Outcomes in Anterior Uveitis Associatedwith the HLA-B27 Haplotype

William J. Power, FRCS, FRCOphth, Alejandro Rodriguez, MD, Miguel Pedroza–Seres, MD,C. Stephen Foster, MD

Objective: This study aimed to test the hypothesis that patients presenting with anterior uveitis who areHLA-B27 positive, either with or without associated systemic disease, have a less-favorable outcome than dopatients with idiopathic anterior uveitis who are HLA-B27 negative.

Design: Retrospective case-controlled series.Participants: Ninety-seven patients who were HLA-B27 positive with no systemic disease, 94 patients who

were HLA-B27 positive with systemic disease, and 72 patients who were HLA-B27 negative who presented withanterior uveitis were studied.

Main Outcome Measures: Ocular complications (e.g., secondary glaucoma, cataract formation, pupillarysynechiae, vitritis, cystoid macular edema, and optic disc edema), medical and surgical treatment, number ofrecurrent attacks, and final visual acuity were recorded for all patients.

Results: The patients who were HLA-B27 positive, either with or without systemic disease, experienced agreater number of complications than did the patients who were HLA-B27 negative. Periocular corticosteroids,systemic corticosteroids, and systemic immunosuppressive chemotherapy were required in a far greater numberof HLA-B27-positive patients than in HLA-B27-negative patients (60% vs. 11%, 53% vs. 7%, and 18% vs. 1%,respectively; P , 0.001). The percentage of legally blind eyes was significantly greater in the HLA-B27-positivegroup, both with and without systemic disease, when compared with the HLA-B27-negative group (11% vs. 2%;P , 0.005).

Conclusions: The prognosis of anterior uveitis associated with the HLA-B27 haplotype, either with or withoutassociated systemic disease, is less favorable when compared with that of HLA-B27-negative patients withidiopathic anterior uveitis. Ophthalmology 1998;105:1646–1651

ris

om

ith

cied

scif-

utolo

andlica-isual27-tionnocia-that

27thic

thefir-lyzedereveitis,

Anterior uveitis is the most common form of intraoculainflammation, accounting for approximately 50% of uveitcases seen in large referral centers.1–3 The group of HLA-B27-associated uveitic syndromes is the second most cmon cause of anterior uveitis.1,2 In one study, a positiveHLA-B27 haplotype had been found in 43% of patients wacute anterior uveitis.4

The clinical characteristics of the anterior uveitis assoated with the HLA-B27 haplotype have been reviewextensively in the literature.4–6 A clear distinction has beenmade of the clinical features between patients with a potive HLA-B27 haplotype and those without such an assoation. But, whether the HLA-B27 association carries dferent prognostic implications still is controversial.5,7–9

Although some authors have found a more favorable ocome in HLA-B27-positive patients when comparedthose with anterior uveitis and a negative HLA-B27 haptype,6–8,10 others have found the opposite.5,9

tionrentions,ana-

eousic

Originally received: October 16, 1997.Revision accepted: February 27, 1998. Manuscript no. 97730

From Immunology & Uveitis Service, Massachusetts Eye & Ear InfirmaryHarvard Medical School, Boston, Massachusetts.

Reprint requests to William J. Power, FRCS, FRCOphth, The RoyaVictoria Eye & Ear Hospital, Adelaide Road, Dublin 2, Ireland.

1646

-

-

i-i-

t-

-

We analyzed the long-term clinical course of the uveitisindependently compared the prevalence of ocular comptions, the medical and surgical management, and the voutcome between a large cohort of patients with HLA-Bpositive uveitis with and without systemic disease associaand those with HLA-B27-negative anterior uveitis withprimary ocular uveitic condition or systemic disease assotion. These comparisons allowed us to test the hypothesispatients presenting with anterior uveitis who are HLA-Bpositive have a worse prognosis than those with idiopaanterior uveitis who are HLA-B27 negative.

