Free
Retina  |   December 2014
Quantitative Assessment of the Human Retinal Microvasculature With or Without Vascular Comorbidity
Author Affiliations & Notes
  • Priscilla E. Z. Tan
    Centre for Ophthalmology and Visual Science, University of Western Australia, Perth, Australia
    Lions Eye Institute, University of Western Australia, Perth, Australia
  • Paula K. Yu
    Centre for Ophthalmology and Visual Science, University of Western Australia, Perth, Australia
    Lions Eye Institute, University of Western Australia, Perth, Australia
  • Stephen J. Cringle
    Centre for Ophthalmology and Visual Science, University of Western Australia, Perth, Australia
    Lions Eye Institute, University of Western Australia, Perth, Australia
  • Dao-Yi Yu
    Centre for Ophthalmology and Visual Science, University of Western Australia, Perth, Australia
    Lions Eye Institute, University of Western Australia, Perth, Australia
  • Correspondence: Dao-Yi Yu, Centre for Ophthalmology and Visual Science and Lions Eye Institute, University of Western Australia, 2 Verdun Street, Nedlands, WA 6009, Australia; dyyu@lei.org.au
Investigative Ophthalmology & Visual Science December 2014, Vol.55, 8439-8452. doi:10.1167/iovs.14-15056
  • Views
  • PDF
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Priscilla E. Z. Tan, Paula K. Yu, Stephen J. Cringle, Dao-Yi Yu; Quantitative Assessment of the Human Retinal Microvasculature With or Without Vascular Comorbidity. Invest. Ophthalmol. Vis. Sci. 2014;55(12):8439-8452. doi: 10.1167/iovs.14-15056.

      Download citation file:


      © ARVO (1962-2015); The Authors (2016-present)

      ×
  • Supplements
Abstract

Purpose.: To determine whether morbidity and mortality in patients with cardiovascular comorbidities, but no known ocular disease, is related to demonstrable quantitative changes in the retinal microvasculature.

Methods.: Eleven eyes from 8 donors with cardiovascular comorbidities as a diseased group were compared with 16 eyes from 14 donors free from vascular disease as a control group. All eyes had no known ocular disease. The retina was perfusion-fixed and labeled for endothelial f-actin using micro-cannulation techniques. The retinal microvasculature 3 mm superior to the optic disc was imaged with confocal scanning laser microscopy. Quantitative measurements of capillary diameter and density were obtained using two-dimensional image reconstructions. Pathological vascular changes in other regions of the retinal vasculature found in the diseased group were identified and reconstructed in two or three dimensions.

Results.: Capillary densities were significantly different between each capillary network in the diseased group. There was a significant decrease in density between both the nerve fiber layer and retinal ganglion cell layer of the diseased group when compared with those layers in the control eyes. There were pathological vascular changes including microaneurysms and tortuous, dilated venules identified in the diseased group.

Conclusions.: Cardiovascular comorbidities may be associated with changes to the capillary density within the human retinal microvasculature, before the manifestation of known ocular diseases. These differences in capillary density may have important correlations with neuronal function and facilitates the basis of understanding pathogenic mechanisms in retinal vascular disease.

