Outcomes of Vogt-Koyanagi-Harada Disease: A Subanalysis From a Randomized Clinical Trial of Antimetabolite Therapies


To report outcomes of Vogt-Koyanagi-Harada (VKH) disease from a clinical trial of antimetabolite therapies.


Subanalysis from an observer-masked randomized clinical trial for noninfectious intermediate, posterior, and panuveitis.


setting : Clinical practice at Aravind Eye Hospitals, India. patient population : Forty-three of 80 patients enrolled (54%) diagnosed with VKH. intervention : Patients were randomized to either 25 mg oral methotrexate weekly or 1 g mycophenolate mofetil twice daily, with a corticosteroid taper. main outcome measures : Primary outcome was corticosteroid-sparing control of inflammation at 5 and 6 months. Secondary outcomes included visual acuity, central subfield thickness, and adverse events. Patients were categorized as acute (diagnosis ≤3 months prior to enrollment) or chronic (diagnosis >3 months prior to enrollment).


Twenty-seven patients were randomized to methotrexate and 16 to mycophenolate mofetil; 30 had acute VKH. The odds of achieving corticosteroid-sparing control of inflammation with methotrexate were 2.5 times (95% CI: 0.6, 9.8; P = .20) the odds with mycophenolate mofetil, a difference that was not statistically significant. The average improvement in visual acuity was 12.5 Early Treatment Diabetic Retinopathy Study (ETDRS) letters. On average, visual acuity for patients with acute VKH improved by 14 more ETDRS letters than those with chronic VKH ( P < .001), but there was no difference in corticosteroid-sparing control of inflammation ( P = .99). All 26 eyes with a serous retinal detachment at baseline resolved, and 88% achieved corticosteroid-sparing control of inflammation.


The majority of patients treated with antimetabolites and corticosteroids were able to achieve corticosteroid-sparing control of inflammation by 6 months. Although patients with acute VKH gained more visual improvement than those with chronic VKH, this did not correspond with a higher rate of controlled inflammation.

Vogt-Koyanagi-Harada (VKH) disease is a bilateral granulomatous panuveitis often associated with neurologic findings, such as tinnitus, encephalopathy, and meningeal involvement. The acute uveitic phase is typically characterized by exudative retinal detachments (also known as serous retinal detachments), optic disc hyperemia, and choroidal lesions. If not adequately treated, VKH disease can progress to a chronic anterior uveitis, often accompanied by a “sunset glow fundus,” and eventually develop sequelae such as cataract, glaucoma, subretinal neovascularization, and subretinal fibrosis. The first step in treatment is to suppress intraocular inflammation with high-dose systemic corticosteroids followed by a slow taper over 6 months. Although appropriate corticosteroid therapy may adequately control the condition in some cases, recurrent and chronic disease are very common.

Corticosteroid-sparing immunosuppressive agents have been suggested to be beneficial in the treatment of VKH disease. A number of observational studies on azathioprine, cyclophosphamide, cyclosporine, methotrexate, and mycophenolate mofetil have demonstrated good clinical results. Only a few small case series have demonstrated efficacy with biologics, including infliximab and adalimumab. Some studies suggest that using immunosuppressive therapy (IMT) along with corticosteroids results in better visual outcomes compared to using corticosteroids as monotherapy or with delayed addition of immunosuppressants. Overall, VKH studies have had varied results, and prospective studies comparing IMT are incredibly scarce. There are several drawbacks to retrospective and nonrandomized prospective studies, such as indication bias. Randomized clinical trials with masked outcome assessments overcome these issues. Here we present a subanalysis of patients with VKH treated with either oral methotrexate or mycophenolate mofetil and corticosteroid taper from a randomized clinical trial for noninfectious uveitis.


