcinolone acetonide in eyes with
cystoid macular edema associated with central retinal vein occlusion.
Am.J.Ophthalmol. 2003;136:419-25.
22. Bashshur ZF, Ma'luf RN, Allam S, Jurdi FA, Haddad RS, Noureddin BN.
Intravitreal triamcinolone for the management of macular edema due to
nonischemic central retinal vein occlusion. Arch.Ophthalmol. 2004;122:1137-40.
23. Ip MS, Gottlieb JL, Kahana A, Scott IU, Altaweel MM, Blodi BA et al. Intravitreal
triamcinolone for the treatment of macular edema associated with central retinal
vein occlusion. Arch.Ophthalmol. 2004;122:1131-6.
24. Jonas JB, Hayler JK, Panda-Jonas S. Intravitreal injection of crystalline cortisone
as adjunctive treatment of proliferative vitreoretinopathy. Br.J.Ophthalmol.
2000;84:1064-7.
25. Sakamoto T, Miyazaki M, Hisatomi T, Nakamura T, Ueno A, Itaya K et al.
Triamcinolone-assisted pars plana vitrectomy improves the surgical procedures
and decreases the postoperative blood-ocular barrier breakdown. Graefes
Arch.Clin.Exp.Ophthalmol. 2002;240:423-9.
26. Degenring RF, Jonas JB. Serum levels of triamcinolone acetonide after
intravitreal injection. Am.J.Ophthalmol. 2004;137:1142-3.
literature survey. Br.J.Ophthalmol. 1996;80:844-8.
frequency of blindness in patients with intraocular inflammatory disease.
Br.J.Ophthalmol. 1996;80:332-6.
duration, and causes of visual loss in uveitis. Br.J.Ophthalmol. 2004;88:1159-
62.
frequency of blindness in patients with intraocular inflammatory disease.
Br.J.Ophthalmol. 1996;80:332-6.
119

literature survey. Br.J.Ophthalmol. 1996;80:844-8.
32. Degenring RF, Jonas JB. Intravitreal injection of triamcinolone acetonide as
treatment for chronic uveitis. Br.J.Ophthalmol. 2003;87:361.
33. Jonas JB. Intravitreal triamcinolone acetonide for treatment of sympathetic
ophthalmia. Am.J.Ophthalmol. 2004;137:367-8.
34. Ozdemir H, Karacorlu M, Karacorlu S. Intravitreal triamcinolone acetonide in
sympathetic ophthalmia. Graefes Arch.Clin.Exp.Ophthalmol. 2005.
35. Shah GK, Stein JD, Sharma S, Sivalingam A, Benson WE, Regillo CD et al. Visual
outcomes following the use of intravitreal steroids in the treatment of
postoperative endophthalmitis. Ophthalmology 2000;107:486-9.
36. Das T, Jalali S, Gothwal VK, Sharma S, Naduvilath TJ. Intravitreal
dexamethasone in exogenous bacterial endophthalmitis: results of a prospective
randomised study. Br.J.Ophthalmol. 1999;83:1050-5.
37. Martidis A, Duker JS, Puliafito CA. Intravitreal triamcinolone for refractory
cystoid macular edema secondary to birdshot retinochoroidopathy.
38. Sorensen TL, Haamann P, Villumsen J, Larsen M. Intravitreal triamcinolone for
macular oedema: efficacy in relation to aetiology. Acta Ophthalmol.Scand.
2005;83:67-70.
39. Mason JO, III, Somaiya MD, Singh RJ. Intravitreal concentration and clearance
of triamcinolone acetonide in nonvitrectomized human eyes. Retina
2004;24:900-4.
40. Audren F, Tod M, Massin P, Benosman R, Haouchine B, Erginay A et al.
Pharmacokinetic-pharmacodynamic modeling of the effect of triamcinolone
acetonide on central macular thickness in patients with diabetic macular edema.
Invest Ophthalmol.Vis.Sci. 2004;45:3435-41.
41. Beer PM, Bakri SJ, Singh RJ, Liu W, Peters GB, III, Miller M. Intraocular
concentration and pharmacokinetics of triamcinolone acetonide after a single
intravitreal injection. Ophthalmology 2003;110:681-6.
42. Jonas JB. Intraocular availability of triamcinolone acetonide after intravitreal
injection. Am.J.Ophthalmol. 2004;137:560-2.
43. Jager RD, Aiello LP, Patel SC, Cunningham ET, Jr. Risks of intravitreous
injection: a comprehensive review. Retina 2004;24:676-98.
44. Jonas JB, Kreissig I, Degenring RF. Endophthalmitis after intravitreal injection of
triamcinolone acetonide. Arch.Ophthalmol. 2003;121:1663-4.
45. Moshfeghi AA, Scott IU, Flynn HW, Jr., Puliafito CA. Pseudohypopyon after
intravitreal triamcinolone acetonide injection for cystoid macular edema.
Am.J.Ophthalmol. 2004;138:489-92.
46. Nelson ML, Tennant MT, Sivalingam A, Regillo CD, Belmont JB, Martidis A.
Infectious and presumed noninfectious endophthalmitis after intravitreal
triamcinolone acetonide injection. Retina 2003;23:686-91.
47. Roth DB, Chieh J, Spirn MJ, Green SN, Yarian DL, Chaudhry NA. Noninfectious
endophthalmitis associated with intravitreal triamcinolone injection.
120
Chapter 9
48. Moshfeghi DM, Kaiser PK, Scott IU, Sears JE, Benz M, Sinesterra JP et al. Acute
endophthalmitis following intravitreal triamcinolone acetonide injection.
Am.J.Ophthalmol. 2003;136:791-6.
49. Ta CN. Minimizing the risk of endophthalmitis following intravitreous injections.
Retina 2004;24:699-705.
50. Aiello LP, Brucker AJ, Chang S, Cunningham ET, Jr., D'Amico DJ, Flynn HW, Jr.
et al. Evolving guidelines for intravitreous injections. Retina 2004;24:S3-19.
51. Gillies MC, Simpson JM, Billson FA, Luo W, Penfold P, Chua W et al. Safety of an
intravitreal injection of triamcinolone: results from a randomized clinical trial.
52. Wingate RJ, Beaumont PE. Intravitreal triamcinolone and elevated intraocular
pressure. Aust.N.Z.J.Ophthalmol. 1999;27:431-2.
53. Smithen LM, Ober MD, Maranan L, Spaide RF. Intravitreal triamcinolone
acetonide and intraocular pressure. Am.J.Ophthalmol. 2004;138:740-3.
54. Jonas JB, Degenring RF, Kreissig I, Akkoyun I, Kamppeter BA. Intraocular
pressure elevation after intravitreal triamcinolone acetonide injection.
55. Helm CJ, Holland GN. The effects of posterior subtenon injection of triamcinolone
acetonide in patients with intermediate uveitis. Am.J.Ophthalmol. 1995;120:55-
64.
56. Lafranco DM, Tran VT, Guex-Crosier Y, Herbort CP. Posterior sub-Tenon's steroid
injections for the treatment of posterior ocular inflammation: indications,
efficacy and side effects. Graefes Arch.Clin.Exp.Ophthalmol. 1999;237:289-95.
57. Singh IP, Ahmad SI, Yeh D, Challa P, Herndon LW, Allingham RR et al. Early
rapid rise in intraocular pressure after intravitreal triamcinolone acetonide
injection. Am.J.Ophthalmol. 2004;138:286-7.
58. Agrawal S, Agrawal J, Agrawal TP. Vitrectomy as a treatment for elevated
intraocular pressure following intravitreal injection of triamcinolone acetonide.
Am.J.Ophthalmol. 2004;138:679-80.
59. Kaushik S, Gupta V, Gupta A, Dogra MR, Singh R. Intractable glaucoma
following intravitreal triamcinolone in central retinal vein occlusion.
Am.J.Ophthalmol. 2004;137:758-60.
60. Jonas JB, Kreissig I, Degenring R. Intraocular pressure after intravitreal injection
of triamcinolone acetonide. Br.J.Ophthalmol. 2003;87:24-7.
61. Jonas JB, Degenring R, Kreissig I, Akkoyun I. Safety of intravitreal high-dose
reinjections of triamcinolone acetonide. Am.J.Ophthalmol. 2004;138:1054-5.
62. Baudouin C, Chassain C, Caujolle C, Gastaud P. Treatment of cytomegalovirus
retinitis in AIDS patients using intravitreal injections of highly concentrated
ganciclovir. Ophthalmologica 1996;210:329-35.
63. Penfold PL, Gyory JF, Hunyor AB, Billson FA. Exudative macular degeneration
and intravitreal triamcinolone. A pilot study. Aust.N.Z.J.Ophthalmol.
1995;23:293-8.
64. Jonas JB, Kreissig I, Degenring RF. Cataract surgery after intravitreal injection of
triamcinolone acetonide. Eye 2004;18:361-4.
65. Gillies MC, Kuzniarz M, Craig J, Ball M, Luo W, Simpson JM. Intravitreal
triamcinolone-induced elevated intraocular pressure is associated with the
121
development of posterior subcapsular cataract. Ophthalmology 2005;112:139-

