The Multicenter Uveitis Steroid Treatment Trial: Rationale, Design

●

Ma●

tcsi●

ai((aotclpfl●

(u(g

A

H

S(DsEotUCPmoDSmM

avjOOe

5

The Multicenter Uveitis Steroid Treatment Trial:Rationale, Design, and Baseline Characteristics

THE MULTICENTER UVEITIS STEROID TREATMENT TRIAL RESEARCH GROUP*

eu●

nmathewOA

Ubsstbambbmp

crehavuart

sHmtcdc

PURPOSE: To describe the design and methods of theulticenter Uveitis Steroid Treatment (MUST) trial

nd the baseline characteristics of enrolled patients.DESIGN: Baseline data from a 1:1 randomized, parallel

reatment design clinical trial at 23 clinical centersomparing systemic corticosteroid therapy (and immuno-uppression when indicated) with fluocinolone acetonidemplant placement.

METHODS: Eligible patients had active or recentlyctive noninfectious intermediate uveitis, posterior uve-tis, or panuveitis. The study design had 90% power2-sided type I error rate, 0.05) to detect a 7.5-letter1.5-line) difference between groups in the mean visualcuity change between baseline and 2 years. Secondaryutcomes include ocular and systemic complications ofherapy and quality of life. Baseline characteristics in-lude demographic and clinical characteristics, quality ofife, and reading center gradings of lens and fundushotographs, optical coherence tomography images, anduorescein angiograms.RESULTS: Over 3 years, 255 patients were enrolled

481 eyes with uveitis). At baseline, 50% of eyes withveitis had best-corrected visual acuity worse than 20/4016% worse than 20/200). Lens opacities (39% ofradeable phakic eyes), macular edema (36%), and

Supplemental Material available at AJO.com.ccepted for publication Nov 13, 2009.*Writing Committee: John H. Kempen, Michael M. Altaweel, Janet T.olbrook, Douglas A. Jabs, and Elizabeth A. Sugar.From the Ocular Inflammation Service, Department of Ophthalmology/

cheie Eye Institute, University of Pennsylvania, Philadelphia, PennsylvaniaJ.H.K.); the Center for Preventive Ophthalmology and Biostatistics,epartment of Ophthalmology/Scheie Eye Institute, University of Penn-

ylvania, Philadelphia, Pennsylvania (J.H.K.); the Center for Clinicalpidemiology and Biostatistics, Department of Biostatistics and Epidemi-logy, University of Pennsylvania, Philadelphia, Pennsylvania (J.H.K.);he Fundus Photograph Reading Center, Department of Ophthalmology,niversity of Wisconsin, Madison, Wisconsin (M.M.A.); the Center forlinical Trials, The Johns Hopkins University Bloomberg School ofublic Health, Baltimore, Maryland (J.T.H., D.A.J., E.A.S.); the Depart-ent of Epidemiology, The Johns Hopkins University Bloomberg School

f Public Health, Baltimore, Maryland (J.T.H., D.A.J., E.A.S.); theepartment of Biostatistics, The Johns Hopkins University Bloombergchool of Public Health, Baltimore, Maryland (E.A.S.); and the Depart-ents of Ophthalmology and Medicine, The Mount Sinai School ofedicine, New York, New York (D.A.J.).Inquiries to John H. Kempen, Center for Preventive Ophthalmology

nd Biostatistics, Department of Ophthalmology, University of Pennsyl-ania, 3535 Market Street, Suite 700, Philadelphia, PA 19104; e-mail:[email protected]; and Douglas A. Jabs, MUST Chairman’sffice, Department of Ophthalmology, Mount Sinai School of Medicine,

pne Gustave L. Levy Place, Box 1183, New York, NY 10029-6574;

-mail: [email protected]

© 2010 BY ELSEVIER INC. A50

piretinal membrane (48%) were common. Mean healthtility was 74.1.CONCLUSIONS: The MUST trial will compare fluoci-

olone acetonide implant versus systemic therapy foranagement of intermediate uveitis, posterior uveitis,

nd panuveitis. Patients with intermediate uveitis, pos-erior uveitis, or panuveitis enrolled in the trial had aigh burden of reduced visual acuity, cataract, maculardema, and epiretinal membrane; overall quality of lifeas lower than expected based on visual acuity. (Am Jphthalmol 2010;149:550–561. © 2010 by Elsevier Inc.ll rights reserved.)

VEITIS, OR INTRAOCULAR INFLAMMATION, IS AN

important cause of visual loss in the developedworld,1–3 reported as causing 10% of cases of

lindness in the United States2 and as being the fifth,ixth, or seventh leading cause of blindness in varioustudies.3,4 Uveitis has a disproportionately high impact inerms of years of potential vision lost and economic effectsecause it often strikes at a younger age than commonge-related eye disorders such as cataract, age-relatedacular degeneration, and glaucoma.1 The proportion of

lindness caused by uveitis may be declining,3 presumablyecause of improving treatment. However, most patientsanaged in tertiary clinics experience visual loss at some

oint during their clinical course.5

On the basis of clinical examination, uveitis can belassified into anterior uveitis, intermediate uveitis, poste-ior uveitis, or panuveitis—based on which portion of theye is inflamed.6,7 The risk of vision loss is progressivelyigher along this spectrum.5,8 In developed countries, suchs the United States, most intermediate uveitis and panu-eitis cases and approximately one half of the posteriorveitis cases treated at uveitis practices, are presumed to beutoimmune, with no evidence of infection and a salutaryesponse to corticosteroid and other anti-inflammatoryherapies.9 –20

Noninfectious uveitides encompass a wide variety ofpecific syndromes, each with specific diagnostic features.owever, for noninfectious cases, corticosteroids are theainstay of treatment in most instances, regardless of

he specific syndrome diagnosed.21–23 Use of systemicorticosteroids—with immunosuppressive drugs when in-icated—historically has been the primary method advo-ated for control of severe cases of intermediate uveitis,
osterior uveitis, and panuveitis.22 The fluocinolone ace-
LL RIGHTS RESERVED. 0002-9394/10/$36.00doi:10.1016/j.ajo.2009.11.019

http://AJO.com

mailto:[email protected]

mailto:[email protected]

tNbtlrh

sactusiitp

T

tmattsnpcstttasdac

ta(ta(le9eEdt(

Tauc

aa(ed(Wirlm

oaIttccdwaotwaEitis

saw1mubswwddm(cdte

V

onide implant (0.59 mg; Bausch & Lomb, Inc, Rochester,ew York, USA),24,25 introduced in 2005, is designed to

e effective in controlling uveitis for 2.5 to 3 years,26 andhus offers an alternative paradigm for the medium- orong-term management of these cases of uveitis. Theelative efficacy and side-effect profiles of these alternativesave not been compared.To provide evidence on the relative effectiveness and

afety of systemic therapy with respect to fluocinolonecetonide implant therapy, we undertook a randomized,ontrolled clinical trial directly comparing these alterna-ives for the management of noninfectious intermediateveitis, posterior uveitis, or panuveitis. This report de-cribes the design of the trial and the baseline character-stics of the patients enrolled into the trial, providing newnformation about the demographic and clinical charac-eristics of intermediate uveitis, posterior uveitis, andanuveitis patients managed in tertiary uveitis practices.

METHODS

HE MULTICENTER UVEITIS STEROID TREATMENT (MUST)

rial is a randomized, partially masked, comparativeulticenter clinical trial comparing the effectiveness

nd safety of local therapy with the fluocinolone ace-onide implant (Bausch & Lomb, Inc) versus systemicherapy with oral corticosteroids and immunosuppres-ive drugs when indicated22 for patients with severeoninfectious intermediate uveitis, posterior uveitis, oranuveitis. The specific aims of the trial are: (1) toompare visual outcomes between implant therapy andystemic therapy; (2) to compare effectiveness of thesereatment strategies for controlling uveitis and avoidinghe accumulation of complications thereof over time; (3)o compare rates of both ocular and systemic side effects;nd (4) to compare quality of life between treatmenttrategies. Cost-effectiveness analysis27 also will be con-ucted. The overall goal is to ascertain the relative benefitsnd risks of the alternative therapies and the incrementalost-effectiveness of the 2 treatment approaches.

A detailed description of the study methods—includinghe rationale for the specific methodologies and treatmentpproaches selected—is provided as supplemental textavailable online at www.AJO.com). In brief, the MUSTrial is designed to detect a difference between randomlyssigned treatment groups of 7.5 letters (1.5 lines) in meaneye-specific) change in logMAR visual acuity from base-ine to follow-up at 2 years. Sample size calculationsstimated that enrollment of 250 patients would provide0% power to detect this difference, allowing for a type 1rror probability of 0.05, assuming 10% losses to follow-up.ligible patients had active or recently active (within 60ays) intermediate uveitis, posterior uveitis, or panuveitishat was judged to require systemic corticosteroid therapy
eligibility criteria are given in Supplemental Table 1). d
MUST TOL. 149, NO. 4

hese are the forms of uveitis for which the fluocinolonecetonide implant therapy is indicated and are forms ofveitis for which systemic therapy as studied in the trial isommonly indicated.22

Because one of the alternative treatments is systemicallydministered, and the goal of the trial was to comparelternative treatment strategies at the level of the patientincluding quality-of-life analysis), patients (rather thanyes) were randomized to the alternative treatments. Ran-om treatment assignments to implant or systemic therapyallocation ratio, 1:1) were stratified by clinical center.

ithin each clinical center, randomization also was strat-fied by the type of uveitis (intermediate uveitis vs poste-ior uveitis or panuveitis) because visual acuity outcomesikely differ between these disease categories8 (see Supple-

ental Figure 1).Implant therapy (see Supplemental Figure 2) consisted

f treatment with the commercially available fluocinolonecetonide 0.59-mg implant (Retisert; Bausch & Lomb,nc) in each eye with uveitis of sufficient severity to justifyreatment with systemic corticosteroids. After quieting ofhe anterior chamber to less than grade 1� of anteriorhamber cells,7 implantation was performed by a study-ertified surgeon using a standard technique25 within 28ays of randomization. When applicable, the second eyeas implanted within 28 days thereafter using the samepproach. Eyes initially receiving systemic corticosteroidr immunosuppressive therapy were tapered off of suchherapy after implant placement, generally within severaleeks, but slowly enough to avoid complications related todrenal suppression resulting from corticosteroid therapy.yes in which a reactivation of uveitis developed after

mplantation were reimplanted using the same approach;he choice to remove or leave in place a preexistingmplant is left to the best medical judgment of the treatingurgeon.

Systemic therapy consisted of oral corticosteroid therapy—upplemented when indicated by immunosuppression—inccordance with expert panel recommendations.22 Patientsith active uveitis at baseline were started on prednisonemg/kg per day up to a maximum adult oral dose of 60g/day (see Supplemental Figure 3), which was continued

ntil the uveitis was controlled or until the patient hadeen receiving this dose of prednisone for 4 weeks. Afteruppression of uveitis was achieved, oral corticosteroidsere tapered, according to specified guidelines. Patientshose uveitis already had been controlled within the 60ays before enrollment started at their current prednisoneose and tapered their prednisone dose in the sameanner. When immunosuppressive therapy was indicated

see Supplemental Figure 3 and Supplemental Text), thehoice of which immunosuppressive drug to use was notictated by the trial to permit selection of the agent withhe best side-effect profile for an individual patient. How-ver, the treatments selected were administered in accor-
ance with expert panel guidelines, including laboratory
RIAL 551

http://www.AJO.com

mtT

pjmbofl

ucstrmsZv

or bo

5

onitoring for potential toxicities.22 Many additionalreatment protocol details are given in the Supplementalext (available at www.AJO.com).Because the primary goal of therapy for uveitis is to

reserve vision, the primary outcome by which success isudged in the MUST trial is best-corrected visual acuity, aseasured by standard logarithmic (Early Treatment Dia-

etic Retinopathy Study) visual acuity charts.38 Otherutcomes evaluated pertain to control of intraocular in-

TABLE 1. Characteristics of Pa

Characteristics Total

Participants, n (%) 255

Demographics

Age at enrollment (yrs)

Mean � SD 46.3 � 15.0

Median (25th to 75th percentile) 47 (34 to 56)

Male gender, n (%) 64 (25%)

Race/ethnicity, n (%)

White 142 (56%)

Hispanic or Latino (any race) 33 (13%)

Black 66 (26%)

Other 14 (5%)

Quality of life

Health utilityb (0 to 100)

Mean � SD 74.1 � 20.0

Median (25th to 75th percentile) 80 (67 to 90)

Missing 3 (1%)

Clinical characteristics

Uveitis, n (%)

Bilateral 226 (89%)

Unilateral 29 (11%)

Time since diagnosis (yrs)

Mean � SD 6.1 � 7.2

Median (25th to 75th percentile) 3.8 (1.2 to 8.0)

Missing 4 (2%)

Systemic disease, n (%)

None 186 (73%)

Present 69 (27%)

Behçet’s disease 8 (3%)

Sarcoidosis 26 (10%)