Patients and Methods

The clinical records of all patients with anterior uveitis seen atImmunology & Uveitis Service, Massachusetts Eye & Ear Inmary, between January 1982 and December 1993 were anathroughout patients’ entire follow-up and the following data wobtained: gender, race, age at presentation, age at onset of ufollow-up time, clinical diagnosis or systemic disease associaor both, clinical presentation, number and duration of recurinflammatory attacks, prevalence of severe ocular complicatvisual outcome, and management. The ocular complicationslyzed were the presence of persistent and extensive ($180°) pos-terior iris synechiae, significant cataract formation ($21 opacity),secondary glaucoma, severe and diffuse or organized vitrinflammation ($21 cells), clinically significant macular, and optdisc edema as shown by fluorescein angiography.

.

,

l

ud-tieionityalld

me

in-pom

crovin

ropd ian

gnsueithysw

ulaor

ds;ksnddiugat-e onpyts

idewitvepsn-

sso-zedriate,

ients27emicthy or

(122eitispno

es) in

sum-

with4).:1.2)o

lmostithgroupand

nsetwithars;ange,set ofars;ar-

.327-–188emicas no

Table 1. General Characteristics of Patients with Anterior Uveitis

Study GroupHLA-B27Haplotype

No. ofPatients

Males[no. (%)]

Females[no. (%)]

Mean Ageat Onset

(yrs)Median Follow-up

Time (mos)

Idiopathic 2 72 30 (42) 42 (58) 38.6 15.4No systemic disease 1 97 45 (46) 52 (54) 34.1 14.6Systemic disease 1 94 60 (64) 34 (36) 36.5 19.3

Power et al z Anterior Uveitis and the HLA-B27 Haplotype

A comprehensive medical and ophthalmologic history incling an extensive review of systems was obtained from each paduring the initial consultation. The ophthalmologic examinatperformed at every visit included best-corrected visual acubiomicroscopy, tonometry, and indirect ophthalmoscopy inpatients. Anterior segment and fundus photography, visual fiefluorescein angiography, and ultrasonography were perforwhen indicated.

Initial laboratory investigations performed on all patientscluded a complete blood cell count, urinalysis, fluorescent trenemal antibody absorption test, angiotensin-converting enzyand HLA-B27 typing. Radiographic investigations such as sailiac spine x-rays were performed for patients suspected of haan associated seronegative spondyloarthropathy.

Systemic disease association (seronegative spondyloarththy or inflammatory bowel disease or both) was documenteevery possible case. The diagnoses of ankylosing spondylitisReiter’s syndrome were made according to the Dougados et al11

and the Lee et al12 criteria, respectively. In cases of inflammatorybowel disease (ulcerative colitis and Crohn’s disease), the diasis was confirmed by histopathologic features of biopsied tisPatients with a positive or negative HLA-B27 typing and wclear evidence of a specific primary ocular uveitic entity or stemic disease association other than those mentioned aboveexcluded from the study.

All patients were treated according to the degree of intraocinflammation (cells in the anterior chamber or vitreous cavityboth) using a stepladder steroid-sparing therapeutic algorithm.13

Most patients initially were treated with topical corticosteroiperiocular corticosteroids were used for severe recurrent attacinflammation, for cystoid macular edema or papillitis or both, afor patients in whom systemic corticosteroids were contraincated. Long-term systemic nonsteroidal anti-inflammatory dr(NSAIDs) were used if active inflammation recurred withtempted topical steroid withdrawal and to prevent recurrencinflammation. Short course systemic corticosteroids were givethe inflammation could not be controlled with previous theraSystemic immunosuppressive agents were used in patienwhom previous therapeutic strategies had failed.

For descriptive and statistical analyses, patients were divinto three major groups. The first group consisted of patientsa negative HLA-B27 haplotype who presented with anterior uitis of unknown etiology (idiopathic). The second and third grouwere comprised of HLA-B27-positive patients suffering from a

Table 2. Systemic Disease Association in Pa

Disease/EtiologyNo. of

PatientsMales

[no. (%)]

Ankylosing spondylitis 51 34 (67)Reiter’s syndrome 26 21 (81)Psoriatic arthritis 5 1 (20)Ulcerative colitis 5 3 (60)Crohn’s disease 7 1 (14)

nt

,ls,d

-e,-g

a-nd

o-.