43.
66. Modarres M, Parvaresh MM, Peyman GA. Accidental subretinal injection of
triamcinolone acetonide. Ophthalmic Surg.Lasers 1998;29:935-8.
67. Garcia-Arumi J, Boixadera A, Giralt J, Martinez-Castillo V, Gomez-Ulla F,
Corcostegui B et al. Comparison of different techniques for purification of
triamcinolone acetonide suspension for intravitreal use. Br.J.Ophthalmol.
2005;89:1112-4.
122
Chapter 9
Chapter 10
Rapid progression of diabetic

retinopathy in eyes with
posterior uveitis
Judie Knol1 MD, Bram van Kooij1 MD, Harold de Valk2 MD, PhD, Aniki
Rothova1 MD, PhD.
1
FC Donders Institute of Ophthalmology, and Department of Internal
Medicine,2 University Medical Center Utrecht The Netherlands.
123
Rapid progression of DRP in eyes with posterior uveitis
Abstract
Purpose
To report on two patients, who developed rapid progression of
asymmetric diabetic retinopathy (DRP) in eyes affected by posterior
uveitis, in contrast to their fellow eyes not affected by uveitis.
Design
Observational case report.
Methods
Two patients with diabetes mellitus (DM) and unilateral uveitis
underwent repeated ophthalmologic examinations and fluorescein
angiography.
Results
Two patients with DM and unilateral posterior uveitis developed
proliferative DRP in eyes with previous uveitis within 3 months after the
uveitis subsided. In contrast, the retinal findings of non-uveitic eyes
remained unchanged on the follow-up of several years.
Conclusions
Since the pathogenesis of intraocular inflammation and diabetic
retinopathy acts through similar biochemical mediators and pathways,
it is feasible that posterior uveitis accelerates the progression of
diabetic retinopathy. Our results support this hypothesis and point out
a risk for rapid retinopathy development in eyes affected with posterior
uveitis.
124
Chapter 10
Introduction
The pathogenetic agents involved in the development of DRP are not
entirely clarified. In addition to hyperglycemia and retinal hypoxia a
number of vasoactive factors promote pathologic changes in the
microvasculature. This paper reports on rapid progression of
asymmetric DRP in 2 patients with posterior uveitis.
Figure 1. Left eye of patient 1, four weeks (Left panel) and 4 months (Right panel) after
the onset of toxoplasmic retinitis. Left panel, old hyperpigmented scar in satellite
formation to an active yellowish toxoplasmic lesion located adjacent to optic disc.
Associated inflammation of arterial wall is visible. No signs of diabetic retinopathy or
neovascularizations are present. Right panel, 3 months later, extensive neovascular
membrane radiating from optic disc and associated with preretinal hemorrhages has
developed.
Case 1
A 27-year-old female presented in with toxoplasmic chorioretinitis OS.
Patient had diabetes mellitus type I since the age of 19 years and was
treated with insulin. She had no chronic complications of her diabetes,
co-morbidity or co-medication. Ocular examination OD was without
abnormalities. Patient was treated with multiple antiparasitic drugs and
within 6 weeks the lesion became scarified; visual acuity returned to
20/16 and no DRP was visible. Three months later vitreous hemorrhage
125
and optic disc neovascularizations developed (Figure 1 and 2). Despite

continuous laser treatment, vitreous bleeding with retinal traction
recurred. Visual acuity dropped to light perception despite pars plana
vitrectomy with silicone oil injection, retinotomy and endolaser
treatment. After the follow-up of 8 years she still had full visual acuity
OD and no manifest DRP changes.
Figure 2. Retinal photographs of patient 1, four months after the onset of toxoplasmic
retinitis in the left eye. Left, retina of the right eye not affected by toxoplasmic retinitis
exhibits no signs of diabetic retinopathy, and severe proliferative diabetic retinopathy in
the left eye associated with old retinal scar adjacent to optic disc and preretinal
hemorrhage (right).
Case 2
A 58-year-old male presented with acute retinal necrosis in OD caused
by a herpes zoster infection. This patient had an insuline-requiring type
II diabetes mellitus of 18 years’ duration complicated by clinical
polyneuropathy. He was known with a minimal background diabetic
retinopathy in both eyes. The infection in OD was treated with
intravenous acyclovir and his retinal lesions scarified and visual acuity
increased to 20/60. However, after this initial favorable reaction, the
patient developed severe proliferative DRP with multiple
126
Chapter 10
neovascularizations and massive vitreous hemorrhage in the OD within

3 months, while the minimal background retinopathy in OS remained
unchanged (Figure 3). Despite panretinal laser photocoagulation and
vitrectomy combined with retinal cryocoagulation, the progression of
the DRP failed to stop and his useful visual acuity OD was lost to light
perception. However, his OS remained stable and exhibited unchanged
features of mild background DRP.
Figure 3. Asymmetric development of diabetic retinopathy in patient 2 affected by Acute

Retinal Necrosis (ARN) in the left eye. Left panel, right eye not affected by ARN and no
visible diabetic retinopathy in contrast to severe background diabetic retinopathy, which
developed within 3 months after the initial beneficial response to ARN treatment. Note
the extensive capillary non-perfusion and start of laser coagulation treatment in the
lower area of the retina (right panel).
Discussion
We report on 2 patients with a rapid progression of DRP in the eyes
affected by posterior uveitis in contrast to fellow non-uveitic eyes which
remained stable on long-term follow-up. Diverse modifying ocular
factors related to asymmetric development of DRP were previously
described and included presence of endophthalmitis, carotid occlusive
disease and cataract surgery.1-4 On the contrary, protective factors
against DRP consisted of optic atrophy, extensive chorioretinal scarring,
high myopia and retinitis pigmentosa.2 Contradictory reports examined
the effect of intraocular inflammation on the development of DRP and
127
included a study of rapid progression of DRP following endophthalmitis

