Grading of age-related maculopathy for epidemiological studies: is digital imaging as good as 35-mm film?

Abstract

Purpose: To compare stereo digital images with stereo 35-mm color transparencies as to the quality and reliability of grading age-related maculopathy (ARM) in the context of a multicenter European epidemiologic study (the EUREYE Study).

Design: Instrument validation study.

Participants: Ninety-one subjects (137 eyes) with varying degrees of ARM, including no ARM.

Methods: From both eyes of the participants, 35-mm film and digital stereoscopic fundus images were obtained with two identical Topcon fundus cameras. Two experienced graders classified all signs of ARM according to the International Classification System. Agreement between imaging techniques and between graders was calculated using the weighted kappa statistic.

Main outcome measures: Signs of ARM (number, size, and morphologic characteristics of drusen; pigmentary changes; geographic atrophy; and neovascular macular degeneration) as well as an overall staging system of increasing ARM severity.

Results: The weighted kappa value for between-technique agreement ranged from 0.41 for number of drusen <63 microm to 0.79 for drusen type and total area occupied by drusen. The kappa values for atrophic and neovascular end-stage ARM were 0.87 and 0.94, respectively. The between-technique agreement on stages of ARM was approximately 0.76. The agreement between graders was largely the same for both techniques of imaging.

Conclusions: In the described setting, digital images were as good as 35-mm film for the grading of ARM. Considering the practical advantages of digital imaging, this technique may serve well in epidemiologic studies of ARM.