Purpose
To evaluate accuracy and efficacy of drusen quantification by spectral domain optical coherence tomography compared to conventional photographic severity grading method.
Methods
This study was retrospective chart review of patients who had underwent optical coherence tomography (Cirrus OCT) macular cube protocol for early and intermediate age-related macular degeneration (AMD) in 420 eyes of 420 patients except advanced AMD such as geographic atrophy and neovascular AMD. Using automated algorithm (RPE analysis tool), quantification of Drusen area, volume (3mm & 5mm) were measured. For comparison, grading of color fundus photograph were estimated by conventional method through AREDS severity scale. Photographs were graded for drusen characteristics (size, type, area), pigmentary abnormalities (increased pigment, depigmentation, geographic atrophy).
Results
Mean grading through conventional method was 2.76±0.52 (minimum: 1, maximum: 3) and mean measurement through RPE analysis were 0.30±0.67 (drusen area in 3mm, minimum: 0, maximum: 5), 0.47±1.02 (drusen area in 5mm, minimum: 0, maximum: 9), 0.02±0.07(drusen volume in 3mm, minimum: 0, maximum: 1) and 0.02±0.09 (drusen volume in 5mm, minimum: 0, maximum: 1). There were significant positive correlation between conventional grading (CG) and quantification of drusen area through RPE analysis (CG vs drusen area in 3mm : p=0.029, Pearson correlation coefficient = 0.149, CG vs drusen area in 5mm : P=0.023, Pearson correlation coefficient = 0.156). but there were no significant correlation between CG and quantification of drusen volume (CG vs drusen volume in 3mm : P=0.232, CG vs drusen volume in 5mm : P=0.210).
Conclusions
SD-OCT using automated algorithm (RPE analysis tool) demonstrated high correlation with conventional AREDS severity scale grading.
Keywords: 504 drusen •
412 age-related macular degeneration •
550 imaging/image analysis: clinical