See

discussions, stats, and author profiles for this publication at: http://www.researchgate.net/publication/43048702

Nonsteroidal anti-inflammatory therapy and

recurrent acute anterior uveitis
ARTICLE in OCULAR IMMUNOLOGY AND INFLAMMATION APRIL 2010
Impact Factor: 1.44 DOI: 10.3109/09273941003587558 Source: PubMed

CITATIONS

DOWNLOADS

VIEWS

27

93

3 AUTHORS, INCLUDING:
c. Stephen Foster
Harvard Medical School
700 PUBLICATIONS 15,548 CITATIONS
SEE PROFILE

Available from: c. Stephen Foster

Retrieved on: 27 July 2015

Ocular Immunology & Inflammation, 18(2), 116120, 2010

Copyright 2010 Informa Healthcare USA, Inc.
ISSN: 0927-3948 print/ 1744-5078 online
DOI: 10.3109/09273941003587558

ORIGINAL ARTICLE

Nonsteroidal Anti-inflammatory Therapy and

Recurrent Acute Anterior Uveitis
Vanessa M. B. Fiorelli, MD, Pooja Bhat, MD, and C. Stephen Foster, MD, FACS, FACR
Massachusetts Eye Research and Surgery Institution, Cambridge, Massachusetts, USA

ABSTRACT
Objective: To investigate oral nonsteroidal anti-inflammatory drug (NSAID) therapy in the
prevention of recurrences of uveitis in patients with recurrent nongranulomatous, idiopathic, or
HLA-B27-associated acute anterior uveitis (AAU).
Methods: Retrospective case series of 59 patients with recurrent AAU treated with celecoxib or
diflunisal.
Results: The average duration of NSAID therapy was 21.25.7 months. The average number of
relapses for all patients prior to systemic NSAID therapy was 2.84 per person-year follow-up. These
relapses declined to 0.53 per person-year follow-up with NSAID therapy (p <.001). The relapse
rates prior to and after treatment in the HLA-B27-positive group (n=21) were compared with the
relapse rates prior to and after treatment in the HLA-B27-negative group (n=38) and were also
statistically significant (p <.001).
Conclusion: Morbid attacks and the cumulative exposure to corticosteroids can be prevented with
systemic NSAID therapy in patients with recurrent AAU.
Keywords: anterior uveitis; nonsteroidal anti-inflammatory drugs; recurrent uveitis; inflammatory eye disease

Recurrent acute anterior uveitis is the most common

form of intraocular inflammation and accounts for
8.2 new cases per 100,000 person-years.1 It has been
defined as recurrent iritis or iridocyclitis that completely resolves within 3 months. From an etiologic
standpoint, this disease is one of the most difficult to
diagnose. Acute anterior uveitis is also challenging
from a therapeutic standpoint. With the introduction
of corticosteroids in 1949, the treatment of inflammatory eye disease was completely revolutionized.2
Topical corticosteroids have been so efficacious that
they continue to remain the first line of therapy for
acute inflammatory eye diseases. However, the side
effects of prolonged or repeated steroid use, especially
cataract and glaucoma, have been well documented.35

Prior to the emergence of corticosteroids, nonsteroidal anti-inflammatory drugs (NSAIDs), particularly

aspirin, were employed in the treatment of severe
ocular inflammation.6,7 A review of the major clinical
trials on the use and therapeutic efficacy of NSAIDs
in ophthalmology was initially published by Flach.8
The vast majority of clinical publications related to
NSAIDs and ocular inflammation deal with postoperative inflammation prophylaxis9; few address
the clinical and experimental models of uveitis and
its therapy with NSAIDs.10-15 Our experience at the
Massachusetts Eye Research and Surgery Institution
(MERSI) indicates that oral NSAIDs are particularly
useful in the long-term management of patients with
nongranulomatous, acute idiopathic, or HLA-B27associated recurrent anterior uveitis, substantially
reducing the amount of corticosteroid required to
maintain inflammatory quiescence and enabling
patients, in many cases, to remain in remission once
the steroids have been discontinued.16,17 We present a
retrospective analysis of the use of prophylactic oral

Received 09 October 2009; revised 21 December 2009;

accepted 30 December 2009
Correspondence: C. Stephen Foster, MD, FACS, FACR, Massachusetts Eye Research and Surgery Institution, 5 Cambridge
Center, 8th floor, Cambridge, MA 02142, USA. E-mail: foster@
uveitis.org

116

NSAIDs and Recurrent Acute Anterior Uveitis 117

NSAIDs in the prevention of recurrences of uveitis in
patients with recurrent anterior uveitis.