-ere

r

of

-s

fif.in

dh-

terior uveitis with and without evidence of systemic disease aciation, respectively. The distribution of counts was analyusing Fisher’s exact test and chi-square test where appropcorrecting for multiple comparisons by requirement ofP , 0.001on individual tests for a net confidence. 95%.

Results

A total of 191 consecutive patients, representing 28.2% of patwith anterior uveitis, referred to our service tested HLA-Bpositive and had no other primary ocular diagnosis and/or systdisease association other than seronegative spondyloarthropainflammatory bowel disease or both. Seventy-two patientseyes) tested HLA-B27 negative and suffered from anterior uvof unknown etiology (idiopathic). The HLA-B27-positive grouwas further divided into 97 (50.8%) patients (143 eyes) withsystemic disease association and 94 (49.2%) patients (148 eywhom a systemic disease association was found.

The general characteristics of the three study groups aremarized in Table 1. The majority (94.2%) of patients were white.There was a slight female predominance among patientsHLA-B27-negative idiopathic uveitis (male:female ratio, 1:1.Males and females were almost equally represented (ratio, 1in the group of patients with HLA-B27-positive uveitis with nsystemic disease. Conversely, the male-to-female ratio was a2:1 in patients with HLA-B27-positive uveitis associated wsystemic disease. The male predominance observed in thiswas because of patients with ankylosing spondylitis (ratio, 2:1)Reiter’s syndrome (ratio, 4:1) (Table 2).

The HLA-B27-negative patients had an older mean age at oof uveitis (38.6 years; range, 18–70 years) than did thoseHLA-B27-positive uveitis without systemic disease (34.1 yerange, 10–67 years) or with systemic disease (36.4 years; r15–77 years). In this last group, the youngest mean age at onuveitis was seen in patients with Reiter’s syndrome (28.9 yerange, 15–52 years) and the oldest in patients with psoriaticthritis (45.8 years; range, 40–53 years) (Table 2).

The median follow-up time of HLA-B27-negative patients (1years; range, 1–82 months) was similar to that of HLA-Bpositive patients without systemic disease (1.2 years; range, 1months) and shorter than in those with an associated systdisease (1.6 years; range, 1–141 months). However, there w

tients with HLA-B271 Anterior Uveitis

Females[no. (%)]

Mean Ageat Onset

(yrs)Median Follow-up

Time (mos)

17 (33) 36.7 13.55 (19) 28.9 14.74 (80) 45.8 24.22 (40) 31.3 18.96 (86) 40.0 21.3

1647

ong

27byas

veodlo-

eah ahehy27(hadNotio

rage,asenpa-–19).am-ecificeasetion

uttions

7-Thispa-

pec-A-aseand

orega-ntt in

tivearyral

yesentsanywasic-

e-eg-een

innys-asso-

withe re-ntspa-smo-sso-

itistivelyys-

er of27

ualoupease

Table 4. Recurrence of Inflammation in Patientswith Anterior Uveitis

Inflammation

HLA-B272

Idiopathic(n 5 72)[no. (%)]

HLA-B271 NoSystemicDisease

(n 5 97)[no. (%)]

HLA-B271

SystemicDisease

(n 5 94)[no. (%)]

Single episode 51 (72) 31 (32) 24 (26)Recurrent episode 21 (28) 66 (68) 70 (74)Mean no. of recurrences

(range) 1.4 (1–12) 3.6 (1–22) 5.2 (1–19)Mean duration of episode

(wks; range) 3.9 (2–10) 5.0 (2–24) 4.2 (2–14)

Table 3. Clinical Characteristics of Patientswith Anterior Uveitis

Clinical Characteristic

HLA-B272

Idiopathic(n 5 72)[no. (%)]

HLA-B271

NoSystemicDisease

(n 5 97)[no. (%)]

HLA-B271

SystemicDisease

(n 5 94)[no. (%)]