Introduction
Previous studies have shown a positive correlation between cardiovascular disease and retinal microvascular abnormalities, but this has usually been in the presence of clinically known retinopathy.13 It is unknown as to whether the morphometrically organized capillary network in the retina is altered before observed clinical retinopathy in patients with cardiovascular comorbidities. Understanding retinal microvascular changes before clinical manifestation will provide important insights into the prevention of retinal vascular diseases and thus prevent resultant vision loss. 
To date there has been only one study that has documented quantitative changes in perifoveal capillary networks in patients with vascular comorbidities.4 We previously have documented the quantitative characteristics of human retinal capillary networks using novel micro-cannulation techniques developed in our laboratory.58 Four capillary networks were identified in the normal human retina, with different quantitative and morphometric characteristics in each of the four retinal capillary layers.5,8 Despite these informative studies, there has been no study to determine whether the representative area of retina also would reveal quantitative changes in capillary networks in patients with cardiovascular comorbidities, which may in turn lead to the development of clinical retinopathy. 
The purpose of this study was to describe whether patients with cardiovascular comorbidities reveal measurable alterations on retinal vasculature with no known ocular disease, leading to greater insight into ophthalmic disease processes. In addition to this, our findings may provide beneficial insights into previous reports that have determined a link between cardiovascular comorbidities and retinal vascular disease. 
Materials and Methods
This study was approved by the human research ethics committee at The University of Western Australia. All human tissue was handled according to the tenets of the Declaration of Helsinki. 
Definitions
In this study, “Control” has been defined as being free from known ocular disease or vascular disease. They have been used in this study as healthy, control eyes. “Diseased” has been defined as having a known history of cardiovascular disease or risk factors, such as ischemic heart disease, stroke, and hypertension,9 and are free from known ocular disease. “Capillary diameter” has been defined as the perpendicular distance across the maximum transverse measurement of each vessel. “Capillary density” has been defined as the percentage quantity of area occupied by vessel lumens. “Laminar” is used to describe a single-planar orientation that is predominantly confined to a single retinal layer with minimal projections along the z-axis. “Retinal vascular pathological changes” are defined as an abnormal appearance of the retinal vasculature. “Three-dimensional” is used to describe the capillary networks that traversed prominent projections along the vertical z-axis and were not confined to a single plane. 
Human Donor Eyes
All human donor eyes used in this article had no known history of ocular disease. A total of 27 human eyes from 22 donors were studied. Of these, 11 eyes from 8 donors were carefully chosen based on having a known history of cardiovascular disease, and 16 eyes from 14 donors were used as the control group. The control eyes used in this study also were used in our previous work.5 In the diseased group, patients had suffered from one or more of the following cardiovascular comorbidities: ischemic heart disease, hypertension, atherosclerosis, cerebral vascular accident, cardiomyopathy, atrial fibrillation, hypercholesterolemia, and vascular disease (Table 1). Eyes were acquired after removal of corneal buttons for transplantation from the Lions Eye Bank (Lions Eye Institute, Western Australia). Each donor's demographic data, cause of death, and postmortem time to cannulation are presented in Table 1 (donors A to V). 
Table 1
 
Donor Demographic Details: Age, Sex, Eye, Vascular Comorbidities, Cause of Death, and Time to Cannulation for Donor Eyes
Table 1
 
