The purpose of this study was to determine classification criteria for multiple evanescent white dot syndrome (MEWDS).
Machine learning of cases with MEWDS and 8 other posterior uveitides.
Cases of posterior uveitides were collected in an informatics-designed preliminary database, and a final database was constructed of cases achieving supermajority agreement on diagnosis using formal consensus techniques. Cases were split into a training set and a validation set. Machine learning using multinomial logistic regression was used in the training set to determine a parsimonious set of criteria that minimized the misclassification rate among the infectious posterior, or panuveitides. The resulting criteria were evaluated in the validation set.
A total of 1,068 cases of posterior uveitides, including 51 cases of MEWDS, were evaluated by machine learning. Key criteria for MEWDS included: 1) multifocal gray-white chorioretinal spots with foveal granularity; 2) characteristic imaging on fluorescein angiography (“wreath-like” hyperfluorescent lesions) and/or optical coherence tomography (hyper-reflective lesions extending from retinal pigment epithelium through ellipsoid zone into the retinal outer nuclear layer); and 3) absent to mild anterior chamber and vitreous inflammation. Overall accuracy for posterior uveitides was 93.9% in the training set and 98.0% (95% confidence interval: 94.3-99.3) in the validation set. Misclassification rates for MEWDS were 7% in the training set and 0% in the validation set.
The criteria for MEWDS had a low misclassification rate and appeared to perform sufficiently well for use in clinical and translational research.
I n 1984, Jampol and associates described a new posterior uveitis which they named multiple evanescent white dot syndrome (MEWDS). The disease occurred in young people (mean age: 28 years), predominantly in women (90%), and was characterized by unilateral, 100-200-µm gray-white dots at the level of the retinal pigment epithelium or outer retina and a foveal granularity. Other common features included posterior vitreous cells, disc swelling with fluorescein staining, and less often vascular sheathing. The white spots had a characteristic wreath-like appearance on fluorescein angiography. The disease spontaneously remitted over ∼2 months with recovery of normal or near normal acuity (20/30 or better) in all cases. No systemic disease was evident, and no treatment appeared warranted.
Subsequent case series have confirmed the clinical features of the disease. The disease presents in young adults; approximately 80% of cases are in women; there is no evident racial or ethnic predilection. Rare cases of bilateral disease and recurrent disease have been reported, but most cases are unilateral with a self-limited disease. ,
MEWDS is a rare disease. The incidence has been estimated at 0.22 per 100,000 population per year, an incidence on the same order of magnitude as acute posterior multifocal placoid pigment epitheliopathy (APMPPE). The cause is unknown. A post-viral autoimmune or auto-inflammatory pathogenesis has been postulated, as case series suggest that ∼50% of patients with MEWDS will have an antecedent influenza-like illness. , However, those case series show recall bias and lack of a control group, making inferences about pathogenesis speculative.
Multimodal imaging is helpful in evaluating the disease. , Fluorescein angiography demonstrates early hyperfluorescence of the multifocal white spots and a “wreath-like” pattern in ∼80% of cases. , Indocyanine green (ICG) angiography demonstrates early to mid-phase hypofluorescence of the white dots and a peripapillary zonal hypofluorescence, the latter of which correlates with the enlarged blind spot often present in patients with MEWDS. , Optical coherence tomography (OCT) imaging demonstrates disruption of the ellipsoid zone and either dome-shaped hyperreflectivity over the retinal pigment epithelium or vertical linear hyper-reflectivity, or both, involving the ellipsoid zone and outer nuclear layer, the latter being seen in ∼80% of cases evaluated with OCT. , , Although there is debate about the exact pathogenesis and extent of the involvement of the choroid, these outer retinal findings may help to distinguish MEWDS from other multifocal choroidopathies. Fundus autofluorescence, evaluated in a limited number of cases, demonstrates hyperautofluorescence of the lesions. ,
Typically, no treatment is given as eyes spontaneously recover in ∼10 weeks with 95% of eyes achieving 20/25 acuity or better. , ,
The Standardization of Uveitis Nomenclature (SUN) Working Group is an international collaboration, which is developing classification criteria for 25 of the most common uveitides by using a formal approach to development and classification. Among the diseases studied was MEWDS.
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.
As previously described, the consensus-based informatics phase permitted development of a standardized vocabulary and 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”). ,
The final database was randomly separated into a training set (∼85% of the cases) and a validation set (∼15% of the cases) for each disease, as described in the accompanying article. Machine learning was used in the training set to determine criteria that minimized misclassification. The criteria then were tested in 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 compared to the consensus diagnosis. For MEWDS, the diseases against which it was evaluated included APMPPE; birdshot chorioretinitis (BSCR); multifocal choroiditis with panuveitis (MFCPU); punctate inner choroiditis (PIC); serpiginous choroiditis; sarcoidosis-associated posterior uveitis; syphilitic posterior uveitis; and tubercular uveitis.
The study adhered to the principles of the Declaration of Helsinki. Institutional Review Boards (IRBs) at each participating center reviewed and approved the study; the study typically was considered either minimal risk or exempt by the individual IRBs.
Ninety-five cases of MEWDS were collected, and 51 (54%) achieved supermajority agreement on the diagnosis during the “selection” phase and were used in the machine learning phase These cases of MEWDS were compared to cases of posterior uveitides, including 82 cases of APMPPE; 207 cases of BSCR; 122 cases of serpiginous choroiditis; 138 cases of MFCPU; 144 cases of PIC; 12 cases of sarcoid posterior uveitis; 35 cases of syphilitic posterior uveitis; and 277 cases of tubercular posterior/panuveitis (including 96 cases of serpiginous-like tubercular choroiditis). Details of the machine learning results for those diseases are outlined in the accompanying article. The characteristics of cases with MEWDS are listed in Table 1 , and the classification criteria developed after machine learning are listed in Table 2 . Key features of the criteria include multifocal white dots ( Figure 1 ), the characteristic “wreath-like” hyperfluorescent lesions on fluorescein angiogram ( Figure 2 ), and the hyper-reflective lesions extending from the retinal pigment epithelium inward on OCT ( Figure 3 ). The overall accuracies for posterior uveitides were 93.9% in the training set and 98.0% (95% confidence interval: 94.3-99.3) in the validation set. The misclassification rate for MEWDS in the training set was 7% and 0% in the validation set.
|Age, median, years (25 th 75 th percentile)||27 (22, 34)|
|Asian, Pacific Islander||6|
|Uveitis course (%)|
|Keratic precipitates (%)|
|Anterior chamber cells (%)|
|Anterior chamber flare (%)|
|Intraocular pressure (IOP), involved eyes|
|Median, mm Hg (25 th , 75 th percentile)||14 (12, 16)|
|Proportion patients with IOP>24 mm Hg either eye (%)||0|
|Vitreous cells (%)|
|Vitreous haze (%)|
|Chorioretinal lesion characteristics|
|Lesion number (%)|
|Unifocal (1 lesion)||0|
|Lesion shape & character (%)|
|Ameboid or serpentine||0|
|Oval or round||85|
|Lesion location (%)|
|Posterior pole involved||47|
|Mid-periphery and periphery only||53|
|Typical lesion size (%)|
|Other features (%)|
|Retinal vascular sheathing||8|
|Retinal vascular leakage||14|
|“Wreath-like” staining of spots on fluorescein angiogram *||65|
|Hyperreflective lesions extending from retinal pigment epithelium into or through ellipsoid zone on optical coherence tomography †||88|