Fundus Autofluorescence Patterns in Stargardt Disease Over Time

Abstract

The recent article by Cukras et al¹ that demonstrated interesting centrifugal “trajectories” for hyperautofluorescent flecks in Stargardt disease enhances our understanding of the progression of this disease and bears on fundamental questions about the role of lipofuscin. Given the importance of these issues, I suggest that the reference therein to related quantitative work² should also be amplified. We specifically investigated the spatial predictive value of hyperautofluorescent lesions for the subsequent development of hypoautofluorescence or atrophy in the same location with precise registration and segmentation of serial images as well as statistical calculation of spatial predictive values. To our knowledge, this is the only such study, and the results were surprising. If toxic effects to the retinal pigment epithelium by excess lipofuscin drives Stargardt disease, then flecks, with high lipofuscin levels, should be more likely to turn atrophic than neighboring background. However, the predictive value of hyperautofluorescent lesions for future hypoautofluorescence or atrophy at a given location was actually less than background, suggesting that an elevated level of lipofuscin, while an important marker for Stargardt disease, may not be a major cause.