Donor Demographic Details: Age, Sex, Eye, Vascular Comorbidities, Cause of Death, and Time to Cannulation for Donor Eyes
Donor ID Age, y Sex Eye Cardiovascular Comorbidities Cause of Death Time to Cannulation, h
A 68 M L Ischemic heart disease, hypertension Cerebral vascular accident 11.5
B 41 M R Atherosclerosis Intracranial hemorrhage 17.0
C 51 M L + R Ischemic heart disease Myocardial infarction 19.0
D 65 M L + R Cerebral vascular accident Intracranial hemorrhage 14.0
E 64 M L Ischemic heart disease Cardiogenic shock 14.0
F 65 M R Cardiomyopathy Heart failure 6.5
G 75 F L* Hypertension, AF, vascular disease Cerebral vascular accident 18.0
H 72 M L* + R* Hypertension, hypercholesterolaemia Myocardial infarction 12.0
I 22 M L MVA 15.0
J 32 M L MVA 20.0
K 49 F L + R Cancer 5.5
L 67 M L Cancer 9.5
M 23 M L Suicide 22.0
N 66 M L + R Cancer 15.0
O 22 M L Suicide 15.0
P 53 M L MVA 14.0
Q 60 M L Collapse 18.0
R 72 M R Drowning 15.0
S 72 F L Sepsis 3.0
T 60 M R Liver disease 3.5
U 59 M R Melanoma 12.0
V 39 M L Sepsis 20.0
Perfusion Labeling of Retinal Vasculature
Identical methodology from our previous reports for targeted retinal endothelial labeling has been used in this study.6,10 Notably, this study follows directly from a previous article, which researched the normal microvasculature of the human retina.5 
Briefly, after cannulation of the central retinal artery, blood was washed out from the retinal vasculature with Ringer's solution and 1% BSA through cannulation of the central retinal artery. It was then fixed with 4% paraformaldehyde in 0.1M phosphate buffer, washed, perfused with 0.1% Triton-X-100 in buffer, washed and lastly labeled over the course of 2 hours, and then had a final wash. We used Phalloidin conjugated to Alexa Fluor 546 (30 U; Invitrogen, Carlsbad, CA, USA) to label the f-actin microfilaments; VE-cadherin (1:50; sc-6458; Santa Cruz Biotechnology, Dallas, TX, USA) to label adhesion molecules at endothelial cell junctions; Concanavalin A (20 μg/mL; Invitrogen) to label glycoglycans; Claudin-5 (1:50; 1 mg/mL; Sigma-Aldrich, St. Louis, MO, USA) to label the transmembrane tight junction proteins; Monoclonal Anti-Glial Fibrillary Acidic Protein (1:80; Sigma-Aldrich) to stain astrocytes and glial cells; Bisbenzamide (H33258; 1.2 μg/mL; Sigma-Aldrich) or iodide dye (YO-PRO-1; 6.6 μM; Invitrogen), to label the cell nuclei; and goat anti-cellular retinaldehyde-binding protein (1:50; sc18757; Santa Cruz Biotechnology) to label the Muller cells. 
Tissue Preparation and Flat Mounting of the Retina
After labeling, the human eye was dissected at the equator to allow viewing of the posterior retina. The posterior retina was carefully dissected out and a few cuts made to the retina to enable flat mounting in glycerol before cover-slipping and imaging. 
Microscopy
Initially, low-magnification epifluorescence microscopy (E800; Nikon, Tokyo, Japan) images were taken of the whole-mount retina to ensure the precise location of the studied region, which was typically 3 mm superior to the optic disc. Confocal imaging was then performed using two Nikon C1 confocal systems and a series of z-stacks were captured to visualize all vascular layers of the retina at this studied region. Other areas of the retinal vasculature also were studied in some eyes in which retinal vascular abnormalities were noticed. 
Image Preparation
The confocal images were processed and analyzed using both ImagePro Plus (version 7.1, Media Cybernetics, Rockville, MD, USA) and Image J version 1.43 (http://imagej.nih.gov/ij/; provided in the public domain by the National Institutes of Health, Bethesda, MD, USA). Also used were Adobe Photoshop (version 12.1; Adobe Systems, Inc., San Jose, CA, USA) and Adobe Illustrator CS5 (version 12.1.0; Adobe Systems, Inc.) for preparation of images for this manuscript. Confocal images were pseudo-colored with ImagePro Plus. Reconstruction of three-dimensional capillary networks were performed using Imaris software (version 7.4.2; Bitplane, Zurich, Switzerland). 
Capillary Network Topography and Quantitative Analysis
For comparison consistency, the methodology used to study the capillary topography and quantitate the capillary networks was identical to the first article on normal vasculature in the human retina.5 Using the movie-sequence function in Image J, we were able to view the z-stack sequentially and determine if the capillary networks were laminar or three-dimensional. This function also enabled viewing of the capillary morphology simultaneously with the surrounding nuclei, and thus confirmed the location of the capillary networks within the retina. 
Once the confocal images had been prepared into a single z-projected image for each of the capillary networks, we were able to obtain measurements of the diameters and density of the capillary networks. 
Statistical Analysis
Sigmastat (Sigmastat, version 3.1; SPSS, Chicago, IL, USA) was used to calculate all data in terms of mean and SEs. R (R Foundation for Statistical Computing, Vienna, Austria)11 was used to analyze multiple measurements from diseased eyes with data from single eyes, as well as those with measurements from the right and left eyes of the same individual. To compare measurements between the layers, one-way ANOVA testing was performed. Our analysis model included “Right” or “Left” nested within “Eye donor” as random effects, and used linear mixed modeling to test measurement differences between the retinal layers. “Eye donor” was assigned as a random effect and used to account for the effects of intra-“eye” correlation as well as “Right” and “Left” to account for right and left eye correlation. We also performed a one-way ANOVA to analyze the differences between control and diseased eyes for both diameter and density, using the same random effects model as described. 
Results
General
The mean donor age of the control group was 49.7 ± 4.9 years, taken from 11 left and 5 right eyes from 2 female and 12 male donors. The average postmortem time before eyes were perfused was 13.39 ± 1.68 hours. 
The mean donor age of the diseased group was 62.6 ± 3.39 years, taken from six left and five right eyes from one female and seven male donors. There was no statistical difference in the mean age of the two groups (P = 0.09). The average postmortem time before perfusion of eyes was 14.3 ± 1.20 hours. 
After cannulation and perfusion of the central retinal artery, all orders of the retinal microvasculature were examined to ensure excellent perfusion and washout of blood from the retinal vasculature, as well as clear labeling of the endothelial cells, nuclei, and smooth muscle cells. 
General Comparisons in Capillary Topography in the Control and Diseased Human Retina Capillary Networks
Our previous report highlights the consistent locations where morphometrically different capillary networks were found in the normal human retina.5 Vasculature networks of the diseased group were found in identical locations as the control group, and were located at nerve fiber layer (NFL), retinal ganglion cell layer (RGC), border of inner plexiform layer (IPL) and superficial boundary of the inner nuclear layer (INL), and boundary of deep INL and outer plexiform layer (OPL). 
Figure 1 compares confocal images of capillary networks in the retina from both the control and diseased groups. In comparison with the control vasculature networks, the same vascular distribution pattern could be observed in the diseased group. 
Figure 1
 
