Adaptive optics-assisted identification of preferential erythrocyte aggregate pathways in the human retinal microvasculature

Abstract

Purpose: To characterize human parafoveal blood flow using adaptive optics scanning laser ophthalmoscopy (AO-SLO).

Methods: In 5 normal subjects, erythrocyte aggregate distributions were analyzed on 3 different days. Erythrocyte aggregates were described as a "dark tail" in AO-SLO. The characteristics of the pathways with dark tail flow in the parafovea were measured. Additionally, the tendency for dark tail flow before and after bifurcations was analyzed to study the blood flow in detail.

Results: Average velocity in parent vessels with dark tail flow was 1.30±0.27 mm/s. Average velocity in daughter vessels with dark tail flow was 1.12±0.25 mm/s, and the average velocity of plasma gaps in daughter vessels without dark tail flow was 0.64±0.11 mm/s. Downstream from the bifurcations, the velocity in vessels with dark tail flow was higher than that in those without it (p<0.001), and the branching angles of vessels with dark tail flow were smaller than those of vessels without it (p<0.001).

Conclusions: Images from the AO-SLO noninvasively revealed pathways with and without dark tail flow in the human parafovea. Pathways with dark tail flow in the daughter vessels generally had faster flow and smaller bifurcation angles than daughter vessels without dark tail flow. Thus, AO-SLO is an instructive tool for analyzing retinal microcirculatory hemodynamics.

Figure 1. Consecutive AO-SLO images from a normal subject.

Bright particles and dark regions were identified in vessel shadows on the cone mosaic. Bright particles are thought to represent leukocytes and plasma gaps. Dark regions represent erythrocyte aggregates, described as a “dark tail.” The red dot circles and yellow dot circles highlight bright particles and dark tails, respectively. Blood flow direction was from the right to the left side.

Figure 2. Extraction of parafoveal network data.

(A) Montage of the parafoveal capillary network. The center black region represents the foveal avascular zone (FAZ). (B) Doughnut-shaped region within 150 µm of the FAZ edge. Vessel lengths were measured within the doughnut-shaped region and on the border of the FAZ. Scale bar represents 150 µm.

Figure 3. Interpretation and characteristics of spatiotemporal (ST) images.

(A) Vessels before bifurcation were defined as parent vessels. Vessels after bifurcation were defined as daughter vessels. (B) Interpretation of ST images. Vessels with a dark tail, a white band, and a dark band were observed. The narrow white band corresponded to trajectories of bright moving objects, and the wide black band corresponded to trajectories of the dark tail. Dark tail velocity was calculated as the slope of the red line placed halfway between white and dark bands. As velocity decreased, the slope steepened. (C) An ST image of a vessel without a dark tail. Curved velocity changes were observed. Unlike vessels with a dark tail, velocities were not straight lines and seemed to change periodically (velocity varied between 0.10 and 0.68 mm/s over 2 s in this ST image). Numbers on the right side are yellow slope velocities showing changes every moment. (D) Combined ST images of dark tail vessels before and after bifurcation. (E) Combined ST images of a dark tail parent vessel and a non-dark tail daughter vessel. Sudden velocity decreases were observed at bifurcations.

Figure 4. Daily variance of dark tail flow distribution.

(A) Changes in the lengths of pathways with or without dark tail on three different days. Note that the lengths show little daily variance. Alphabet A, B, C, D, and E represent each subject. (B) Blood flow distributions of dark tails in the parafovea were nearly identical on each of 3 different days, with only minimal changes in 4 of 5 subjects. Red lines represent the same distribution of dark tails on the 3 different days. The yellow, blue, and green lines represent differences in distribution between days.