Disease onsetAcute 43 (60) 86 (90) 86 (92)Insidious 29 (40) 11 (10) 8 (8)

Eye involvementUnilateral 22 (30) 51 (53) 40 (42)Bilateral alternating 4 (6) 25 (26) 31 (33)Bilateral simultaneous 46 (64) 21 (21) 23 (25)

Type of inflammationNongranulomatous 60 (83) 93 (96) 92 (98)Granulomatous 12 (17) 4 (4) 2 (2)

AC inflammatoryreactionFibrinous 0 (0) 24 (25) 23 (24)Hypopyon 0 (0) 15 (16) 12 (13)

AC 5 anterior chamber.

Ophthalmology Volume 105, Number 9, September 1998

significant difference in the median follow-up observed ampatients in the three groups (P . 0.2).

The most common systemic disease association in HLA-Bpositive patients was ankylosing spondylitis (54.2%) followedReiter’s syndrome (27.6%) and inflammatory bowel dise(12.7%) (Table 2). There was a clear distinction in the clinicalpresentation between patients with HLA-B27-positive and HLA-B27-negative anterior uveitis. The majority of HLA-B27-positipatients presented with acute onset (90%), recurrent epis(71.2%) of unilateral or bilateral alternating (76.9%), nongranumatous (96.8%) iridocyclitis (Tables 3 and 4). But more strikingwas the exclusive occurrence of anterior chamber fibrinous rtion (24.6%) and hypopyon formation (14.1%) in patients witpositive HLA-B27 haplotype. No significant differences in tclinical characteristics and in the presence of fibrinous orpopyon inflammation or both were observed between HLA-Bpositive patients with or without associated systemic diseaseP .0.2). Conversely, HLA-B27-negative patients more frequentlya single inflammatory episode (72%) of bilateral uveitis (64%).anterior chamber fibrinous and/or hypopyon inflammatory reacwas noted in any of these patients.

There was a significant difference in the mean recurrenceof inflammation between HLA-B27-negative (1.4 attacks; ran1–12) and HLA-B27-positive patients with no systemic dise(3.6 attacks; range, 2–22;P , 0.001). This difference was evegreater if the comparison was made with HLA-B27-positivetients with systemic disease association (5.2 attacks; range, 2No significant differences in the mean number of recurrent inflmatory episodes were observed between patients with spseronegative spondyloarthropathies or inflammatory bowel disor both. There was no significant difference in the mean duraof recurrent attacksof inflammation among thethreegroups(Table4; P . 0.2).

Patients who were HLA-B27 positive, either with or withosystemic disease, experienced a greater number of complicathan those who were HLA-B27 negative (Tables 5 and 6). Themost frequent ocular complication in patients with HLA-B2negative uveitis was cataract formation, seen in 10% of eyes.percentage was at least twice as great in HLA-B27-positive

1648

-

e

es

c-

--

n

te

tients with and without systemic disease (21% and 24%, restively). The most common ocular complication observed in HLB27-positive patients, both with and without systemic diseassociation, was cystoid macular edema, observed in 32%29% of eyes, respectively. This complication was five times mfrequent in HLA-B27-positive patients than in those with a netive HLA-B27 haplotype (P , 0.001). Extensive and persistepupillary synechiae were almost three times more frequenHLA-B27-positive patients (20.2%) than in those with a negaHLA-B27 haplotype (7.3%). Pupillary seclusion with secondpupillary block glaucoma, requiring laser or surgical peripheiridectomy, was observed in three HLA-B27-positive uveitic e(2%) with no systemic disease and in nine eyes (6%) of patiwith systemic disease. This complication was not observed ineye of HLA-B27-negative patients. Secondary glaucoma alsomore frequent in patients with HLA-B27-positive uveitis, partularly among those associated with a systemic disease (Table 5).Finally, the two ocular complications with the most striking frquency difference between HLA-B27-positive and HLA-B27-native patients were severe vitritis and papillitis. Vitritis was sten times more frequently in HLA-B27-positive patients thanHLA-B27-negative patients (P , 0.001), and papillitis was sevetimes more frequent in HLA-B27-positive patients with no stemic disease and ten times more frequent in those with anciated systemic condition than in HLA-B27-negative patients (P ,0.001) (Table 5).

