Vogt-Koyanagi-Harada Disease

Vogt-Koyanagi-Harada (VKH) disease is an autoimmune disorder characterized in its complete form by the occurrence of bilateral uveitis accompanied by dermatologic as well as neurologic or auditory symptoms or signs.1–6 Four stages of VKH disease are recognized, including prodromal, acute uveitic, convalescent, and chronic/recurrent uveitic. A history of penetrating ocular trauma or surgery prior to the onset of uveitis precludes the diagnosis. Most patients with VKH disease who are diagnosed and treated promptly never develop dermatologic changes and as such fail to manifest the complete form of the disease. Patients who develop uveitis consistent with VKH disease in the absence of other identifiable causes of uveitis and who have either dermatologic or neurologic/auditory findings, but not both, are said to have incomplete disease, whereas those with uveitis alone are said to have probable disease.7 Somewhat ironically, recent clinic-based studies have suggested that half or more of all patients with ocular findings consistent with acute disease have uveitis alone, or probable VKH.8–12 Rao et al. studied the frequency of distinguishing clinical features of VKH disease observed in patients with bilateral uveitis who presented over a 3-month period at 10 uveitis centers in the United States, Europe, Northern Africa, India and Asia.13 Although a number of symptoms and signs were relatively more common in patients with VKH disease as compared to those with non-VKH uveitis, the findings most predictive of VKH disease were exudative retinal detachment during the acute uveitic phase (likelihood ratio 41.8; 95% CI 5.9–295.1) and choroidal depigmentation, or “sunset glow fundus,” during the chronic/recurrent phase of the disease (likelihood ratio 50.2; 95% CI 22.6–111.3). Yang et al. studied 410 consecutive patient with VKH disease seen at a tertiary referral center in Guangzhou, China, and identified a general evolution over time from early choroidal inflammation where disc edema and serous retinal detachment were the main findings, to recurrent, granulomatous anterior inflammation seen in patients with incompletely treated chronic/recurrent disease. This latter group was characterized by a particularly high rate of development of sunset-glow fundus, Dalen-Fuchs nodules, cataract and ocular hypertension.14 While current diagnostic criteria for VKH disease require the presence of bilateral uveitis,7 it should be noted that unilateral or delayed fellow eye involvement can occur – albeit quite rare.15 Use of the terms “Harada’s disease” to describe inflammation limited to the eye – a misnomer since Dr. Einosuke Harada himself described uveitis in association with both dermatologic and neurologic findings, or of “atypical VKH” are generally discouraged.7

The prevalence of VKH disease varies by region and race, but in general is more common in Asians, Asian Indians, American Indians, Mexican Mestizos, and those from the Middle East. In fact, the development of VKH disease in Caucasians and blacks from sub-Saharan Africa is distinctly uncommon. Women develop the disease slightly more frequently than men. Such epidemiologic observations, together with the identification of associated MHC class II antigens, strongly support an underlying genetic predisposition in the pathogenesis of the disorder. Various triggers of VKH disease have been suggested, including antecedent viral infection or skin trauma, and most evidence points to a melanin-associated antigen as the primary immune target. The precise pathogenesis remains unidentified in the vast majority of patients, however.1–6 While corticosteroids are the mainstay of therapy and are required to control active inflammation during episodes of acute or recurrent inflammation, cumulative evidence supports the first-line use of corticosteroid-sparing immunosuppressive agents to reduce the risk of recurrence, decrease the development of late complications, and improve long-term vision.16–20 Multimodal imaging studies, particularly fluorescein angiography and spectral domain-ocular coherence tomography (SD-OCT), are important adjuncts for initial diagnosis and early monitoring of treatment response.5 Indocyanine green angiography (ICGA) appears to be especially useful for detecting suboptimal or incomplete control of choroidal inflammation in patients with long-standing or chronic/recurrent disease.21,22 Two original articles23,24 and two letters to the editor25,26 in this issue of Ocular Immunology & Inflammation present important findings relevant to the pathogenesis, clinical presentation, diagnosis and management of VKH disease.

Ozdal et al. described the demographic and clinical features of 32 Turkish patients with VKH disease seen in a uveitis tertiary referral center in Ankara, Turkey, over 15 years.23 These 32 patients constituted 2.7% of their total uveitis population seen over that time period. The mean age was 33.6 years (range 12–65 years) and three of the four patients were female. The uveitis was bilateral at presentation in all but one patient (96.6%); that patient developed fellow eye involvement 2 months later. Sixteen patients (50%) presented with acute uveitic disease, four of whom (25%) had recurrent inflammation. Of these sixteen patients, 75% had both serous retinal detachment and optic disc edema/hyperemia, whereas 25% had optic disc involvement alone. Less than one-third of patients (31.2%) had the complete form of the disease, including ocular, neurologic and dermatologic changes. Nearly one in five (18.8%) had uveitis alone, or probable VKH disease. Follow-up ranged from 5 months to nearly 8 years, with an average of just over 3 years. All patients with active uveitis were treated with intravenous and or oral methylprednisolone. Non-corticosteroid immunosuppressive agents were used in 19 patients (59.4%). Late ocular complications included retinal pigment epithelial (RPE) disruption and sunset glow fundus in approximately two thirds of eyes (68.8% and 62.5%, respectively), nummular chorioretinal scars in the periphery in just under half of eyes (46.9%), and severe RPE clumping in nearly one in five eyes (18.8%). Visual acuity at last visit was 20/40 or better in 73.4% of eyes, and 20/200 or worse in 17.2% of eyes. Twenty-three patients (71.9%) experienced no recurrences. Whereas 70% of patients with cutaneous involvement were seen after developing chronic disease, it is noteworthy that nearly one in five (18.8%) patients who presented in the acute uveitic phase of VKH disease and who were treated early and aggressively still developed cutaneous manifestations of the condition. These findings were generally consistent with reports of other clinic-based cohorts, including an unrelated cohort from Turkey.9