This subanalysis reports outcomes of patients with VKH disease from a randomized clinical trial comparing methotrexate and mycophenolate mofetil as corticosteroid-sparing treatments for noninfectious intermediate, posterior, and panuveitis ( ClinicalTrials.gov registration no. NCT01232920 ). Patients were enrolled at Aravind Eye Hospitals in Madurai and Coimbatore, India between October 2011 and June 2012, and were followed monthly for 6 months. Institutional Review Board/Ethics Committee approval was obtained at the University of California, San Francisco and at Aravind Eye Hospitals. All patients provided written informed consent. Details of the trial, including enrollment criteria and results, have been previously published. Patients were randomized to receive either 25 mg oral methotrexate weekly or 1 g mycophenolate mofetil twice daily. Patients were required to start on >15 mg of systemic corticosteroids and corticosteroids were tapered according to guidelines with the goal of reducing to ≤10 mg daily by the 5-month visit. Oral corticosteroids were allowed to be tapered faster if there was intolerability or for other medical reasons. The diagnosis of VKH was based on the Revised International Diagnostic Criteria with all patients at a minimum fulfilling the criteria for “probable” VKH disease.


Clinical eye examinations, visual acuity measurements, and optical coherence tomography (OCT) scans were conducted every month for up to 6 months. Study ophthalmologists, visual acuity examiners, and OCT operators were masked. Anterior chamber cells were measured according to the Standardization of Uveitis Nomenclature (SUN) guidelines. Vitreous cells were assessed using a scale derived from the Multicenter Uveitis Steroid Treatment Trial. Vitreous haze was graded by the standardized National Eye Institute scoring system. Study-certified masked refractionists measured best-corrected visual acuity using a logarithm of the minimal angle of resolution (logMAR) tumbling “E” chart at 4 meters.

Patients were defined as having acute VKH if they had been diagnosed ≤3 months prior to enrollment; chronic VKH was defined as having been diagnosed >3 months prior to enrollment. Clinical features are classically used to differentiate between acute and chronic VKH ; however, no standard exists for using duration of disease for differentiating acute vs chronic VKH. Prior studies vary widely in how the duration for acute VKH is defined, ranging from 1 month to 6 months.

Subretinal fluid was determined by the Stratus time-domain OCT machine (Carl Zeiss Meditec Model 3000; software version 4.0) or the Cirrus HD-OCT (Carl Zeiss Meditec, Dublin, California, USA; software version 3.0). Presence of a serous retinal detachment (SRD) was defined by fluid separating the neurosensory retina from the retinal pigment epithelium on the OCT with a central subfield macular thickness >260 μm on the Stratus OCT3 machine. Cirrus measurements were transformed to concord with Stratus OCT values.

Outcome Measures

The primary outcome of treatment success was corticosteroid-sparing control of inflammation in both eyes at the 5- and 6-month study visits. Corticosteroid-sparing control of inflammation was defined by the following: (1) ≤0.5+ anterior chamber cells, ≤0.5+ vitreous cells, ≤0.5+ vitreous haze, and no active retinal/choroidal lesions; (2) ≤10 mg of oral prednisone daily and ≤2 drops of prednisolone acetate 1% (or equivalent) a day; and (3) no declaration of treatment failure because of intolerability or safety concerns.

Secondary outcomes were visual acuity, central subfield thickness, time to resolution of SRD, characteristics of inflammation in treatment failures, and adverse events. Outcomes were also analyzed based on duration of VKH (acute vs chronic). Visual acuity data were analyzed by change in logMAR and by number of Early Treatment Diabetic Retinopathy Study (ETDRS) letters read. Change in central subfield thickness was assessed by calculating the percent change in central subfield thickness at study completion compared to baseline. Resolution of SRD was defined as a reduction of macular thickness to ≤260 μm and absence of fluid on OCT. Foveal atrophy was defined as having a central subfield macular thickness <172 μm, or 2 standard deviations lower than the normative value. All outcomes were prespecified and measured at every study visit.

Statistical Methods

Patients who were lost to follow-up were excluded from the analysis. Patient demographics were analyzed using Fisher exact test. Eye clinical characteristics were analyzed using generalized estimating equations to account for clustering of eyes. Comparisons of outcomes were made using mixed-effects linear and logistic regression models to adjust for within-person correlation between eyes. Mixed-effects linear regression models adjusting for treatment arm were used to determine the association between change in retinal thickness and change in visual acuity. The Kaplan-Meier estimator was used to calculate time to resolution of SRDs, and the log-rank test was used to compare time to resolution by treatment arm. Analyses were performed using Stata 13 (StataCorp LP, College Station, Texas, USA) and R (The R Project for Statistical Computing, Version 3.0.2; available at: http://www.r-project.org/ ).


