Subclinical Peripheral Vascular Leakage in Demyelinating Disease-Associated Uveitis.

Abstract

To evaluate the cross-sectional association between clinical examination findings and angiographic disease activity in demyelinating disease-associated uveitis, and to describe longitudinal trends in clinical, ultrawidefield fluorescein angiography (UWF-FA) leakage, and optical coherence tomography (OCT) markers. Retrospective cohort study with cross-sectional and longitudinal components. A total of 58 eyes from 30 patients with confirmed demyelinating disease (predominantly multiple sclerosis) and concurrent uveitis evaluated at a tertiary referral center. Eyes were evaluated at three timepoints: initial presentation, acute uveitis flare, and most recent encounter. Clinical inflammation, graded per Standardization of Uveitis Nomenclature criteria, was compared with OCT parameters and UWF-FA leakage (scored by anatomic Zones 1-3). Linear mixed models and generalized estimating equations were used, accounting for intereye correlation and repeated measures. Clinical measures (logMAR visual acuity, anterior chamber [AC] and vitreous cell grades); macular fluid dynamics (central subfield thickness, intraretinal fluid [IRF], and subretinal fluid); and UWF-FA leakage. AC and vitreous inflammation significantly improved following acute flare (P < .001 and P = .01, respectively), as did macular IRF (P = .001) and subretinal fluid (P < .001). Total anatomic burden of angiographic leakage was highest at initial presentation and did not change significantly between acute flare and most recent encounter (P = .78). Among 48 paired clinical-angiographic encounters, 68.8% of eyes classified as clinically inactive (0 AC cells, 0 vitreous cells, no active retinal/choroidal lesions) demonstrated ongoing vascular leakage on UWF-FA. In a multivariable generalized estimating equations model, far-peripheral (Zone 3) leakage did not significantly predict concurrent IRF (OR = 2.33, 95% CI 0.62-8.73, P = .21). Cumulative structural damage, including epiretinal membrane and macular atrophy, increased significantly over time (P < .05 and P = .04, respectively). The high prevalence of subclinical peripheral vascular leakage on UWF-FA in clinically inactive eyes demonstrates clinical examination and OCT may not capture the full spectrum of angiographic abnormalities in demyelinating disease-associated uveitis. This observational clinical-angiographic discordance suggests that UWF-FA may provide complementary information for disease staging and monitoring. Prospective studies are needed to determine whether angiographic findings should influence therapeutic endpoints.