Abstract
Purpose :
OCT-Angiography (OCT-A) studies reported decreased peripapillary vessel density (pVD) in glaucoma (G), and its correlation with RNFL defects. Conversely, the macular VD (mVD) has not been described. The purpose of this study is to evaluate mVD and pVD in glaucoma suspects (GS) and G, and their relationship with ganglion cell-inner plexiform layer (GCIPL) and retinal nerve fiber layer (RNFL) defects.
Methods :
The study cohort consisted of healthy controls (C), GS, and G patients. Macular and peripapillary 6x6 mm scans were acquired with OCT (Cirrus, Zeiss) to get GCIPL, macular and peripapillary RNFL (mRNFL, pRNFL, respectively) thickness. Swept Source (SS) OCT-A (Angioplex, Cirrus, Zeiss) was used to acquire macular and peripapillary 6x6 mm scans to get mVD and pVD. OCT-A scans were segmented to get full-thickness, superficial and deep VD (FT-, SUP-, DEEP-VD), exported to Image J software, and binarized using the “mean” autothreshold tool. Two region of interest (ROI) were used to evaluate the VD: 1. ellipsoid ROI centered on the fovea (0.5x0.6mm vertical x horizontal inner radius; 2.0x2.4mm vertical x horizontal outer radius); 2. annular ROI centered on the optic disc (1.95x3.45mm inner x outer radius). ANOVA and t-test were performed stratifying patients first by diagnosis (C, GS, G), then by mGCIPL/pRNFL hemifield defects in order to evaluate the relationship with mVD and pVD.
Results :
Twenty, 15, and 20 eyes from C, GS and G patients – respectively, were included in the analysis. Stratification by groups: thinning among groups was found in all the GCIPL sectors (p<0.001), average (avg), supero-temporal (ST), infero-nasal (IN), inferior (I), infero-temporal (IT), inferior-hemifield (inf-hem) mRNFL sectors (p<0.05), and S/I pRNFL quadrants (p<0.001). Avg, S and I FT-pVD (but not SUP and DEEP-pVD) was decreased among groups (p<0.05). FT-, SUP-, DEEP-mVDs were not significantly decreased. Stratification by Sup/Inf-hemifield GCIPL defects or Sup/Inf-quadrant pRNFL defects: no mVD differences were found in all sectors; FT-, SUP- and DEEP-pVDs were significantly reduced (p<0.01), according with the GCIPL/pRNFL defects.
Conclusions :
Superficial and deep pVD seems reduced in glaucoma (but not in GS), showing good relationship with pRNFL defects. Conversely, mVD may be spared both in GS and glaucoma, regardless GCIPL/RNFL defects.
This is an abstract that was submitted for the 2017 ARVO Annual Meeting, held in Baltimore, MD, May 7-11, 2017.