Abstract
Purpose :
To quantitatively evaluate the relationship between drusen area from photographs and RPE and drusen volume from SDOCT volume scans with respect to baseline association, AREDS Severity Scale, and change in these measurements over time.
Methods :
In a set of 88 eyes with non-advanced AMD (median AREDS Severity Scale = 7), drusen area from color photographs was compared to RPE and drusen volume derived from SDOCT volume scans at baseline and one year follow up. Drusen area within the ETDRS macular grid was assessed using AREDS grading procedures at a reading center. Drusen volume within the OCT macular grid was measured using a custom segmentation algorithm written in MatLab: after automated segmentation, any errors were manually corrected. Change in area and volume within each modality was analyzed using Wilcoxon’s signed rank test and comparisons between baseline and change between modalities was analyzed using ANOVA. Intergrader agreement was assessed by Bland Altman correlations.
Results :
81 eyes had gradable baseline and one year images from both modalities. There was an increase between baseline and one year in drusen volume (p<0.001), but no significant changes in drusen area or change in AREDS Severity Score. However, there was a significant association between drusen volume and drusen area between eyes (p<0.01) irrespective of visit. The association between drusen volume and AREDS Severity Scale was not significant, nor was association between change in volume and change in area or AREDS Scale, perhaps in part due to the minimal changes in these variables at one year. Mean difference between the two graders was 0.02 mm (CI 0.02 0.08) in drusen volume.
Conclusions :
Drusen area from color photographs and RPE and drusen volume from OCT are correlated in eyes with non advanced AMD. Over a one year period, the increase in drusen volume was small but significant while drusen area and AREDS Severity Scale demonstrated little change in this cohort. The utility of drusen volume change as an anatomic endpoint in non-adanced AMD requires further study.
This is an abstract that was submitted for the 2017 ARVO Annual Meeting, held in Baltimore, MD, May 7-11, 2017.