Multiple sclerosis 4 (2%)

Juvenile idiopathic arthritis 5 (2%)

Bone density, n (%)d

Normal 132 (53%)

Osteopenia 96 (39%)

Osteoporosis 21 (8%)

Missing 6 (2%)

SD � standard deviation; yrs � years.aUnless otherwise specified, P values compare the distribution obHealth utility reflects the EuroQol visual analog scale score,40 wcP value indicates a comparison of a single category across stradOsteopenia is defined as a T score between �1 and �2.49 inclu

defined as a T score of �2.5 or worse at the spine, femoral neck,

ammation, the occurrence of ocular complications of m

AMERICAN JOURNAL OF52

veitis or of therapy, the occurrence of systemic compli-ations of therapy, and self-reported quality of life. Imagingtudies used and the representative features evaluated byhem include: lens photography (cataract), fundus photog-aphy (macular edema, epiretinal membranes, optic nerveorphologic features, chorioretinal lesions, vascular occlu-

ions), Stratus optical coherence tomography ([Carleiss Meditec, Inc, Jena, Germany]; macular edema,itreoretinal interface abnormalities including epiretinal

with Uveitis at Randomization

Disease Stratum

P ValueaIntermediate Posterior Uveitis or Panuveitis

97 (38%) 158 (62%)

45.4 � 15.1 46.9 � 15.0 .46

46 (33 to 55) 47 (35 to 58)

28 (29%) 36 (23%) .28

60 (62%) 82 (52%) �.01

5 (5%) 28 (18%)

29 (30%) 37 (23%)

3 (3%) 11 (7%)

72.1 � 21.0 75.3 � 19.4 .20

75 (60 to 87) 80 (70 to 90)

2 (2%) 1 (1%)

83 (86%) 143 (91%) .23

14 (14%) 15 (9%)

6.0 � 5.0 6.2 � 8.2 .07

5.0 (2.0 to 8.8) 3.4 (0.9 to 7.9)

1 (1%) 3 (2%)

77 (79%) 109 (69%) .07

20 (21%) 49 (31%)

0 (0%) 8 (5%) .03c

7 (7%) 19 (12%) .29c

3 (3%) 1 (1%) .16c

2 (2%) 3 (2%) �.99c

53 (56%) 79 (51%) .61

33 (35%) 63 (41%)

9 (9%) 12 (8%)

2 (1%) 4 (2%)

rimary categories except for the 2 missing strata.

0 reflects death and 100 reflects perfect health.

at the spine or femoral neck (whichever is worse). Osteoporosis is

th.37

tients

f all p

here

ta.

sive

embrane), and fluorescein angiography (macular edema,

OPHTHALMOLOGY APRIL 2010

http://www.AJO.com

V

TABLE 2. Ocular Characteristics of Eyes with Uveitis at Baseline


Disease Stratum


No. of participants 255 97 158Eyes with uveitis at enrollment, n (%) 481 180 (37%) 301 (63%)Vitreous haze, n (%)b

0 141 (31%) 49 (29%) 92 (32%) .261� 205 (45%) 71 (42%) 134 (47%)2� 77 (17%) 36 (21%) 41 (14%)3� 20 (4%) 9 (5%) 11 (4%)4� 4 (1%) 0 (0%) 4 (1%)Cannot assess 8 (2%) 6 (3%) 2 (1%)Missing 26 (5%) 9 (5%) 17 (6%)

Anterior vitreous cells, n (%)c

0 75 (17%) 29 (17%) 46 (16%) .820.5� 109 (24%) 34 (20%) 75 (27%)1� 146 (32%) 61 (36%) 85 (30%)2� 90 (20%) 35 (21%) 55 (19%)3� 25 (6%) 6 (4%) 19 (7%)4� 6 (1%) 3 (2%) 3 (1%)Cannot assess 28 (6%) 12 (7%) 16 (5%)Missing 2 (1%) 0 (0%) 2 (1%)

IOP (mm Hg)Mean � SD 14.0 � 4.7 13.7 � 4.0 14.3 � 5.0 .14Median (25th to 75th percentile) 14.0 (11.5 to 17.0) 14.0 (11.5 to 16.0) 14.5 (12.0 to 17.0)Missing 4 (0%) 0 (0%) 4 (1%)

Visual acuityEyes with uveitis at enrollment

Mean � SD 61.4 � 26.4 61.8 � 25.8 61.2 � 26.9 .92Median (25th to 75th percentile) 70.1 (49.1 to 80.1) 69.1 (49.1 to 81.1) 70.1 (50.1 to 80.1)Worse than 20/40, n (%) 237 (50%) 94 (53%) 143 (48%) .34d

Worse than 20/200, n (%) 74 (15%) 27 (15%) 47 (16%) .88d

Missing 6 (1%) 1 (1%) 5 (2%)Better eyese

Mean � SD 73.4 � 18.0 73.5 � 19.6 73.4 � 16.9 .48Median (25th to 75th percentile) 78.1 (66.1 to 86.1) 80.1 (64.1 to 87.1) 77.1 (66.1 to 85.1)Worse than 20/40, n (%) 79 (31%) 32 (33%) 47 (30%) .68d

Worse than 20/200, n (%) 12 (5%) 5 (5%) 7 (4%) .77d

Missing 1 (0.3%) 0 (0%) 1 (1%)Worse eyese

Mean � SD 50.5 � 30.5 52.2 � 29.2 49.5 � 31.4 .58Median (25th to 75th percentile) 59.1 (30.1 to 75.1) 61.1 (35.1 to 75.1) 57.1 (29.1 to 75.1)Worse than 20/40, n (%) 165 (65%) 64 (66%) 101 (64%) .89Worse than 20/200, n (%) 67 (26%) 24 (25%) 43 (27%) .66Missing 1 (0.3%) 0 (0%) 1 (1%)

Visual field automated perimetryMean deviationMean � SD �7.8 � 7.5 �6.4 � 6.5 �8.7 � 8.0 �.01Median (25th to 75th percentile) �5.2 (�9.5 to �3.0) �4.0 (�7.6 to �2.8) �5.8 (�11.2 to �3.2)Missing 22 (5%) 6 (3%) 16 (5%)

IOP � intraocular pressure; SD � standard deviation.aP values are calculated using linear, logistic, or multinomial regression adjusting for multiple measurements per individual with robust

standard errors when appropriate. Unless otherwise indicated, the P value represents the comparison between strata across all categories

except for missing or cannot assess.bVitreous haze measurements are based on the scale developed by Nussenblatt and associates and affirmed by the SUN Working Group.cAnterior vitreous cell measurements are based on the following scale, based on cells present in a 1 � 0.5-mm beam: 0 � no cells; 0.5� � 1

to 5 cells; 1� � 6 to 10 cells; 2� � 11 to 20 cells; 3� � 21 to 50 cells; and 4� � 51 cells or more.dP value indicates a comparison of a single category across strata.e
Visual acuity for better and worse eyes are calculated based on all eyes, regardless of whether they have uveitis.
MUST TRIALOL. 149, NO. 4 553

5

TABLE 3. Image Characteristics in Eyes with Uveitis at Enrollment


Disease Stratum


No. of participants 255 97 158Eyes with uveitis at enrollment, n (%) 481 180 301Lens images, n 464 180 284Lens opacities, n (%)

Absent 134 (61%) 56 (67%) 78 (57%) .21b

Present 87 (39%) 27 (33%) 60 (43%)Aphakic/pseudophakic 191 (41%) 75 (41%) 116 (41%)Cannot gradec 52 (11%) 22 (12%) 30 (10%)

Optical coherence tomography, n 454 171 283Retinal thickness at center (�m)

Mean � SD 268.0 � 185.0 288.2 � 188.8 256.0 � 181.9 .09Median (25th to 75th percentile) 198.5 (162.0 to 296.5) 211.0 (168.0 to 334.0) 192.0 (158.0 to 270.0)�200 223 (51%) 71 (44%) 152 (56%) .05201 to 239 57 (12%) 26 (15%) 31 (11%)�240 156 (36%) 66 (41%) 90 (33%)Cannot gradec 18 (4%) 8 (4%) 10 (3%)

Cystoid spaces, n (%)Absent 265 (62%) 97 (61%) 168 (62%) .81Definite 166 (38%) 63 (39%) 103 (38%)

Not at center 58 (13%) 15 (9%) 43 (16%)At center 108 (25%) 48 (30%) 60 (22%)

Cannot gradec 23 (5%) 11 (6%) 12 (4%)Color fundus images, n 436 165 271

Epiretinal membrane, n (%)Absent 198 (52%) 60 (42%) 138 (57%) .01Present 185 (48%) 82 (58%) 103 (43%)

Cellophane reflex/subtle 139 (36%) 66 (46%) 73 (30%)Definite 46 (12%) 16 (11%) 30 (13%)

Cannot gradec 53 (12%) 23 (14%) 30 (11%)Subretinal fibrosis, n (%)

Absent 367 (96%) 141 (99%) 226 (94%) .05Definite 17 (4%) 2 (1%) 15 (6%)

Not at center 14 (4%) 2 (1%) 12 (5%)At center 3 (1%) 0 (0%) 3 (1%)

Cannot gradec 52 (12%) 22 (13%) 30 (11%)Cup-to-disc ratio (vertical)

Mean � SD 0.31 � 0.12 0.29 � 0.13 0.33 � 0.11 .10Median (IQR) 0.30 (0.21 to 0.39) 0.27 (0.19 to 0.37) 0.32 (0.24 to 0.40)Cannot gradec 82 (18%) 30 (18%) 52 (19%)

Papillary swellingAbsent 378 (96%) 149 (99%) 229 (94%) .03Definite 16 (4%) 1 (1%) 15 (6%)Cannot gradec 42 (10%) 15 (9%) 27 (10%)

Pigment disturbance within the macular aread

Absent 263 (69%) 117 (84%) 146 (61%) �.01Definite 116 (31%) 23 (16%) 93 (39%)

Not at center 70 (19%) 15 (11%) 55 (23%)At center point 46 (12%) 8 (6%) 38 (16%)

Cannot gradec 57 (13%) 25 (15%) 32 (12%)Fluorescein angiography, n 394 142 254

Cystoid changes in the maculaAbsent 256 (70%) 80 (60%) 176 (76%) �.01Present 109 (30%) 53 (40%) 56 (24%)Cannot gradec 31 (8%) 9 (6%) 22 (9%)

Continued on next page

AMERICAN JOURNAL OF OPHTHALMOLOGY54 APRIL 2010

vvaEgSqIampgcif

●

Piaocpvaaratdo

B

prpuapattuvb7ptpps1pwlda(o

V

ascular leakage, perfusion abnormalities, choroidal neo-ascularization). Health-related quality-of-life data alsore assessed, including health utility as measured by theuroQol 5-dimension and Visual Analog Scale scores,39,40

eneral health-related quality of life as measured by thehort Form 36-item questionnaire,41,42 and vision-relateduality of life as measured by the 25-item National Eyenstitute Visual Function Questionnaire.43 Masking ispplied to the determination of visual acuity (beginning 6onths after randomization, to avoid unmasking by visible

ostoperative changes), those outcomes based on photo-raphic reading, and diagnosis of glaucoma by an outcomesommittee. Because sham surgery is not performed, mask-ng during ascertainment of the other outcomes is noteasible.

STATISTICAL METHODS FOR BASELINE ANALYSIS:

atient, ocular, and imaging characteristics at random-zation were summarized for the population as a wholend were stratified by uveitis type at enrollment basedn data collected as of October 1, 2009. For patientharacteristics, differences between the strata were com-ared using the Wilcoxon rank-sum test for continuousariables and the chi-square test— or Fisher exact test ifppropriate—for categorical variables. When the unit ofnalysis was the eye, linear, logistic, or multinomialegression models were used to compare strata afterdjusting for the excess correlation between eyes fromhe same individual using generalized estimating equations-erived methods for continuous, binary, and multinomial

TABLE 3. Image Characteristics in Ey


Capillary loss within the macular areac

Absent 197 (95%)

Present 11 (5%)

Missing 188 (47%)

Leakage within the macular aread

Mean � SD 3.5 � 5.3

Median (IQR) 0.6 (0.0 to 4.