in contrast to two patients with unilateral chronic uveitis without DRP
manifestations and severe DRP in their fellow eyes without uveitis.3,5,6
The rapid course of the unilateral DRP progression and the stable
disease in the fellow eyes suggest that the asymmetric DRP may have
occurred due to the effects of inflammatory mediators on retinal
vasculature. Multiple biochemical mechanisms have been proposed to
explain the pathogenesis of DRP. A major factor consists of VEGF,
which is known to be a potent pro-angiogenic and permeability factor
and has been implicated in the development of retinal
neovascularizations.7 The expression of many inflammatory cytokines is
increased in ocular fluid of patients with DRP. Likewise, high intraocular
levels of VEGF were found in eyes with uveitis.7 Because intraocular
inflammation and DRP may act through similar biochemical mediators
and pathways, it is possible that the elevated levels of VEGF in uveitis
might have provoked the rapid development of DRP in our patients. In
addition, the vascular wall changes in the retina in posterior uveitis
associated with increased leakage might have also contributed to the
aggressive development of DRP. Further studies are needed to clarify
our findings and hypotheses.
To conclude, our results support the hypothesis that inflammation
can accelerate progression of diabetic retinopathy and point out a risk
for rapid retinopathy development in eyes affected with posterior
uveitis.
128
Chapter 10
References
1. Dogru M, Inoue M, Nakamura M, Yamamoto M. Modifying factors related to
asymmetric diabetic retinopathy. Eye 1998; 12:929-933.
2. Browning DJ, Flynn HW, Blankenship GW. Asymmetric retinopathy in patients
with diabetes mellitus. Am J Ophthalmol 1988; 105:584-589.
3. Dev S, Pulido JS, Tessler HH, et al. Progression of diabetic retinopathy after
endophthalmitis. Ophthalmology 1999 Apr; 106:774-781.
4. Hauser D, Katz H, Pokroy R, Bukelman A, Shechtman E, Pollack A. Occurrence
and progression of diabetic retinopathy after phacoemulsification cataract
surgery. J Cataract Refract Surg 2004 Feb; 30:428-432.
5. Murray DC, Sung VCT, Headon MP. Asymmetric diabetic retinopathy associated
with Fuchs’ heterochromic cyclitis. Br J Ophthalmol 1999 Aug; 83:988-989.
6. Scialdone A, Menchini U, Pietroni C, Brancato R. Unilateral proliferative diabetic
retinopathy and uveitis in the fellow eye: report of a case. Ann Ophthalmol
1991; 23:259-261
7. Vinores SA, Youssri AI, Luna JD, et al. Upregulation of vascular endothelial
growth factor in ischemic and non-ischemic human and experimental retinal
diseases. Histol Histopathol 1997 Jan; 12:99-109.
129
130
Chapter 10
Chapter 11
Summary and Conclusions
131
Summary and conclusions
The aim of this thesis was to elucidate the possible role of vascular
disease in the development of inflammatory cystoid macular edema
(CME) and to explore the new treatment options for inflammatory CME.
Chapter 2 describes a cross-sectional study of 529 patients with

uveitis in which we evaluated the impact of CME on visual acuity. CME
was noted in 33% of all uveitis patients, of whom 44% had low vision
(visual acuity equal to or less than 20/60) in at least one eye.
Moreover, out of all uveitis patients, 35% developed low vision in at
least one eye, which was caused by CME in 42% of the patients. Poor
visual acuity in patients with CME was associated with the advanced
age of the patients, chronic inflammation and various specific uveitis
entities. Visual loss in patients with pan- and intermediate uveitis was
caused in the majority of cases by CME (59% and 85%, respectively).
In conclusion, we established that CME is a major complication causing
visual loss in patients with uveitis, and is especially severe among
elderly patients with chronic disease.
In chapter 3 we determined the breakdown of the blood-retinal barrier

(BRB) by measuring the leakage of immunoglobulin G (IgG) into the
aqueous humor of 112 patients with uveitis. We demonstrated that the
intraocular IgG concentration was elevated in patients with uveitis
compared to non-uveitis controls (P < .001) and was especially high in
patients with CME and retinal ischemia compared to those without
these characteristics (P = .017 and P = .001, respectively). An
association was found between CME and choroidal leakage (P = .043)
but not between choroidal leakage and IgG concentrations, suggesting
that the outer BRB might be selectively permeable for fluid, but not for
large molecules like IgG. In conclusion, we found high intraocular IgG
levels in patients with uveitis, particularly in those with CME and/or
retinal ischemia. Our study revealed a possible selective permeability of
the outer BRB for fluid, but not for large proteins.
In chapter 4 we determined the global cytokine and chemokine

expression pattern by a multiplex immunoassay in the aqueous humor
of 31 uveitis patients and related them to clinical features. Uveitis
samples had higher levels of interleukin-6, interleukin-8, soluble
intercellular adhesion molecule, soluble vascular cell adhesion molecule
132
Chapter 11
(sVCAM), and interferon-inducible protein-10 (IP-10) than non-uveitis

controls. Active uveitis samples had higher levels of interleukin-6,
interleukin-10, interferon-γ, sVCAM, RANTES, and IP-10 than quiescent
uveitis samples. Infectious uveitis was associated with higher levels of
interleukin-10 than non-infectious uveitis (P<.03 for all subgroups). No
significant differences were found between CME and non-CME samples.
We showed that specific mediators were elevated in active and in
infectious uveitis.
Chapter 5 examines the role of age and cardiovascular diseases (and

their risk factors) on the development of CME, in which the data from
97 patients with uveitis obtained by an extensive questionnaire were
related to ophthalmologic features. Multivariate logistic regression
showed that the development of CME was strongly associated with the
advanced age of patients, independent of the duration of uveitis.
Patients older than 50 had a 3.8-fold higher risk of developing CME
than younger patients. In contrast, in this study of 97 persons, no clear
association between cardiovascular diseases, their risk factors and the
development of CME was proven. Our study points out that elderly
patients or patients with onset of uveitis at an advanced age carry an
increased risk of developing CME.
In chapter 6 we assessed the role of cardiovascular morbidity, its risk

factors, and microalbuminuria in the development of inflammatory
CME. In this matched case-control study we included 24 consecutive
patients with uveitis and CME. Twenty-four uveitis patients without
CME, matched for age and duration of uveitis, served as controls.
Patients and controls were interviewed for the presence of
cardiovascular risk factors and cardiovascular morbidity. In addition,
morning urinary albumin concentration was measured in all
participants. We found a positive association between trace- and/or
microalbuminuria and inflammatory CME (P = .001; odds ratio 13.0),
but no relation between CME and cardiovascular morbidity or its risk
factors. In conclusion, we found a positive association between
inflammatory CME and trace microalbuminuria. The presence of trace
microalbuminuria indicates the presence of early systemic vascular
disease in patients with inflammatory CME. This finding suggests an
133
increased permeability of retinal and choroidal vessels and therefore

the early development of CME.
Chapter 7 describes a randomized, double-blind, placebo-controlled

cross-over trial to analyze the effect of lisinopril on inflammatory CME
and visual acuity. Each included patient (n = 40) received either
lisinopril (10 mg daily) or a placebo for three months. After 2 months
of a lisinopril/placebo free wash-out period, the groups received the
reverse study medication for 3 months. Fluorescein angiography was
performed to evaluate the retina, specifically the macular area. We
found that lisinopril had no effect on visual acuity, CME, papillary
leakage, retinal vasculitis and choroidal leakage. We observed a
decrease in blood pressure (P = .02) and a decrease in morning urinary
albumin excretion (P = .003). In conclusion, we found no effect of a 3-
month course of lisinopril on inflammatory CME of long duration and
visual acuity, although a positive effect on the vascular system was
observed.
Chapter 8 describes the role of systemic non-steroidal anti-

inflammatory drugs (NSAIDs) in the treatment of inflammatory CME. In
this study we included 66 patients with inflammatory CME and treated
them with naproxen (2 x 250 mg daily; n = 28) or rofecoxib (1 x 25
mg daily; n = 38). After 4 months of therapy, visual acuity and
inflammation activity were measured. We evaluated the grade of CME,
retinal vasculitis and papillary leakage with fluorescein angiography. No
clear effect of NSAIDs on inflammatory CME, visual acuity or
intraocular inflammation was observed. Drop out after 4 months of
treatment was 52% because of adverse effects and/ or inefficacy of the
treatment. We concluded that systemic NSAIDs have a limited role (if
any) in the treatment of inflammatory CME.
Chapter 9 includes a literature review on the use, efficacy and

complications of intravitreal triamcinolone acetonide (IVTA) injections
in non-infectious uveitis and inflammatory CME (14 and 41 patients
respectively). In most of the patients, IVTA applications brought a
rapid improvement in vision. The drawbacks of this treatment modality
included the temporary duration of the effect with the need for
repeated injections which re-exposed patients to the risk of
134
Chapter 11
complications. The risk of bacterial endophthalmitis was 0.5% and was