The medical treatment of all patients is summarized in Table 7.More than 90% of patients from the three groups were treatedtopical steroids. Periocular and systemic corticosteroids werquired in a far greater number of HLA-B27-positive patie(59.6% and 53.4%, respectively) than in HLA-B27-negativetients (11% and 7%, respectively;P , 0.001). The difference wamore striking for the use of systemic immunosuppressive chetherapy among these groups, particularly for those with an aciated systemic disease.

Surgical intervention (Table 8) was required with significantlygreater frequency among patients with HLA-B27-positive uve(45 eyes) when compared with those with an HLA-B27-negahaplotype (5 eyes) (P , 0.001). This difference was particularmarked in HLA-B27-positive patients without an associated stemic condition (Table 8).

The visual outcome for all three groups is listed in Table 9. Itcan be seen that there is a marked difference in the numblegally blind eyes between those patients who were HLA-Bnegative and those who were HLA-B27 positive (P , 0.001). Twopercent of eyes in the HLA-B27-negative group had a final visacuity of 20/200 or worse, whereas 11% of eyes in the HLA grpositive without systemic disease and 11% with systemic dishad a final visual acuity of 20/200 or worse.

itsp-

tlylo

pear

tiv

ithauA-ithy-st

ei

ve

riotheseeveintheecet’sown

ons7-ssi-oftedive

nssi-neofndiord inratebeupdis-ith a

7-pri-i.e.,

Table 7. Medical Treatment of Patients with Anterior Uveitis

HLA-B272

Idiopathic

HLA-B271 NoSystemicDisease

HLA-B271

SystemicDisease

Table 5. Ocular Complications in Patients with Anterior Uveitis

Ocular Complication

HLA-B272Idiopathic

(n 5 122)*[no. (%) of eyes]

HLA-B271 NoSystemic Disease

(n 5 143)*[no. (%) of eyes]

HLA-B271Systemic Disease

(n 5 148)*[no. (%) of eyes]

Secondary glaucoma 6 (5) 14 (10) 21 (14)Cataract formation ($ 21) 12 (10) 30 (21) 36 (24)Pupillary synechiae ($ 180°) 9 (7) 28 (20) 31 (21)Severe vitritis ($ 21 cells) 2 (2) 32 (22) 30 (20)Cystoid macular edema 8 (6) 41 (29) 48 (32)Papillitis 3 (2) 22 (15) 30 (20)Total no. (%) of affected eyes 40 167 196

* Total no. of involved eyes.

Power et al z Anterior Uveitis and the HLA-B27 Haplotype

Discussion

Anterior uveitis is the most common form of uveitis, andassociation with the HLA-B27 haplotype in the white poulation has been estimated to be approximately 50%.14,15

Patients with HLA-B27-positive anterior uveitis frequenpresent with unilateral or bilateral alternating, nongranumatous, recurrent iridocyclitis in association with nonscific arthritic symptoms or with a seronegative spondylothropathy.9 These clinical characteristics have made thisgroup of patients distinguishable from those with a negaHLA-B27 haplotype.7,9 The data from our series confirmsuch a distinction. D’Alessandro et al16 found that themajority of patients with anterior uveitis who presented whypopyon had a positive HLA-B27 haplotype. Thesethors found a prevalence of hypopyon of 14.5% in HLB27-positive patients compared to only 2.2% in those wa negative HLA-B27 haplotype. This prevalence of hpopyon uveitis in HLA-B27-positive patients was almoidentical to that found in our series (14.1%).