Absent 148 (40%)

Present 218 (60%)

Cannot gradec 30 (8%)

IQR � interquartile range; SD � standard deviation.aUnless otherwise specified, P values are calculated to compare p

involvement.bComparing absent and present.cImage quality insufficient to allow grading, usually because of odThe macular area refers to the 16-disc area–region centered on

utcomes, respectively.35 i


RESULTS

ETWEEN DECEMBER 6, 2005, AND DECEMBER 9, 2008, 255

atients were enrolled at 23 clinical centers and wereandomized to implant (129 patients) or systemic (126atients) therapy. Ninety-seven patients (180 eyes withveitis) were enrolled in the intermediate uveitis stratum,nd 158 (301 eyes with uveitis) were enrolled in theosterior uveitis or panuveitis stratum (see Table 1). Thege and sex distributions were similar between groups, buthe race or ethnicity distribution differed significantly inhat Hispanic persons were more likely to have posteriorveitis or panuveitis than intermediate uveitis. EuroQolisual analog scale40 health utility scores were similaretween groups: the mean overall score (74.1; 95% CI,1.6 to 76.6) was reduced with respect to normativeopulation values59 (normative score, 80.0; standardized tohe MUST population for age group and sex). Eighty-nineercent of patients had bilateral uveitis, with a similarroportion in both strata. In total, 21% of patients had aystemic inflammatory disease in association with uveitis,5% in the intermediate uveitis group versus 25% in theosterior uveitis or panuveitis group (P � .05). Patientsith posterior uveitis or panuveitis were substantially more

ikely to have Behçet’s disease than patients with interme-iate uveitis (5% vs 0%; P � .03). Osteoporosis, defined asT score less than �2.5 on DEXA scanning of the spine

L2 to L4) and left femoral neck,37 was present among 8%f patients in both groups at baseline.Among 481 eyes with uveitis, 180 (37%) were in the

ith Uveitis at Enrollment (Continued)

Disease Stratum


74 (96%) 123 (94%) .50

3 (4%) 8 (6%)

65 (46%) 123 (48%)

3.5 � 5.0 3.6 � 5.5 .90

1.1 (0.0 to 4.9) 0.5 (0.0 to 4.6)

48 (36%) 100 (43%) .29

85 (64%) 133 (57%)

9 (6%) 21 (8%)

y categorizations excluding cannot grade, missing, or central point

scarring preventing adequate imaging.

ovea.

es w

9)

rimar

cular

the f

ntermediate uveitis stratum and 301 (63%) were in the

RIAL 555

pTiggpsehvwwueubetd7

wo3ati�micsmgmpdnh(wtS(ci

T

woaod

tibiSemiipaldp

ataotesegmloaomsifp

apptdrswpptaw

ouiqpo

5

osterior uveitis or panuveitis stratum (see Table 2).he distributions of severity of vitreous haze and vitreous

nflammatory cells were similar at baseline between the 2roups. Intraocular pressure also was similar in the 2roups, with most patients having average intraocularressure at baseline; patients with high intraocular pres-ures and uncontrolled glaucoma had been excluded fromnrollment (see Supplemental Table 1). Approximatelyalf of eyes with uveitis had low vision (best-correctedisual acuity worse than 20/40), and approximately 16%ere legally blind (best-corrected visual acuity of 20/200 ororse), with a similar distribution across intermediateveitis versus posterior uveitis or panuveitis cases. Consid-ring the better eye of each patient, including eyes withoutveitis, 31% had low vision and 5% were legally blind ataseline. As expected, visual field reductions were morextensive in the posterior uveitis or panuveitis group thanhe intermediate uveitis group (average Humphrey meaneviations, �8.7 vs �6.4 respectively; P � .01). Overall,5% of patients had a mean deviation of �3.0 or worse.Regarding structural ocular abnormalities associated

ith inflammation, as ascertained by image grading, 41%f eyes were pseudophakic at baseline in each group, and9% of the remainder were judged to have a lens opacityt baseline (see Table 3). The mean macular thicknessended to be greater in the intermediate uveitis group thann the posterior uveitis or panuveitis group (288 vs 256m; P � .09). Cystoid edema was present in the centralacula in 38% of eyes, with a similar distribution across

ntermediate uveitis and posterior uveitis and panuveitisases. Fluorescein angiographic grading demonstrated aimilar degree of macular leakage in the 2 groups, with aean of 3.5 � 5.3 disc areas. By grading of color photo-

raphs, epiretinal membranes were significantly more com-on in the intermediate uveitis group than the posterior or

anuveitis group (57% vs 42%; P � .01), although theifference arose from a greater frequency of subtle epireti-al membranes in the former group, whereas both groupsad a similar proportion with definite epiretinal membrane11% vs 13%; P � .63). Macular pigmentary disturbancesere more common in posterior uveitis or panuveitis cases

han intermediate uveitis cases (39% vs 16%; P � .01).ubretinal fibrosis (6% vs. 1%; P � .05) and disc swelling6% vs 1%; P � .03) were infrequent, but were moreommon in the posterior uveitis and panuveitis group thann the intermediate uveitis group.

DISCUSSION

HE 2005 INTRODUCTION OF THE FLUOCINOLONE IMPLANT,

hich is effective for controlling severe posterior segmentcular inflammation over long periods,24,25 raised questionsbout the relative merits of local therapy with the fluocin-lone acetonide implant versus systemic therapy for interme-
iate uveitis, posterior uveitis, and panuveitis. Implant 2

herapy is expected to be highly effective for control ofnflammation while the implant continues to deliver drug,ut is expected to result in a high incidence of cataract andntraocular pressure elevation requiring treatment.24,25,28

ystemic therapy is likely to have more systemic sideffects22 and more relapses of inflammation—because theinimum suppressive prednisone dose is determined by

terative tapering—but is likely to result in a lowerncidence of local complications of corticosteroid thera-y60,61 and may have systemic or ocular benefits byddressing more comprehensively the autoimmunity thatikely underlies many of these cases. The MUST trial isesigned to provide guidance as to whether one of thesearadigms is superior.Given that the proportion of eyes with bilateral uveitis

nd the number of patients actually enrolled was higherhan expected (see Supplemental Text, available at www.jo.com), the trial will have greater statistical power thanriginally estimated, and therefore is expected to answerhe questions of interest effectively. The trial is designed tovaluate both the benefits and the side effects (local andystemic) of the alternative treatments. Because local sideffects are expected to be more frequent in the implantroup24,25,28 and systemic side effects are expected to beore frequent in the systemic therapy group,22 quality-of-

ife results may be highly informative regarding the meritsf the alternative treatments. Cost and cost-effectivenessnalyses27 will be performed to determine the relative costsf each treatment approach and to determine the incre-ental cost of each incremental gain in benefit with the

uperior treatment. Enrollment of the trial was completedn December 2008, and the primary results at the 2-yearollow-up can be expected shortly after that follow-upoint is reached (early 2011).In baseline observations from the MUST trial, intermedi-

te uveitis was diagnosed in a lower proportion of Hispanicersons than white or black persons; because clinical trialatients are not necessarily sampled in a manner representa-ive of the general population, it is unclear whether thisifference in the distribution of site of inflammation byace/ethnicity is meaningful. The results also suggest thatystemic inflammatory disease is less commonly associatedith intermediate uveitis than with posterior uveitis oranuveitis, confirming clinical impressions. Nearly 10% ofatients had osteoporosis at baseline, emphasizing the impor-ance of following bone mineral density testing and providingppropriate preventive and remedial treatment in patients forhom systemic corticosteroid therapy will be used.22,37

At enrollment, patients and eyes had a high prevalencef reduced visual acuity and of ocular complications ofveitis, and patients reported a substantially reduced qual-ty of life, emphasizing the significant adverse conse-uences of intermediate uveitis, posterior uveitis, andanuveitis. Using a time tradeoff methodology, the lossesf health utility associated with a reduction of vision to the
0/60 to 20/100 range and to 20/200 or worse have been

http://www.ajo.com

http://www.ajo.com

eqhivapas

ipbgoirLptacktimpiOu

tlttelaepqtit

hrcpoLiahahaauaecscipppiswtfo

tipdiEp2Mplq

SAYsuNoscRMi

V

stimated as being a 28% and a 39% reduction in theuality of life, respectively.62 Our patients were observed toave health utility similar to that associated with visual acuity

n the 20/60 to 20/100 range, although only 31% had lowision in their better eye, suggesting that uveitis may havedditional health impact over and above its effect on vision,erhaps via symptoms of inflammation unmeasured by visualcuity, side effects of treatment, the impact of associatedystemic disease, or a combination thereof.

Among eyes with uveitis, visual acuity was similar in thentermediate uveitis group and the posterior uveitis andanuveitis group. Overall visual sensitivity as measuredy automated visual field testing was reduced in bothroups—likely reflecting the impact of macular edema andther complications of uveitis—but was significantly lowern the posterior uveitis and panuveitis group, probablyeflecting the additional impact of chorioretinal lesions.ens opacities were common in both groups, with 80% eitherseudophakic or with lens opacity at enrollment, indicatinghat there is a high risk of cataract in severe uveitis cases,lthough likely in part reflecting a greater willingness toonsider randomization to implant treatment in pseudopha-ic or cataractous eyes, given that implant therapy is expectedo cause cataract in nearly all phakic eyes.25,63 Eyes withntermediate uveitis tended to have a higher prevalence ofacular edema and epiretinal membrane, which may reflect a

ractice pattern of using the occurrence of such events as anndication for more aggressive anti-inflammatory therapy.ther complications of posterior segment inflammation were

nusual in both groups.In addition to addressing the clinical questions regarding

reatment approaches, the MUST trial will provide the firstarge, prospective data set of long-term, longitudinal observa-ions on the outcomes of patients with severe uveitis, poten-ially yielding information regarding the clinicalpidemiologic characteristics of complications of uveitis. Thearge number of patients at baseline with reduced visualcuity, and other uveitis-related problems such as maculardema and epiretinal membrane, suggests that the study willrovide substantial power to evaluate several important openuestions for severe cases of these types of uveitis, includinghe risk of both improvement in and loss of visual acuity, thencidence of posterior segment complications of uveitis, and
he incidence of ocular surgery, among others. b

Limitations of the study design arise primarily frometerogeneity of the forms of uveitis involving the poste-ior segment. Posterior uveitis and panuveitis particularlyonsist of a wide variety of inflammatory entities, whichotentially may be heterogeneous in their response to oner both of the treatments. Results from the Bausch &omb BLP 415–001 trial of the fluocinolone acetonide

mplant suggest that implant therapy is highly effectivecross a broad array of uveitis patients,24,25 suggesting thateterogeneity is not likely to be relevant for the implantrm. The benefits of systemic anti-inflammatory therapyave been indicated in several specific disease states,61,64–66

lso suggesting a broad usefulness of this therapeuticpproach. Because patients are being enrolled at tertiaryveitis centers, where treatments of this nature often aredministered, it is likely that more severe cases werenrolled than might have been enrolled at less specializedenters. However, because qualitative treatment–diseaseeverity interactions are unlikely, the relative risk ofhange in vision likely also should be sufficiently general-zable to be useful in determining which treatment ap-roach is best in a less specialized setting. Finally, becauseatients enrolled in clinical trials usually are not typicalatients, the experience of MUST trial patients may differn some ways from that of future patients. However, thetrengths of the randomized clinical trial methodologyith respect to alternative designs outweigh this limita-

ion, leaving the clinical trial approach as the mostavorable design available for evaluating the relative meritsf alternative treatments for specified disease diagnoses.In summary, the MUST trial is a phase 4 effectiveness

rial that aims to evaluate whether fluocinolone acetonidemplant therapy or systemic corticosteroid therapy (sup-lemented when indicated by use of immunosuppressiverugs) is superior for the management of noninfectiousntermediate uveitis, posterior uveitis, and panuveitis.nrollment was completed in December 2008, and therimary 2-year follow-up results can be expected in early011. Observations from the baseline characteristics of theUST patients suggest that these patients have a high

revalence of visual loss, cataract or pseudophakia, macu-ar edema, and epiretinal membrane, and a reduced overalluality of life greater in degree than would be expected
ased on loss of visual acuity alone.
UPPORTED BY COLLABORATIVE AGREEMENTS EYU10EY014655 (DR JABS), EYU10EY014660 (DR HOLBROOK), AND EY014656 (DRltaweel) from the National Eye Institute, Bethesda, Maryland. Additional support was provided by Research to Prevent Blindness, Inc, New York, Nework, and by the Paul and Evanina Mackall Foundation, New York, New York. Bausch & Lomb Inc., Rochester, New York, provided support to the

tudy in the form of donation of a limited number of fluocinolone implants for patients randomized to implant therapy who were uninsured or otherwisenable to pay for implants. Dr Kempen is a Research to Prevent Blindness James S. Adams Special Scholar Award recipient. A representative of theational Eye Institute participated in the conduct of the study, including the study design and the collection, management, analysis, and interpretation

f the data, as well as in the review and approval of this manuscript. Dr Kempen has served as a consultant to Alcon and Lux Biosciences; Dr Jabs haserved as a consultant to Abbott, Allergan, Applied Genetic Technologies, Genzyme, Glaxo Smith Kline, Novartis, and Roche. Involved in design andonduct of study (MUST Research Group; J.H.K., M.M.A., J.T.H., D.A.J.); Collection, management, analysis, and interpretation of data (MUSTesearch Group; J.H.K., M.M.A., J.T.H., D.A.J., E.A.S.); and preparation, review, or approval of the manuscript (MUST Research Group; J.H.K.,.M.A., J.T.H., D.A.J., E.A.S.). The study has been registered on clinicaltrials.gov (identifier NCT00132691) and was approved by the governing

nstitutional review boards of all participating institutions.

