Assessing patients who are treated with intravitreal injections by ultra-wide field imaging versus slit lamp biomicroscopy

Abstract

Purpose

To evaluate the agreement between a virtual clinic model based on ultra-widefield imaging (UWFI) and spectral-domain optical coherence tomography (SD-OCT) and conventional face-to-face (F2F) slit-lamp fundus examination with SD-OCT for treatment decision-making in patients receiving intravitreal injections.

Methods

In this retrospective masked paired comparative study, consecutive patients receiving intravitreal injections underwent F2F evaluation by a retina specialist using slit-lamp biomicroscopy and SD-OCT. F2F examination was predefined as the reference standard. During the same clinical encounter, UWFI was obtained. A second retina specialist, masked to the clinical findings and decisions, independently reviewed the SD-OCT and UWFI images in a virtual setting and made management decisions.

Results

A total of 426 eyes from 304 patients were included. Of these, 217 eyes (50.94%) had neovascular age-related macular degeneration (NVAMD), 151 (35.45%) diabetic macular edema (DME), and 56 (13.15%) retinal vein occlusion (RVO; 36 branch RVO, 14 central RVO, and 6 hemi-retinal RVO). One eye (0.23%) had myopic choroidal neovascularization (CNV), and one (0.23%) had Sorsby macular dystrophy with CNV.

The virtual assessment demonstrated 98.12% agreement with the F2F examination for treatment decisions (Cohen's κ = 0.90; 95% CI, 0.83–0.97). Recognition of NVAMD-associated macular hemorrhages was comparable between modalities. UWFI identified additional cases of neovascularization of the disc (NVD) and neovascularization elsewhere (NVE), all of which were subsequently confirmed on slit-lamp examination. The mean virtual review time was more than twofold shorter than the F2F evaluation (P < 0.001).

Conclusions

Virtual assessment using ultra-widefield imaging demonstrated high agreement with F2F slit-lamp examination for treatment decision-making in patients receiving intravitreal injections. This approach may represent an efficient alternative for selected follow-up visits in which anterior segment evaluation is not required.