Methods
The clinical records of 59 patients with a diagnosis
of recurrent acute anterior uveitis who presented to
MERSI between May 2005 and April 2008 were evaluated. All patients with recurrent, acute, and painful
attacks of uveitis and a follow-up period of at least
1 year prior to and after beginning the oral NSAID
therapy were included.
All 59 patients underwent standard screening tests
that are performed at MERSI on all patients with
uveitis. These investigations included tests for syphilis and HLA-B27 typing. All patients were examined
by the principal investigator (CSF) at all visits. Classification of uveitis was performed according to the
International Uveitis Study Group recommendations.18
Anterior chamber inflammation was graded as defined
by Foster and Vitale.19 According to the SUN working
group descriptions of uveitis, the attack was considered acute if it was sudden in onset and was limited
(less than 3 months) in duration.20 Patients with signs,
symptoms, and laboratory workup suggestive of
rheumatologic diseases were further evaluated by a
rheumatology consultation.
Multiple variables were assessed, including age at
onset, number and duration of attacks, associated systemic diseases, time and duration of topical corticosteroid use, and the time and duration of systemic NSAID
use. Recurrences of inflammation prior to commencing
NSAID therapy were recorded from the documentation
within chart notes of the patients obtained from the
referring physician, while those after the administration of NSAID therapy were recorded on examination
at MERSI as were the side effects while on NSAIDs.
Remission was defined as trace or absent inflammatory
cells in the anterior chamber while on systemic NSAID
therapy but off topical/systemic corticosteroid therapy
for at least 6 months. Systemic NSAIDs evaluated in
this study were celecoxib (Celebrex, Pfizer, New York,
NY) and diflunisal (Dolobid, Merck, Rahway, NJ).
Statistical analysis was performed using the Wilcoxon test to compare the differences of the relapses
prior to and during NSAID treatment. Mann-Whitney
test was used to compare the differences in follow-up

and remission between the following groups: celecoxib and diflunisal; HLA-B27 positive and negative;
and male and female groups. A p value of less than
.05 was considered to be statistically significant. This
study was approved by the Institutional Review Board
of the Massachusetts Eye and Ear Infirmary and was
performed in concordance with the Declaration of
Helsinki.

Results
The average age at presentation was 4311.7 years.
There were 26 male and 33 female patients. All 59
patients received systemic nonsteroid anti-inflammatory therapy for an average of 21.25.7 months.
Systemic autoimmune diseases were observed in 13
patients (11 female and 2 male): ankylosing spondilitis (n=4), juvenile idiopathic arthritis (n=2), psoriasis
(n=2), fibromyalgia (n=1), Hashimoto thyroiditis
(n=1), rheumatoid arthritis (n=1), and Crohn disease
(n=1).
All patients had a follow-up of at least 1 year prior
to commencing NSAID therapy. For the 59 patients, the
average number of relapses prior to systemic NSAID
therapy was 2.84 per person-year follow-up. Relapses
declined to 0.53 per person-year follow-up while on
NSAID therapy. The difference in the relapse rate was
highly statistically significant at p <.001. These data are
shown in Table 1.
The study patient group was divided on the basis
of gender (males=26; females=33) and reanalyzed. A
relapse rate of 2.73 in males and 2.94 in females prior
to NSAID therapy was found. This reduced to 0.53
in males and 0.57 in females after systemic NSAID
therapy was commenced. All patients remained in
remission for an average of 18.22 months. These data
are shown in Table 2.
The results were also analyzed based on the
patients who received celecoxib (n=30) versus those
who received diflunisal (n=29). Of the 30 patients who
received celecoxib, 26 received a 20-mg po bid dose
and the remaining 4 patients received 100mg po bid.
All of the patients on diflunisal received 500mg po
bid. The average follow-up for the celecoxib group
was 21.9 months. The rate of relapse prior to celecoxib therapy in that group was 2.73. This relapse
rate declined to 0.36 while the group was on cele-