influenced by the specific IVTA preparation. IOP should be monitored
carefully because of the risk of elevated IOP in 30 to 43% of eyes.
Cataract developed in 29% of the patients’ eyes. IVTA application
brought a rapid improvement in vision in the majority of cases, even in
patients with otherwise intractable and prolonged CME. IVTA might
therefore shorten the time interval needed to achieve improvement or
remission in those patients otherwise treated with slowly acting
medications. To diminish the risk of endophthalmitis in our institution,
we chose ready-for-use IVTA-injections prepared by our pharmacy
department in which 90% of the toxic additives were removed and the
dispensed dose of triamcinolone acetonide was validated for the
amount of triamcinolone present. In our own patients (n = 19) treated
with IVTA, none developed endophthalmitis. We concluded that IVTA
might especially be of value in the acute presentations and/or severe
recurrences of non-infectious intraocular inflammation.
Chapter 10 describes 2 patients with diabetes mellitus and rapid

progression of unilateral diabetic retinopathy in eyes recovering from
an attack of infectious posterior uveitis. Since the pathogenesis of
intraocular inflammation and diabetic retinopathy acts through similar
biochemical mediators and pathways, it is feasible that posterior uveitis
accelerates the progression of diabetic retinopathy. Our results support
this hypothesis and point out a risk for rapid retinopathy development
in eyes affected by posterior uveitis.
135
Conclusions and future research

In this thesis we established that CME is a major complication of
uveitis, which has an especially severe course in elderly patients. CME
was noted in 33% of all uveitis patients, of whom 44% had low vision
(visual acuity equal to or less than 20/60) in at least one eye. Of all
uveitis patients with low vision, 42% was caused by CME. These
findings underline the need for improved management of this
complication in this patient population.
CME and microvascular damage

The new key insight into the pathogenesis of CME consists of an
association between inflammatory CME and generalized microvascular
damage. Until now, CME has been considered to represent a
complication of uveitis associated with leaking ocular vessels and was
believed to represent an ocular disorder only. However, our findings
linked microalbuminuria, which is typical for a general vascular
disorder, to inflammatory CME. Although an association between
cardiovascular disease and CME due to diabetes mellitus, retinal vein
occlusions and post intraocular surgery was reported, we found no
clear association between inflammatory CME and cardiovascular
morbidity or its risk factors. However, the numbers of patients in our
series were very limited (n = 48) and therefore our data preclude
definitive conclusions. A study to reveal a possible association between
cardiovascular disease and inflammatory CME would require several
hundred patients of various ages. Since uveitis is a relatively rare
disease, the inclusion of sufficient numbers of patients would be
difficult and it is questionable whether such a complicated study is
really necessary. Instead, a more detailed study of vascular wall
function, especially of the endothelium, might clarify the pathogenesis
of inflammatory CME and indicate a novel basis for future therapeutic
interventions.
If the pathological vascular conditions constitute a high risk for
developing CME, we would expect that the vessel protecting drugs
might have a beneficial effect on inflammatory CME. We found no
effects of short-term lisinopril medication on inflammatory CME.
However, other promising options are available and could be
investigated in the future such as, for example, the effects of
136
Chapter 11
angiotensin II receptor antagonists, various statins or heparin

analogues on CME. The option of a randomized clinical trial with a
combination of these vessel-protecting medications could be very
attractive because a maximal beneficial effect on the vascular wall
could be expected in this setting. Up till now the studies have been
performed with patients with severe and usually long standing CME and
the retinal vessels might have already developed the irreversible
changes on which vascular-protecting drugs have no effect or need a
longer period to be effective. Therefore, a study including patients with
CME of limited duration might reveal other results.
Various cytokines and factors influencing the endothelial function
play a role in the increased vascular permeability of patients with CME.
Recent advances in the technique of the multiplex immunoassay allow
us to measure a variety of cytokines in small volumes as in intraocular
fluid samples. The understanding of the function that these mediators
have in various clinical conditions could lead to the development of
novel therapeutic interventions for diverse uveitis entities and possibly
also CME.
Treatment
The findings of the poor visual prognosis of inflammatory CME resulted
in a changed treatment policy and a treatment has recently been
recommended for all patients with inflammatory CME, even for those in
an early stage and with full visual acuity. However, standard
treatments are frequently reported to be inefficient and associated with
(systemic) side effects. The need for randomized clinical trials is
therefore very important. Yet we found no effect of systemic NSAIDs
on visual acuity and inflammatory CME, only a high drop out rate
because of side effects (26%) of this medication. Therefore, we
concluded that the role of systemic NSAIDs in the treatment of
inflammatory CME is limited and these specific results together with
previous reports on NSAIDs are not encouraging and probably do not
warrant a randomized clinical trial. In contrast, the use of intravitreal
triamcinolone acetonide (IVTA) injections has given ophthalmologists a
powerful tool in the treatment of CME with no –or very limited-
systemic side effects. The role of IVTA might especially be of value in
those patients with acute presentations and/or severe recurrences of
137
non-infectious intraocular inflammation and patients with intractable

inflammatory CME. IVTA might shorten the time interval needed to
achieve improvement or remission in patients otherwise treated with
slow acting medications. Further studies with longer follow-up and a
higher number of patients of various uveitis entities are needed to
determine the exact place of IVTA in the treatment of uveitis and
inflammatory CME. Studies on inflammatory CME can be controlled
nowadays by optical coherence tomography, a useful and non-invasive
procedure in which objective measurements such as macular thickness
can be monitored.
Future research
As indicated above, many questions concerning the development and
treatment of inflammatory CME remain unanswered and still more new
questions are coming forward, which might lead us to (hopefully
useful) answers in the future research. The potential future research
projects based on this thesis might include:
• The investigation of the vessel-protecting drugs on the

development of CME.
• The investigation of various inflammatory mediators and their
effects on the vascular permeability and the vascular
endothelium.
• The investigation of intraocular application of inflammatory
mediators and their blockers as anti-VEGF drugs on
inflammatory CME.
• The investigation of the IVTA injections and intraocular
corticosteroid devices in the management of non-infectious
uveitis and inflammatory CME.
In this thesis we have proven that systemic microvascular disease,

plays a role in the development of inflammatory CME. Hopefully, will
our findings lead not only to new insights in the pathogenesis of
inflammatory CME but also open up new avenues in the treatment of
this blinding disorder.
138
Hoofdstuk 11
139
Het doel van dit proefschrift was het ophelderen van de mogelijk rol
van vaatziekten in de ontwikkeling van cystoid macula oedeem (CME)
bij patiënten met uveitis en om nieuwe behandelingsmogelijkheden
voor CME bij patiënten met uveitis te onderzoeken.
Hoofdstuk 2 beschrijft een cross-sectionele studie bij 529 patiënten

met uveitis waarin we de invloed van CME op de visus evalueren. CME
werd gezien bij 33% van alle uveitis patiënten, wat in 44% van de
gevallen leidde tot een slechte visus (visus gelijk of minder dan 20/60)
in tenminste één oog. Verder vonden we dat van alle uveitis patiënten
35% een slechte visus had in tenminste één oog, wat in 42% van de
gevallen werd veroorzaakt door CME. Slechte visus bij patiënten met
CME was geassocieerd met oudere leeftijd van de patiënten, chronische
ontstekingen en verschillende uveitis entiteiten. Visusdaling bij
patiënten met pan- en intermediaire uveitis werd in de meeste gevallen
veroorzaakt door CME (59% en 85%, respectievelijk). Concluderend,
we hebben vastgesteld dat CME een belangrijke complicatie is bij
patiënten met uveitis, leidend tot visusdaling en dat het vooral ernstig
is bij oudere patiënten met een chronische ontsteking.
In hoofdstuk 3 bepalen we de verstoring van de bloed-retina barrière