Comparison of the capillary network in control and diseased groups. (AD) Representative images of retinal vasculature from control group ([A] donor J; [B] donor M; [C] and [D] donor K). (EH) Representative images of retinal vasculature from diseased group (donor D). Whole-mount confocal microscopy images capture both vascular and nuclei information at 3 mm superior to the optic disc. Phalloidin labeled the endothelial f-actin, and Hoechst or YO-PRO-1 labeled the nuclei. (A) The vasculature of the NFL from the control group demonstrates a higher density of vasculature compared with the vasculature from the diseased group at the same region (E). This is also shown in the comparison of the vasculature in the RGC layer from the control group (B) and the diseased group (F) in the same region. In the RGC layer, venules and arterioles also are observed in addition to capillaries. (C) The control capillary network that lies between the IPL and the superficial INL; (G) the capillary network in this region from the diseased group. (D) and (H) display the deepest capillary network of the control and diseased groups respectively, and lie at the border of the deep INL and OPL. Scale bar: 100 μm.
Figure 1
 
Comparison of the capillary network in control and diseased groups. (AD) Representative images of retinal vasculature from control group ([A] donor J; [B] donor M; [C] and [D] donor K). (EH) Representative images of retinal vasculature from diseased group (donor D). Whole-mount confocal microscopy images capture both vascular and nuclei information at 3 mm superior to the optic disc. Phalloidin labeled the endothelial f-actin, and Hoechst or YO-PRO-1 labeled the nuclei. (A) The vasculature of the NFL from the control group demonstrates a higher density of vasculature compared with the vasculature from the diseased group at the same region (E). This is also shown in the comparison of the vasculature in the RGC layer from the control group (B) and the diseased group (F) in the same region. In the RGC layer, venules and arterioles also are observed in addition to capillaries. (C) The control capillary network that lies between the IPL and the superficial INL; (G) the capillary network in this region from the diseased group. (D) and (H) display the deepest capillary network of the control and diseased groups respectively, and lie at the border of the deep INL and OPL. Scale bar: 100 μm.
The NFL capillary network in the diseased group was similar to the control group, in that it was laminar in structure, and also orientated parallel to the RGC axon bundles linearly, and appeared to have fewer anastomoses compared with the other capillary networks. The main difference that could be observed was the marked decrease in capillary density of the network in the diseased group. 
The RGC layer presented with the largest density of vasculature, and both retinal arterioles and venules were seen. Similar to the control group, capillaries were observed to not only project along a single plane but also were seen to run at varying angles obliquely along the z-plane, and also form anastomoses with capillary networks in all capillary layers, forming a three-dimensional structure. It also was observed in this layer that there was a significant decrease in capillary density (P < 0.05) in the diseased group compared with the control group. 
In the diseased group, the IPL and superficial portion of the INL capillary network displayed a three-dimensional vascular configuration and was consistently the least dense capillary network compared with all other layers in a projected single plane, similar to the control group. Occasional capillary closed loops were seen in this region. 
At the plane of the horizontal cells in the INL, the deepest capillary network could be found, for both control and diseased groups. This layer was laminar in morphology, and most capillaries were seen to project along a single plane. Closed loops were observed at this layer to be more numerous than any other layer. 
Quantitative Analysis of Capillary Diameter
Mean capillary diameter for all networks was 8.92 ± 0.05 μm (n = 1980). Table 2 provides the mean capillary diameter for each capillary network in patients with cardiovascular comorbidities. Diameters were not significantly different in each of the capillary networks (P > 0.05). 
Table 2
 
Quantitative Capillary Diameter and Density Data for Each Network of the Diseased Human Eyes
Table 2
 
Quantitative Capillary Diameter and Density Data for Each Network of the Diseased Human Eyes
Capillary Network NFL RGC IPL/Superficial INL Deep INL/OPL
Capillary diameter, μm 8.86 ± 0.09 (495) 8.93 ± 0.09 (495) 8.93 ± 0.10 (495) 8.96 ± 0.10 (495)
Total capillary density, % 8.15 ± 0.01 (11) 21.42 ± 0.02 (11) 9.69 ± 0.01 (11) 14.19 ± 0.01 (11)
Figure 2 and Table 3 show the comparison between the control and diseased networks. Although capillary diameters in the diseased group were all larger than their control equivalents for all capillary networks, there was no statistically significant difference when comparing each of the four capillary networks of the diseased eyes against each of the four equivalent capillary networks of the control group (P > 0.05). 
Figure 2
 
