This prospective, single-center reliability and validity analysis was approved by the Institutional Review Board of the Cleveland Clinic Foundation and adhered to the tenets of the Declaration of Helsinki. All patients were evaluated at the Cole Eye Institute, Cleveland Clinic. Written informed consent was obtained from all participants.

Eligible eyes were those diagnosed with uveitis and evaluated in the clinics of the treating physicians (S.S. and S.K.S.) between February 2025 and June 2025. All included eyes underwent both ultra-widefield FA and macula-centered 12 × 12-mm spectral-domain OCT (SD-OCT) on the same day. Eyes were excluded if they exhibited severe media opacities, retinal detachment, epiretinal membrane with traction, vitreomacular traction, or any ocular conditions that could confound image interpretation, such as severe nonproliferative or proliferative diabetic retinopathy, inherited retinal degeneration, or a history of pars plana vitrectomy. All segmentation profiles of OCT images were visually inspected by a retinal specialist (Y.M.). Manual correction of every individual B-scan was not performed, as that could introduce artificial steps or discontinuities in the final reconstruction. Instead, scans with signal strength less than 7 or with severe artifacts or segmentation errors were excluded to ensure reliable image evaluation.

Ultra-widefield FA was performed using the Optos California (Optos PLC, Dunfermline, Scotland, United Kingdom), and same day SD-OCT imaging was obtained with the CIRRUS 6000 (Carl Zeiss Meditec, Dublin, California, USA). OCT data from the CIRRUS 6000 were exported to and analyzed in the ZEISS FORUM Viewer (software version 4.4.2; Carl Zeiss Meditec AG, Jena, Germany), including the Retina Workplace module (version 2.8.0.445).

FA-defined retinal vascular activity was defined as the presence of intraocular inflammation together with vascular wall staining and/or vascular leakage on late phase (after 5 minutes) ultra-widefield FA. Evaluation was limited to the macula-centered 12 × 12-mm region corresponding to the OCT scan area. Intraocular inflammation was defined as the presence of ≥ 0.5 + anterior chamber cells based on the Standardization of Uveitis Nomenclature (SUN) criteria and/or ≥ 0.5 + vitreous haze according to the Nussenblatt scale. ^,

Structural changes associated with FA-defined inflammatory retinal vascular activity are characterized by vascular dilation and perivascular thickening or perivascular hyperreflective changes associated with compression of the outer nuclear layer (ONL) on OCT. To visualize these changes, we hypothesized that slabs centered around the outer plexiform layer (OPL)–ONL transition would best capture these structural alterations, based on prior OCT B-scan observations of these structural changes demonstrating vascular dilation and perivascular changes extending toward the transition. During study development and before formal analysis, the feasibility of this anatomically motivated approach was qualitatively assessed in a small number of cases. Based on this approach, structural en face OCT images were generated by extracting a slab extending from 35 µm below the inner plexiform layer (IPL) to 35 µm below the OPL using the built-in segmentation in the Zeiss Forum Retina Workplace software, a range that generally corresponds to the OPL-ONL transition zone.

Under normal conditions, this region represents a relatively vessel-free zone on structural OCT, providing a low-signal background against which pathological vascular extension and perivascular hyperreflective changes can be more readily identified. Because retinal vessels are normally located internal to this predefined slab, vascular structures appear dark in these en face images due to vessel shadowing. In contrast, when vascular dilation is present, the affected vessels extend into the predefined slab and their central portions are visualized as hyperreflective signals on en face OCT. Furthermore, when accompanied by perivascular hyperreflective changes, indistinct hyperreflectivity is observed extending beyond the hyporeflective lines of the vessel margins, occasionally causing these lines to become obscured ( Figure 1 ). On structural OCT B-scans, a normal vessel appears as a central hyperreflective lumen with a surrounding hyporeflective border. ^, When perivascular changes are present, additional indistinct hyperreflective areas emerge around these vascular signals, resulting in blurring of the vessel boundaries, corresponding to the en face findings described above.

Structural correlates of FA-defined retinal vascular activity on en face OCT. Case 1 (A-C). Right eye of a 61-year-old woman with undifferentiated uveitis. (A) FA demonstrates venous fluorescein leakage. (B) En face OCT image obtained using customized segmentation extending from 35 µm below the inner plexiform layer (IPL) to 35 µm below the outer plexiform layer (OPL) shows hyperreflective vascular changes and irregular perivascular hyperreflectivity extending beyond the vessel shadow with indistinct margins, involving a broader area both proximal and distal to the sites of fluorescein leakage observed on FA. (C) Corresponding OCT B-scan demonstrates, at the arrowed region, structural changes associated with FA-defined inflammatory retinal vascular activity characterized by localized retinal thickening, displacement of vessels toward the outer nuclear layer (ONL), and perivascular hyperreflective changes. Case 2 (D-F). Right eye of a 35-year-old woman with undifferentiated uveitis. (D) FA demonstrates venous fluorescein leakage. (E) At sites corresponding to FA-defined activity, the en face OCT image (arrow) shows hyperreflective vascular and perivascular changes extending beyond the vessel shadow with indistinct margins, involving a broader area distal to the sites of leakage observed on FA. In contrast, at sites without fluorescein leakage on FA, the en face OCT image (arrowhead) shows central intravascular hyperreflectivity with relative hyporeflectivity at the vessel margins and no perivascular hyperreflective changes extending beyond the vessel shadow. (F) On the OCT B-scan corresponding to the FA-defined site of activity (arrow), hyperreflective changes extending horizontally from the vessel are observed, consistent with the perivascular changes seen on en face OCT.

An en face image was graded as positive when at least part of a vessel appeared hyperreflective with indistinct margins or exhibited hyperreflective changes extending beyond its hyporeflective boundaries ( Figure 1 ). Grading was performed using en face OCT images alone to evaluate the diagnostic utility of this modality; the corresponding B-scans were not reviewed during the grading process.

Image grading was performed independently by 2 retinal specialists (Y.M. and G.B) in a masked and randomized fashion. Discrepancies were resolved by consensus.

Statistical analysis was conducted using SPSS vs.23.0 (IBM-SPSS, Chicago, Illinois, USA). Diagnostic performance was evaluated on an image-by-image basis to reflect visit-level assessment in routine uveitis follow-up. Each en face imaging result was compared with FA findings as the reference standard using 2 × 2 contingency tables. Sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and F1 score were calculated with 95% CIs. Because multiple images from the same eye and inclusion of both eyes introduced within-patient correlation, 95% CIs were estimated using cluster bootstrap resampling at the patient level. Inter-grader agreement was assessed using Cohen’s κ coefficient. κ was interpreted according to Landis and Koch.