Purpose
To determine classification criteria for Vogt-Koyanagi-Harada (VKH) disease.
Design
Machine learning of cases with VKH disease and 5 other panuveitides.
Methods
Cases of panuveitides were collected in an informatics-designed preliminary database, and a final database was constructed of cases achieving supermajority agreement on the diagnosis, using formal consensus techniques. Cases were split into a training set and a validation set. Machine learning using multinomial logistic regression was used on the training set to determine a parsimonious set of criteria that minimized the misclassification rate among the panuveitides. The resulting criteria were evaluated on the validation set.
Results
One thousand twelve cases of panuveitides, including 156 cases of early-stage VKH and 103 cases of late-stage VKH, were evaluated. Overall accuracy for panuveitides was 96.3% in the training set and 94.0% in the validation set (95% confidence interval 89.0, 96.8). Key criteria for early-stage VKH included the following: (1) exudative retinal detachment with characteristic appearance on fluorescein angiogram or optical coherence tomography or (2) panuveitis with ≥2 of 5 neurologic symptoms/signs. Key criteria for late-stage VKH included history of early-stage VKH and either (1) sunset glow fundus or (2) uveitis and ≥1 of 3 cutaneous signs. The misclassification rates in the learning and validation sets for early-stage VKH were 8.0% and 7.7%, respectively, and for late-stage VKH 1.0% and 12%, respectively.
Conclusions
The criteria for VKH had a reasonably low misclassification rate and seemed to perform sufficiently well for use in clinical and translational research.
I n 1906 Vogt and independently in 1929 Koyanagi described a disorder characterized by chronic anterior uveitis, alopecia, vitiligo, and dysacusis. In 1929 Harada described a disorder characterized by bilateral serous retinal detachments, chronic posterior uveitis, and cerebrospinal fluid pleocytosis. Subsequently it was recognized that these anterior and posterior segment inflammatory conditions were manifestations of the same disease process, and the disease was named Vogt-Koyanagi-Harada (VKH) disease.
VKH disease is a well-delineated disorder that classically follows an evolutionary disease progression. The disease starts with a prodromal phase characterized by a “flu-like” illness, headache, and meningismus, followed by bilateral choroiditis with serous retinal detachments (early-stage disease, previously termed “acute”). Typically these detachments are multiple with multiple, early pinpoint leaks and late dye pooling on fluorescein angiogram; occasionally they may evolve into bullous detachments. Although the detachments can subside spontaneously, untreated disease typically evolves into a chronic anterior uveitis or panuveitis. The early stage often, though not always,
is accompanied by neurologic symptoms of tinnitus and dysacusis; lumbar puncture, if performed, demonstrates cerebrospinal fluid pleocytosis. Several months after disease onset, late-stage disease (previously termed “chronic”) occurs with a “sunset glow” fundus, often with peripapillary atrophy, foveal granular pigment deposition, and peripheral, depigmented, atrophic chorioretinal spots, typically in the inferior periphery. Active late-stage disease has a chronic anterior uveitis or a panuveitis with choroidal inflammatory lesions, similar to those seen in sympathetic ophthalmia and sometimes termed “Dalen Fuchs–like nodules.” Late-stage disease also may be accompanied by cutaneous lesions, including alopecia, poliosis, and vitiligo. Ocular complications of late-stage disease include choroidal neovascularization and subretinal fibrosis.
VKH disease occurs most often in individuals of East Asian or South Asian heritage but also is common in the Middle East. , In Japan, VKH is the most common uveitic disease seen in tertiary care ophthalmology referral clinics. In the United States, it is seen most often among persons of Hispanic or Native American heritage. The HLA-DR4 genotype is a risk factor, in particular HLA-DRB1*0405.
Treatment of early-stage VKH typically consists of high-dose oral or pulse intravenous corticosteroids. , Early corticosteroid treatment (within 2 weeks of onset of symptoms) is associated with a marked reduction in progression to late-stage disease, but corticosteroid treatment over 6 months in duration is required. Late-stage disease seems to do better with immunosuppression than with corticosteroids alone, and early-stage disease with a delay in treatment initiation may do better with immunosuppression as well.
The Standardization of Uveitis Nomenclature (SUN) Working Group is an international collaboration that has developed classification criteria for 25 of the most common uveitic diseases using a formal approach to development and classification. Among the diseases studied was VKH disease.
Methods
The SUN Developing Classification Criteria for the Uveitides project proceeded in 4 phases as previously described: (1) informatics, (2) case collection, (3) case selection, and (4) machine learning.
Informatics
As previously described, the consensus-based informatics phase permitted the development of a standardized vocabulary and the development of a standardized, menu-driven hierarchical case collection instrument.
Case Collection and Case Selection
De-identified information was entered into the SUN preliminary database by the 76 contributing investigators for each disease, as previously described. Cases in the preliminary database were reviewed by committees of 9 investigators for selection into the final database, using formal consensus techniques described in the accompanying article. , Because the goal was to develop classification criteria, only cases with a supermajority agreement (>75%) that the case was the disease in question were retained in the final database (ie, were “selected”).
