Age-Related Macular Degeneration Staging by Color Fundus Photography vs. Multimodal Imaging-Epidemiological Implications (The Coimbra Eye Study-Report 6).

Cláudia Farinha,João Figueira,Sandrina Nunes,Dalila Alves,José Cunha-Vaz,Isabel Pires,Amélia Martins,Rita Coimbra,Maria Luz Cachulo,João Pedro Marques,Isa Sobral,José Costa,Luisa Ribeiro,Inês Laíns,Rufino Silva,Patrícia Barreto

doi:10.3390/jcm9051329

Abstract

Epidemiology of age-related macular degeneration (AMD) is based on staging systems relying on color fundus photography (CFP). We aim to compare AMD staging using CFP to multimodal imaging with optical coherence tomography (OCT), infra-red (IR), and fundus autofluorescence (FAF), in a large cohort from the Epidemiologic AMD Coimbra Eye Study. All imaging exams from the participants of this population-based study were classified by a central reading center. CFP images were graded according to the International Classification and Grading System for AMD and staged with Rotterdam classification. Afterward, CFP images were reviewed with OCT, IR, and FAF and stage update was performed if necessary. Early and late AMD prevalence was compared in a total of 1616 included subjects. In CFP-based grading, the prevalence was 14.11% for early AMD (n = 228) and 1.05% (n = 17) for late AMD, nine cases (0.56%) had neovascular AMD (nAMD) and eight (0.50%) geographic atrophy (GA). Using multimodal grading, the prevalence increased to 14.60% for early AMD (n = 236) and 1.61% (n = 26) for late AMD, with 14 cases (0.87%) of nAMD and 12 (0.74%) of GA. AMD staging was more accurate with the multimodal approach and this was especially relevant for late AMD. We propose that multimodal imaging should be adopted in the future to better estimate and compare epidemiological data in different populations.

Highlights

Age-related macular degeneration (AMD) is the leading cause of central vision loss in the elderly populations of industrialized countries [1], and this burden is expected to increase as recent estimates point to a projected number of 288 million affected individuals by 2040 [2]
The purpose of the present report is to compare the staging of AMD when using color fundus photography (CFP) grading only to grading using a multimodal approach with CFP, spectral-domain optical coherence tomography (SD-OCT), infra-red (IR), and fundus autofluorescence (FAF) imaging, and to analyze the consequent epidemiologic changes in the set of patients from the coastal town of Mira that participated in the AMD Incidence Coimbra Eye Study
Concerning late AMD, we found that stage 4 was misclassified as a lower stage in nine participants when using only CFP, and this was found to be related with early cases of neovascular AMD (nAMD) or geographic atrophy (GA) not clearly seen in fundus images, or due to CFP image quality preventing discrimination of fine details

Summary

Introduction

Age-related macular degeneration (AMD) is the leading cause of central vision loss in the elderly populations of industrialized countries [1] , and this burden is expected to increase as recent estimates point to a projected number of 288 million affected individuals by 2040 [2]. We witness a change in AMD grading from the conventional color fundus based classification approach to a multimodal one, capable of more accuracy in disease diagnosis and staging [17,18,19,20] Implementation of this multimodal approach to epidemiologic studies will probably be the necessary step, in order to precisely stage AMD and to achieve true comparability between populations. The extent of the discrepancies that would arise when implementing a multimodal classification, compared to the conventional color fundus photography (CFP) based grading, is not known in epidemiologic analysis

Objectives

Methods

Results

Conclusion