(BRB) door de lekkage van immuunglobuline G (IgG) te meten in het
voorste oogkamer water (VOK) van 112 patiënten met uveitis. Wij
toonden aan dat de intraoculaire IgG concentratie verhoogd was bij
patiënten met uveitis vergeleken met controle patiënten zonder uveitis
(P < .001) en dat deze concentratie vooral verhoogd was bij patiënten
met CME en retinale ischemie vergeleken met patiënten zonder deze
karakteristieken (P = .017 and P = .001, respectievelijk). We vonden
een associatie tussen CME en choroidale lekkage (P = .043) maar niet
tussen choroidal lekkage en IgG concentraties. Dit suggereert dat de
buitenste BRB selectief doorlaatbaar kan zijn voor vocht, maar niet
voor grote moleculen zoals IgG. Concluderend, we vonden hoge
intraoculaire IgG concentraties bij patiënten met uveitis en dan vooral
in diegene met CME en/of retinale ischemie. Onze studie brengt een
mogelijke selectieve doorlaatbaarheid van vocht, maar niet voor grote
eiwitten door de buitenste BRB aan het licht.
140
Hoofdstuk 11
In hoofdstuk 4 bepalen we het globale cytokine and chemokine

expressie patroon met een multiplex immunoassay in het voorste
oogkamer water van 31 uveitis patiënten. Uveitis monsters hadden
hoger concentraties van interleukine (IL)-6, IL-8, soluble intercellular
adhesion molecule, soluble vascular cell adhesion molecule (sVCAM),
and interferon-inducible protein-10 (IP-10) dan controles zonder
uveitis. Monsters van patiënten met actieve uveitis hadden hogere
concentraties van IL-6, IL-10, interferon-γ, sVCAM, RANTES, and IP-10
dan rustige uveitis. Infectieuze uveitis was geassocieerd met hoger
concentraties van IL-10 dan niet-infectieuze uveitis (P<.03 voor alle
subgroepen). Er werden geen significante verschillen gevonden tussen
patiënten met CME en zonder CME. In deze studie laten we zien dat
specifieke mediatoren verhoogd zijn bij patiënten met actieve en
infectieuze uveitis.
Hoofdstuk 5 onderzoekt de rol van de leeftijd en cardiovasculaire

ziekten (en z’n risico factoren) op de ontwikkeling van CME. De data
van 97 patiënten met uveitis, verkregen via een uitgebreide vragenlijst
werden gerelateerd aan oogheelkundige kenmerken. Multivariate
logistische regressie liet zien dat de ontwikkeling van CME sterk was
geassocieerd met de oudere leeftijd van de patiënt, onafhankelijk van
de duur van de uveitis. Patiënten ouder dan 50 jaar hadden een 3.8
keer hoger risico op het ontwikkelen van CME dan jongere patiënten.
In deze studie met 97 personen vonden we geen duidelijk bewijs voor
een associatie tussen cardiovasculaire ziekten, z’n risico factoren en de
ontwikkeling van CME. Onze studie laat zien dat oudere patiënten of
patiënten die uveitis op een latere leeftijd krijgen een verhoogd risico
hebben op het ontwikkelen van CME.
In hoofdstuk 6 onderzoeken we de rol van cardiovasculaire

morbiditeit, z’n risico factoren en microalbuminurie op de ontwikkeling
van CME bij uveitis patiënten. In deze matched case-control studie
hebben we 24 opeenvolgende patiënten met uveitis en CME
geïncludeerd. Als controle dienden 24 patiënten met uveitis zonder
CME, gematched voor leeftijd en duur van de uveitis. Patiënten en
controles werden gevraagd naar de aanwezigheid van cardiovasculaire
risico factoren en cardiovasculaire morbiditeit. Verder werd de
ochtendurine albumine excretie gemeten bij alle deelnemers. We
141
vonden een positieve associatie tussen trace en/of microalbuminurie en

inflammatoir CME (P = .001; odds ratio 13.0), maar geen relatie
tussen CME en cardiovasculaire morbiditeit of z’n risico factoren.
Concluderend vonden we een positieve associatie tussen inflammatoir
CME en trace microalbuminurie. De aanwezigheid van trace
microalbuminurie wijst op de aanwezigheid van vroeg systemische
vasculaire ziekten bij patiënten met inflammatoir CME. Deze bevinding
suggereert een verhoogde permeabiliteit van de retinale en choroidale
vaten en zodoende een vroege ontwikkeling van CME.
Hoofdstuk 7 beschrijft een gerandomiseerde dubbel blinde,

placebogecontroleerde cross-over studie om het effect van lisinopril op
inflammatoir CME en de visus te analyseren. Elke geïncludeerde patiënt
(n = 40) kreeg of lisinopril (10 mg per dag) of een placebo gedurende
3 maanden. Na 2 maanden van een lisinopril/placebo vrije wash-out
periode kregen de groepen de tegengestelde studie medicatie
gedurende 3 maanden. Fluoresceïne angiografie werd verricht om de
retina, in het bijzonder het maculagebied te beoordelen. We vonden
dat lisinopril geen effect had op de visus, CME, papillaire lekkage,
retinale vasculitis and choroidale lekkage. We zagen een afname in
bloeddruk (P = .02) en een afname in ochtendurine albumine excretie
(P = .003). Concluderend, vonden we geen effect van een 3 maanden
durende behandeling met lisinopril op langdurig inflammatoir CME en
de visus, alhoewel een positief effect werd gezien op het vasculaire
systeem.
Hoofdstuk 8 beschrijft de rol van systemische NSAID’s in de

behandeling van inflammatoir CME. In deze studie hebben we 66
patiënten met inflammatoir CME geïncludeerd en hebben ze behandeld
met naproxen (2 x 250 mg per dag; n = 28) of met rofecoxib (1 x 25
mg per dag; n = 38). Na 4 maanden therapie werd de visus en de
ontstekingsactiviteit gemeten. We beoordeelden de graad van CME,
retinale vasculitis en papillaire lekkage met fluoresceïne angiografie. Er
werd geen duidelijk effect gevonden van NSAID’s op inflammatoir CME,
visus en intraoculaire ontsteking. Het percentage uitvallers na 4
maanden was 52%, vanwege bijwerkingen en/of ineffectiviteit van de
behandeling. We concludeerden dat systemische NSAID’s een beperkte
rol (zoniet geen) hebben in de behandeling van inflammatoir CME.
142
Hoofdstuk 11
Hoofdstuk 9 bevat een literatuur review naar het gebruik, effectiviteit

en de complicaties van intravitreaal triamcinolon acetonide (IVTA)
injecties in niet-infectieuze uveitis and inflammatoir CME (14 en 41
patiënten respectievelijk). Bij de meeste patiënten bracht toediening
van IVTA een snelle verbetering van de visus. Het bezwaar van deze
behandeling bestaat uit de tijdelijke duur van het effect met als gevolg
het nodig zijn van herhaalde injecties die de patiënt blootstellen aan
het risico op complicaties. Het risico op bacteriële endophthalmitis was
0.5% en werd beïnvloed door de specifieke IVTA preparatie. De
oogboldruk moet goed in de gaten worden gehouden omdat na een
IVTA injectie een verhoogde oogdruk in 30 tot 43% van de ogen
voorkomt. Cataract ontwikkelde zich in 29% van de ogen. IVTA
toediening bracht in de meeste gevallen een snelle vooruitgang in
visus, zelfs bij patiënten met anders onbehandelbare en/of langdurige
CME. De IVTA injectie kan daarom het tijdsinterval verkorten dat nodig
is om verbetering of remmissie te bereiken in die patiënten die
normaliter worden behandeld met langzaam werkende medicijnen. Om
het risico op endophthalmitis te verminderen heeft ons instituut
gekozen voor klaar-voor-gebruik IVTA injecties, geprepareerd door de
Afdeling Farmacie, waarbij 90% van de toxische toevoegingen is
verwijderd. De toegediende dosis van triamcinolon acetonide werd
gevalideerd voor de hoeveelheid aanwezige triamcinolon. Bij onze
eigen patiënten (n = 19) behandeld met IVTA, ontwikkelde niemand
endophthalmitis. We concludeerde dat IVTA voornamelijk van waarde
kan zijn bij de acute presentatie en/of ernstige herhalingen van niet-
infectieuze intraoculaire onstekingen.
Hoofdstuk 10 beschrijft 2 patiënten met diabetes mellitus en een

snelle progressie van unilaterale diabetische retinopathie in ogen die
aanval van infectieuze uveitis posterior hebben doorgemaakt. Omdat er
in de pathogenese van intraoculaire ontsteking en diabetische
retinopathie dezelfde biochemische mediatoren betrokken zijn, is het
mogelijk dat uveitis posterior de progressie van diabetische
retinopathie versnelt. Onze resultaten ondersteunen deze hypothese en
laten een risico zien voor de snelle ontwikkeling van retinopathie in
ogen die aangedaan zijn door uveitis posterior.
143
Conclusies en toekomstig onderzoek

