Purpose:
To compare measurements and enlargement rates of geographic atrophy (GA) secondary to age-related macular degeneration (AMD) obtained using SDOCT, camera-based fundus autofluorescence (AF), confocal scanning laser ophthalmoscopic (SLO) AF, and fluorescein angiographic (FA) imaging.
Methods:
AMD patients with GA measuring from 1.25 mm2 to 18mm2 based on SDOCT fundus imaging were enrolled in the COMPLement Inhibition with Eculizumab for the Treatment of Non-Exudative Age-Related Macular Degeneration (COMPLETE) Study. Imaging studies were performed at baseline and at follow-up months 3, 6, 9, and 12. These studies included AF imaging with a fundus camera based flash system (TRC-50DX, Topcon Medical Systems; AF Excitation : 535-585 nm Detection :605-715nm), AF and FA imaging with a confocal SLO system (Spectralis, Heidelberg Engineering; AF Excitation :488 nm Detection : > 500 nm), and SDOCT imaging (Cirrus, Carl Zeiss Meditec Inc). The SDOCT system was used to generate OCT fundus image (OFI) and enhanced OFIs, using the 200x200 scan pattern. The OFIs and the enhanced OFIs were manually outlined by two independent graders at the Bascom Palmer Eye Institute, and consensus images were used for measurements. The Topcon AF images were registered to the OFIs using custom-built retinal image registration software. The registered Topcon Color and AF images and the Spectralis AF and FA images were manually outlined by two graders at the Doheny Eye Institute reading center (DIRC), and consensus images were used for quantification. Statistical analysis was performed using paired t-tests, intraclass correlations (ICC), and Bland-Altman analyses.
Results:
A total of 30 patients were enrolled in the GA cohort. Only one eye of each patient was included as a study eye, although 19 fellow eyes meeting entry criteria were evaluated as a secondary endpoint. At baseline, the mean (SD) area of GA was 6.5 (4.2) mm2 with the OFI, 6.6 (4.2) mm2 with the enhanced OFI, 5.9 (4.2) mm2 with Topcon FAF, 5.6 (4.2) with Spectralis AF, and 5.8 (SD=4.1) mm2 Spectralis FA. The differences of 0.6 to 0.9 mm2 between the SDOCT measurements and measurements by AF and FA were highly significant (p<0.001), even though ICCs between all pairs of measurement modalities were excellent at >0.96. For all pairs of modalities, Bland-Altman limits of agreement were within ± 1mm2, except between the SDOCT measurements and Spectralis AF, which were 1.5 mm2.
Conclusions:
While GA can be measured using different imaging modalities, each modality measures a different property of GA. On average, areas of GA measured using AF and FA were smaller than when using SDOCT images. Whether these different imaging modalities yield similar enlargement rates remains to be determined
Clinical Trial:
http://www.clinicaltrials.gov NCT00935883