Forty-three of the 80 patients (54%) enrolled in the trial carried a diagnosis of VKH. The median time since diagnosis was 37 days (interquartile range [IQR]: 5, 678 days). Of the 43 patients with VKH, 27 were randomized to methotrexate (63%) and 16 to mycophenolate mofetil (37%). Baseline demographics and clinical characteristics at enrollment are summarized in Table 1 . The amount of corticosteroid exposure was similar between treatment arms ( P = .81). In the 90 days before enrollment, the median highest oral prednisone dose (or equivalent) was 53 mg (IQR: 30, 60 mg). At enrollment, the median oral prednisone dose was 40 mg (IQR: 30, 50 mg). Thirty-eight of 43 patients (88%), 23 on methotrexate and 15 on mycophenolate mofetil, completed the study and were included in the analysis.

Table 1

Baseline Demographics and Clinical Characteristics of Patients With Vogt-Koyanagi-Harada Disease Enrolled in a Randomized Clinical Trial Comparing Methotrexate With Mycophenolate Mofetil for the Treatment of Noninfectious Uveitis

Patient Demographics/Clinical Characteristics Methotrexate
N (%)
27 (100)
Mycophenolate Mofetil
N (%)
16 (100)
P Value a
Mean age, y (SD) 37.2 (9.0) 40.4 (8.2) .25
Female 20 (74) 10 (63) .50
Median visual acuity, logMAR (IQR) 0.30 (0.06–0.50) 0.26 (0–0.98) .76
Median days since diagnosis (IQR) 29 (6–1752) 42 (4–323) .66
Corticosteroids (prednisone, mg)
Median at baseline (IQR) 40 (30–50) 40 (35–55) .54
Median highest dose in past 90 days (IQR) b 53 (30–60) 57 (40–60) .81
Location of inflammation at enrollment .28
Anterior chamber only 3 (11) 0 (0)
Posterior segment (vitreous, retina, choroid) 24 (89) 16 (100)
Increased retinal thickness (in at least 1 eye) from:
Serous retinal detachment 12 (44) 8 (50) .76
Cystoid or diffuse macular edema 3 (11) 1 (6) .99
Highest level of inflammation past 90 days (either eye)
Anterior chamber cells .95
0 4 (15) 4 (25)
0.5+ 5 (19) 4 (25)
1+ 8 (30) 3 (19)
≥2+ 10 (37) 6 (38)
Vitreous haze .66
0 16 (59) 9 (56)
0.5+ 2 (7) 1 (6)
1+ 3 (11) 3 (19)
≥2+ 4 (15) 1 (6)
Could not assess 2 (7) 2 (13)
Anterior vitreous cells .74
0 11 (41) 6 (38)
0.5+ 1 (4) 2 (13)
1+ 4 (15) 4 (25)
≥2+ 9 (33) 4 (25)
Could not assess 2 (7) 0 (0)
Active retinal/choroidal lesions 21 (78) 16 (100) .14
Lens status .44
Normal or trivial opacities 21 (78) 13 (81)
Cortical changes 0 (0) 1 (6)
Posterior subcapsular 5 (19) 1 (6)
Pseudophakic 1 (4) 1 (6)

IQR = interquartile range; SD = standard deviation.

a P value from Fisher exact test for categorical data and Wilcoxon rank-sum test for continuous data.

b Corticosteroids in the past 90 days included oral, subcutaneous, and intravenous and were adjusted to equivalent calculations of oral prednisone.

Of the 38 patients with VKH who completed the study, 25 (66%) met the definition of treatment success. In comparison, 48% of patients with a diagnosis other than VKH achieved treatment success in the trial. By medication assignment, 17 of 23 patients (74%) taking methotrexate were treatment successes compared with 8 of 15 patients (53%) taking mycophenolate mofetil; the odds of success with methotrexate were 2.5 times the odds with mycophenolate mofetil (95% CI: 0.6, 9.8; P = .20). Of the 6 patients who failed on methotrexate treatment, 4 had uncontrolled inflammation and/or were unable to taper corticosteroids to ≤10 mg/day, and 2 failed because of intolerable side effects. Of the 7 patients who failed mycophenolate mofetil, 6 had uncontrolled inflammation and/or were unable to taper corticosteroids to ≤10 mg/day, and 1 failed because of intolerable side effects.