Table 1 Follow-up, remission, and relapses prior to and during NSAID therapy for all patients
Relapses (prior to
Patients (N=59)
Follow-up (months)
NSAIDs)
Remission (months)
Average
21.23
2.84
18.22
STDEV
5.72
1.65
6.28
Note. STDEV, standard deviation. Relapses prior to and during NSAIDs therapy: p <.001.
2010 Informa Healthcare USA, Inc.

Relapses (with NSAIDs)

0.53
0.79

118 V. M. B. Fiorelli etal.

Table 2 Follow-up, remission, and relapses prior to and during NSAID therapy by gender
Relapses prior to
Relapses during
Follow-up (months)
NSAID
Remission (months)
NSAID
Male
Average
20.61
2.73
18
0.53
(n=26)
STDEV
6.3
1.64
5.9
0.75
Female
Average
21.9
2.94
18.36
0.57
(n=33)
STDEV
5.2
1.67
6.65
0.79
Note. STDEV, standard deviation. Remission during NSAIDs therapy by gender: p=.113. Relapses during NSAIDs therapy by
gender: p=.256. Relapses prior to and during NSAIDs therapy: p <.001.

Table 3 Follow-up, remission, relapses, and side effects in patients with celecoxib and diflunisal
Follow-up
Relapses prior
Remission
Relapses dur(months)
to NSAIDs
(months)
ing NSAIDs
Celecoxib
Average
21.9
2.73
21.0
0.36
n=30
STDEV
4.87
1.46
5.50
0.67
Diflunisal
Average
18.56
3
15.34
0.7
n=29
STDEV
5.30
0.66
5.78
0.87
Statistic analysis
p=.027
p=.727
p <.001
p=.165
Note. STDEV, standard deviation; GI, gastrointestinal side effect.

8: GI
1: dysuria
1: drowsiness

Table 4 Details of NSAID therapy in HLA-B27-positive and HLA-B27-negative patients

Relapses prior to
Follow-up (months)
NSAIDs
Remission (months)
HLA B27+
Average
20.43
2.24
17.10
n=21
STDEV
5.78
1.04
5.6
Average
21.84
2.97
18.84
HLA B27
n=38
STDEV
5.68
1.55
6.61
Statistic analysis
p=.479
p=.017
p=.256
Note. STDEV, standard deviation. Relapses prior to and during NSAIDs therapy in both groups: p <.00l.

coxib therapy. For the diflunisal group, the average

follow-up was 18.56 months. The rate of relapse prior
to initiation of diflunisal therapy was calculated at
3.0. This reduced to 0.7 while they were on diflunisal
therapy. The difference between the relapse rate while
on celecoxib therapy versus diflunisal therapy was not
found to be statistically significant (p=.165). However,
patients on celecoxib therapy remained in remission
longer (215.50 months) than patients on diflunisal
therapy (15.345.78 months). This difference was
found to be statistically significant at p <.001.
One patient developed gastrointestinal side effects
while on celecoxib therapy. The symptoms resolved
on decreasing the celecoxib dose and drug therapy
was not discontinued. Ten patients developed side
effects in the diflunisal group. Eight patients developed gastrointestinal side effects, 1 patient developed dysuria, and 1 patient complained of persistent
drowsiness. Diflunisal therapy had to be discontinued
in 5 patients, demonstrating a difference but without
sufficient numbers to allow statistical testing. These
data are shown in Table 3.
The study was also divided on the basis of the
HLA-B27 status. The average follow-up in the HLA-

Discontinued
NSAIDs
No

Side effects
1: GI

Yes
(n=5)

Relapses during
NSAIDs
0.24
0.44
0.66
0.88
p=.027

B27-positive group was 20.43 months and the relapse

rates prior to treatment with NSAID was found to
be 2.24. These patients remained in remission for 17
months while on NSAID therapy and the relapse rate
while on NSAID medications was 0.24. The average
follow-up in the HLA-B27-negative group was 21.84
months with relapse rates prior to treatment being
2.97. The duration of remission while on treatment
was 18.84 months. The relapse rate in this group was
0.66 with treatment. The difference in the relapse rate
prior to and during NSAID therapy in both groups
was highly statistically significant at p <.001. These
data are shown in Table 4.