In dit proefschrift stelden we vast dat CME een belangrijke complicatie
is van uveitis en dat het vooral een ernstig beloop heeft bij oudere
patiënten. CME werd gezien bij 33% van alle uveitis patiënten, van wie
44% slechtziend was (visus gelijk of minder dan 20/60) in tenminste 1
oog. Van alle patiënten met uveitis die slechtziend waren, werd 42%
veroorzaakt door CME. Deze bevindingen onderstrepen de noodzaak
voor verbeterde behandeling van deze complicatie in deze patiënten
populatie.
CME en microvasculaire afwijkingen

Het belangrijkste nieuwe inzicht in de pathogenese van CME bestaat uit
een associatie tussen inflammatoir CME en gegeneraliseerde
microvasculaire schade. Tot dusver werd CME beschouwd als een
complicatie van uveitis, geassocieerd met lekkende oculaire vaten en
werd verondersteld alleen een oogheelkundige afwijking te zijn. Echter,
onze bevindingen linken microalbuminurie, wat typisch is voor
gegeneraliseerde vaatziekten, met inflammatoir CME. Alhoewel een
associatie tussen cardiovasculaire ziekten en CME ten gevolge van
diabetes mellitus, retinale vene occlusie en na intraoculaire operaties is
gerapporteerd, vonden wij geen duidelijke associatie tussen
inflammatoir CME en cardiovasculaire morbiditeit en zijn risico factoren.
Echter, het aantal patiënten in ons onderzoek was erg laag (n = 48) en
daarom sluiten onze data definitieve conclusies uit. Een studie om een
mogelijke associatie tussen cardiovasculaire ziekten en inflammatoir
CME aan het licht te brengen zou uit honderden patiënten van
verschillende leeftijden moeten bestaan. Omdat uveitis een relatief
zeldzame ziekte is, zou de inclusie van geschikte patiënten moeilijk
zijn en is het de vraag of zo’n gecompliceerde studie echt nodig is. In
plaats daarvan zou met een meer gedetailleerde studie naar de functie
van de vaatwand en dan vooral het endotheel, de pathogenese van het
inflammatoir CME opgehelderd kunnen worden en kunnen leiden naar
een nieuwe basis voor toekomstige therapeutische interventies.
Als de pathologische vasculaire condities een hoger risico vormen
voor het ontwikkelen van CME, zouden we kunnen verwachten dat
medicijnen die de functie van de vaatwand verbeteren, en dan met
name van het endotheel, een gunstig effect hebben op inflammatoir
144
Hoofdstuk 11
CME. We vonden geen effect van een korte behandeling met lisinopril
op inflammatoir CME. Echter, er zijn andere veelbelovende opties
beschikbaar die in de toekomst onderzocht kunnen worden,
bijvoorbeeld, het effect van angiotensine II receptor antagonisten,
verschillende statines of heparine analogen op CME. De optie van een
gerandomiseerde klinische trial met een combinatie van deze bloedvat
beschermende medicijnen zou erg aantrekkelijk zijn, omdat in deze
setting een maximaal nuttig effect op de vaatwand kan worden
verwacht. Tot nu toe zijn de studies uitgevoerd bij patiënten met
ernstig en meestal langdurig CME en de retinale vaten zouden al
onomkeerbare veranderingen hebben ontwikkeld waarop de bloedvat
beschermende medicijnen geen effect hebben of een langere periode
nodig hebben om effectief te zijn. Daarom zou een studie waarin
patiënten met kortdurend CME worden geïncludeerd andere resultaten
aan het licht kunnen brengen.
Verschillende cytokines en factoren die de endotheel functie
beïnvloeden spelen een rol bij de verhoogde vasculaire permeabiliteit
bij patiënten met CME. Recente vooruitgangen in de techniek van de
multiplex immunoassay stellen ons in staat om een verscheidenheid
aan cytokines te meten in kleine volumina zoals bij intraoculaire
samples. Het begrijpen van de functie die deze mediatoren hebben in
verschillende klinische condities kan leiden tot de ontwikkeling van
nieuwe therapeutische interventies voor verschillende uveitis entiteiten
en mogelijk ook voor CME.
Behandeling
De bevinding van een slechte visuele prognose bij inflammatoir CME
resulteerde in een veranderd behandelingsbeleid en een behandeling
wordt tegenwoordig aangeraden voor alle patiënten met inflammatoir
CME, zelfs voor diegene die in een vroeg stadium nog een volledige
visus hebben. Echter, er wordt vaak gerapporteerd dat de standaard
behandelingen inefficiënt zijn en geassocieerd zijn met (systemische)
bijwerkingen. Daarom is de noodzaak voor het verrichten van
gerandomiseerde klinische studies erg belangrijk. We vonden geen
effect van systemische NSAID’s op de visus en inflammatoir CME. We
vonden alleen een hoog percentage afvallers door bijwerkingen (26%)
van deze medicatie. Daarom concludeerde we dat de rol van
systemische NSAID’s bij de behandeling van inflammatoir CME beperkt
145
is. Onze resultaten, en de resultaten van eerder verschenen studies

zijn niet bemoedigend en rechtvaardigen waarschijnlijk geen
gerandomiseerde klinische studie. In tegenstelling tot de NSAID’s geeft
het gebruik van intravitreaal triamcinolon acetonide (IVTA) injecties de
oogarts een krachtige optie in de behandeling van CME met geen – of
hele beperkte- systemische bijwerkingen. De rol van IVTA kan vooral
van waarde zijn in die patiënten met acute presentaties en/of ernstige
exacerbaties van niet-infectieuze intraoculaire ontstekingen en
patiënten met onbehandelbaar inflammatoir CME. IVTA kan de
tijdsduur verkorten die nodig is om verbetering of remissie te bereiken
bij patiënten die anders behandeld worden met langzaam inwerkende
medicijnen. Verdere studies zijn nodig, met een langere follow-up en
een grotere hoeveelheid patiënten met verschillende uveitis entiteiten
om de exacte plaats van IVTA in de behandeling van uveitis en
inflammatoir CME te bepalen. Bij studies naar inflammatoir CME kan
tegenwoordig gebruik gemaakt worden van optical coherence
tomografie, een bruikbare en niet invasieve methode waarbij de
macula dikte objectief kan worden gemeten en gevolgd.
Toekomstig onderzoek
Zoals boven uiteengezet blijven veel vragen betreffende de
ontwikkeling en behandeling van inflammatoir CME onbeantwoord en
dienen zich steeds meer nieuwe vragen aan, die hopelijk worden
beantwoord in toekomstige onderzoeken. De mogelijke toekomstige
onderzoeksprojecten gebaseerd op dit proefschrift kunnen bestaan uit:
• Het onderzoeken van het effect van bloedvat beschermende

medicatie op de ontwikkeling van CME.
• Het onderzoeken van het effect van verschillende
inflammatoire mediatoren op de vasculaire permeabiliteit en
het vasculair endotheel.
• Het onderzoeken van intraoculaire toediening van
inflammatoire mediatoren en blokkers zoals anti-VEGF
medicijnen op inflammatoir CME.
• Het onderzoeken van IVTA injecties en intraoculaire
corticosteroïden devices in de behandeling van niet-infectieuze
uveitis and inflammatoir CME.
146
Hoofdstuk 11
In dit proefschrift hebben we bewezen dat systemische microvasculaire

