Purchase this article with an account.
D. A. VanNasdale, A. E. Elsner, A. Weber, M. Miura, B. P. Haggerty, Y. Zhao, J. A. Papay, M. J. Cannon, C. J. Kenrick, B. D. Hansel; Foveal Localization and Polarimetric Image Comparisons Between Age-Related Macular Degeneration and Normals. Invest. Ophthalmol. Vis. Sci. 2008;49(13):890.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
To localize the fovea in subjects with non-exudative age-related macular degeneration (AMD) and age-matched controls using scanning laser polarimetry. To quantify tissue disruption using the variability of scattered light in AMD and control subjects.
We used a scanning laser polarimeter (GDx, CZM) to image the macula in 19 eyes with non-exudative AMD and 19 age-matched controls. Each image series has 40 raw images: 20 input polarizations and 2 detectors, one parallel to the input polarization and one perpendicular (crossed). We computed 18 images with differing polarization content. A maximum phase image, indicating both the magnitude and orientation of corneal/Henle fiber birefringence, was used to determine the foveal location. These images also showed potential birefringence changes between drusen and adjacent retina . The depolarized light image estimates scattered light at each pixel, since multiply scattered light does not modulate with input polarization. We calculated the minimum of the crossed detector output as a function of input polarization. Then, the scattered light ratio was computed as 2 X the depolarized light image/the average image. The scattered light ratio was computed at 4 regions of interest (85 x 85 microns) 1 deg eccentric from the fovea. The variation in light scatter was normalized by calculating the coefficient of variation (SD/mean) for each region of interest, and compared between AMD cases and controls using ANOVA.
The foveal location was easily determined from the birefringence images for all subjects, despite significant disruption to the central macula in many AMD cases. Drusen displayed differing reflection and polarization properties, which could be arranged into three main categories: primarily specularly reflecting, primarily scattering, and center/surround differences in reflecting vs. scattering. Most drusen did not provide orientation changes in their birefringence, compared with adjacent retina, unlike large regions of atrophy. ANOVA analysis indicated the coefficient of variation measurements taken at 1 degree eccentricity were significantly greater in patients with AMD than in controls (p=0.010).
Scanning laser polarimetry can be used to localize the fovea, even in the presence of moderate AMD pathology. Disruption of the highly organized retina in AMD leads to larger variation in light scattering.
This PDF is available to Subscribers Only