Graphs showing the averaged data of capillary diameter and density in the control and diseased groups in each of the capillary networks of the retina at 3 mm superior to the optic disc. Layer 1 = NFL, layer 2 = RGC layer, layer 3 = border of IPL and superficial INL, layer 4 = border of deep INL and OPL. The diameter bar graph shows there were no significant differences in each of the capillary networks (P > 0.05). Density box plots show the mean value and the SEs. Significant differences (P < 0.05) were noted between the densities of the control versus diseased eyes in layer 1 and layer 2, as indicated by the asterisk.
Figure 2
 
Graphs showing the averaged data of capillary diameter and density in the control and diseased groups in each of the capillary networks of the retina at 3 mm superior to the optic disc. Layer 1 = NFL, layer 2 = RGC layer, layer 3 = border of IPL and superficial INL, layer 4 = border of deep INL and OPL. The diameter bar graph shows there were no significant differences in each of the capillary networks (P > 0.05). Density box plots show the mean value and the SEs. Significant differences (P < 0.05) were noted between the densities of the control versus diseased eyes in layer 1 and layer 2, as indicated by the asterisk.
Table 3
 
Morphometric Comparisons Between Control and Diseased Groups
Table 3
 
Morphometric Comparisons Between Control and Diseased Groups
Capillary Network NFL RGC IPL/Superficial INL Deep INL/OPL
Control Diseased Control Diseased Control Diseased Control Diseased
Capillary diameter, μm 8.47 ± 0.05 (540) 8.86 ± 0.09 (495) 8.01 ± 0.05 (540) 8.93 ± 0.09 (495) 7.99 ± 0.06 (540) 8.93 ± 0.10 (495) 8.56 ± 0.06 (540) 8.96 ± 0.10 (495)
Total capillary density, % 13.69 ± 0.01 (12) 8.15 ± 0.01 (11) 26.74 ± 0.01 (12) 21.42 ± 0.02 (11) 11.28 ± 0.01 (12) 9.69 ± 0.01 (11) 16.12 ± 0.01 (12) 14.19 ± 0.01 (11)
<60yo >60yo <60yo >60yo <60yo >60yo <60yo >60yo
Capillary diameter, μm 8.48 ± 0.05 (540) 8.85 ± 0.09 (495) 8.27 ± 0.06 (540) 8.65 ± 0.09 (495) 8.25 ± 0.06 (540) 8.65 ± 0.10 (495) 8.58 ± 0.06 (540) 8.94 ± 0.10 (495)
Total capillary density, % 12.02 ± 0.01 (12) 9.96 ± 0.02 (11) 26.43 ± 0.01 (12) 21.76 ± 0.02 (11) 10.50 ± 0.01 (12) 10.54 ± 0.01 (11) 15.51 ± 0.01 (12) 14.86 ± 0.01 (11)
Figure 3 and Table 3 show the comparisons for diameter between the ages of younger than and older than 60 years. There also was no statistically significant difference in diameter when comparing each of the four capillary networks of the eyes younger than 60 years with those older than 60 years (P > 0.05). 
Figure 3
 
Graphs showing the average capillary diameter and density in the young (<60 years old) and elderly (>60 years old) groups in each of the different capillary networks. Layer 1 = NFL, layer 2 = RGC layer, layer 3 = border of IPL and superficial INL, layer 4 = border of deep INL and OPL. The diameter bar graph shows there were no significant differences in each of the capillary networks (P > 0.05) in those older than or younger than 60 years old. The density box plot shows the mean value and the SEs. There were no significant differences (P > 0.05) between all capillary layers.
Figure 3
 
Graphs showing the average capillary diameter and density in the young (<60 years old) and elderly (>60 years old) groups in each of the different capillary networks. Layer 1 = NFL, layer 2 = RGC layer, layer 3 = border of IPL and superficial INL, layer 4 = border of deep INL and OPL. The diameter bar graph shows there were no significant differences in each of the capillary networks (P > 0.05) in those older than or younger than 60 years old. The density box plot shows the mean value and the SEs. There were no significant differences (P > 0.05) between all capillary layers.
Quantitative Analysis of Capillary Density
Table 2 also provides the mean density measurements for each capillary network in patients with cardiovascular comorbidities. The capillary density was greatest in the RGC layer, followed by the deep INL/OPL, then IPL/superficial INL, and last the NFL. There was a significant difference between the capillary density of the NFL and RGC (P < 0.001), the NFL and deep INL/OPL (P < 0.002), the RGC and IPL/superficial INL (P < 0.001), the RGC and deep INL/OPL (P < 0.003), and the IPL/superficial INL and deep INL/OPL (P < 0.006). 
Figure 3 and Table 3 provide a comparison of capillary networks between the control and diseased groups. The average density measurements of all capillary networks in the diseased group were less than their control counterparts. There was a significant decrease in density between the NFL of the diseased group when compared with the NFL of the control eyes (P < 0.004), as well as a significant decrease when comparing the RGC layers of the diseased to the control groups (P < 0.02). 
Figure 3 and Table 3 reveal the comparisons of densities in each capillary layer for those who were younger than and older than 60 years. There was no statistically significant difference between the densities of the four capillary networks between the different sets of ages (P > 0.05). 
Pathological Vascular Changes
There were three eyes that were identified to have pathological vascular changes corresponding to clinical appearances of microaneurysms, tortuous dilated venules, and retinal vascular abnormalities in the diseased group (Figs. 415526). There were no vascular pathologies found in the control group studied. 
Figure 4
 
