Purpose
To determine classification criteria for serpiginous choroiditis.
Design
Machine learning of cases with serpiginous choroiditis and 8 other posterior uveitides.
Methods
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 on the training set to determine a parsimonious set of criteria that minimized the misclassification rate among the infectious posterior uveitides / panuveitides. The resulting criteria were evaluated on the validation set.
Results
One thousand sixty-eight cases of posterior uveitides, including 122 cases of serpiginous choroiditis, were evaluated by machine learning. Key criteria for serpiginous choroiditis included (1) choroiditis with an ameboid or serpentine shape; (2) characteristic imaging on fluorescein angiography or fundus autofluorescence; (3) absent to mild anterior chamber and vitreous inflammation; and (4) the exclusion of tuberculosis. 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. The misclassification rates for serpiginous choroiditis were 0% in both the training set and the validation set.
Conclusions
The criteria for serpiginous choroiditis had a low misclassification rate and seemed to perform sufficiently well for use in clinical and translational research.
S erpiginous choroiditis is a chronic posterior uveitis characterized by ameboid or serpentine choroidal inflammatory lesions with a characteristic fluorescein angiographic appearance, typically, though not always, extending from the edge of the optic nerve into adjacent retina. Untreated, it is a waxing and waning progressive disease, which results in chorioretinal damage and atrophy. Several other names have been used in the past, including peripapillary choroidal sclerosis, helicoid peripapillary chorioretinal degeneration, geographic choroiditis, geographical choroidopathy, geographic helicoid peripapillary choroidopathy, and serpiginous choroidopathy. The active lesions appear large, paucifocal, or less often multifocal; involve the choroid; and are gray-white or yellow. There is no to minimal anterior chamber and vitreous inflammation. Active lesions resolve over 1-2 months, leaving areas of retinal, retinal pigment epithelial (RPE), and choroidal atrophy. Newly active lesions appear at the border of the previous atrophic lesions. The development of contiguous active lesions at seemingly haphazard locations along the chorioretinal atrophy/scars gives rise to a serpent-like appearance, resulting in its name. Although the majority of lesions are connected and abut the optic disc, separate lesions, not adjacent to the disc, also occur. The fluorescein angiogram demonstrates a classic appearance with early diffuse hypofluorescence of the lesions and late hyperfluorescence of the borders of the lesions (previously termed a “block early, stain late at the borders” appearance), which distinguishes serpiginous choroiditis from most other choroidopathies, such as acute posterior multifocal placoid pigment epitheliopathy (APMPPE), in which the fluorescein angiogram has a characteristic early diffuse hypofluorescence of the lesions and late diffuse hyperfluorescence of the lesions (originally described as “block early, stain late diffusely” appearance). , , Choroidal neovascularization is an uncommon, but well-recognized, complication of serpiginous choroiditis.
Serpiginous choroiditis is a rare entity, accounting for <5% of cases of posterior uveitis in case series from tertiary care referral centers. The etiology is unknown, but an autoinflammatory or autoimmune pathogenesis has been presumed. The reported pathology is that of choroidal lymphocytic inflammation in the areas of active disease with atrophy of the choroid and RPE and photoreceptor loss in the clinically atrophic areas. Scarring may extend into the retina, and focal areas of RPE hypertrophy correspond to pigment clumping sometimes seen at the lesion borders. The course characterized by spread into contiguous choroid has led to the speculation that it may be infectious in nature, as has the detection by polymerase chain reaction of herpes simplex virus and varicella zoster virus in the aqueous humor, each in a single case of serpiginous choroiditis. However, the reported success with immunosuppression, including alkylating agent therapy (which would worsen an infectious process, not control it), strongly suggests that it is not an infectious disease. A similar-appearing disease, known as serpiginous-like tubercular choroiditis, occurs in the context of evidence of infection with tuberculosis (TB) and is treated with antituberculous therapy.
Other imaging modalities may be of value in the management of serpiginous choroiditis. Indocyanine green angiography demonstrates hypofluorescent choroidal lesions throughout the angiogram and may show late hyperfluorescent borders. Indocyanine green angiography may demonstrate more extensive lesions than those seen clinically. , Fundus autofluorescence has been reported to be valuable in the management of serpiginous choroiditis. It demonstrates hypo-autofluorescent lesions in the area of scarring/atrophy; active lesions have hyper-autofluorescent borders, which resolve with treatment. Optical coherence tomographic angiography of active lesions demonstrates an absence of choriocapillaris with variable outer retinal and RPE thickening, whereas in inactive lesions it demonstrates partial reappearance of the choriocapillaris. In atrophic lesions, these structures cannot be imaged.
Because it is a rare disease, most treatment evidence comes from case series. Some investigators believe that oral corticosteroids shorten the duration of acute exacerbations but do not affect the long-term course. Conversely, immunosuppression seems to control the disease and prevent exacerbations. One comparative case series reported significantly better disease control with immunosuppression. Discontinuation of nonalkylating agent immunosuppression may lead to recurrent disease, and the optimal duration of therapy is uncertain. , , Alkylating agent therapy has been reported to induce long-term, sustained, drug-free remissions, but this approach now is used sparingly owing to concerns about increased risks of malignancy associated with alkylating agent therapy.
The Standardization of Uveitis Nomenclature (SUN) Working Group is an international collaboration that has developed classification criteria for 25 of the most common uveitides using a formal approach to development and classification. Among the diseases studied was serpiginous choroiditis.