Xu et al. examined either serum interleukin (IL)-25 levels or the effect of a physiologic concentration of IL-25 on lipopolysaccharide (LPS)-induced IL-1β, IL-6, and TNF-α release from peripheral blood mononuclear cells (PBMC) in separate subsets of 22 patients with active VKH disease, including 15 patients in the acute uveitic phase and seven with chronic/recurrent inflammation.24 Twelve of the patients with VKH disease were seen prior to treatment, whereas 10 were treated with low dose systemic prednisone within 2 months of testing. The authors did not identify which of these 22 patients constituted the subgroups tested in each experiment. Serum samples from 10 to 12 normal individuals were used as controls in each experiment, although the demographic characteristics of the control groups and whether the control patients were the same in each experiment was not reported. Serum levels of IL-25 tested in 10 patients with VKH disease were, on average, 3.4 times lower than levels in controls (p = 0.001; Mann–Whitney test). PBMCs collected from patients with active VKH disease released more IL-1β, IL-6, and TNF-α in response to LPS compared to PBMCs from normal controls, and the release of all three cytokines from PBMCs collected from patients with active VKH disease was decreased when physiologic concentrations of recombinant IL-25 (rIL-25) were added together with LPS (nominal p-values all <p = 0.05; paired-samples t-test). While the correlations appear clear, and could suggest a role for decreased IL-25 in active VKH disease, the extent to which such correlations support causality remain to be studied. It would be particularly interesting to test for such differences in other forms of uveitis and to know whether the differences noted in patients with VKH disease persisted following treatment and prolonged control of the inflammation, since persistent differences might suggest an underlying or intrinsic difference in IL-25 regulation in patients with VKH disease.

Rani and Chandra described a 36-year-old Asian Indian woman with mild bilateral panuveitis, optic disc hyperemia, and serous retinal detachments producing retinal folds.25 Initial fluorescein angiogram prior to treatment with corticosteroids revealed multiple pinpoint leaks under the detachment space in the right eye and progressive leakage from the optic disc in each eye typical for VKH disease, but also showed a smokestack pattern of leakage in the right eye and several small pigment epithelial detachments (PED) in the left eye, more common in central serous chorioretinopathy (CSCR). Ocular coherence tomography confirmed the presence of both serous retinal and PEDs and revealed irregular undulations of the retinal pigment epithelium (RPE) due to acute inflammatory thickening of the choroid and typical for the acute uveitic phase of VKH disease.5,27–30 Neither enhanced depth imaging nor indocyanine green angiography was performed. The patient responded to treatment with high dose systemic prednisone followed by a transition to azathioprine with reversal of all ocular complications and had no recurrences with 1 year of follow-up. While CSCR as a complication of corticosteroid treatment has been reported in patients with active uveitis, including two patients with VKH disease,31,32 the patient described by Rani and Chandra denied prior corticosteroid use. Arantes et al. have described CSCR misdiagnosed as VKH disease, but that patient had no evidence of uveitis.33 CSCR is a common misdiagnosis in patients with VKH disease, especially early during the acute uveitic phase of the condition when vitreous and anterior chamber inflammation are either absent or mild.14

Finally, Garza Leon et al. described a 37-year-old Mexican woman of undisclosed race who was noted to have a persistent tonic pupil and absent lower extremity reflexes, or Holmes-Adie syndrome, concurrent with the acute uveitic phase of VKH disease.26 Salient initial ocular findings included moderately severe bilateral panuveitis, as well as disc edema and hyperemia with serous retinal detachment in each eye. The patient also had tinnitus, and eventually developed nummular chorioretinal scars, sunset glow fundus, alopecia and vitiligo. Laboratory evaluations were unrevealing and we are left to assume that there was no prior history of eye surgery or penetrating trauma – thus confirming the diagnosis of complete VKH disease. The inflammation responded to use of topical cycloplegic/mydriatic agents together with local and systemic corticosteroids, followed by systemic azathioprine. The anisocoria persisted despite the absence of posterior synechiae formation and discontinuation of all topical medications 1 month following the onset of symptoms. The enlarged pupil showed vermiform movements in response to light and little or no response to accommodation. Absent knee and arm tendon reflexes together with a positive response to 0.125% pilocarpine confirmed the diagnosis of Holmes-Adie syndrome. While others have reported development of bilateral tonic, or Adie, pupils in patients with VKH disease,34–36 the authors appear to have been the first to note unilateral involvement and to have documented absent lower extremity tendon reflexes required to diagnose Holmes-Adie syndrome. The cause of parasympathetic denervation resulting in a tonic pupil in what appears to be a minority of patients with VKH disease was unclear.

Together, these four studies highlight both common and uncommon ocular manifestations of VKH disease, emphasize the increasing importance of multimodal imaging in diagnosis and monitoring, and suggest future avenues of inquiry into the pathogenesis and treatment of this condition.

Declaration of interest

This study was supported in-part by The Pacific Vision Foundation (ETC) and The San Francisco Retina Foundation (ETC). The authors have no financial conflicts.

Acknowledgements

The authors thank Dr. Peizeng Yang and Dr. Manuel Garza for thoughtfully commenting on an early version of this editorial.