Comparison of retinal vascular caliber measured by adaptive optics and from fundus photographs softwar.

Abstract

Purpose: Retinal vascular caliber analysis has been linked with increased cardiovascular risk and is predictive of cardiovascular pathology. The main purpose of this study was to compare 2 methods to evaluate retinal vascular caliber: IVAN® software from fundus photography and retinal vascular diameters from adaptive optics optics (AO) imaging. Secondary objective was to assess the relationships of each retinal vascular variables with systolic blood pressure (BP).

Methods: 25 treatment-naive individuals underwent both AO imaging and fundus photography. Central Retinal artery equivalent (CRAE) was measured from fundus photographs using semi-automated standardized IVAN® software. Adaptive optics imaging of the superotemporal retinal artery was performed and semi-automated segmentation allowed extracting parietal thickness and lumen diameter. Pearson's correlation coefficient and Bland-Altman method were used to assess relationship between both technics of retinal vascular caliber assessment.

Results: Mean (+/- standard deviation [SD]) CRAE from IVAN® software was 135.37 +/- 14.0 µm and mean arterial diameter from adaptive optics was 76.41+/- 11.7 µm. Mean systolic BP was 130.2 +/- 20 mmHg. Correlation coefficient between arterial diameter using AO and CRAE was 0.66 (p=0.0003). The biais between both technic (95% confidence interval) according to the Bland-Altman definition was 58.8 (54.7-63.2) µm. Inverse correlation was found between CRAE and wall to lumen ratio (r=-0.52, p=0.008). Systolic BP was significantly correlated with arterial parameters (r=-0.41, p=0.04; r=-0.50, p=0.01 and r=0.71, p<0.001; for CRAE, arterial diameter from AO and wall to lumen ratio from AO respectively).

Conclusions: Our study shows correlation between adaptive optics measures (both diameters and wall to lumen ratio) and central retinal artery equivalent from fundus photography. All parameters were correlated with systolic BP. Larger studies are necessary to confirm these results and to compare variables derived from retinal arterial caliber with cardiovascular outcome.