Abstract
Purpose: :
To determine if drusen area or its change over time is a good predictor for a subsequent neovascular event.
Methods: :
Using photographs from the AREDS Pittsburgh site, the PTAMD Study’s Pittsburgh site and 50 additional AMD patients being followed in the principal investigator’s practice. Participants were followed for 3 visits at least a 2–year period (baseline, 12, 24 months). At each visit, visual acuity and digital fundus images were obtained. The digitized images were evaluated by a trained technician using the Drusen AnalyzerTM program in which autodetection of drusen was performed within a 1000 and 3000 microns area from the center of the foveola. Two groups of participants were established based on the occurrence of CNV over time. Visual acuity, drusen area within the 1000 and 3000 microns from foveola area were compared between the two groups. Logistic regression was used to determine if a choroidal neovascular event could be predicted by visual acuity, drusen area calculations, or the presence of atrophy or pigment.
Results: :
Out of a total sample of 974 eyes, 59 events were recorded. At baseline, there were significant differences between eyes where a future event occurred compared to eyes where no event occurred in visual acuity (t=4.80, P<.0001), total drusen size at 1000 microns (t=–3.84, p<.001) and total drusen size at 3000 microns (t=–3.47, p=.001). Additionally, the presence of pigment at 1000 mircons (x2=19.1, p<.001) and 3000 microns (x2=36.8, p<.001) or atrophy at 1000 microns (x2=11.4,p.007) and 3000 microns (x2=9.0, p=.003) was also significantly different between event and non–event eyes. Logistic regression analyses show that drusen area size at 1000 or 3000 microns at baseline was not predictive of a future event, eyes with the presence of pigment at both 1000 and 3000 mircons were 4.2 (95% CI: 2.3,7.7) and 4.8 (95% CI:2.8,8.3) were more likely to experience a future event, respectively. Eyes with atrophy at 1000 and 3000 microns were 4.1 (95% CI:1.4,12.0) and 2.4 (95% CI:.9, 6.0) more likely to experience an event, respectively. Also, we found that there are statically significant changes from baseline to event in visual acuity (t=–3.56, p=.0006) and total drusen area size at 3000 mircons (t=2.15, p=.03), but not at 1000 microns (t=1.62, p=.11).
Conclusions: :
We are able to predict a future event based on pigment abnormalities and atrophy. Drusen AnalyzerTM may be a good objective method of determining the likelihood of a visual acuity event associated with the presence of pigment abnormalities, atrophy, or a drusen area change.
Keywords: age-related macular degeneration • choroid: neovascularization • drusen