Although the clinical features of the HLA-B27-positivanterior uveitis have been a matter of extensive reviewthe literature, the prognosis of the disease still is controsial. In contrast to our findings, Wakefield et al8 havesuggested that patients with HLA-B27-negative anteuveitis have a poorer prognosis, perhaps becausepatients have a more chronic clinical course. We belithat the poor outcome in the HLA-B27-negative grouptheir report was caused by the etiologic heterogeneity ofHLA-B27-negative group. Two of their HLA-B27-negativpatients had herpes simplex keratouveitis, one had Behdisease, and another one had syphilis. These are well-kn

Table 6. Number of Patients with Complications

No. of Complications HLA-B272

HLA-B271

SystemicDisease

HLA-B271

NoSystemicDisease

0 45 8 101 20 36 292 6 36 4031 1 17 15Total 72 97 94

---

e

-

nr-

r

causes of uveitis with serious prognostic implications.17–19

Furthermore, the percentage, if any, of ocular complicatiascribed to each of these patients within the HLA-B2negative group was not specified, leaving open the pobility that these four patients, with a higher likelihoodhaving ocular complications develop, may have contributo the poorer outcome reported for the HLA-B27-negatgroup of patients.

More recently, Linssen and Meenken20 also reported thatthere was no significant difference in ocular complicatioand visual outcome between patients with HLA-B27-potive and HLA-B27-negative acute anterior uveitis. Oprominent finding in their study was the high rate (60%)ocular complications in the HLA-B27-negative group, athis figure was exclusive of patients in whom postersegment complications developed who were not includethe analysis. One possible reason for the relatively highof complications in the HLA-B27-negative group maythat 25% of the patients in the HLA-B27-negative grohad an associated systemic disease including Behcet’sease and sarcoidosis, which frequently are associated wless-than-favorable prognosis.

In contrast to these two previous reports, the HLA-B2negative group in the current study had no associatedmary uveitic condition or systemic disease association (

Therapeutic Agent(n 5 72)[no. (%)]

(n 5 97)[no. (%)]

(n 5 94)[no. (%)]

Topical steroids 68 (94) 97 (100) 92 (98)Topical NSAIDs 10 (14) 30 (31) 31 (33)Systemic NSAIDs 33 (46) 73 (75) 81 (86)Periocular steroids 8 (11) 52 (54) 62 (66)Systemic steroids 5 (7) 50 (52) 52 (55)Methotrexate 1 (1) 8 (8) 16 (17)Azathioprine 0 (0) 7 (7) 12 (13)Cyclosporine 0 (0) 3 (3) 3 (3)Chlorambucil 0 (0) 5 (5) 1 (1)Cyclophosphamide 0 (0) 1 (1) 4 (4)

NSAIDs 5 nonsteroidal anti-inflammatory drugs.

1649

ats2iti

ghh ahao

endca

rinonmin

veke, wa-mpsiduae

itis

seg-this

7-se,27-

asantthevegestpa-

s-oretrol

calA-ac-gnifi-nd

si-g-o-

usede-is-and

LA-

Table 8. Surgical Management of Patients with Anterior Uveitis

Surgical Procedure

HLA-B272Idiopathic(n 5 122)

[no. (%) of eyes]

HLA-B271 NoSystemic Disease

(n 5 143)[no. (%) of eyes]

HLA-B271Systemic Disease

(n 5 148)[no. (%) of eyes]

Cataract extraction 4 (3) 15 (10) 9 (6)Pars plana vitrectomy 0 (0) 9 (6) 5 (3)Filtering procedure 1 (1) 4 (3) 1 (1)Retinopexy 0 (0) 1 (1) 1 (1)

Ophthalmology Volume 105, Number 9, September 1998

their uveitis was truly idiopathic in nature). We believe ththis more homogenous study population better allows uassess what effect, if any, the presence of the HLA-Bhaplotype has on patients who present with anterior uve

Rothova et al9 found that HLA-B27-positive patientshave a more severe clinical course and, hence, a hiincidence of ocular complications than do patients witnegative HLA-B27 haplotype. These authors found tvisual outcome did not differ significantly between the twgroups. They also found no clear differences when patiwith HLA-B27-positive anterior uveitis were subdivideaccording to the presence of systemic disease, specifiwith ankylosing spondylitis.4,6 Our findings are in accor-dance with those of Rothova et al, in that patients suffefrom anterior uveitis and in whom an HLA-B27 associatiis found have a significantly higher rate of recurrent inflamatory attacks and, hence, a higher likelihood of havsevere ocular complications develop.