Discussion
Over the last 3 decades much has been discovered
about the mechanisms of action of a class of agents
collectively known as nonsteroidal anti-inflammatory
agents. These agents are widely prescribed in general
medicine for the treatment of rheumatologic diseases
and are principally used topically in ophthalmology for
the treatment and reduction of cystoid macular edema,
Ocular Immunology & Inflammation

NSAIDs and Recurrent Acute Anterior Uveitis 119

intraoperative miosis, and postoperative inflammation.8, 21 Our study demonstrates the effectiveness of
systemic NSAID use in the treatment of patients with
recurrent acute anterior uveitis.
NSAIDs possess anti-inflammatory, analgesic, and
anti-pyretic properties by virtue of their ability to
inhibit prostaglandin synthesis via the cyclooxygenase (COX) pathway. When tissue is damaged, either
by injury or inflammation, tissue phospholipids are
released and are acted upon by enzyme phospholipase A2 with the resultant release of arachidonic acid
(AA). COX acts upon AA to produce endoperoxidases
PG-G2 and PG-H2, which are precursors of prostaglandins (PG) in ocular and nonocular tissues.17, 22
Within the eye, PGs can disrupt the bloodaqueous
barrier. An active transport system for PGs has been
demonstrated in the ciliary body area and a transvitreal pathway from the anterior segment to the retina
has been identified. In addition, PGs enhance vascular
permeability of the bloodocular barrier, producing
conjunctival hyperemia, intraocular pressure changes,
and increasing inflammation.8 NSAID-mediated PG
synthesis inhibition forms the mainstay of NSAID
use in ophthalmology. Evidence also suggests that
NSAIDs have free radical scavenger activity and
anti-chemotactic activity, which modulate humoral
and cellular events during inflammatory reactions.23,
24
These effects formed the basis of our employment of
systemic NSAID therapy in the care of patients with
recurrent acute anterior uveitis who are analyzed in
this report.
The NSAIDs used in this study were celecoxib and
diflunisal. The specific mechanism of action of celecoxib is primarily via inhibition of cyclooxygenase-2
(COX-2). The most frequent side effects associated
with it are abdominal pain, diarrhea, and dyspepsia.
Diflunisal is a difluorophenyl derivative of salicylic
acid and is a nonselective COX inhibitor. The most frequent side effects include nausea, vomiting, abdominal
pain, diarrhea, constipation, and dyspepsia.
Rothova and associates in 1986 determined that
attacks of uveitis usually recur within 100 weeks from
onset for HLA-B27+ patients and within 58.3 weeks
in HLA B27 patients. These are similar to findings
by Chung in Chinese patients with HLA-B27+ anterior uveitis where recurrences were noted within 78
weeks.25 Therefore, it is quite common for patients
tp be treated repeatedly with corticosteroids, which
have been the mainstay of therapy for ocular inflammations4 despite the problems associated with its
chronic use. Our study demonstrates that recurrent
attacks can be reduced considerably and significantly and patients can thereby avoid corticosteroid
therapy through the employment of systemic NSAID
therapy.
2010 Informa Healthcare USA, Inc.

