Purchase this article with an account.
S. B. Reed, S. Gangaputra, A. Domalpally, Q. Peng, M. Altaweel, MUST Study Group; Morphological Assessment of Fundus Color Images for the Multicenter Uveitis Steroid Treatment (MUST) Trial. Invest. Ophthalmol. Vis. Sci. 2010;51(13):328.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
To describe the uveitis grading program and report the reproducibility of evaluation procedures for color fundus images in the MUST Trial. The MUST trial is the first large multi-center randomized clinical trial comparing the effectiveness and side-effects of standard systemic therapy versus fluocinolone acetonide intraocular implant for the treatment of severe non-infectious uveitis. A grading program was developed to assess morphological features attributed to uveitis at baseline and subsequent changes.
Stereoscopic 3 Field images of the fundus are obtained by certified photographers at study visits, capturing the optic disc and macula. Images are evaluated longitudinally by certified ocular disease evaluators using standardized procedures. The features of various types of uveitis are evaluated by determining the type, location and activity of chorioretinal lesions. The posterior pole is evaluated for macular edema - its presence, area and proximity to fovea; vitreoretinal interface abnormalities such as epiretinal membrane; macular pigment disturbance or atrophy; and optic disc assessment for glaucomatous changes. Global assessment includes retinal evaluation for evidence of neovascularization, vascular occlusion, vessel wall sheathing, and tractional retinal changes. A randomly selected subset of images is regraded for quality control. The regrade is compared with the original grade of record for percentage agreement on a categorical scale, e.g. presence vs. absence.
Quality control evaluation of 66 images was performed. The percentage agreement between the two grades for presence of chorioretinal lesions was 91%. Inter-grader agreement was 92% for presence of macular edema and 88% for presence of epiretinal membrane. Vertical cup to disc ratio evaluations had an intra-class correlation (ICC) of 0.74.
The MUST system for evaluation of stereoscopic color fundus photographs for features of uveitis is highly reproducible and can be used for longitudinal evaluation of uveitis in clinical trials.
Clinical Trial: :
This PDF is available to Subscribers Only