In the study by Rothova et al,9 the authors only ana-lyzed patients with acute (,3 months’ duration) inflam-mation and excluded all patients with HLA-B27-negatiuveitis with a chronic clinical course. In an effort to maconclusive remarks with respect to disease prognosisanalyzed HLA-B27-positive and HLA-B27-negative ptients throughout their entire clinical course and copared the rate of ocular complications between grou

All the ocular complications, but particularly cystomacular edema, were important contributors to poor visoutcome in HLA-B27-positive patients in this study. Wpreviously have emphasized that HLA-B27-positive uve

ent-

A-m ofhisc-cy-is

sero-is-

teriorisg-

ithlo-the, anden-

Table 9. Final Visual Acuity in Patients with Anterior Uveitis

Visual Acuity Range

HLA-B272

Idiopathic(n 5 122)*[no. (%) of

eyes]

HLA-B271 NoSystemicDisease

(n 5 143)*[no. (%) of

eyes]

HLA-B271

SystemicDisease

(n 5 148)*[no. (%) of

eyes]

20/20–20/40 109 (89) 106 (74) 116 (78)20/50–20/100 11 (9) 24 (17) 19 (13)20/200–20/400 2 (2) 9 (6) 9 (6)CF–LP 0 (0) 3 (2) 4 (3)NLP 0 (6) 1 (1) 0 (0)No. (%) of legally

blind eyes 2 (2) 13 (11) 13 (11)

CF 5 counting fingers; LP 5 light perception; NLP 5 no light perception.

* Total no. of involved eyes.

1650

to7s.

er

t

ts

lly

g

-g

e

-.

l

may be related to severe, sight-threatening posteriorment manifestations in up to 17.4% of patients and thatis an under-recognized phenomenon.13

In the current study, only 2% of eyes in the HLA-B2negative group had a final visual acuity of 20/200 or worwhereas the percentage of legally blind eyes in HLA-Bpositive patients with and without systemic disease wsignificantly greater (11% in both groups). The significdifference in the rate of ocular complications andmarked shifts of vision experienced by HLA-B27-positipatients during the clinical course of the disease sugthat, if treated inappropriately or left untreated, thesetients would have a poorer visual outcome.

The less-favorable clinical course in the HLA-B27-poitive group is further supported by the requirement of maggressive therapeutic strategies in this group to coninflammation. The more frequent use (three-fold) of topinonsteroidal anti-inflammatory drugs in patients with HLB27-positive uveitis reflects the higher rate of cystoid mular edema among these patients. There also was a sicantly increased requirement of periocular (four-fold) asystemic corticosteroids (eight-fold) in the HLA-B27-potive patients when compared to those with HLA-B27-neative uveitis. Finally, systemic immunosuppressive chemtherapy, especially methotrexate and azathioprine, wasin all but one case in patients with HLA-B27-positive uvitis, particularly in those with an associated systemic dease. These differences indicate, with aggressive steroidchemotherapy as proxy measures of severity, that HB27-positive patients clearly suffer more severe treatmresistant uveitis than do HLA-B27-negative patients.

The intraocular inflammation associated with the HLB27 haplotype represents the second most frequent foranterior uveitis preceded only by idiopathic uveitis. Tform of uveitis represents a distinct clinical entity charaterized by recurrent attacks of nongranulomatous iridoclitis. In approximately 50% of the patients, the uveitisassociated with a specific systemic disease, namely anegative spondyloarthropathy or inflammatory bowel dease or both. Our data suggest that the prognosis of anuveitis associated with the HLA-B27-positive haplotypeless favorable when compared with that of HLA-B27-neative patients with idiopathic anterior uveitis. A patient wanterior uveitis associated with a positive HLA-B27 haptype should be informed of the recurrent nature ofdisease, the probable association with systemic diseasethe not-too-infrequent possibility of serious sight-threating ocular complications.

rrahal

50. In

on8.r

ve

ior

o-d.onmleub

27

cal03

rioA

oc

ee-

lal-

s--.-

d

cer.

o--8.

gy

dl

Power et al z Anterior Uveitis and the HLA-B27 Haplotype

References

1. Rodriguez A, Calonge M, Pedroza–Seres M, et al. Refepatterns of uveitis in a tertiary eye care center. Arch Ophtmol 1996;114:593–9.