RIAL 557

http://www.clinicaltrials.gov

1

1

1

1

1

1

1

1

1

1

2

2

2

2

2

2

2

2

2

2

3

3

3

3

3

3

3

5

REFERENCES

1. Durrani OM, Meads CA, Murray PI. Uveitis: a poten-tially blinding disease. Ophthalmologica 2004;218:223–236.

2. Nussenblatt RB. The natural history of uveitis. Int Ophthal-mol 1990;14:303–308.

3. ten Doesschate J. Causes of blindness in The Netherlands.Doc Ophthalmol 1982;52:279–285.

4. National Advisory Eye Council. Vision Research. A Na-tional Plan, 1983–1987. Bethesda, MD: National Institutesof Health, Public Health Service, US Department of Healthand Human Services; 1983:13.

5. Durrani OM, Tehrani NN, Marr JE, Moradi P, Stavrou P,Murray PI. Degree, duration, and causes of visual loss inuveitis. Br J Ophthalmol 2004;88:1159–1162.

6. Bloch-Michel E, Nussenblatt RB. International UveitisStudy Group recommendations for the evaluation of intraoc-ular inflammatory disease. Am J Ophthalmol 1987;103:234–235.

7. Jabs DA, Nussenblatt RB, Rosenbaum JT. Standardization ofuveitis nomenclature for reporting clinical data. Results ofthe First International Workshop. Am J Ophthalmol 2005;140:509–516.

8. Rothova A, Suttorp-van Schulten MS, Frits Treffers W,Kijlstra A. Causes and frequency of blindness in patientswith intraocular inflammatory disease. Br J Ophthalmol1996;80:332–336.

9. Henderly DE, Genstler AJ, Smith RE, Rao NA. Changingpatterns of uveitis. Am J Ophthalmol 1987;103:131–136.

0. McCannel CA, Holland GN, Helm CJ, Cornell PJ, WinstonJV, Rimmer TG. Causes of uveitis in the general practice ofophthalmology. UCLA Community-Based Uveitis StudyGroup. Am J Ophthalmol 1996;121:35–46.

1. Mercanti A, Parolini B, Bonora A, Lequaglie Q, TomazzoliL. Epidemiology of endogenous uveitis in north-eastern Italy.Analysis of 655 new cases. Acta Ophthalmol Scand 2001;79:64–68.

2. Merrill PT, Kim J, Cox TA, Betor CC, McCallum RM, JaffeGJ. Uveitis in the southeastern United States. Curr Eye Res1997;16:865–874.

3. Palmares J, Coutinho MF, Castro-Correia J. Uveitis innorthern Portugal. Curr Eye Res 1990;9(Suppl):31–34.

4. Perkins ES, Folk J. Uveitis in London and Iowa. Ophthal-mologica 1984;189:36–40.

5. Pivetti-Pezzi P, Accorinti M, La Cava M, Colabelli GisoldiRA, Abdulaziz MA. Endogenous uveitis: an analysis of 1,417cases. Ophthalmologica 1996;210:234–238.

6. Rodriguez A, Calonge M, Pedroza-Seres M, et al. Referralpatterns of uveitis in a tertiary eye care center. ArchOphthalmol 1996;114:593–599.

7. Rothova A, Buitenhuis HJ, Meenken C, et al. Uveitis andsystemic disease. Br J Ophthalmol 1992;76:137–141.

8. Santin M, Badrinas F, Mascaro J, et al. Uveitis: an etiologicalstudy of 200 cases following a protocol [in Spanish]. MedClin (Barc) 1991;96:641–644.

9. Smit RL, Baarsma GS, de Vries J. Classification of 750consecutive uveitis patients in the Rotterdam Eye Hospital.
Int Ophthalmol 1993;17:71–76.

0. Tran VT, Auer C, Guex-Crosier Y, Pittet N, Herbort CP.Epidemiological characteristics of uveitis in Switzerland. IntOphthalmol 1994;18:293–298.

1. Foster CS, Vitale AT. Diagnosis and Treatment of Uveitis.Philadelphia: W. B. Saunders Company; 2002.

2. Jabs DA, Rosenbaum JT, Foster CS, et al. Guidelines for theuse of immunosuppressive drugs in patients with ocularinflammatory disorders: recommendations of an expert panel.Am J Ophthalmol 2000;130:492–513.

3. Nussenblatt RB, Whitcup SM, Palestine AG. Uveitis. Fun-damentals and Clinical Practice. St. Louis: Mosby; 1996.

4. Callanan DG, Jaffe GJ, Martin DF, Pearson PA, ComstockTL. Treatment of posterior uveitis with a fluocinoloneacetonide implant: three-year clinical trial results. ArchOphthalmol 2008;126:1191–1201.

5. Jaffe GJ, Martin D, Callanan D, Pearson PA, Levy B,Comstock T. Fluocinolone acetonide implant (Retisert) fornoninfectious posterior uveitis: thirty-four-week results of amulticenter randomized clinical study. Ophthalmology 2006;113:1020–1027.

6. Jaffe GJ, Ben Nun J, Guo H, Dunn JP, Ashton P. Fluocin-olone acetonide sustained drug delivery device to treat severeuveitis. Ophthalmology 2000;107:2024–2033.

7. Gold MR, Siegel JE, Russell LB, Weinstein MC. Cost-effectiveness in health and medicine. New York: OxfordUniversity Press; 1996.

8. Goldstein DA, Godfrey DG, Hall A, et al. Intraocularpressure in patients with uveitis treated with fluocinoloneacetonide implants. Arch Ophthalmol 2007;125:1478–1485.

9. Beck RW, Maguire MG, Bressler NM, Glassman AR, Lind-blad AS, Ferris FL. Visual acuity as an outcome measure inclinical trials of retinal diseases. Ophthalmology 2007;114:1804–1809.

0. Ferris FL III, Kassoff A, Bresnick GH, Bailey I. New visualacuity charts for clinical research. Am J Ophthalmol 1982;94:91–96.

1. Gragoudas ES, Adamis AP, Cunningham ET Jr, Feinsod M,Guyer DR. Pegaptanib for neovascular age-related maculardegeneration. N Engl J Med 2004;351:2805–2816.

2. Verteporfin in Photodynamic Therapy Study Group. Verte-porfin therapy of subfoveal choroidal neovascularization inage-related macular degeneration: two-year results of a ran-domized clinical trial including lesions with occult with noclassic choroidal neovascularization—verteporfin in photo-dynamic therapy report 2. Am J Ophthalmol 2001;131:541–560.

3. Photodynamic therapy of subfoveal choroidal neovascular-ization in age-related macular degeneration with verteporfin:one-year results of 2 randomized clinical trials—TAP report.Treatment of age-related macular degeneration with photo-dynamic therapy (TAP) Study Group. Arch Ophthalmol1999;117:1329–1345.

4. Laser photocoagulation of subfoveal neovascular lesions inage-related macular degeneration. Results of a randomizedclinical trial. Macular Photocoagulation Study Group. ArchOphthalmol 1991;109:1220–1231.

5. Zeger SL, Liang KY. Longitudinal data analysis for discreteand continuous outcomes. Biometrics 1986;42:121–130.

6. Ryan P. sxd1_4: random allocation of treatments in blocks.
Stata J 2008;8:146.

3

3

3

4

4

4

4

4

4

4

4

4

4

5

5

5

5

5

5

5

5

5

5

6

6

6

6

6

6

6

V

7. American College of Rheumatology Ad Hoc Committee onGlucocorticoid-Induced Osteoporosis. Recommendations forthe prevention and treatment of glucocorticoid-induced osteo-porosis: 2001 update. Arthritis Rheum 2001;44:1496–1503.

8. Ferris FL III, Bailey I. Standardizing the measurement of visualacuity for clinical research studies: guidelines from the Eye CareTechnology Forum. Ophthalmology 1996;103:181–182.

9. Brooks R. EuroQol: the current state of play. Health Policy1996;37:53–72.

0. EuroQol Group. EuroQol—a new facility for the measure-ment of health-related quality of life. Health Policy 1990;16:199–208.

1. Brazier JE, Harper R, Jones NM, et al. Validating the SF-36health survey questionnaire: new outcome measure for pri-mary care. BMJ 1992;305:160–164.

2. Jenkinson C, Wright L, Coulter A. Criterion validity andreliability of the SF-36 in a population sample. Qual Life Res1994;3:7–12.

3. Mangione CM, Lee PP, Gutierrez PR, Spritzer K, Berry S,Hays RD. Development of the 25-item National Eye Insti-tute Visual Function Questionnaire. Arch Ophthalmol 2001;119:1050–1058.

4. Hogan MJ, Kimura SJ, Thygeson P. Signs and symptoms ofuveitis. I. Anterior uveitis. Am J Ophthalmol 1959;47:155–170.

5. Mangione CM, Lee PP, Pitts J, Gutierrez P, Berry S, HaysRD. Psychometric properties of the National Eye InstituteVisual Function Questionnaire (NEI-VFQ). NEI-VFQ FieldTest Investigators. Arch Ophthalmol 1998;116:1496–1504.

6. Coons SJ, Rao S, Keininger DL, Hays RD. A comparativereview of generic quality-of-life instruments. Pharmacoeco-nomics 2000;17:13–35.

7. Musch DC, Lichter PR, Guire KE, Standardi CL. TheCollaborative Initial Glaucoma Treatment Study: study de-sign, methods, and baseline characteristics of enrolled pa-tients. Ophthalmology 1999;106:653–662.

8. Asman P, Heijl A. Glaucoma Hemifield Test. Automatedvisual field evaluation. Arch Ophthalmol 1992;110:812–819.

9. Sommer A, Tielsch JM, Katz J, et al. Relationship betweenintraocular pressure and primary open angle glaucoma amongwhite and black Americans. The Baltimore Eye Survey. ArchOphthalmol 1991;109:1090–1095.

0. Age-Related Eye Disease Study Research Group. The age-related eye disease study (AREDS) system for classifyingcataracts from photographs: AREDS report no. 4. Am JOphthalmol 2001;131:167–175.

1. Expert Committee on the Diagnosis and Classification ofDiabetes Mellitus. Report of the Expert Committee on theDiagnosis and Classification of Diabetes Mellitus. DiabetesCare 1997;20:1183–1197.

2. Osteoporosis: review of the evidence for prevention, diagno-sis and treatment and cost-effectiveness analysis. StatusReport. Osteoporos Int 1998;8(Suppl 4):S3–S6.

3. Miller PD. Bisphosphonates for the prevention and treat-ment of corticosteroid-induced osteoporosis. Osteoporos Int2001;12(Suppl 3):S3–S10.

4. Kanis JA. Assessment of fracture risk and its applicationto screening for postmenopausal osteoporosis: synopsis of aWHO report. WHO Study Group. Osteoporos Int 1994;4:368–381.

5. Executive Summary of The Third Report of The National
Cholesterol Education Program (NCEP) Expert Panel on

Detection, Evaluation, And Treatment of High BloodCholesterol In Adults (Adult Treatment Panel III). JAMA2001;285:2486–2497.

6. Lloyd-Jones DM, Evans JC, Larson MG, O’Donnell CJ, LevyD. Differential impact of systolic and diastolic blood pressurelevel on JNC-VI staging. Joint National Committee onPrevention, Detection, Evaluation, and Treatment of HighBlood Pressure. Hypertension 1999;34:381–385.

7. The sixth report of the Joint National Committee onprevention, detection, evaluation, and treatment of highblood pressure. Arch Intern Med 1997;157:2413–2446.

8. American Psychiatric Association. Diagnostic and StatisticalManual of Mental Disorders. Washington, DC: AmericanPsychiatric Association, 2000.

9. Hanmer J, Lawrence WF, Anderson JP, Kaplan RM, FrybackDG. Report of nationally representative values for thenoninstitutionalized US adult population for 7 health-relatedquality-of-life scores. Med Decis Making 2006;26:391–400.

0. Kacmaz RO, Kempen JH, Newcomb CW, et al. Ocularinflammation in Behçet’s disease: incidence of ocular com-plications and of loss of visual acuity. Am J Ophthalmol2008; 146:828 – 836.

1. Thorne JE, Jabs DA, Peters GB, Hair D, Dunn JP, KempenJH. Birdshot retinochoroidopathy: ocular complications andvisual impairment. Am J Ophthalmol 2005;140:45–51.

2. Brown MM, Brown GC, Sharma S, Busbee B. Quality of lifeassociated with visual loss: a time tradeoff utility analysiscomparison with medical health states. Ophthalmology2003;110:1076 –1081.

3. Jaffe GJ, McCallum RM, Branchaud B, Skalak C, Butuner Z,Ashton P. Long-term follow-up results of a pilot trial of afluocinolone acetonide implant to treat posterior uveitis.Ophthalmology 2005;112:1192–1198.

4. BenEzra D, Cohen E, Chajek T, et al. Evaluation of conven-tional therapy versus cyclosporine A in Behcet’s syndrome.Transplant Proc 1988;20:136–143.