afwijkingen een rol spelen in de ontwikkeling van inflammatoir CME.
Hopelijk zullen onze bevindingen niet alleen leiden tot nieuwe inzichten
in de pathogenese van inflammatoir CME, maar ook nieuwe wegen
openen in de behandeling van deze blindmakende ziekte.
147
148
Dankwoord
Na drieëneenhalf jaar patiënten dossiers doorploegen, dokters
handschriften ontcijferen, artikelen schrijven en lezen, patiënten
onderzoeken, FAG’s prikken, bloed prikken, urine verzamelen en uit
m’n handen laten vallen ligt hier dan het eindresultaat van mijn
promotieonderzoek. Uiteraard was dit nooit gelukt zonder de hulp en
steun van anderen.
Allereerst wil ik mijn promotor, Prof. dr. Aniki Rothova bedanken. Ik

heb ontzettend veel aan jouw begeleiding gehad. Je snelheid van
stukken nakijken heeft me altijd verbaasd en was uiteraard erg prettig
voor mij. Ik kon altijd even bij je langskomen om dingen te bespreken
en altijd nam je de tijd, ondanks de hectiek van de poli, het professor
zijn, het onderzoek in Thailand, het begeleiden van andere promovendi
en keuze-co’s enz. enz. Niet alleen op het gebied van het onderzoeken,
maar ook op klinisch gebied heb ik veel van jou geleerd.
Dr. Rob Fijnheer, voor jouw begeleiding wil ik je ook ontzettend
bedanken. Als hematoloog heb je vaak een frisse kijk op de
oogheelkundige problemen. Ook al werk je 4 dagen in Den Bosch, op
vrijdag kwam je (bijna) altijd naar onze bespreking daarnaast kon ik je
tussendoor altijd bellen. Aniki en Rob, ik vond het heel erg prettig dat
we het persoonlijk ook goed met elkaar kunnen vinden. Dit heeft altijd
erg stimulerend gewerkt.
Prof. dr. W.F. Treffers en Prof. dr. J.S. Stilma, bedankt voor de
gelegenheid die jullie mij hebben geboden om dit promotieonderzoek te
kunnen uitvoeren. Daarnaast wil ik jullie bedanken voor de adviezen en
interesse die jullie altijd hebben gehad in mijn onderzoek.
Dank aan de leden van de beoordelingscommissie, Prof. dr. Ph.G.
de Groot, Prof. dr. J.S. Stilma, Prof. dr. M.D. de Smet, Prof. dr. J.E.E.
Keunen en Dr. J. de Boer, voor de tijd die u heeft genomen voor het
beoordelen van het manuscript.
Natuurlijk gaat er ook grote dank uit naar alle patiënten die hun
medewerking hebben verleend aan de diverse studies.
Mensen van de uveitis groep, Jolanda de Groot, Harold de Valk,
Annemarie Weersink, Gina Postma en Anneke Jansen dank voor de
adviezen, de prettige overleggen en dat ik mijn praatjes bij jullie mocht
oefenen. Ninette ten Dam en Joke de Boer bedankt voor alle hulp bij
149
het vinden van geschikte patiënten, het beoordelen van FAG’s, het
beantwoorden van zinnige en onzinnige vragen en de lol die we gehad
hebben bij het maken van de PowerPoint presentatie voor Aniki.
Speciaal voor jullie ben ik vandaag geschoren, geknipt en sta ik hier
met gepoetste schoenen. Wolter de Loos, je bent nu niet meer hier,
maar ongetwijfeld ben je op de een of andere manier bij ons. In het
begin van mijn onderzoek hebben we vaak van gedachten gewisseld,
wat zeker een goede bijdrage heeft geleverd aan mijn onderzoek.
Karen Sijssen en Dagmar Bockholts, de uren achter de computer
werden met jullie een stuk leuker dankzij de huizenjacht, het bijna
dagelijkse bezoekje aan de websites van Ophthalmology en AJO (hoe
ver is jouw artikel al?) en niet te vergeten de ontzettende leuke week
in Chicago. Dagmar dankzij jou heb ik mij al enigszins kunnen
voorbereiden op het leven als arts-assistent in de oogheelkunde.
Karen, ontzettend leuk dat jij mijn paranimf wil zijn, kan je alvast een
beetje oefenen voor over twee jaar.
Alle andere assistenten wil ik bedanken voor de leuke momenten
tijdens congressen, borrels of gewoon op de poli of in de gang. Ik heb
erg veel zin om te beginnen met mijn opleiding en verheug me erop
om met jullie te gaan samenwerken.
Ook wil ik de leden van de Landelijke Uveitis Werkgroep bedanken
voor de nuttige opmerkingen tijdens de vergaderingen.
Alle medeauteurs, voor zover nog niet eerder genoemd wil ik ook
bedanken. Imke Bartelink, Philip de Vries, Mark Roest, David van de
Vijver, Charlotte Lardenoye, Kiki Probst, Ger Rijkers, Annemarie
Weersink en Tos Berendschot bedankt voor de commentaren,
statistische hulp, logistieke hulp en het meedenken over de vorm en
inhoud van de artikelen.
Dank aan alle stafleden die in de loop der jaren mijn vragen
hebben beantwoord of mij op de een of andere manier hebben
geholpen.
Birgit Wilms, Hiske Ypma en Marja Wilmans bedankt voor alle hulp
op het secretariële vlak en voor de gezellige koffie pauzes, dit geldt
natuurlijk ook voor Martha Bastiaanse en Woudie Ensie. Peter Klein
Gunnewiek en Bertus de Jager bedankt voor het oplossen van (bijna)
alle computerproblemen.
De zusterpost (en af en toe broeder Jos) en de rode balie wil ik
ook bedanken voor al die keren dat jullie mij piepten als er weer een
150
patiënt voor me was, voor het helpen zoeken naar onderzoekskamers,
voor het tussendoor meten van oogdruk en bloeddruk. In het bijzonder
wil ik Henny Hölzken bedanken voor de woensdagmiddagen als we de
Zestril-patiënten zagen. Dankzij jou liep dit erg gesmeerd. Ik vond het
erg leuk om dit samen met jouw te doen.
Dames en heren van het poli secretariaat, vaak zijn jullie op mijn
kamer geweest opzoek naar dossiers. Van mij zullen jullie op dat
gebied geen last meer hebben, maar er komt ongetwijfeld weer iemand
anders. Ook heel veel dank voor het zoeken naar niet te vinden
dossiers op al die verschillende plekken.
Mensen van de fotografie en de OCT, Gerard de Graaf, Jos
Aalbers, Hilda Schenk, Celia de Ruiter en Dagmar Gültzau bedankt voor
jullie flexibiliteit als er weer eens een patiënt even tussendoor moest.
Uiteraard ook bedankt voor de gezelligheid.
Boezemvrienden, Marc en Willem, Jaarclub en andere vrienden en