The median visual acuity at enrollment was 0.29 (IQR 0.02, 0.54) logMAR, or Snellen equivalent 20/40 (IQR 20/20, 20/70). At the final visit, the median visual acuity was 0 (IQR −0.08, 0.10) logMAR, Snellen equivalent 20/20 (IQR 20/17, 20/25), reflecting an average improvement of 12.5 ETDRS letters (95% CI: 3.1, 21.9 letters; P = .009). Forty-five of 74 eyes (61%) had 20/20 or better vision at the end of the trial. However, 10 of these eyes (22%) were in 6 patients who were declared treatment failures owing to the inability to taper corticosteroids or to control inflammation. There was no significant difference in improvement of visual acuity by treatment (mean difference: 0.1 ETDRS letters, 95% CI: −12.0, 12.1 letters; P = .99). Fourteen eyes (19%) in 9 patients had 20/50 or worse vision at the end of the study, 2 eyes because of posterior subcapsular cataract and 12 eyes because of uncontrolled inflammation. Four eyes (5%) in 3 patients had 20/200 or worse vision. Of these patients, 1 patient (2 eyes) exited the study after 2 weeks because of intolerability to immunosuppressive therapy and did not continue with the study medication, 1 had a dense posterior subcapsular cataract, and the third had 2+ vitreous haze.

Outcomes of Acute and Chronic Vogt-Koyanagi-Harada Disease

Thirty patients (70%) were classified as having acute VKH and 13 patients (30%) had chronic VKH. Table 2 displays baseline clinical characteristics of patients by acute and chronic disease. Nineteen patients with acute disease (63%) had evidence of SRDs on OCT at baseline; no patients with chronic disease presented with SRDs. Six patients with acute disease (20%) had evidence of foveal atrophy at baseline. Twenty-six of 30 patients with acute VKH (87%) and 12 of 13 patients with chronic VKH (92%) completed follow-up.

Table 2

Baseline Clinical Characteristics of Patients With Acute Versus Chronic Vogt-Koyanagi-Harada Disease Enrolled in a Clinical Trial Comparing Methotrexate With Mycophenolate Mofetil for Noninfectious Uveitis

Clinical Characteristics Acute VKH (≤3 Months Since Diagnosis)
N (%)
30 (100)
Chronic VKH (>3 Months Since Diagnosis)
N (%)
13 (100)
P Value
Median days since diagnosis (IQR) 12 (3-39) 2240 (1752-2581) <.001
Treatment assignment
Methotrexate 18 (60) 9 (69) .74
Mycophenolate mofetil 12 (40) 4 (31)
Corticosteroids (prednisone, mg)
Median at baseline (IQR) 40 (40-60) 30 (20-40) .02
Median highest dose in past 90 days (IQR) a 60 (53-60) 30 (20-60) .01
Serous retinal detachment (either eye) b 19 (63) 0 (0) .04
Posterior synechiae (either eye) 2 (7) 5 (38) .001
Highest level of inflammation past 90 days (either eye)
Anterior chamber cells .07
0 7 (23) 1 (8)
0.5+ 6 (20) 2 (15)
1+ 8 (27) 3 (23)
≥2+ 9 (30) 7 (54)
Vitreous haze .006
0 22 (73) 3 (23)
0.5+ 1 (3) 2 (15)
1+ 4 (13) 2 (15)
≥2+ 2 (7) 3 (23)
Could not assess 1 (3) 3 (23)
Anterior vitreous cells .19
0 15 (50) 2 (15)
0.5+ 1 (3) 2 (15)
1+ 5 (17) 3 (23)
≥2+ 9 (30) 4 (31)
Could not assess 0 (0) 2 (15)
Active retinal or choroidal lesions 30 (100) 7 (54) <.001
Visual acuity at baseline E (%) 60 (100) E (%) 24 c (100)
Better than 20/50 36 (60) 15 (63) .56
20/50 to 20/200 18 (30) 5 (21)
Worse than 20/200 6 (10) 4 (17)

Only gold members can continue reading. Log In or Register to continue

Stay updated, free articles. Join our Telegram channel

Jan 6, 2017 | Posted by in OPHTHALMOLOGY | Comments Off on Outcomes of Vogt-Koyanagi-Harada Disease: A Subanalysis From a Randomized Clinical Trial of Antimetabolite Therapies

Full access? Get Clinical Tree

Get Clinical Tree app for offline access