5. Thorne JE, Wittenberg S, Jabs DA, et al. Multifocal choroiditiswith panuveitis incidence of ocular complications and of loss ofvisual acuity. Ophthalmology 2006;113:2310–2316.

6. Yazici H, Pazarli H, Barnes CG, et al. A controlled trial ofazathioprine in Behcet’s syndrome. N Engl J Med 1990;322:281–285.

MULTICENTER UVEITIS STEROIDTREATMENT TRIAL PARTICIPATINGCLINICAL CENTERS (ORDERED BY

NUMBER OF PATIENTS RECRUITED)

● United Kingdom Institute of Ophthalmology, Lon-don, United Kingdom: Susan L. Lightman (Director),Lavnish Joshi, Patricio Pacheco, Timothy Stubbs, SimonR. Taylor, Hamish M. A. Towler. Former Member: KateEdwards.

● Johns Hopkins University, Baltimore, Maryland:James P. Dunn (Director), Jeff Boring, Diane M. Brown,Alison G. Livingston, Yavette Morton, Kisten Nolan,Terry Reed, Priscilla Soto, Michelle Tarver-Carr, Yue
Wang. Former Members: Marie-Lynn Belair, Stephen G.
RIAL 559

5

Bolton, Joseph B. Brodine, Lisa M. Brune, Anat Galor,Meera Kapoor, Sanjay Kedhar, Stephen Kim, Henry A.Leder, George B. Peters, Ricardo Stevenson.

● Vitreoretinal Consultants, Houston, Texas: Rosa Y.Kim (Director), Matthew S. Benz, David M. Brown,Eric Chen, Richard H. Fish, Shayla Hay, LauraShawver, Tien P. Wong. Former Members: RebeccaDe La Garza, Karin Mutz.

● University of Michigan, Ann Arbor, Michigan: SusanG. Elner (Director), Melissa Bergeron, Rebecca Brown,Linda Fournier, Julie R. Gothrup, Richard Hackel,Robert Prusak, Stephen J. Saxe. Former Members: ReneéBlosser, Carrie Chrisman-McClure, Deanna Sizemore.

● Washington University, St. Louis, Missouri: P. Ku-mar Rao (Director), Kevin J. Blinder, Rhonda Weeks.Former Members: Rajendra S. Apte, Pam Light, Gau-rav K. Shah, Russell VanGelder.

● Massachusetts Eye Research and Surgery Institute,Cambridge, Massachusetts: C. Stephen Foster (Di-rector), Linda Bruner, Angelica Contrero, KayleighFitzpatrick, David M. Hinkle, Jyothir Johnson,Danielle Marvell. Former Members: Sarah Acevedo,Fahd Anzaar, Tom Cesca, Karina Q. Lebron, ChandraMorgan, Nita Patel, Janet Sprague, Taygan Yilmaz.

● University of Southern California, Los Angeles,California: Narsing A. Rao (Director), LawrenceChong, Lupe Cisneros, Jackie Douglass, Dean Eliott,Amani Fawzi, Margaret Padilla. Former Members:Alexia Aguirre, Rahul Khurana, Jennifer Lim, RachelMead, Julie H. Tsai.

● Duke University, Durham, North Carolina: GlennJ. Jaffe (Director), Brenda Branchaud, Joyce Bryant,Michael Cusick, Justis P. Ehlers, Cindy Skalak. FormerMembers: Claxton Allen Baer, Sai H. Chavala, PouyaN. Dayani, Muge R. Kesen, Alex Melamud, Jawad A.Qureshi, Adrienne Williams Scott, Robert FrancisSee, Robert Keith Shuler.

● Emory University, Atlanta, Georgia: Sunil Srivas-tava (Director), Alicides Fernandes, Deborah Gibbs.Former Members: Daniel F. Martin.

● Royal Victorian Eye & Ear Hospital, East Mel-bourne, Australia: Richard J. Stawell (Director), LisaBreayley, Carly D’Sylva, Julie Ewing, Lauren Hodg-son, Ignatios Koukouras, Lyndell Lim, Cecilia Ling,Rachel McIntosh, Andrew Newton, Richard Small-wood, Ehud Zamir. Former Members: Nicola Hunt,Lisa Jones, Suzanne Williams.

● University of California at San Diego, La Jolla,California: William R. Freeman (Director), DenineE. Cochran, Igor Kozak, Megan E. Loughran, Luzan-dra Magana, Vivian Nguyen, Stephen F. Oster, Deb-bie Powell. Former Members: Nichole Brumley, TomClark, Joshua Hedaya, Tiara Kemper, Jacqueline M.LeMoine, Francesca Mojana, Victoria Morrison.

● University of Miami, Miami, Florida: Janet L. Davis
(Director), Thomas Albini, Jaclyn L. Kovach, Clau-

dia Teran. Former Members: Daniela Castaño, MacyHo, Susan Pineda, Richard C.-S. Lin, Jackie K.-D.Nguyen, Kimberly E. Stepien.

● University of Pennsylvania, Philadelphia, Pennsyl-vania: John H. Kempen (Director), Jim Berger, JoanDuPont, Sheri Grand, Dawn McCall, Albert M.Maguire, Laurel Weeney, Wei Xu. Former Members:Tim Hopkins, Monique McRay, Daniel Will.

● University of Utah, Salt Lake City, Utah: Albert T.Vitale (Director), Paul S. Bernstein, Bonnie Carl-strom, James Gilman, Sandra Hanseen, Paula Morris,Diana Ramirez, Kimberley Wegner.

● New York Eye & Ear Infirmary, New York, NewYork: Paul A. Latkany (Director), Jenny Gallardo,Monica Lorenzo-Latkany, Robert Masini, Susan Mo-rell, Ann Nour, Meredith Sanchez, Kate Steinberg.Former Members: Kenneth M. Boyd, Jacek Jarczynski,Mirjana McGrosky.

● University of California at San Francisco, San Fran-cisco, California: Todd P. Margolis (Director), NishaAcharya, David Clay, Claire M. Khouri, Salena Lee,Mary Lew, Jay Stewart, Ira G. Wong.

● National Eye Institute, Bethesda, Maryland: H.Nida Sen (Director), Denise Cunningham, ChloeGottlieb, Darryl Hayes, Dessie Koutsandreas, RichardMercer, Robert B. Nussenblatt, Dominic Obiyor,Patty R. Sherry, Gregory L. Short. Former Members:Allison Bamji, Hanna Coleman, Geetaniali Davuluri,Lisa Faia, Guy V. Jirawuthiworavong, Julie C. Lew,Cheryl H. Perry, Natalia Potapova, Eric Weichel,Keith J. Wroblewski, Steven Yeh.

● University of California at Los Angles, Los Angeles,California: Gary N. Holland (Director), Robert D.Almanzor, Jose T. Castellanos, Jean Pierre Hubschman,Ann K. Johiro, Michael A. Kapamajian, Ralph D.Levinson, Susan S. Ransome. Former Members: Chris-tine R. Gonzales, Anurag Gupta, Peter J. Kappel.

● Virginia Eye Consultants, Norfolk, Virginia: JohnD. Sheppard (Director), Nancy Crawford, Stephen V.Scoper, Jeanette Fernandez, Lisa Felix-Kent, RebeccaMarx, Joseph Webb. Former Members: R. DeniseCole, Gregory Schultz, Pamela Yeager, Jen Martin.

● Rush University Medical Center, Chicago, Illinois:Pauline T. Merrill (Director), Pam Hulvey, ElaineKernbauer, Scott Toennessen, Denise L. Voskuil-Marre. Former Members: Bruce Gaines, ChristinaGiannoulis, Heena S. Khan, Sarah J. Levine.

● Texas Retina Associates, Dallas, Texas: Robert C.Wang (Director), Hank Aguado, Sally Arceneaux,Jean Arnwine, Kimberly Cummings, Gary E. Fish,Susie Howden, Diana Jaramillo, Karin Mutz, BrendaSanchez. Former Members: David Callanan.

● University of Illinois at Chicago, Chicago, Illinois:Debra A. Goldstein (Director), Marcia Niec, MiselRamirez, Howard H. Tessler. Former Members: Cathe-
rine E. Crooke, Melody Huntley, Dimitry Pyatetsky.

V

● University of South Florida, Tampa, Florida: PeterReed Pavan (Director), JoAnn Leto, Lori Mayor,Maria Ortiz, Scott E. Pautler, Wyatt Saxon, JudySoto. Former Members: Dee Dee Szalay.

MULTICENTER UVEITIS STEROIDTREATMENT TRIAL COMMITTEES

● Executive Committee: Douglas A. Jabs (Study Chair-man), Michael A. Altaweel (Reading Center Director),Janet T. Holbrook (Coordinating Center Director),John H. Kempen (Study Vice-Chairman), NatalieKurinij (NEI Project Officer).

● Steering Committee: Douglas A. Jabs (Chair), RobertD. Almanzor, Michael M. Altaweel, Diane Brown,James P. Dunn, James Gilman, Janet T. Holbrook,Gary N. Holland, John H. Kempen, Rosa Y. Kim,Natalie Kurinij, Nancy Prusakowski, Jennifer Thorne.Former Members: Stephen G. Bolton, Lisa M. Brune,Tom Clark, Larry D. Hubbard, Daniel F. Martin,Robert B. Nussenblatt.

● Data, Safety and Monitoring Committee: Janet Wit-tes (Chair), Michael M. Altaweel*, William E. Bar-low, Marc Hochberg, Janet T. Holbrook*, Douglas A.Jabs*, Natalie Kurinij, Alice T. Lyon, Alan G. Pales-tine, Lee S. Simon, Harmon Smith. Former Members:James T. Rosenbaum. *Nonvoting members.

● Surgical Quality Assurance Committee: John H.Kempen (Chair), James P. Dunn, Glenn J. Jaffe.Former Members: Daniel F. Martin.

● Medical Therapy Quality Assurance Committee:Jennifer E. Thorne (Chair), Nisha Acharya, DouglasA. Jabs, John H. Kempen, Paul A. Latkany, Albert T.Vitale. Former Members: Robert B. Nussenblatt, Rus-sell Van Gelder.

● Visual Function Quality Assurance Committee:Robert D. Almanzor (Chair), Judith Alexander, Jef-frey A. Boring, Deborah Gibbs, Salena Lee, JenniferE. Thorne (Advisor). Former Members: Wai Ping Ng.

● Glaucoma Committee: David S. Friedman (Chair),Anna Adler, Alyce Burke, Janet T. Holbrook, JoanneKatz, John H. Kempen, Nancy Prusakowski, SusanReed, Jennifer E. Thorne. Former Members: NicholasCohen, Sanjukta Modak, Wai Ping Ng.

● Statistical Analysis Committee: Elizabeth Sugar
(Chair), Lea Drye, Kevin Frick, Janet T. Holbrook,

Joanne Katz, Thomas Louis, Mark L. Van Natta.Former Members: Sanjukta Modak, David Shade.

MULTICENTER UVEITIS STEROIDTREATMENT TRIAL RESOURCE CENTERS

● Chairman’s Office, Mount Sinai School of Medicine,New York, New York: Douglas A. Jabs (Chairman),Melissa A. Nieves, Karen Pascual, Jill S. Slutsky.Former Members: Yasmin Hilal. University of Pennsyl-vania, Philadelphia, Pennsylvania: John H. Kempen(Vice-Chairman). Johns Hopkins University, Balti-more, Maryland: Former Members: Colby Glomp, MariaStevens.

● Coordinating Center, Johns Hopkins UniversityBloomberg School of Public Health, Baltimore,Maryland: Janet Holbrook (Director), Anna Adler,Judith Alexander, Jeff Boring, Alyce Burke, PaulChen, Karen Collins, John Dodge, Lea Drye, Cath-leen Ewing, Kevin Frick, David Friedman, RosettaJackson, Joanne Katz, Andrea Lears, Hope Living-ston, Thomas Louis, Curtis Meinert, Jill Meinert,Deborah Nowakowski, Nancy Prusakowski, DaveShade, Virginia Shen, Rochelle Smith, ElizabethSugar, Jennifer Thorne, Ada Tieman, Mark VanNatta, Richard Zheng. Former Members: NicholasCohen, Sanjukta Modak, Wai Ping Ng, CharlesShiflett.

● Fundus Photograph Reading Center, University ofWisconsin, Madison, Wisconsin: Michael A. Al-taweel (Director), Wendy K. Benz, Geoffrey Cham-bers, Debra J. Christianson, Amitha Domalpally,Jacquelyn Freund, Vonnie Gama, Sapna Gangaputra,Anne Goulding, Kathleen E. Glander, Jeffrey M.Joyce, Christina N. Kruse, Susan Reed, Amy Remm,James L. Reimers, Lori La See, Ruth A. Susman,Dennis Thayer, Erika Treichel, Kelly J. Warren,Sheila M. Watson, James K. White, Tara Wilhelm-son. Former Members: Margaret A. Fleischli, DennisHafford, Susan E. Harris, Larry D. Hubbard, KristineA. Johnson, Dawn J. Myers, Lauren Nagle, Gwyn E.Padden-Lechten, Alyson Pohlman, Peggy Sivesind,Mary K. Webster, Tong Wu.