familie, bedankt voor jullie interesse en als jullie vroegen: Wat houd
zo’n promotie nou precies in en wat doe je de hele dag? Hier ligt het
eindresultaat, hier heb ik meer dan 3 jaar aan gewerkt, nu kunnen
jullie het zien en lezen, maar of jullie het begrijpen……?
Evert-Kees bedankt voor je adviezen en inspanningen bij het
vinden van een drukker en een ontwerpbureau. Ook jouw adviezen
over het omslagontwerp heb ik erg gewaardeerd.
Lieve papa en mama, weer een mijlpaal in mijn leven waar jullie
getuige van zijn, zonder jullie was deze nooit bereikt, bedankt voor alle
steun en interesse. Lieve Aukje, ik ben erg blij dat jij mijn paranimf wil
zijn, we hadden bijna in hetzelfde schuitje gezeten, nu maak je het
toch nog een beetje mee. Lieve Bonnie, nu ik arts-assistent ben, kom
ik je weer eens tegen op de OK, gezellig!
Lieve Hanneke, Ajo liefie…. zeggen we vaak tegen elkaar. Zou dat
de reden zijn dat zoveel van mijn artikelen in AJO (American Journal of
Ophthalmology) zijn gepubliceerd? Wat hebben we gedurende mijn
promotieonderzoek al veel meegemaakt, eerst samenwonen, toen
trouwen en nu een geweldig huis gevonden. Bedankt voor al je steun,
liefde en dat je me erop wees dat ik mijn onderzoek toch wat simpeler
moest uitleggen anders snapte niemand er wat van.
151
152
Curriculum Vitae
De auteur van dit proefschrift werd geboren op 22 mei 1975 te
Amersfoort. Na het behalen van zijn VWO diploma in 1993 aan het
Eemland College Zuid in Amersfoort studeerde hij Geneeskunde aan de
Universiteit Utrecht. Deze studie onderbrak hij één jaar voor het
bestuur van studentenvereniging C.S. Veritas. Het doctoraal en het
artsexamen werden in respectievelijk 2000 en 2002 behaald. Het keuze
co-schap werd verricht op de afdeling oogheelkunde van het UMC
Utrecht onder begeleiding van Prof. dr. A. Rothova. Vanaf september
2002 werkte hij als arts-onderzoeker aan zijn proefschrift op de
afdeling oogheelkunde van het UMC Utrecht, onder begeleiding van
Prof. dr. A. Rothova, Prof. dr. W.F. Treffers en Dr. R. Fijnheer. Een van
zijn artikelen (Hoofdstuk 6 van dit proefschrift) kreeg in 2005 de
Editor’s Choice Award van The American Academy of Ophthalmology.
Vanaf maart 2006 is hij als assistent in opleiding tot oogarts werkzaam
op de afdeling oogheelkunde van het UMC Utrecht met als opleider
Prof. dr. J.S. Stilma.
153
List of publications
B. van Kooij, M. Canninga - van Dijk, J. de Boer, V. Sigurdsson, A.
Rothova. Is granuloma annulare related to intermediate uveitis with
retinal vasculitis? British Journal of Ophthalmology 2003
Jun;87(6):763-6.
B. van Kooij, S.F. Thijsen, E. Meijer, H.G. Niesters, J.W. van Esser,
J.J. Cornelissen, L.F. Verdonck, A.M. van Loon. Sequence analysis of
EBV DNA isolated from mouth washings and PBMCs of healthy
individuals and blood of EBV-LPD patients. Journal of Clinical Virology
2003 Sep;28(1):85-92.
A. Rothova, T.T. Berendschot, K. Probst, B. van Kooij, G.S. Baarsma.

Birdshot chorioretinopathy: long-term manifestations and visual
prognosis. Ophthalmology 2004 May;111(5):954-9.
B. van Kooij, R. Fijnheer, M. Roest, A. Rothova. Trace

microalbuminuria in inflammatory cystoid macular edema. American
Journal of Ophthalmology 2004;138:1010–1015.
B. van Kooij, J. de Boer, N. ten Dam-van Loon, R. Fijnheer, A.

Rothova. The effect of non-steroidal anti-inflammatory drugs on
inflammatory cystoid macular edema. American Journal of
Ophthalmology 2005;140:563–564.
B. van Kooij, Ph. de Vries, A. Rothova. The pro’s and contra’s of

intravitreal triamcinolone injections for uveitis and inflammatory
macular edema. Ocular Immunology and Inflammation, in press.
B. van Kooij, R. Fijnheer, J. de Boer, N. ten Dam – van Loon, I.

Bartelink, M. Roest, A. Rothova. A randomized, masked, cross-over
trial of lisinopril for inflammatory macular edema. American Journal of
Ophthalmology, in press.
J.A Knol, B. van Kooij, H.W de Valk, A. Rothova. Rapid progression of

diabetic retinopathy in eyes with posterior uveitis. American Journal of
Ophthalmology, in press.
154
C.W.T.A. Lardenoye, B. van Kooij, A. Rothova. The impact of macular
edema on visual acuity in patients with uveitis. Submitted for
publication.
K. Probst, B. van Kooij, R. Fijnheer, M. Roest, W. de Loos†, A.

Rothova. Prognosticators of cystoid macular edema in uveitis patients.
Submitted for publication.
B. van Kooij, A.J.L. Weersink, D.A.M.C van de Vijver, J.D.F. de Groot

– Mijnes, A. Rothova. IgG concentration in the aqueous humor of
patients with uveitis and cystoid macular edema. Submitted for
publication.
B. van Kooij, A. Rothova, G.T. Rijkers, J.D.F. de Groot – Mijnes.

Distinct cytokine and chemokine profiles in the aqueous of patients
with uveitis and cystoid macular edema. Submitted for publication.
155

Full

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

Full

Uploaded by

Copyright:

Available Formats

Bram W.

Cystoid macular edema

Cover: Twin design, Culemborg, The Netherlands

Printed by: Torendruk, Nijkerk, The Netherlands

Cystoid macula oedeem in uveitis

ter verkrijging van de graad van doctor

Bram Wouter van Kooij

Co-promotor: Dr. R. Fijnheer, Universiteit Utrecht

This thesis was supported by the Dr. F.P. Fisher Foundation

Chapter 2 The impact of macular edema on visual acuity 19

Chapter 3 IgG concentration in the aqueous humor of 31

Chapter 4 Distinct cytokine and chemokine profiles in the 41

Chapter 5 Prognosticators of cystoid macular edema in 49

Chapter 6 Trace microalbuminuria in inflammatory cystoid 61

Chapter 7 A randomized, masked, cross-over trial of 75

Chapter 8 The effect of non-steroidal anti-inflammatory 89

Chapter 9 The pro’s and contra’s of intravitreal 99

Chapter 11 Summary and conclusions 131

Curriculum Vitae 153

List of publications 154

diabetic and 9 with inflammatory CME compared to 28 patients without

One additional factor contributing to the development of CME in

recommended, even for patients with (still) undamaged visual acuity.35

Aim of the Thesis

Outline of the Thesis

treated with rofecoxib and 28 patients treated with naproxen) in a

15. Funatsu H, Yamashita H, Sakata K, Noma H, Mimura T, Suzuki M, Eguchi S, Hori

31. Romundstad S, Holmen J, Hallan H, Kvenild K, Kruger O, Midthjell K.

The impact of macular edema

Material and methods

The relationships between visual loss and CME for different

Table 1: General characteristic of patients with uveitis

Total Non-CME* CME*

Patients 529 354 (67%) 175 (33%)

Patients with visual acuity (VA)

Anterior 213 23 (11%) 49 (23%) 7 (14%) 7/23 (30%)

Total 529 175 (33%) 185(35%) 77 (42%) 77/175 (44%)

Patients with visual acuity (VA)

HLA-B27-associated 58 7 (12%) 6 (10%) 2 (33%) 2/7 (29%)

seen in viral infections of the anterior segment (11%; Table 3) and

18% of patients retained CME despite the treatment.4 The differences

The poor visual acuity in inflammatory CME was clearly associated

(submitted for publication)

Material and methods

Sample collection and processing

Determination of intraocular IgG production concentrations

Table 1: General characteristics of uveitis patients

Number of patients 112

Number of patients with duration of uveitis (%):

Anatomic location of uveitis (%):

Patients with cystoid macular edema (CME) (%) 34 (30%)

Table 2: Geometric mean IgG concentrations and correlations of general and

Geometric mean IgG (mg/mL) Multivariate

Cystoid macular edema duration:

The geometric mean IgG concentrations in aqueous humor and their

P-value (2-sided)* Linear regression

* chi-square or Fisher exact test.

Previously, others noted an association between the prolonged

Distinct cytokine and

(submitted for publication)

Cytokines, chemokines and soluble adhesion molecules modulate

Geometric Mean Concentration, Median, (pg/ml)

Geometric Mean Concentration, Median, (pg/ml)

Our findings confirm and extend previous findings of elevated levels of