Abstract
Purpose: :
Prior studies of geographic atrophy (GA) secondary to age-related macular degeneration (AMD) have primarily used either color fundus photos (CFP) or fundus autofluorescence (FAF) images to measure size and progression of atrophy. It is not known whether one method is superior. Here we assess the inter-observer reliability when grading GA using digital CFP and digital FAF images.
Methods: :
Patients with GA secondary to AMD with both CFP (Visupac, Carl Zeiss Meditec) and FAF (HRA2, Heidelberg Engineering, Germany) obtained on the same day at Duke Eye Center between 1/1/05 and 12/14/06 were identified. Eyes with GA extending outside the arcades, history of choroidal neovascularization or other macular disease, or those with unreadable image quality were excluded. If both eyes of one patient met the inclusion criteria, the left eye was arbitrarily chosen for grading. 72 eyes (72 patients) were included in the analysis. Three graders calculated total GA area on the digital images using the Zeiss and Heidelberg camera software. Graders were masked to each other, and were also masked to the corresponding FAF or CFP image. The study was approved by the Duke University Institutional Review Board.
Results: :
The average age was 80.4 years (range 56-96) and 46 patients (64%) were female. The agreement for graders 1, 2, and 3 for FAF images was 0.99 (95% CI, 0.98-0.99; intraclass correlation). The agreement for graders 1, 2, and 3 for CFP was 0.94 (95% CI, 0.92-0.96; intraclass correlation). The agreement between FAF and CFP images for graders 1, 2, and 3 were 0.93, 0.88, and 0.87, respectively (intraclass correlation). The average difference between measurements of GA using CFP and FAF images was small although the difference tended to increase as the size of GA increased for all graders. For larger areas of atrophy, the measured GA tended to be larger when measured on CFP. Conversely, for smaller areas of atrophy, the measured GA tended to be larger when measured on FAF.
Conclusions: :
The results indicate that inter-observer agreement for CFP and FAF images was very high when measuring GA area. However, agreement between graders was slightly higher for FAF images. The agreement between CFP and FAF was also high indicating that both techniques are consistent and reliable methods of measuring GA. Combining the use of color and FAF images may be the best approach for following GA progression.
Keywords: age-related macular degeneration • imaging methods (CT, FA, ICG, MRI, OCT, RTA, SLO, ultrasound) • retinal pigment epithelium