Two prior studies in the past have shown that

systemic NSAID therapy can reduce inflammation
and allow a reduction in corticosteroid use. The first
was in children with chronic iridocyclitis14 and the
second in patients with juvenile rheumatoid arthritisassociated iridocyclitis. 15 One of the studies14 evaluated the adjunctive use of NSAIDs in the treatment
of 14 patients with chronic iridocyclitis. In all cases,
iridocyclitis was attenuated with the addition of
NSAIDs to the treatment regimen, and in some cases
the reduction of steroids was possible. Our study provides additional data on NSAID use for patients with
recurrent anterior uveitis and supports the notion that
such therapy is effective, steroid sparing, and clearly
worthy of consideration in the quest for steroid-free
remission of uveitis.
A recent study by Braakenburg and associates26
on long-term follow-up and gender differences in
patients with HLA-B27-associated anterior uveitis
demonstrated that evolution to chronic uveitis
increased from 14% at 1 year of follow-up to 19% at
10-year follow-up. The prevalence of bilateral uveitis
also increased from 5% at 1-year follow-up to 21%
at 10-year follow-up. Additionally, in time, posterior synechiae developed in approximately 50% of
the patients. Periocular steroid injections had been
used in 54% of the patients at 10-year follow-up.
Systemic NSAID therapy was not employed in any
of the patients. The fact that these patients required
frequent ophthalmologic examinations and repeated
institution of steroid therapy coupled with the time
and money spent to receive treatment, the hours of
productivity lost due to poor vision during attacks
(either due to inflammation or mydriatics) and during
visits to physicians produced considerable burden on
patients and on society. Our study demonstrates that
the relapse rate of uveitis is significantly reduced, not
only in such patient populations but also in patients
who do not possess the HLA-B27 gene, with systemic
NSAID therapy providing respite from more frequent
morbid attacks.
The side-effect profile of celecoxib and diflunisal is
different; our data show that celecoxib was much better
tolerated than was diflunisal, although the therapeutic
effect obtained by each was similar. Patient compliance
plays a very important role in successful treatment,
but in our experience, due to insurance approval constraints, celecoxib could not be readily prescribed and
was used predominantly in patients who could not
tolerate diflunisal. It is important to mention that there
are several studies in the literature focusing on the
safety of long-term use of selective and nonselective
NSAIDs. Serious gastrointestinal risks, such as gastric
and duodenal ulcers with subsequent perforation and
hemorrhage, are associated with nonselective NSAIDs

120 V. M. B. Fiorelli etal.

like diflunisal. The selective COX-2 inhibitor celecoxib
has been associated with serious cardiovascular risks
like myocardial infarction and stroke. However, this
association has been demonstrated to be dose and
duration dependent. The patients in our study did not
experience any of the serious side effects described
above. Nonetheless, our study was not designed to
assess the safety of long-term NSAID use.27, 28
Our results indicate that a randomized controlled
trial is required to further evaluate the efficacy of
NSAIDs for the treatment of recurrent AAU and to
test its safety for long-term use. But until such a trial
is performed, our data indicate that systemic NSAID
therapy provides an intermediate step between corticosteroid therapy and long-term immunosuppressive
therapy in the paradigm of care in patients with recurrent acute anterior uveitis.
Declaration of interest: The authors report no conflicts
of interest. The authors alone are responsible for the
content and writing of the paper.

References
[1] Rothova A, van Veenendaal WG, Linssen A, Glasius E,
Kijstra, de Jong PTVM. Clinical features of acute anterior
uveitis. Am J Ophthalmol. 1987;103:137145.
[2] Gordon DM. Prednisone and prednisolone in ocular disease.
Am J Ophthalmol. 1956;21:8290.
[3] Gallant C, Kenny P. Oral glucocorticoids and their complications: a review. J Am Acad Dermatol. 1986;14:161177.
[4] Soukiasian SH, Foster CS, Raizman MB. Treatment strategies for scleritis and uveitis associated with inflammatory
bowel disease. Am J Ophthalmol. 1994;118:601611.
[5] Foster CS. Medical Treatment of intermediate uveitis
[review]. Dev Ophthalmol. 1992;23:156157.
[6] Tuovinen E, Esila R, Liesmaa M. Experience of the use of
indomethacin in inflammatory eye diseases. Acta Ophthalmol. 1966;44:585589.
[7] Perkins ES, MacFaul PA. Indomethacin in the treatment
of uveitis: a double blind trial. Trans Ophthalmol Soc UK.
1965;85:5358.
[8] Flach AJ. Cyclo-oxygenase inhibitors in ophthalmology.
Surv Ophthalmol. 1992;36:259284.
[9] Manthey KF. Nonsteroid anti-inflammatory drugs in the
treatment of intermediate uveitis [review]. Dev Ophthalmol.
1992;23:158162.
[10] Melluish JW, Brooks CD, Rouff G, Cross CJ, Sanborn, EC.
Ibuprofen and visual function: prospective evaluation. Arch
Ophthalmol. 1975;93:781782.
[11] Rosenbaun JT, Robertson JE Jr. Recognition of posterior
scleritis and its treatment with indomethacin. Retina.
1993;13:1721.