(A) Confocal projected image of 24 optical sections that show capillary microaneurysms within donor H's retinal vasculature from the diseased group using the ×20 objective lens. Saccular microaneurysms (arrowheads) were predominantly noted. (B) Confocal projected image of 17 optical sections with the ×60 objective lens shows a microaneurysm that is seen as saccular capillary outpouching containing proliferated endothelial cells. (C) A z-series of 110 optical sections with the ×60 objective lens captures a round-shaped saccular microaneurysm among a background of numerous nuclei of neural cells in the INL. (D) A magnified image of (C) with fewer stacked images of 72 optical sections, to allow identification of the proliferated endothelial cells on the wall. The f-actin staining on the wall of this microaneurysm shows remarkably uneven intensity. The cell boundary of most endothelial cells cannot be clearly defined except an endothelial cell (outlined in yellow). The difference in nuclei size and density staining of the endothelial cells also is found in this microaneurysm. (E) and (F) are schematic diagrams of images (C) and (D), respectively. Both (E) and (F) show the outline of microaneurysms and highlight the difference in nuclei size and density. Some nuclei are relatively larger with a speckled chromatin-filled appearance shown in purple (F), and some nuclei show relatively dark staining (pyknotic-like appearance) shown in blue (F). These changes of endothelial cells may indicate varying stages of endothelial proliferation. Scale bar: (A) 100 μm, (B) 17.5 μm, (C) 33.3 μm, (D) 15.9 μm.
Figure 4
 
(A) Confocal projected image of 24 optical sections that show capillary microaneurysms within donor H's retinal vasculature from the diseased group using the ×20 objective lens. Saccular microaneurysms (arrowheads) were predominantly noted. (B) Confocal projected image of 17 optical sections with the ×60 objective lens shows a microaneurysm that is seen as saccular capillary outpouching containing proliferated endothelial cells. (C) A z-series of 110 optical sections with the ×60 objective lens captures a round-shaped saccular microaneurysm among a background of numerous nuclei of neural cells in the INL. (D) A magnified image of (C) with fewer stacked images of 72 optical sections, to allow identification of the proliferated endothelial cells on the wall. The f-actin staining on the wall of this microaneurysm shows remarkably uneven intensity. The cell boundary of most endothelial cells cannot be clearly defined except an endothelial cell (outlined in yellow). The difference in nuclei size and density staining of the endothelial cells also is found in this microaneurysm. (E) and (F) are schematic diagrams of images (C) and (D), respectively. Both (E) and (F) show the outline of microaneurysms and highlight the difference in nuclei size and density. Some nuclei are relatively larger with a speckled chromatin-filled appearance shown in purple (F), and some nuclei show relatively dark staining (pyknotic-like appearance) shown in blue (F). These changes of endothelial cells may indicate varying stages of endothelial proliferation. Scale bar: (A) 100 μm, (B) 17.5 μm, (C) 33.3 μm, (D) 15.9 μm.
Figure 5
 
(A) A large sausage-shaped, dilated, and tortuous retinal venule with irregularities in f-actin staining and shape in the vessel wall among sparse vasculature from donor G in the diseased group. It comprises 19 optical sections to complete a confocal projected image using the ×20 objective lens. Phalloidin labels the endothelial f-actin, and Hoechst labels the nuclei. (B) The lower portion of this vascular abnormality magnified and captured using the ×60 objective lens. (C) and (D) are schematic diagrams of the vasculature in images (A) and (B), respectively. These schematics highlight with clarity the tortuosity of the vasculature. Scale bar: (A) 100 μm, (B) 33.3 μm.
Figure 5
 
(A) A large sausage-shaped, dilated, and tortuous retinal venule with irregularities in f-actin staining and shape in the vessel wall among sparse vasculature from donor G in the diseased group. It comprises 19 optical sections to complete a confocal projected image using the ×20 objective lens. Phalloidin labels the endothelial f-actin, and Hoechst labels the nuclei. (B) The lower portion of this vascular abnormality magnified and captured using the ×60 objective lens. (C) and (D) are schematic diagrams of the vasculature in images (A) and (B), respectively. These schematics highlight with clarity the tortuosity of the vasculature. Scale bar: (A) 100 μm, (B) 33.3 μm.
Figure 6
 
