Directional kinetics analysis of the progression of geographic atrophy.


Doheny Image Reading Center, Doheny Eye Institute, 1355 San Pablo Street, Suite 211, Los Angeles, CA, 90033, USA. [Email]


OBJECTIVE : To investigate the influence of baseline geographic atrophy (GA) size on the rate of GA progression by using both distance and area measurements.
METHODS : Thirty-five eyes from 24 patients with GA due to age-related macular degeneration were obtained from anonymized datasets available at the Doheny Image Reading Center. Baseline and month 12 fundus autofluorescence (FAF) images were used for this analysis. Borders of GA lesions were semiautomatically segmented by certified reading center graders to create masks of the GA lesion. The masks from the two visits were registered and overlaid to allow the differences in area as well as the differences in the position of GA border between the visits to be computed. Distance measurements were performed using a Euclidean distance map. Sectoral (clock hour)/directional GA progression rates with respect to the foveal center were also calculated.
RESULTS : GA progressed 1.6 ± 0.9 mm2 in area and 92.9 ± 64.9 μm in distance over the 12 months. Smaller GA lesions were associated with more rapid progression when measured using distance (P = 0.0004, R = - 0.554). In contrast, there was no significant correlation in this cohort between baseline GA area and the progression measured in area (P = 0.406). In the sectoral/directional GA progression analysis, progression speed differed among clockwise directions, when progression was evaluated by using area measurements. However, this difference was not found, when evaluated by using distance measurements.
CONCLUSIONS : Use of linear distance-based measurements enables evaluation of GA progression which is not confounded by baseline lesion size.


Euclidean distance map,Fundus autofluorescence,Geographic atrophy,Progression rate,