● National Eye Institute, Bethesda, Maryland: Natalie
Kurinij.
RIAL 561

5

SUPPLEMENTAL TEXT: DESIGN DETAILSFOR THE MULTICENTER UVEITIS

STEROID TREATMENT TRIAL

● Eligibility: Eligibility criteria are given in Supplemen-tal Table 1. Eligible patients had active or recentlyactive (within 60 days) intermediate uveitis, posterioruveitis, or panuveitis—the forms of uveitis for whichthe fluocinolone acetonide implant therapy is indi-cated. These are forms of uveitis for which systemictherapy commonly is indicated.22 Because of safetyconsiderations, patients who were pregnant or breast-feeding, those younger than 13 years, those withuncontrolled glaucoma or diabetes, those with aller-gies to required study medications, those with scleritisor ocular toxoplasmosis, and those with known immu-nodeficiency syndromes were excluded. Because someepisodes of substantially elevated intraocular pressure(IOP) were expected to occur in the implant arm,28

patients with advanced glaucoma who might experi-ence irreversible damage from such an IOP spike alsowere excluded. Patients with associated systemic in-flammatory diseases likely to require systemic therapyfor a nonocular indication were excluded on thegrounds that systemic therapy would be preferable toimplant therapy. Patients were required to have avisual acuity of hand movements or better in at least1 eye with uveitis, because eyes with visual acuityworse than this level may not benefit enough fromtreatment to justify the potential risks of the alterna-tive therapies.

● Sample Size: The primary analysis for the MUST trialis to compare the mean eye-specific change in loga-rithm of the minimum angle of resolution visualacuity from baseline to follow-up at 2 years betweenthe randomly assigned treatment groups. Meanchange in visual acuity was chosen as the primaryoutcome because of an increasing consensus amongthe ophthalmology clinical trials community in favorof designing clinical trials to detect a difference in themean visual acuity rather than the proportion achiev-ing a given number of lines of visual acuity lost orgained 29 Based on considerations that previous clin-ical trials demonstrating a difference of approximately7.5 letters (equivalent to 1.5 lines on a logarithmicvisual acuity chart)30 in mean visual acuity betweentreatment groups have resulted in the widespreadadoption of the new treatments,31–34 the MUST trialis designed to detect at least a 7.5-letter difference invisual acuity between groups, with at least 90% power.

For the sample size calculation, the number ofindependent observations needed (assuming only 1eye from each patient) was adjusted to take intoaccount the percent of patients expected to havebilateral disease at baseline (estimated as 67%) and
the expected between-eye correlation in acuity
AMERICAN JOURNAL OF61.e1

changes for involved eyes (conservatively estimatedto be 0.4).35 We estimated the standard deviation ofchange in visual acuity to be 16 letters (3.2 lines).Based on these assumptions, and allowing for a cross-

SUPPLEMENTAL TABLE 1. Eligibility Criteria for the MulticenterUveitis Steroid Treatment (MUST) Trial

Inclusion criteria

1. Age 13 years or older

2. Diagnosis of noninfectious intermediate uveitis, posterior

uveitis, or panuveitis by a MUST-certified ophthalmologist

3. Active uveitis of a degree for which systemic

corticosteroid therapy is indicated in the judgment of a

MUST-certified ophthalmologist or such uveitis active within

the last 60 days as determined either by examination by a

MUST-certified ophthalmologist or by review of ophthalmic

medical records by a MUST-certified ophthalmologist

4. Uveitis with or without an associated systemic disease is

acceptable; however, the systemic disease must not be

sufficiently active that it dictates therapy with oral

corticosteroids or immunosuppressive agents at the time of

study entry

5. Best-corrected visual acuity of hand movements or better

in at least 1 eye with uveitis

6. Baseline intraocular pressure of 24 mm Hg or less in all

eyes with uveitis

7. Collection of required baseline data within 10 days before

randomization

8. Signed informed consent

Exclusion criteria

1. Use of a fluocinolone acetonide implant within the last 3

years

2. Diabetes mellitus that is inadequately controlled,

according to best medical judgment

3. A known allergy to a required study medication

4. Uncontrolled glaucoma

5. Advanced glaucomatous optic nerve injury meeting the

following criteria: (1) for patients able to undertake a

Humphrey visual field analysis, depression of 2 points or

more within 10 degrees of fixation by at least 10 dB, mean

deviation worse than –15 dB, or both; (2) for patients unable

to undertake a Humphrey visual field analysis, vertical cup-

to-disc ration � 0.9

6. A history of scleritis (because of concerns regarding the

potential for scleral melting with local corticosteroid therapy)

7. Presence of an ocular toxoplasmosis scar

8. Pregnancy

9. Current breastfeeding

10. Known human immunodeficiency virus infection or other

immunodeficiency disease for which corticosteroid therapy

would be contraindicated according to best medical

judgment

11. Patients for whom participation in the trial would

constitute a risk exceeding the potential benefits of study

participation, in the judgment of the treating physician

12. Medical problems or drug or alcohol dependence

problems sufficient to prevent adherence to treatment and

study procedures

over rate of 10%, a sample size of 250 (125 patients


S

V

per group) was projected to have a power of 91% witha 2-sided type 1 error rate of 5% to detect at least a7.5-letter difference in visual acuity between groups.

● Randomization: Uveitis may affect one or both eyes.Implant therapy is given to one eye at a time, butsystemic treatment affects both eyes simultaneously.Because one of the alternative treatments is systemi-cally administered, and because the goal of the trialwas to compare alternative treatment strategies at thelevel of the patient (including quality-of-life analysis),patients (rather than eyes) were randomized to thealternative treatments. Random treatment assign-ments (allocation ratio, 1:1) were stratified by thetype of uveitis (intermediate uveitis vs posterior uve-itis or panuveitis) because visual acuity outcomeslikely differ between these disease categories8 and alsoby clinical center to balance any potential effects ofvarying protocol implementation or population differ-ences at the clinics across treatment arms (see Sup-

UPPLEMENTAL FIGURE 1. Randomization schema used f

plemental Figure 1). The Rallocate procedure36 in


Stata software version 9 (StataCorp, College Station,Texas, USA) was used to generate permuted blockrandomization tables with variable block sizes for eachstratum level. The treatment assignment procedurewas implemented using a web-based system thatreturned treatment assignments in real time aftercompletion of eligibility checks. Randomization wasconducted after obtaining informed consent and en-tering required baseline data. Before randomization,institutional review board approval was obtained byeach participating center. The MUST trial is regis-tered at clinicaltrials.gov (identifier NCT00132691).

● Implant Treatment Protocol: Implant therapy (seeSupplemental Figure 2) consisted of treatment withthe commercially available fluocinolone acetonide0.59-mg implant (Retisert; Bausch & Lomb, Inc,Rochester, New York, USA) in each eye with uveitisof sufficient severity to justify treatment with systemiccorticosteroids. After quieting of the anterior cham-

e Multicenter Uveitis Steroid Treatment Trial.
or th
ber to a grade of less than 1� anterior chamber cells7

RIAL 561.e2

http://clinicaltrials.gov

ST

5

(using topical and/or periocular corticosteroids,and/or systemic corticosteroid therapy when the first 2approaches are judged insufficient), the protocolcalled for surgical implantation of an implant by aMUST-certified surgeon using a standard technique25

under conventional anesthesia within 28 days ofrandomization in the first eye. When applicable, thesecond eye was implanted within 28 days thereafterusing the same approach. For second eyes that ini-tially did not meet criteria for implantation, butsubsequently worsened to the point where systemiccorticosteroid therapy would be justified, implanta-tion was performed at the time the eye met suchcriteria in the same manner. Eyes in which a reacti-vation of uveitis developed after implantation werereimplanted (regardless of when the reactivation oc-curred with respect to the original surgery); eyes inwhich a second reactivation developed within 18months after reimplantation were managed accordingto best medical judgment. Eyes also were managedaccording to best medical judgment when any of thefollowing occurred: (1) the patient required systemicprednisone or immunosuppressive therapy for sys-

UPPLEMENTAL FIGURE 2. Fluocinolone acetonide implanrial.

temic indications; (2) toxicity of implant therapy


occurred; or (3) inflammation was not controlledinitially after implantation.

Eyes initially receiving systemic corticosteroid orimmunosuppressive therapy were tapered off of suchtherapy after implant placement—generally withinseveral weeks, but slowly enough to avoid complica-tions related to adrenal suppression from corticoste-roid therapy. Eyes for which additional intraocularsurgery—such as cataract surgery—was needed atbaseline (and so noted before randomization on anenrollment form) were permitted under the protocolto have the other intraocular surgery combined withimplant surgery. However, vitrectomy surgery to clearvitreous opacities was not performed on the groundsthat implant therapy was expected to clear suchopacities.

● Systemic Treatment Protocol: Systemic therapy con-sisted of oral corticosteroid therapy—supplementedwhen indicated by immunosuppression—in accor-dance with expert panel recommendations.22 Patientswith active uveitis at baseline were started on pred-nisone 1 mg/kg per day up to a maximum adult oraldose of 60 mg/day (see Supplemental Figure 3), which

tment algorithm for the Multicenter Uveitis Steroid Treatment
t trea
was continued until the uveitis was controlled, or


S

V

until the patient had been receiving this dose ofprednisone for 4 weeks.

After suppression of uveitis was achieved, oralcorticosteroids were tapered, according to the follow-ing guidelines: (1) 10 mg/day decrements each weekto 40 mg daily; (2) 5 mg/day decrements each week to20 mg daily; (3) 2.5 mg/day decrements every week to10 mg daily; (4) 2.5 mg/day decrements every 2 weeksuntil off prednisone. Patients whose uveitis alreadyhad been controlled within the 60 days before enroll-ment started tapering from their current prednisonedose and tapered their prednisone dose in the samemanner.

Immunosuppressive therapy was used when: (1)uveitis was not controlled on corticosteroids alone;(2) maintenance of suppression of uveitis required aprednisone dose of more than 10 mg/day or a dosesufficient to cause intolerable side effects; (3) forspecific diseases known to respond poorly absentinitial immunosuppressive therapy, or (4) a combina-tion thereof.22 The choice of which immunosuppres-sive drug to use was not dictated by the trial to permit

UPPLEMENTAL FIGURE 3. Systemic treatment algorithm

selection of the agent with the best side-effect profile


for an individual patient. However, the immunosup-pressive treatment(s) selected was administered inaccordance with expert panel guidelines, includinglaboratory monitoring for potential toxicities.22 Be-cause of the potential additive therapeutic effect ofantimetabolites and T-cell inhibitors, a second immu-nosuppressive drug of the alternative class was recom-mended for patients not meeting treatment goals with1 immunosuppressive drug. Alkylating agents (cyclo-phosphamide and chlorambucil), although given inconjunction with prednisone, were not to be com-bined with other immunosuppressive drugs because ofthe potential for substantially increased toxicity withsuch combinations.

When needed (i.e., for patients who had flare-upsof uveitis while still taking oral corticosteroids), ta-pering of prednisone was repeated iteratively: the doseof prednisone was raised to a level sufficient to controlthe inflammation each time a flare-up of inflammationoccurred and then tapered until either uveitis wascontrolled without prednisone or until a minimalsuppressive dose of prednisone (with or without con-

he Multicenter Uveitis Steroid Treatment Trial.
for t
comitant immunosuppressive therapy) was identified

RIAL 561.e4

5

that was effective for maintaining suppression ofinflammation and had an acceptable side-effect pro-file. If 2 iterations of tapering in the presence of animmunosuppressive agent failed to achieve a tolerableminimal suppressive prednisone dose, treatment wasto be directed by best medical judgment. Likewise, ifthe suppressive dose proved to be too high to have afavorable side-effect profile when an alkylating agentwas used as the initial immunosuppressive agent,treatment thereafter was directed by best medicaljudgment.

After 6 months of successful suppression of uveitisusing systemic therapy, an attempt was made to taperthe prednisone further with monthly decrements fol-lowing the steps above. If such tapering resulted inreactivation of uveitis, the prednisone dose was raisedsufficiently to control the uveitis, and then taperedback to the suppressive dose, after which the patientwas to be maintained on suppressive systemic therapyat the suppressive dose for 12 months before furthertapering attempts.

While receiving oral corticosteroids, patientsreceived calcium and vitamin D supplementationto minimize the incidence of osteoporosis.37 Pa-tients in whom osteoporosis developed were re-ferred for treatment, according to guidelines of theAmerican College of Rheumatology.37 Patients inwhom hypertension or diabetes developed as aresult of systemically administered drugs also werereferred for appropriate management.

● Ancillary Treatment: Patients in both treatmentarms were allowed topical corticosteroid therapy asindicated by best medical judgment, given accordingto study guidelines. Given the severe nature of theuveitis of patients enrolled in this trial, and given thelimited penetration of topical corticosteroids past theanterior chamber to the primary site of inflammationin these patients,22 these drops are unlikely to have asubstantial effect on the major causes of visual loss.

