Abstract
Purpose :
The study described the presence of perifoveal vessels in outer retinal transverse images of diabetics and age-matched controls.
Methods :
Macular-centered en face images and 20×15 deg SD-OCT grids (Spectralis, Heidelberg Engineering) were taken for 17 diabetics and age-matched controls. Each SD-OCT data cube consisted of an average of 133 b-scans, 30 µm apart with 16 frames averaged per b-scan. Transverse images were generated from the SD-OCT data cube by automatically segmenting the following slabs: nerve fiber layer (NFL), ganglion cell layer (GCL), inner plexiform layer (IPL), inner nuclear layer (INL), outer plexiform layer (OPL), outer nuclear layer (ONL), photoreceptor layer (PRL), and retinal pigment epithelium (RPE). To reduce unwanted noise in transverse images, custom software (Matlab, Mathworks) first computed mean, maximum, and minimum intensity for each row of pixels in the b-scan. Next, to smooth out large changes in intensity row to row, a mask was created by first subtracting the maximum in each row from the mean and then added to the minimum. Finally, the absolute value was computed, and the mask was overlaid onto the original image. The images were then investigated for the presence of perifoveal vessels and structural changes. Two graders compared the retinal vessels and structural changes in the en face images and each slab, counting terminal vessel branches.
Results :
Number of vessel branches counted were significantly different between en face and NFL transverse images for diabetics (14.8 +/- 2.4 vs. 13.2 +/- 1.9, p=0.01) but were not significantly different for controls (12.8 +/- 4.2 vs. 12.4 +/- 3.4, p=0.42). The retinal vessels visualized in the en face images were seen in the superficial layers (NFL, and GCL) for controls, but were distributed among superficial (NFL, and GCL) and the deeper layers (ONL, PRL, and RPE) for diabetics. Vessels in ONL and structural changes were confirmed from b-scans.
Conclusions :
We found perifoveal vessels in the outer retina of the macula region of diabetics.
This is an abstract that was submitted for the 2016 ARVO Annual Meeting, held in Seattle, Wash., May 1-5, 2016.