Purpose
The purpose of this study was to determine classification criteria for punctate inner choroiditis (PIC).
Design
Machine learning of cases with PIC 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 by 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 posterior uveitides. The resulting criteria were evaluated in the validation set.
Results
A total of 1,068 cases of posterior uveitides, including 144 cases of PIC, were evaluated by machine learning. Key criteria for PIC included: 1) “punctate”-appearing choroidal spots <250 µm in diameter; 2) absent to minimal anterior chamber and vitreous inflammation; and 3) involvement of the posterior pole with or without mid-periphery. 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 PIC were 15% in the training set and 9% in the validation set.
Conclusions
The criteria for PIC had a reasonably low misclassification rate and appeared to perform sufficiently well for use in clinical and translational research.
P unctate inner choroiditis (PIC), originally termed punctate inner choroidopathy, was first described by Watzke and associates in 1984 as a distinct type of multifocal choroiditis occurring typically in young adult myopic women patients and characterized by small “punctate” lesions of the inner choroid or retinal pigment epithelium or both. Lesions often had overlying subretinal fluid, and on fluorescein angiography in the acute phase were hyperfluorescent and leaked fluorescein dye. Anterior chamber and vitreous inflammation typically were absent. The lesions healed into atrophic scars, and choroidal neovascularization developed in 6 of 10 patients. Although often symptomatic with blurred vision, flashing lights, or paracentral scotomata, vision was minimally affected unless choroidal neovascularization developed.
Punctate inner choroiditis is an uncommon disease, accounting for ≤10% of posterior uveitides presenting to a tertiary care referral ophthalmology center, and its incidence has been estimated at 0.4 cases/1,000,000 population/year. , The cause and pathogenesis of PIC are unknown; PIC is an eye-limited disease unassociated with a systemic disease. It has been speculated that PIC is an autoimmune process, and the association between PIC and polymorphisms in the interleukin-10 and tumor necrosis factor-α genes has been taken as evidence of an autoimmune process. However, it is unclear whether these genetic risk factors are risk factors for the disease itself or for the associated choroidal neovascularization.
Subsequent larger case series , have confirmed the clinical picture described by Watzke and associates. Although there is a wide range of age at presentation, the age ranges from ∼32-33 years. More than 90% of reported cases are women, and ∼85%-92% are myopic. Active lesions are yellow to yellowish white, estimated at 100-300 μm in size, and largely located in the posterior pole. At presentation, ∼55% will have bilateral disease, but bilateral disease has been reported to occur in as many as 88%. Data from 1 study can be used to estimate the rate of bilateral disease as ∼0.03/person-years. More than 90% have no anterior chamber or vitreous inflammation, and other signs of inflammation (eg, posterior synechiae) typically are not present. Approximately 50% of cases will present with choroidal neovascularization in at least 1 eye, , but with follow-up, estimates of choroidal neovascularization run as high ∼75%-80%. , One study estimated the incidence of new and recurrent choroidal neovascularization as 0.02/eye-year (EY) and 0.04/EY, respectively. Other structural complications of uveitis (eg, macular edema, disc edema) typically are not present.
The active “punctate” lesions of PIC are distinct, yellow-white or cream-colored, typically round or oval, and <150 μm in size. There may be an overlying serous elevation, and the lesions may resolve or heal with atrophic scarring. Fluorescein angiography demonstrates early hyperfluorescence of active lesions with late staining. Late-stage atrophic scars are seen as window defects on fluorescein angiography. Indocyanine angiography demonstrates hypofluorescent lesions throughout the angiogram. , More lesions may be evident on imaging than are appreciated on ophthalmoscopy. Optical coherence tomography (OCT) demonstrates focal hyperreflective elevation of the retinal pigment epithelium with corresponding disruption of the inner and outer segment photoreceptor interface. Enhanced depth imaging OCT of acute lesions may demonstrate increased choroidal thickening. Choroidal neovascularization, when present, is seen on fluorescein angiography, OCT, and OCT angiography. , , Fundus autofluorescence demonstrates hyperautofluorescence of active lesions and may be useful in following the response to treatment.
The course of PIC is variable. Active inflammatory lesions may spontaneously involute to small atrophic scars. Bilateral disease may occur simultaneously or asynchronously, with the disease in the second eye occurring years later. Rarely, choroidal neovascularization may spontaneously involute, but now nearly all choroidal neovascularization is treated with anti-vascular endothelial growth factor (VEGF) agents (eg, bevacizumab, ranibizumab, aflibercept). Patients with recurrent disease, chronic disease, bilateral choroidal neovascularization, or choroidal neovascularization requiring multiple injections of anti-VEGF agents typically are treated with oral corticosteroids and immunosuppression. , , Case series suggest that, with appropriate use of immunosuppression, disease control, preservation of vision, and decreased or no need for anti-VEGF injections can be accomplished along with successful corticosteroid-sparing (prednisone, ≤7.5 mg/d) in most patients. , Rates of visual impairment (worse than 20/40) and blindness (20/200 or worse) have been estimated at 0.06/EY and 0.006/EY, respectively, with preservation of good vision typically observed in at least 1 eye.