2. Smit RLMJ, Baarsma GS, deVries J. Classification of 7consecutive uveitis patients in the Rotterdam Eye HospitalOphthalmol 1993;17:71–6.

3. Weiner A, BenEzra D. Clinical patterns and associated cditions in chronic uveitis. Am J Ophthalmol 1991;112:151–

4. Brewerton DA, Caffrey M, Nicholls A, et al. Acute anteriouveitis and HL-A 27. Lancet 1973;2:994–6.

5. Mapstone R, Woodrow JC. HL-A 27 and acute anterior uitis. Br J Ophthalmol 1975;59:270–5.

6. Wakefield D, Montanaro A, McCluskey P. Acute anteruveitis and HLA-B27. Surv Ophthalmol 1991;36:223–32.

7. Rothova A, Kijlstra A, Buitenhuis HJ, et al. HLA-B27 assciated uveitis—a distinct clinical entity? In: Saari KM, eUveitis update: Proceedings of the First Int’l SymposiumUveitis, Hanasaar, Espoa, Finland, May 16–19, 1984. Asterdam; New York: Excerpta Medica; New York, NY: Sadistributors for the Asia and Canada, Elsevier Science PCo., 1984;91–5.

8. Wakefield D, Easter J, Penny R. Clinical features of HLA-Banterior uveitis. Aust J Ophthalmol 1984;12:191–6.

9. Rothova A, van Veenendaal WG, Linssen A, et al. Clinifeatures of acute anterior uveitis. Am J Ophthalmol 1987;1137–45.

10. Beckingsale AB, Guss RB, Rosenthal AR. Acute anteuveitis associated with HLA B27 positive tissue type.

l-

t

-

-

-

.

:

r

comparative study in two populations. Trans Ophthalmol SUK 1982;102:168–70.

11. Dougados M, van der Linden SJEF, Juhlin R, et al. ThEuropean Spondylarthropathy Study Group preliminary critria for the classification of spondylarthropathy. ArthritisRheum 1991;34:1218–27.

12. Lee DA, Barker SM, Daniel Su WP, et al. The clinicadiagnosis of Reiter’s syndrome. Ophthalmic and nonophthmic aspects. Ophthalmology 1986;93:350–6.

13. Rodriguez A, Akova YA, Pedroza–Seres M, Foster CS. Poterior segment ocular manifestations in patients with HLAB27-associated uveitis. Ophthalmology 1994;101:1267–74

14. Brewerton DA, Hart FD, Nicholls A, et al. Ankylosing Spondylitis and HL-A 27. Lancet 1973;1:904–7.

15. Khan MA, Kushner I, Braun WE. Association of HLA-A2with uveitis in HLA-B27 positive patients with ankylosingspondylitis. J Rheumatol 1981;8:295–8.

16. D’Alessandro LP, Forster DJ, Rao NA. Anterior uveitis anhypopyon. Am J Ophthalmol 1991;112:317–21.

17. Rodriguez A, Power WJ, Neves RA, Foster CS. Recurrenrate of herpetic uveitis in patients on long-term oral acycloviDoc Ophthalmol 1995;90:331–40.

18. Faris BM, Foster CS. Behcet’s Disease. In: Albert DM, Jakbiec FA, eds. Principles and Practice of Ophthalmology: Clinical Practice. Philadelphia: WB Saunders, 1994; v. 2, chap. 8

19. Tamesis RR, Foster CS. Ocular Syphilis. Ophthalmolo1990;97:1281–7.

20. Linssen A, Meenken C. Outcomes of HLA-B27-positive anHLA-B27-negative acute anterior uveitis. Am J Ophthalmo1995;120:351–61.

1651