Reappraisal of Geographic Atrophy Patterns Seen on Fundus Autofluorescence Using a Latent Class Analysis Approach

Abstract

The original description of FAF patterns, which is based exclusively on distribution of hyperautofluorescence around GA, ultimately classifies patients according to area of atrophy. These results suggest that FAF patterns are not true phenotypes and that they rather represent different stages of the disease. This may have implications regarding the role of lipofuscin on disease pathogenesis.

Prognostic implications of imaging in atrophic macular degeneration and its use in clinical practice and clinical trial design

“…The ‘banded’ or ‘diffuse’ FAF phenotypes are then consequences of GA enlargement once it reaches the perifoveal areas . At this point, the increased FAF was proposed to be because of disorganized, vertically superimposed lipofuscin‐containing RPE cells instead of the accumulation of lipofuscin in individual cells, as claimed by Rudolf et al …”

“…Biarnes et al . obtained a faster median enlargement rate of 1.65 mm 2 /year, possibly accounted for by sampling differences …”

“…A significant drawback to this progressive theory is that progression in phenotype was not noted in any of their subjects over a 19‐month period, despite the inclusion of 59 subjects (75 eyes) . Similarly, Holz et al .…”

“…A recent study conducted by Biarnes et al . analysed FAF phenotypes with the aim of simplifying their classification and evaluation with latent class analysis (LCA) . LCA evaluates the relationship between a set of observed and unobserved variables to identify another set of unobserved variables that may be linked .…”

“…LCA evaluates the relationship between a set of observed and unobserved variables to identify another set of unobserved variables that may be linked . As the current phenotypic classification of atrophic AMD is based on subjective criteria, the LCA approach aimed to provide new, latent classes devoid of subjective criteria (Table ) …”