(A) Confocal projected image of the beginnings of a small retinal vascular abnormality comprising 61 optical sections using the ×20 objective lens from donor H. Claudin-5 labels the transmembrane tight junction protein, Hoechst labels the nuclei, glial fibrillary acidic protein labels the astrocytes, and cellular retinaldehyde-binding protein labels the Muller cells. (B) Three-dimensional reconstruction of (A) showing the diseased vasculature from the scleral surface. Vasculature is labeled in red, the Muller cell processes in blue, and the astrocytes and nuclei are labeled in green. The white arrow points toward microaneurysms in (A) and correlates with the three-dimensional reconstructions in (B), and the yellow arrows point toward a larger retinal vascular abnormality. The retinal vascular changes are seen to be in closer association with the Muller cells, but a limited association with the astrocytes. Scale bar: (A) 100 μm, (B) 200 μm.
Figure 6
 
(A) Confocal projected image of the beginnings of a small retinal vascular abnormality comprising 61 optical sections using the ×20 objective lens from donor H. Claudin-5 labels the transmembrane tight junction protein, Hoechst labels the nuclei, glial fibrillary acidic protein labels the astrocytes, and cellular retinaldehyde-binding protein labels the Muller cells. (B) Three-dimensional reconstruction of (A) showing the diseased vasculature from the scleral surface. Vasculature is labeled in red, the Muller cell processes in blue, and the astrocytes and nuclei are labeled in green. The white arrow points toward microaneurysms in (A) and correlates with the three-dimensional reconstructions in (B), and the yellow arrows point toward a larger retinal vascular abnormality. The retinal vascular changes are seen to be in closer association with the Muller cells, but a limited association with the astrocytes. Scale bar: (A) 100 μm, (B) 200 μm.
Numerous bulges, which we refer to in this study as microaneurysms, were observed in the retinal capillary network from two of these donor eyes. The bulges were present in all four retinal vascular layers, and extended right out to the periphery of the retina in all four quadrants. More than 100 microaneurysms were seen per quadrant. On randomly sampling 30 microaneurysms, we found the largest external diameter of microaneurysms to range from 29.12 μm to 116.25 μm. Most of these microaneurysms were noted to be lined by endothelial cells with nuclei of different size and density. Some had large nuclei with speckled areas of increased density, and others were small and contiguously dense throughout. In general, a greater proportion of the endothelial cell nuclei were larger, less dense, more immature-appearing cells. No pericytes were seen. Claudin-5, which stained the transmembrane tight junctions, was noted to be particularly intense within the microaneurysms (Fig. 7). F-actin was contained within the vasculature and the nuclei were distinctly separated from each other; however, most of the cell borders were indiscernible. 
Figure 7
 
Whole-mount confocal images capture the four different vascular layers in a diseased eye (donor H) at 6 mm nasal to the optic disc. (A) to (D) show Claudin-5 labeling the transmembrane tight junction proteins and particularly highlight the microaneurysms. (E) to (H) demonstrate labeling of Claudin-5 (green), glial fibrillary acidic protein (blue), and cellular retinaldehyde-binding protein (red). After merging the images, the microaneurysms appear yellow, demonstrating the colocalization of the Claudin-5 and cellular retinaldehyde-binding protein. Microaneurysms are seen to be present in all layers of the retinal vasculature. As they are relatively large compared with the capillaries, some of the microaneurysms are seen to traverse several retinal vasculature layers. (A) and (E) NFL (layer 1); (B) and (F) RGC layer (layer 2); (C) and (G) border of IPL and superficial INL (layer 3); (D) and (H) border of deep INL and OPL (layer 4). Scale bar: 100 μm.
Figure 7
 