The protocol also permitted the use of periocularcorticosteroid injections (triamcinolone acetonide, 40mg) when indicated for the following circumstances:(1) for patients randomized to implant therapy, peri-ocular corticosteroids could be given before implan-tation to suppress intraocular inflammation rapidlybefore surgery; or (2) for persistent or refractorymacular edema in either group. Periocular corticoste-roids could be given either via the posterior sub-Tenon route or via the orbital floor route. Ancillaryintravitreous corticosteroid therapy also could begiven for the same indications when periocular corti-costeroid injections were likely to fail or had failed.The protocol directed that periocular and intraocularcorticosteroid injections should not be given repeti-
tively on a routine basis as a primary form of therapy.

● Data Collection and Follow-up: Patients enrolled inthe MUST trial have visits for data collection atbaseline, at 1 and 3 months after randomization, andthen quarterly until the end of the trial (see Supple-mental Table 2). Other than the visit 1 month afterrandomization (which is excluded from the efficacyanalysis), study visits should not be conducted within1 month of ocular surgery (e.g., implant surgery) toavoid biasing visual acuity measurements downwardin recently operated eyes, which are expected to havereduced visual acuity transiently in the early postop-erative period. Patients generally receive more fre-quent visits for clinical care. Data regarding treatmentchanges and events occurring between study visits arecollected at the subsequent study visit, except foradverse events that meet requirements for expedited

SUPPLEMENTAL TABLE 2. Data Collection Schedule for theMulticenter Uveitis Steroid Treatment (MUST) Trial*

Target week 0 4 13 26 39 52 65 78 91 104

Visit ID BL F1 F2 F3 F4 F5 F6a F7a F8a F9a

Visual acuity x x x x x x x x x x

Goldmann tonometry x x x x x x x x x x

Ophthalmic

examination x x x x x x x x x x

Gonioscopy x x x

Color slit-lamp

lens photographs x x x x

Color fundus photographs

(disc, macula) x x x x x

Fluorescein

angiography x x x x x

Optical coherence

tomography x x x x x

Humphrey 24-2

perimetry x x x

Medical history x x x x x x x x x x

Blood pressure x x x x x x x x x x

Height x x x

Weight x x x x x x x x x x

EuroQol x x x x x x x x x

SF-36, NEI-VFQ x x x x x

Complete blood count x x x

Comprehensive

chemistry panel x x x

Casual plasma glucose x x x x x x x

Fasting plasma glucose x x x

Fasting lipid analysis x x x

Serum pregnancy test xb

Bone densitometry x x x

ID � identification; NEI-VFQ � National Eye Institute Visual

Function Questionnaire; SF � Short Form 36-item question-

naire.aThe F6, F7, F8, and F9 data collection schedules repeat every

12 months until the common study closeout.bFor women who were able to become pregnant.

reporting.


V

● Outcomes: Because the primary goal of therapy foruveitis is to preserve vision, the primary outcome bywhich success is judged in the MUST trial is best-corrected visual acuity. Other outcomes evaluatedpertain to control of intraocular inflammation, theoccurrence of ocular complications of uveitis or oftherapy, the occurrence of systemic complications oftherapy, and self-reported quality of life (QOL).

Masking is applied to the determination of visualacuity, those outcomes based on photographic read-ing, and diagnosis of glaucoma by the GlaucomaOutcomes Committee. Because sham surgery is notperformed, masking during ascertainment of the otheroutcomes is not feasible.

The primary outcome—best-corrected visual acu-ity—is measured every 3 months at study visits understandardized lighting conditions by certified studyexaminers with best refractive correction in place,using standard logarithmic (Early Treatment DiabeticRetinopathy Study [ETDRS]) visual acuity charts.38

At the 6-month visit (by which time the impact ofimplant surgery rarely would be visible in a darkenedroom) and subsequently, these measurements aremade by examiners unaware of treatment assignment.Eye-specific outcomes such as visual acuity are fol-lowed as by-eye variables, an approach that providesmore information than a by-patient analysis andavoids the problem of having to select a study eye. Inthe ETDRS system, 15 letters is equal to a change of3 lines of visual acuity, for example, from 20/40 to20/20 or vice versa, a halving or doubling of the visualangle30; a change of 7.5 letters corresponds to a 25%decrease or 50% increase in the visual angle.

Secondary outcomes for the MUST trial includeevaluation of inflammatory and morphologic charac-teristics of enrolled eyes, which are measured longi-tudinally throughout the course of the study (seeAppendix, below). Assessment is performed by un-masked ophthalmologist investigators and by maskedgrading of imaging studies. The methods used and therepresentative features evaluated by them include:lens photography (cataract), fundus photography(macular edema, epiretinal membranes, optic nervemorphologic features, chorioretinal lesions, vascularocclusions), Stratus optical coherence tomography(macular edema, vitreoretinal interface abnormalitiesincluding epiretinal membrane), and fluorescein an-giography (macular edema, vascular leakage, perfusionabnormalities, choroidal neovascularization). Health-related quality-of-life data also are assessed, includinghealth utility as measured by the EuroQol 5-dimen-sion and Visual Analog Scale scores,39,40 generalhealth-related QOL as measured by the Short Form36-item questionnaire,41,42 and vision-related QOL asmeasured by the 25-item National Eye Institute Vi-
sual Function Questionnaire.43

● Data and Safety Monitoring: The Data and SafetyMonitoring Committee (DSMC)—an independentboard appointed by and advisory to the studyfunding agency (the National Eye Institute)—isresponsible for ongoing review of efficacy and safetydata, policy and ethical issues, and study perfor-mance. Members of the DSMC include 2 ophthal-mologists, 2 rheumatologists, 2 biostatisticians, andan ethicist. The DSMC meets twice yearly with theoption for ad hoc meetings if required. An O’BrienFleming spending function was used to developguidelines for stopping the trial. The DSMC hasplanned for 3 formal analyses of the efficacy data,after 50%, 75%, and 100% of participants havebeen followed up for 2 years. These guidelines callfor stopping the trial early only if large differencesbetween groups in the mean change in visual acuityfrom baseline—12.5 to 15 letters or more—areobserved at the interim analyses. One of the phy-sician members of the DSMC reviews adverse eventdata between DSMC meetings, which are reviewedfirst by a Coordinating Center ophthalmologist andsafety officer.

● Study Organization: The organizational structureof the MUST Trial Research Group consists of 3resource centers—a Chairman’s Office, Coordinat-ing Center, and Reading Center—and 23 clinicalcenters, funded via U10 cooperative agreementswith the National Eye Institute. Surgical treatmentin the MUST trial is administered by surgeons whohave been certified in the protocol procedures by aSurgical Quality Assurance Committee, either bylive observation of surgery or by review of anunedited video of surgery performed on a human oranimal eye. Adherence to the systemic therapyprotocol is assessed at approximately annual sitevisits by expert clinicians serving on the MedicalTherapy Quality Assurance Committee. A VisualFunction Quality Assurance Committee overseesthe training, monitoring, and certification of allvisual function technicians in the trial, ensuringthe quality of visual acuity measurements made inthe trial. Other aspects of protocol enforcementand clinic performance monitoring are conductedby Coordinating Center staff through approxi-mately annual site visits and remote surveillance ofstudy activities. Day-to-day study management isconducted by an Executive Committee consistingof the Study Chair and Vice-Chair, the Coordinat-ing Center Director, the Reading Center Director,and a representative of the National Eye Institute.A Steering Committee consisting of the ExecutiveCommittee and a majority of elected investigatorsis charged with major policy decisions regarding
study implementation and management.
RIAL 561.e6

5

APPENDIX: OUTCOME DEFINITIONSAND PLANNED ANALYSES FOR THE

MULTICENTER UVEITIS STEROIDTREATMENT TRIAL

● Visual Acuity: Best-corrected visual acuity score ismeasured at every study visit under standardizedlighting conditions by certified study examinerswith best refractive correction in place, using log-arithmic (ETDRS) visual acuity charts38 (as de-scribed above). The design outcome is based onchange by 2 years’ follow-up in visual acuity (num-ber of letters read) from baseline in eyes meetingtrial inclusion criteria at baseline.

In addition to the primary analysis, visual acuitydata are to be analyzed in several different ways: (1)calculating the proportion with low vision (worsethan 20/40) and legal blindness (20/200 or worse, orvisual field subtending an angle of 20 degrees or less)over time and (2) assessment of gain or loss of visionwith respect to baseline ETDRS visual acuity on anordinal scale over time: loss of 30 letters or more, lossof 15 to 29 letters, loss of 5 to 14 letters, within 4letters of baseline visual acuity, gain of 5 to 14 letters,gain of 15 to 29 letters, or gain of 30 letters or more.These outcomes are evaluated both by patient (bothfor the worse eye and for the better eye, only eyes withuveitis) and by eye (only eyes with uveitis).

● Intraocular Inflammation: Intraocular inflammationis assessed at every visit by clinical examination.Inflammatory status is based primarily on anteriorchamber cells and vitreous haze, each of which aregraded on ordinal scales.7

Uveitis is judged to be inactive when anteriorchamber cells are graded as 0 or 0.5� and vitreoushaze is graded as 0; gradings of 1� or higher on eitherscale are interpreted as indicated that the uveitis isactive. The examining clinician also is asked to gradethe uveitis as active, mildly active, or inactive basedon her or his own judgment. For rare uveitic condi-tions, such as serpiginous choroidopathy, in whichinflammation may be limited to the choroid andretina, clinical judgment of activity is substituted forthe composite score based on anterior chamber cellsand vitreous haze.

Clinical examination data regarding consequencesof intraocular inflammation also are noted at everyvisit, including anterior chamber flare (graded over anordinal scale of 0, 1�, 2�, 3�, 4�),44 and thepresence or absence of posterior synechiae, preretinalneovascularization, choroidal neovascularization,epiretinal membrane or macular pucker, macularedema, optic nerve swelling, pars plana exudation,and retinal detachment. The presence of peripheralanterior synechiae and angle closure are documented
by annual gonioscopy. If present, the extent (in

degrees) of posterior synechiae and of peripheralanterior synechiae or angle closure are recorded.Cataract grading is performed by the Reading Centerfrom slit-lamp lens photographs and red reflex photo-graphs of the lens plane (see below).

● Retinal Morphologic Features: To document theocular abnormalities of uveitis and its complications,MUST trial clinical centers obtain fundus photo-graphs, fluorescein angiograms, optical coherence to-mography (OCT) scans, and lens photographs usingstandardized protocols generated by the Reading Cen-ter at the University of Wisconsin, Madison. Imagesare graded at the Reading Center according to stan-dardized protocols for status of and change in ocularmorphologic features. The resultant data are summa-rized into analysis variables and are transmitted to theCoordinating Center.

Retinal morphologic features often are alteredpathologically as a result of intermediate uveitis,posterior uveitis, or panuveitis. A common morpho-logic change, and the most common cause of visionloss in uveitis, is the occurrence of macular edema.Therefore, measurement of macular edema is a pri-mary goal of ocular image analysis. The presenceversus absence of macular edema is determined byfluorescein angiography (FA) and OCT. When edemais present, severity of fluorescein leakage and ofcystoid spaces is assessed using a modification of theETDRS protocol. The area of retinal thickening andof cystoid spaces found on concurrently graded FAand color fundus photographs also is noted. Theextent of fluorescein leakage is noted in disc areas.Central macular thickness is measured by OCT, as isthe volume within the ETDRS circle. Other OCTfeatures consistent with macular edema, such as cys-toid spaces, are graded in terms of presence andextent. These morphologic measures are expected tocorrelate with visual acuity.

Additional retinal morphologic features (with theimage from which it is derived in parentheses) beinggraded include vascular nonperfusion (FA), choroidalneovascularization (FA), preretinal neovasculariza-tion (color fundus photographs, FA), retinal detach-ment (fundus photo, FA), optic disc edema (fundusphoto, FA), glaucomatous optic disc changes (en-larged cup-to-disc ratio, notching or thinning of thedisc rim, disc pallor, and disc hemorrhage; stereofundus photographs), vitreoretinal interface abnor-malities (such as epiretinal membrane) and theirrelationship to macular edema (OCT and fundusphotography), and chorioretinal lesions (fundus pho-tography, FA with sweep fields). Morphologic featuresin most cases are measured as dichotomous variables(present or absent), and if present, their extent is
graded.

V

● Quality of Life: As an ocular disease, uveitis isexpected to affect QOL primarily through its impacton vision. However, systemic therapy and associatedsystemic disease may have effects on perceived generalhealth-related QOL and health utility. In addition,because the effect of vision loss on general well-beingis profound, the effect of uveitis on general well-beingmay be substantial despite localization of disease tothe eye.