The Standardization of Uveitis Nomenclature (SUN) Working Group is an international collaboration, which has developed classification criteria for 25 of the most common uveitides by using a formal approach to development and classification. Among the diseases studied was PIC.
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 the formal consensus techniques described in the accompanying articles. , 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 compared to the consensus diagnosis. For PIC, the diseases against which it was evaluated included acute posterior multifocal placoid pigment epitheliopathy (APMPPE); birdshot chorioretinitis (BSCR) multifocal choroiditis with panuveitis (MFCPU); multiple evanescent white dot syndrome (MEWDS); serpiginous choroiditis; sarcoidosis-associated posterior uveitis; syphilitic posterior uveitis: and tubercular (TB) uveitis.
This 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.
RESULTS
A total of 250 cases of PIC were collected, and 144 (58%) achieved supermajority agreement on the diagnosis during the “selection” phase and were used in the machine learning phase. Those cases of PIC were compared to cases of posterior uveitides, including 82 cases of APMPPE, 207 cases of BSCR, 51 cases of MEWDS, 138 cases of MFCPU, 122 cases of serpiginous choroiditis, 12 cases of sarcoid posterior uveitis, 35 cases of syphilitic posterior uveitis, and 277 cases of TB posterior or panuveitis (including 96 cases of serpiginous-like TB choroiditis). Details of the machine learning results for those diseases are outlined in the accompanying article. The characteristics of cases with PIC are listed in Table 1 , and the classification criteria developed after machine learning are listed in Table 2 . Key features of the criteria include: 1) characteristic punctate choroidal lesions ( Figure 1 ); 2) absent or minimal anterior chamber and vitreous inflammation; and 3) posterior pole involvement. The overall accuracies for posterior uveitides were 93.9% in the learning set and 98.0% in the validation set (95% confidence interval: 94.3-99.3). The misclassification rate for PIC in the learning set was 15% and 9% in the validation set. The disease most often confused with PIC was MFCPU.
| Characteristic | Result |
|---|---|
| Number of cases | 144 |
| Demographics | |
| Median IQR (25th,75th) age, y | 32 (25, 39) |
| Men, % | 13 |
| Women, % | 87 |
| Race/ethnicity, % | |
| White, non-Hispanic | 81 |
| Black, non-Hispanic | 6 |
| Hispanic | 1 |
| Asian, Pacific Islander | 2 |
| Other | 4 |
| Missing | 6 |
| Uveitis history | |
| Uveitis course, % | |
| Acute, monophasic | 26 |
| Acute, recurrent | 6 |
| Chronic | 65 |
| Indeterminate | 3 |
| Laterality, % | |
| Unilateral | 41 |
| Unilateral, alternating | 0 |
| Bilateral | 59 |
| Ophthalmic examination | |
| Keratic precipitates, % | |
| None | 100 |
| Anterior chamber cells, % | |
| Grade 0 | 99 |
| ½+ | 1 |
| ≥1+ | 0 |
| Anterior chamber flare, % | |
| Grade 0 | 100 |
| Iris, % | |
| Normal | 100 |
| IOP, involved eyes | |
| Median IQR (25th,75th) mm Hg | 15 (13, 17) |
| Proportion of patients with IOP>24 mm Hg either eye, % | 2 |
| Vitreous cells, % | |
| Grade 0 | 91 |
| ½+ | 9 |
| ≥1+ | 0 |
| Vitreous haze, % | |
| Grade 0 | 99 |
| ½+ | 1 |
| ≥1+ | 0 |
| Chorioretinitis characteristics | |
| Lesion number, % | |
| Unifocal (1 lesion) | 0 |
| Paucifocal (2–4 lesions) | 28 |
| Multifocal (≥5 lesions) | 72 |
| Lesion shape and character, % | |
| Ameboid or serpentine | 0 |
| Oval or round | 18 |
| Placoid | 0 |
| Atrophic | 38 |
| Punctate | 82 |
| Inflammatory lesion/scar location, % a | |
| Posterior pole involved | 78 |
| Posterior pole and periphery/mid-periphery involved | 21 |
| Mid-periphery and periphery only | 1 |
| Typical lesion size, % | |
| <125 μm | 54 |
| 125–250 μm | 33 |
| 250–500 μm | 9 |
| >500 μm | 3 |
| Missing | 1 |
| Other features, % | |
| Retinal vascular sheathing | 3 |
| Retinal vascular leakage | 17 |
| Choroidal neovascularization | 19 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