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 learning 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 on the learning set to determine criteria that minimized misclassification. The criteria then were tested on the validation set; for both the learning 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 serpiginous choroiditis, the diseases against which it was evaluated included APMPPE, birdshot chorioretinitis, multifocal choroiditis with panuveitis (MFCPU), multiple evanescent white dot syndrome, punctate inner choroiditis, sarcoidosis-associated posterior uveitis, syphilitic posterior uveitis, and TB posterior uveitis.
Comparison of Cases With Serpiginous Choroiditis and Serpiginous-like Tubercular Choroiditis
Comparison of the characteristics of cases with serpiginous choroiditis and cases with serpiginous-like TB choroiditis was performed with the χ 2 test for categorical variables or the Fisher exact test when the count of a variable was less than 5. Continuous variables were summarized as medians and compared with the Wilcoxon rank sum test.
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
One hundred fifty-seven cases of serpiginous choroiditis were collected, and 122 (78%) achieved supermajority agreement on the diagnosis during the “selection” phase and were used in the machine learning phase These cases of serpiginous choroiditis were compared to cases of posterior uveitides, including 82 cases of APMPPE, 207 cases of birdshot chorioretinitis, 51 cases of multiple evanescent white dot syndrome, 138 cases of MFCPU, 144 cases of punctate inner choroiditis, 12 cases of sarcoid posterior uveitis, 35 cases of syphilitic posterior uveitis, and 277 cases of tubercular posterior uveitis (including 96 cases of serpiginous-like tubercular choroiditis). The details of the machine learning results for these diseases are outlined in the accompanying article. The characteristics of cases with serpiginous choroiditis are listed in Table 1 , and a comparison between serpiginous choroiditis and serpiginous-like TB choroiditis is listed in Table 2 . Cases with serpiginous-like TB choroiditis involved patients who were younger, more likely to be Asian, less likely to have chronic disease, more likely to be unilateral, more likely to have higher grades of vitreous cells, and more likely to have multifocal lesions. The classification criteria developed after machine learning are listed in Table 3 . Key features of the criteria include the ameboid or serpentine appearance of the lesions ( Figure 1 ) and the characteristic fluorescein angiogram ( Figure 2 ) or fundus autofluorescence demonstrating hypo-autofluorescent lesions with hyper-autofluorescent borders. The overall accuracies for posterior uveitides were 93.9% in the learning set and 98.0% (95% confidence interval 94.3, 99.3) in the validation set. The misclassification rates for serpiginous choroiditis were 0% in both the learning set and the validation set.
| Characteristic | Result |
|---|---|
| Number of cases | 122 |
| Demographics | |
| Age, median, years (25th, 75th percentile) | 50 (38, 58) |
| Sex (%) | |
| Male | 54 |
| Female | 46 |
| Race/ethnicity (%) | |
| White, non-Hispanic | 74 |
| Black, non-Hispanic | 3 |
| Hispanic | 2 |
| Asian, Pacific Islander | 2 |
| Other | 13 |
| Missing | 6 |
| Uveitis history | |
| Uveitis course (%) | |
| Acute, monophasic | 11 |
| Acute, recurrent | 0 |
| Chronic | 88 |
| Indeterminate | 1 |
| Laterality (%) | |
| Unilateral | 14 |
| Unilateral, alternating | 0 |
| Bilateral | 86 |
| Ophthalmic examination | |
| Keratic precipitates (%) | |
| None | 98 |
| Fine | 2 |
| Anterior chamber cells, grade (%) | |
| 0 | 92 |
| ½+ | 5 |
| 1+ | 2 |
| 2+ | 1 |
| 3+ | 0 |
| 4+ | 0 |
| Anterior chamber flare, grade (%) | |
| 0 | 98 |
| 1+ | 2 |
| 2+ | 0 |
| 3+ | 0 |
| 4+ | 0 |
| Iris (%) | |
| Normal | 98 |
| Posterior synechiae | 2 |
| Iris nodules | 0 |
| Iris atrophy (sectoral, patchy, or diffuse) | 0 |
| Heterochromia | 0 |
| IOP, involved eyes | |
| Median, mm Hg (25th, 75th percentile) | 16 (14, 18) |
| Proportion of patients with IOP > 24 mm Hg either eye (%) | 2 |
| Vitreous cells, grade (%) | |
| 0 | 63 |
| ½+ | 21 |
| 1+ | 11 |
| 2+ | 4 |
| 3+ | 1 |
| 4+ | 0 |
| Vitreous haze, grade (%) | |
| 0 | 90 |
| ½+ | 9 |
| 1+ | 1 |
| 2+ | 0 |
| 3+ | 0 |
| 4+ | 0 |
| Chorioretinitis characteristics | |
| Lesion number (%) | |
| Unifocal (1 lesion) | 5 |
| Paucifocal (2-4) | 67 |
| Multifocal (≥5) | 28 |
| Lesion shape and character (%) | |
| Ameboid or serpentine | 100 |
| Oval or round | 0 |
| Placoid | 0 |
| Atrophic | 13 |
| Punctate | 0 |
| Lesion location (%) | |
| Posterior pole involved | 96 |
| Midperiphery and periphery only | 4 |
| Typical lesion size (%) | |
| <125 μm | 0 |
| 125-250 μm | 2 |
| 250-500 μm | 6 |
| >500 μm | 91 |
| Missing | 1 |
| Other features (%) | |
| Retinal vascular sheathing | 1 |
| Retinal vascular leakage | 6 |
| Choroidal neovascularization | 2 |
| Fluorescein angiogram demonstrating early diffuse hypofluorescent lesions and late hyperfluorescent lesion borders a | 100 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