The National Eye Institute Visual Function Ques-tionnaire is an instrument designed to be responsiveto the effects of eye diseases on QOL, particularlyaddressing vision-related QOL. It is based on aspectsof vision and ocular symptoms. The National EyeInstitute Visual Function Questionnaire has beendemonstrated to give valid and reliable results acrossa variety of ocular conditions43,45 and is a widely usedinstrument for measurement of vision-related QOL.The Medical Outcomes Study Short Form HealthSurvey is an instrument measuring general health-related QOL that has been demonstrated to fulfillrigorous validity and reliability standards.41,42 It is theinstrument most commonly used to measure generalhealth-related QOL in clinical studies.46 The Euro-Qol is a generic health index widely used in clinicalresearch to calculate a health utility,39 the measure-ment of which has been recommended for all clinicaltrials by the Panel on Cost Effectiveness in Healthand Medicine.27 A health utility score is useful as asummary indicator of a patient’s self-perceived generalhealth status, which can be used to compare theimpact of uveitis (and its treatment) relative to theimpact of other diseases. Health utility also willprovide the primary effectiveness measure for cost-effectiveness analysis. The MUST trial QOL batteryconsists of these 3 instruments, all available in bothEnglish and Spanish.

● Potential Complications of Study Treatment: Bothrandomized treatment approaches are expected toresult in side effects. The relative occurrence of suchevents are key results of the study, which will play animportant part in interpretation of the overall results.Specific complications to be evaluated prospectivelyare described below. An adverse events reportingsystem also is used to ascertain the occurrence of bothanticipated and unanticipated adverse events.

Ocular side effects of treatment anticipated tooccur with substantial frequency in the MUST trialinclude cataract and glaucoma, which are commonside effects both of uveitis and of corticosteroidtherapy. Occurrence of complications of intraocularsurgery—such as endophthalmitis, vitreous hemor-rhage, and implant extrusion—most likely will beuncommon, but are important. Because ocular adverse
events will be important for comparison of the alter-

native treatments, detailed measurements are beingobtained.

Systemic therapy with corticosteroids or immuno-suppressive (corticosteroid-sparing) agents can be as-sociated with adverse systemic side effects. Becausethe occurrence of systemic side effects with systemictherapy is an important element of the rationale forusing the fluocinolone acetonide implant, detaileddata are collected on anticipated potential complica-tions.

● Elevated Intraocular Pressure and Glaucoma: IOPelevation and glaucoma may be either a complicationof uveitis itself or may be corticosteroid induced. IOPis measured using the protocol followed by the Col-laborative Initial Glaucoma Treatment Study, beforegonioscopy or dilation.47 A certified technician oper-ates a Goldmann applanation tonometer, while a secondobserver records the findings. The second observer thenresets the tonometer, and the measurement is repeated.If the 2 measures differ by 2 mm Hg or more, then a thirdmeasurement is made to adjudicate the discrepancy.Tono-Pen (Mentor Ophthalmics, Norwell, Massachu-setts, USA) IOP measurement is acceptable when Gold-mann tonometry cannot be performed. IOP measurementsof more than 21 mm Hg are considered to be elevated, andmeasurements higher than 30 mm Hg are considered tobe highly elevated. Elevations of IOP and the outcomesof these events are monitored.

Diagnosis of glaucoma is made through the integra-tion of several inputs. Stereo optic nerve photographsare graded by the Reading Center. Humphrey visualfield testing is used to screen for glaucomatous visualfield defects through use of the Glaucoma HemifieldTest.48 Unreliable or abnormal results are evaluatedfurther by repeat testing. Eyes for which photographic,visual field, or clinical evaluation, or a combinationthereof suggests glaucoma may be present undergo amasked review by a Glaucoma Outcomes Committeeled by a glaucoma specialist or ophthalmic epidemi-ologist. Such eyes are followed up to make or excludethe diagnosis of glaucomatous optic nerve damage,applying the methodology used by the Baltimore EyeSurvey.49

● Cataract: The occurrence and progression of cataractis evaluated based on red reflex photographs andslit-lamp photographs, graded using an adaptation ofthe Age-Related Eye Diseases Study cataract gradingprotocol.50 This system grades posterior subcapsularand cortical cataracts based on the percentage of lensinvolvement. Nuclear sclerotic cataracts are gradedon the basis of optical density, using a numerical scalegenerated by the grader’s interpolation of the opticaldensity of the nuclear sclerosis between 4 anchorstandard photographs (1.0, 2.0, 3.0, and 4.0).

Pseudophakia and aphakia are noted based on
clinical examination or photography—such eyes are
RIAL 561.e8

5

excluded from cataract grading. The occurrence ofcataract surgery and of yttrium–aluminum–garnet la-ser capsulotomy or surgical capsulotomy during fol-low-up is noted. Whether sufficient posterior capsularopacification exists to warrant yttrium–aluminum–garnet capsulotomy, based on clinical examinationalone, is noted at each visit.

● Other Ocular Complications: Other ocular compli-cations that are expected to be uncommon, but areimportant with respect to evaluation of the alterna-tive treatments, include: retinal detachment, vitreoushemorrhage, endophthalmitis, and implant extrusion.

● Hyperglycemia and Diabetes Mellitus: Fasting plasmaglucose levels are measured at baseline, 3 months, 12months, and annually thereafter. Casual plasma glu-cose levels are measured at all other visits. Results areinterpreted using the thresholds recommended by theExpert Committee on the Diagnosis and Classi-fication of Diabetes Mellitus (American DiabetesAssociation) for hyperglycemia and diabetes.51 Treat-ment-induced diabetes mellitus is diagnosed when asubject who is euglycemic at baseline: is observed tohave diabetic-level hyperglycemia during follow-up,has medical records demonstrating that the subjectwas diagnosed with diabetes mellitus (with or withoutspecification of treatment relationship) during follow-up, is started on therapy for diabetes mellitus, or acombination thereof. The number of oral hypoglyce-mic medications and use of insulin at each visit areapplied to construct an ordinal scale evaluating thedegree of difficulty in controlling diabetes mellitus:0 � no hypoglycemic therapy; 1 � one oral hypogly-cemic agent; 2 � 2 or more oral hypoglycemic agents;and 3 � daily use of insulin. The baseline value onthis scale is subtracted from the maximum value ofthis scale during follow-up to indicate the change inglycemic status with therapy, allowing this outcometo be evaluated both in subjects who at baseline werenondiabetic and those who were diabetic but notrequiring insulin.

● Osteoporosis: Subjects are screened for osteoporosisusing dual-emission radiograph absorptiometry scan-ning of the spine (L2 to L4) and of the left femoralneck to measure bone mineral density at baseline andannually thereafter. Bone mineral density, the singlebest predictor of osteoporotic fracture risk,52 is wors-ened substantially by corticosteroid therapy.53 Gener-alizing the World Health Organization Study Grouppostmenopausal osteoporosis guidelines54 to the set-ting of corticosteroid-induced bone loss—followingthe adaptation of the American College of Rheuma-tology37—the MUST trial defines osteopenia as a Tscore between �1 and �2.49 inclusive at the spine orfemoral neck (whichever is worse); osteoporosis isdefined as a T score of �2.5 or worse at the spine,
femoral neck, or both. Fracture events, likely to be

infrequent during the follow-up period of the trial,also are noted when confirmed by medical records.

● Hyperlipidemia: Fasting lipid panels—including totalcholesterol, low-density lipoprotein (LDL) choles-terol, high-density lipoprotein cholesterol, and tri-glyceride levels—are obtained at baseline, the3-month follow-up, the 12-month follow-up, andannually thereafter. Because elevated LDL cholesterolis a major cause of coronary heart disease and is theprimary target of cholesterol-lowering therapy,55

changes in LDL cholesterol from baseline are evalu-ated as the main indicator of treatment effects onhyperlipidemia. Based on the categories of the Na-tional Cholesterol Education Program Expert Panel,55

LDL cholesterol levels are categorized ordinally asfollows: � 100 mg/dL, 100 to 129 mg/dL, 130 to 159mg/dL, 160 to 189 mg/dL, and 190 mg/dL or more.The change in ordinal category from baseline isevaluated as an indicator of the effect of alternativetreatments on LDL cholesterol levels. Comparisons ofchange in ordinal National Cholesterol EducationProgram category from baseline for the other lipidparameters are measured. Quantitative change inother lipid levels (e.g., high-density lipoprotein cho-lesterol, total cholesterol) from baseline also areevaluated.

● Hypertension: Blood pressure (BP) measurement isobtained at all clinic visits. According to the methodused in the Framingham Heart Study, BP is measuredtwice in the seated position after a patient has restedat least 5 minutes. Measurements are obtained at least2 minutes apart, on the left arm. The average of the 2readings obtained is used for both systolic and dia-stolic BP.56 Blood pressure stages (an ordinal scale)are evaluated in the same manner as that described forhyperlipidemia above. Based on Joint National Com-mittee on Prevention, Detection, Evaluation, andTreatment of High Blood Pressure recommenda-tions,57 ordinal categories for systolic BP are: � 130mm Hg, 130 to 139 mm Hg, 140 to 159 mm Hg, 160to 179 mm Hg, and 180 mm Hg or more. Ordinalcategories for diastolic BP are: � 85 mm Hg, 85 to 89mm Hg, 90 to 99 mm Hg, 100 to 109, and 110 mm Hgor more. Quantitative changes in systolic BP, diastolicBP, and mean arterial pressure from baseline also areevaluated.

● Weight Gain: Weight is measured using a balancescale. Change in weight is evaluated directly andaccording to the body mass index (in kg/m2). Changefrom baseline and maximum values for these variablesare evaluated.

● Other Potential Systemic Complications of Cortico-steroid Therapy: Other less common potential com-plications of systemic corticosteroid therapy aredeemed to be present when medical records confirm
that a diagnosis has been made. Conditions specifi-

V

cally monitored for include diagnosis of any of thefollowing conditions, when not present at baseline: asystemic infection requiring anti-infectious therapy orhospitalization, an axis I psychiatric disorder,58 pan-creatitis, and ischemic necrosis of bone.

● Potential Systemic Complications of Immunosup-pressive Therapy: Each immunosuppressive agenthas a unique spectrum of potential systemic sideeffects. Some of these agents can cause hypertensionand hyperlipidemia, which also can result from corti-costeroid therapy, and are evaluated as describedabove. Additional side effects associated with thesemedications that are medically important and ex-pected to occur with detectable frequency during thetrial are the following.

Bone marrow suppression. (1) Reversible neutrope-nia is evaluated over time as the proportion having atotal white blood cell count of 2500 cells/�L or fewer,which corresponds to an increased infectious risk andis a common indication for interruption of therapy.Change in white blood cell count from baseline anduse of granulocyte stimulatory factors also are noted.(2) Reversible thrombocytopenia is evaluated overtime as the proportion having a platelet count 100000/�L or fewer, which is a common indication forsuspension of therapy. Change from baseline is noted.Hemorrhagic events, requirement for platelet transfu-sion, or other treatments for thrombocytopenia also isnoted. (3) Reversible anemia is evaluated over time asthe proportion having a hemoglobin level of 10 g/dLor less, a threshold often considered clinically signif-icant. Change from baseline is noted. Use of transfu-sions and of erythropoietin also are noted.

Occurrence of myelodysplasia is noted when confirmedin medical records. (1) Reversible hepatotoxicity is
evaluated over time as the proportion with any of the

following, verified in laboratory reports: aspartateaminotransferase or alanine aminotransferase levelsmore than twice the upper range of normal confirmedby repeat testing, or discontinuation of an immuno-suppressive agent as a result of hepatotoxicity. Casesof persistent elevation of aspartate aminotransferase,alanine aminotransferase, or both after discontinua-tion of the offending agent are noted. (2) Reversiblenephrotoxicity is evaluated over time as the propor-tion with: serum creatinine elevated 30% abovebaseline or to a level of 1.5 mg/dL or higher, ordiscontinuation of an immunosuppressive drug forrenal toxicity. Persistent renal insufficiency is noted.(3) Other potential systemic complications of immu-nosuppressive therapy are deemed to be present whendiagnosis is confirmed by medical records. Occurrenceof the following is noted (if not present at baseline):liver cirrhosis, interstitial pneumonia, malignancy,end-stage renal disease, and hemorrhagic cystitis.

● Mortality: Mortality is not expected to occur at ahigh rate in this study, because neither uveitis nor thetreatments being tested are thought to be associatedwith substantially increased risk of death. However,mortality is evaluated in the trial for safety analyses.Death is considered to have occurred when one of thefollowing criteria are met: a death certificate for theparticipant is obtained, the participant is listed asdeceased in the Social Security Death Index or theNational Death Index, notice of death appears inprint (e.g., an obituary), 2 or more relatives oracquaintances testify that the participant has died andprovide the same date of death. Survival audits areperformed periodically, before publication of primaryMUST trial results, or both. Date of death is ascer-tained from the source reporting that death has
occurred.
RIAL 561.e10

The Multicenter Uveitis Steroid Treatment Trial: Rationale, Design

Documents

Transcript of The Multicenter Uveitis Steroid Treatment Trial: Rationale, Design