Machine Learning
The final database then was randomly separated into a training set (∼85% of cases) and a validation set (∼15% of cases) for each disease, as described in the accompanying article. Machine learning was used on the training set to determine criteria that minimized misclassification. The criteria then were tested on the validation set; for both the training set and the validation set, the misclassification rate was calculated for each disease. The misclassification rate was the proportion of cases classified incorrectly by the machine learning algorithm when compared to the consensus diagnosis. For VKH disease, the diseases against which it was evaluated were Behçet disease uveitis, sympathetic ophthalmia, sarcoidosis-associated panuveitis, syphilitic panuveitis, and tubercular panuveitis. Early-stage and late-stage VKH were evaluated separately, as they have different clinical features.
The study adhered to the principles of the Declaration of Helsinki. Institutional review boards at each participating center reviewed and approved the study; the study typically was considered either minimal risk or exempt by the individual institutional review boards.
Results
Two hundred twenty-four cases of early-stage VKH and 177 cases of late-stage VKH were collected, and 156 (70%) cases of early-stage VKH and 103 (58%) cases of late-stage VKH achieved supermajority agreement on the diagnosis during the “selection” phase and were used in the machine learning phase. These cases of VKH were compared to cases of other uveitides, including 194 cases of Behçet disease, 110 cases of sympathetic ophthalmia, 102 cases of sarcoidosis-associated panuveitis, 70 cases of syphilitic panuveitis, and 277 cases of tubercular panuveitis. The details of the machine learning results for these diseases are outlined in the accompanying article. The characteristics of cases with early-stage VKH are listed in Table 1 and with late-stage VKH in Table 2 . The criteria developed after machine learning for early-stage VKH are listed in Table 3 and for late-stage VKH in Table 4 . Key features of early-stage VKH disease are characteristic serous retinal detachments ( Figures 1 and 2 ) or uveitis with ≥2 of 5 appropriate neurologic findings. Key features of late-stage VKH are sunset glow fundus ( Figure 3 ) or uveitis with ≥1 of 3 characteristic cutaneous findings. The overall accuracy for panuveitides was 96.3% in the training set and 94.0% in the validation set (95% confidence interval 89.0, 96.8). The misclassification rate for early-stage VKH in the training set was 8.0%, and for late-stage VKH 1.0%. In the validation set, the misclassification rates for early-stage VKH and late-stage VKH were 7.7% and 12%, respectively. The diseases with which early-stage and late-stage VKH were most often confused were each other.
| Characteristic | Result |
|---|---|
| Number of cases | 156 |
| Demographics | |
| Age, median, years (25th, 75th percentile) | 39 (28, 51) |
| Sex (%) | |
| Male | 26 |
| Female | 74 |
| Race/ethnicity (%) | |
| White, non-Hispanic | 12 |
| Black, non-Hispanic | 7 |
| Hispanic | 12 |
| Asian, Pacific Islander | 41 |
| Other | 27 |
| Missing | 1 |
| Uveitis history | |
| Uveitis course (%) | |
| Acute, monophasic | 54 |
| Acute, recurrent | 2 |
| Chronic | 35 |
| Indeterminate | 9 |
| Laterality (%) | |
| Unilateral | 1 |
| Unilateral, alternating | 0 |
| Bilateral | 99 |
| Ophthalmic examination | |
| Keratic precipitates (%) | |
| None | 66 |
| Fine | 22 |
| Round | 1 |
| Stellate | 0 |
| Mutton fat | 10 |
| Other | 1 |
| Anterior chamber cells, grade (%) | |
| 0 | 18 |
| ½+ | 13 |
| 1+ | 29 |
| 2+ | 24 |
| 3+ | 12 |
| 4+ | 4 |
| Hypopyon (%) | 0 |
| Anterior chamber flare, grade (%) | |
| 0 | 54 |
| 1+ | 29 |
| 2+ | 16 |
| 3+ | 0 |
| 4+ | 1 |
| Iris (%) | |
| Normal | 87 |
| Posterior synechiae | 13 |
| Sectoral iris atrophy | 0 |
| Patchy iris atrophy | 0 |
| Diffuse iris atrophy | 0 |
| Heterochromia | 0 |
| IOP, involved eyes | |
| Median, mm Hg (25th, 75th percentile) | 13 (12, 16) |
| Proportion of patients with IOP > 24 mm Hg either eye (%) | 0 |
| Vitreous cells, grade (%) | |
| 0 | 47 |
| ½+ | 12 |
| 1+ | 25 |
| 2+ | 10 |
| 3+ | 6 |
| 4+ | 0 |
| Vitreous haze, grade (%) | |
| 0 | 68 |
| ½+ | 12 |
| 1+ | 14 |
| 2+ | 4 |
| 3+ | 1 |
| 4+ | 0 |
| Retinal and choroidal findings (%) | |
| Exudative retinal detachment | 94 |
| Multifocal choroiditis without exudative detachment | 6 |
| Sunset glow fundus (%) | 2 |
| Systemic features (%) | |
| Headache | 63 |
| Tinnitus | 29 |
| Dysacusis | 17 |
| Meningismus | 17 |
| Cerebrospinal fluid pleocytosis a | 28 |
| Vitiligo | 4 |
| Poliosis | 2 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
