Correlation of macular thickness with microvascular density and morphology using spectral domain optical coherence tomography angiography (SD-OCTA) among normal subjects.

Abstract

Purpose : To evaluate the correlation of SD-OCTA metrics of capillary density and morphology with OCT macular thickness (MT).

Methods : Retrospective, observational study of healthy subjects from a tertiary-care center under an approved Insitituitonal Review Board protocol. Subjects were imaged with SD-OCTA AngioPlex^TM (Carl Zeiss Meditec Inc, Dublin, CA). A 3x3-mm SD-OCTA scan centered on the fovea of each eye was segmented into superficial (SRL), deep (DRL), and full-thickness (NS) retinal layers using a semi-automated algorithm. Only scans with signal strength of 7 or higher were used. A custom MATLAB (Mathworks, Natick, MA) algorithm was used to quantify several parameters describing capillary density and morphology including: vessel area density (VAD), vessel skeleton density (VSD), vessel diameter index (VDI), vessel diameter (VD), vessel perimeter index (VPI), and vessel complexity index (VCI). In addition, two measures of flow-impairment regions are described (FIR2 and FIR5). OCT Explorer^TM software (University of Iowa, Iowa City, IA) was used to evaluate macular thickness. The SRL was defined by internal limiting membrane (ILM) to inner nuclear layer (INL). The DRL was defined as INL to outer nuclear layer. The NS layer was defined from the ILM to retinal pigmented epithelium. Pearson’s correlations (p<0.05) were used to determine correlations between each SD-OCTA quantitative parameter and MT.

Results : Twenty Eight subjects (47 eyes) with mean age 37 years and 50% males were identified and included in the study. The mean MT in was 336±14, 112±6, and 68±4 microns for NS, SRL, and DRL, respectively in males and 329 ±17, 112±7, and 68 ±6 microns for NS, SRL, and DRL, respectively in females. There were significant correlations between MT and FIR2, FIR5, Flux, and FSI (ρ= -0.32, -0.32, 0.33, 0.32; p-values=0.031, 0.030, 0.026, 0.031, respectively) in the NS layer for all subjects (Fig 1). In the SRL layer, there was a significant correlation between MT and FIR5 (ρ= -0.377, p-value=0.01). There were no significant correlations in the DRL.

Conclusions : OCTA based metrics of blood flow and physiologic flow voids correlate with MT in a pilot cohort of healthy subjects. Larger studies will show whether demographic variable such as gender may play a role in this study due to the effects on MT.

This is an abstract that was submitted for the 2018 ARVO Annual Meeting, held in Honolulu, Hawaii, April 29 - May 3, 2018.