[12] March W, Coniglione TC. Ibuprofen in the treatment of

uveitis. Ann Ophthalmol. 1985;17:103104.
[13] Hunter PJL, Fowler PD, Wilkinson P. Treatment of anterior
uveitis: comparison of oral oxyphenbutazone and topical
steroids. Br J Ophthalmol. 1973;57: 892896.
[14] Olson NY, Lindsley CB, Godfrey WA. Non-steroidal antiinflammatory drug therapy in chronic childhood iridocyclitis. Am J Dis Child (US). 1988; 142:12891292.
[15] Giordano M. Daverprophylaxe der rezidivierenden
spondylitischen iridocyclitisdurch antimalarica und
nichtsteroidale antiphlogistica. Z Rheumatol. 1982;41:105
106.
[16] Teasely LA, Christen GC, Foster CS. Oral non steroidal
anti-inflammatory agents in the treatment of chronic or
recurrent noninfectious uveitis. Contemp Ophthalmol. 2007;
6:11.
[17] Samiy N, Foster CS. The role of nonsteroidal antiinflammatory drugs in ocular inflammation [review]. Int Ophthalmol
Clin. 1996;36:195206.
[18] Bloch-Michel E, Nussenblatt RB. International Uveitis Study
Group recommendations for the evaluation of intraocular inflammatory diseases. Am J Ophthalmol. 1987; 103:
234235.
[19] Harper SL, Chorich LJ, Foster CS.Diagnosis of uveitis. In:
Foster CS, Vitale AT, eds. Diagnosis and Treatment of Uveitis.
Philadelphia: WB Saunders; 2002:79103.
[20] Jabs DA, Nussenblatt RB, Rosenbaum JT. Standardization
of Uveitis Nomenclature (SUN) Working Group. Standardization of uveitis nomenclature for reporting clinical data:
results of the First International Workshop [review]. Am J
Ophthalmol. 2005;140:509516.
[21] Almeida DR, Johnson D, Hollands H, etal. Effect of prophylactic nonsteroidal antiinflammatory drugs on cystoid
macular edema assessed using optical coherence tomography quantification of total macular volume after cataract
surgery. J Cataract Refract Surg. 2008;34:6469.
[22] Vitale AT, Foster CS.Non-steroidal and anti-inflammatory
drugs. In: Foster CS, Vitale AT, eds. Diagnosis and Treatment
of Uveitis. Philadelphia: WB Saunders; 2002:167176.
[23] Burne K, Glatt M, Graf P: Minireview: mechanisms of
action of anti-inflammatory drugs. Gen. Pharmacol. 1976;
7:2733.
[24] Abramson SB, Weissmann G. The mechanism of action of
non steroidal anti- inflammatory drugs. Arthritis Rheum.
1989; 32: 19.
[25] Chung YM, Yeh TS, Liu JH. Clinical manifestations of HLAB27-positive acute anterior uveitis in Chinese. Chung Hua I
Hsueh Tsa Chih (Taiwan). 1989; 43: 97104.
[26] Braakenburg AM, de Valk HW, de Boer J, Rothova A.
Human leukocyte antigen-B27-associated uveitis: long-term
follow-up and gender differences. Am J Ophthalmol. 2008;
145:472479.
[27] Scheiman JM. Balancing risks and benefits of cyclooxygenase-2 selective nonsteroidal anti-inflammatory
drugs [review]. Gastroenterol Clin North Am. 2009;38:
305314.
[28] White WB, West CR, Borer JS, et al. Risk of cardiovascular events in patients receiving celecoxib: a meta-analysis of randomized clinical trials. Am J Cardiol. 2007;99:
9198.

Ocular Immunology & Inflammation

Copyright of Ocular Immunology & Inflammation is the property of Taylor & Francis Ltd and its content may
not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written
permission. However, users may print, download, or email articles for individual use.