Whole-mount confocal images capture the four different vascular layers in a diseased eye (donor H) at 6 mm nasal to the optic disc. (A) to (D) show Claudin-5 labeling the transmembrane tight junction proteins and particularly highlight the microaneurysms. (E) to (H) demonstrate labeling of Claudin-5 (green), glial fibrillary acidic protein (blue), and cellular retinaldehyde-binding protein (red). After merging the images, the microaneurysms appear yellow, demonstrating the colocalization of the Claudin-5 and cellular retinaldehyde-binding protein. Microaneurysms are seen to be present in all layers of the retinal vasculature. As they are relatively large compared with the capillaries, some of the microaneurysms are seen to traverse several retinal vasculature layers. (A) and (E) NFL (layer 1); (B) and (F) RGC layer (layer 2); (C) and (G) border of IPL and superficial INL (layer 3); (D) and (H) border of deep INL and OPL (layer 4). Scale bar: 100 μm.
Higher proportions of venules observed in the retinal vasculature in the diseased group were more tortuous in appearance and had dilated lumens. One particularly tortuous vessel from donor G was found 3 mm nasal to the optic disc and was 39.47 μm in diameter (Fig. 5). It was observed to be a derivation from the vasculature at the RGC layer (Layer 2), and appeared to be a draining tortuous vein. High magnification showed circumferential arrangement of the f-actin around the dilated vessel. 
Pathological retinal vascular changes were not limited in the location that we selected to perform quantitative assessments of capillary density and diameter. These pathological vascular changes were seen to occur in the different capillary networks and were associated with other cellular changes, particularly glial cells in the retina. In one particular eye (donor H), there were over 1000 retinal vascular abnormalities seen per quadrant. Claudin-5, glial fibrillary acidic protein, and cellular retinaldehyde-binding protein label were used for illustrating the pathological alterations of vascular endothelial cells, astrocytes, and Muller cells in the microaneurysms. Figure 6 is a three-dimensional reconstructed confocal projected image from donor H showing that microaneurysms were located in various capillary networks and associated with alterations of glial cells and possible leakage from the wall of the microaneurysm, forming a relatively large retinal vascular abnormality. Whole-mount confocal images captured (Fig. 7) demonstrate Claudin-5 labeling the transmembrane tight junction proteins, which particularly highlight the microaneurysms. Close relationship between microaneurysms and glial cells was evidenced by co-labeled Claudin-5, glial fibrillary acidic protein, and cellular retinaldehyde-binding protein. Under microscopy, they fluoresced brightly and were different from the bulging outpouchings from retinal vasculature. They had a “burst” appearance with some minor leakage. They may potentially be representative of burst microaneurysms which were leaky. 
Discussion
The major findings from this study are as follows. First, cardiovascular comorbidities were found to have an association with changes in capillary density within the retina. Specifically, there was a significant decrease in capillary density of inner retina in donors with cardiovascular comorbidities when compared with the control group. Second, pathological vascular changes were identified in the vasculature of 3 (27.3%) of 11 donors with cardiovascular comorbidities despite no known ocular disease. Specifically, vascular abnormalities found corresponded to clinical appearances of microaneurysms, tortuous venules, and other retinal changes. Close association of microaneurysms and glial cells in different capillary networks also are demonstrated. 
It is important to note that only in these three eyes were vascular anomalies noted, and there were no other findings of these vascular abnormalities seen in any of the other human donor eyes studied, regardless of age or disease. 
These findings may suggest that retinopathy likely begins much earlier than the onset of clinically detectable disease and we potentially should counsel at-risk patients with multiple cardiovascular comorbidities much earlier to prevent retinopathy. 
Endothelial cells within the morphometrically organized capillary network play an indispensable role in modulating retinal homeostasis.10,1214 The intricate morphometric design of the capillary networks is believed to be strongly associated with the metabolic demands of neuronal subtypes within the region,1317 enabling the heterogenic metabolic demand of the neurons and supporting glia to be met.5,18 Theoretically, patients with cardiovascular comorbidities have the potential to alter the blood-retinal barrier and thus play an important role in influencing retinal nourishment.19,20 By demonstrating quantitative changes in parameters, such as capillary density and diameter, important information regarding how retinal disease processes are formed in patients with cardiovascular disease is more clearly understood.21,22 
Cardiovascular comorbidities, including ischemic heart disease, hypertension, and hypercholesterolemia, have been shown in existing studies to be associated with retinal vascular disease.9,2329 Changes at the cellular level to basement membranes, mural cells, and endothelial tight junctions have been shown in patients with cardiovascular disease and risk factors in histopathologic and electron microscopy studies.30,31 However, previous studies that demonstrated links and alterations in patients already had clinical manifestations of retinal disease, or aimed to induce diseased states. Our study not only supports existing data but also augments previous studies in that we have shown demonstrable quantitative changes before the manifestation of known retinal vascular disease. Once clinical manifestations of retinal disease are present, particularly at the macula, accelerated and permanent vision loss can develop. 
All capillary layer densities were lower in the retina of patients with cardiovascular comorbidities compared with normal retina, and were significantly so in the NFL and the RGC layer. A decrease in capillary density may be the normal progression of retinal vascular disease processes, and similar results have been confirmed in previous studies.32 Although capillary network layer three (border of IPL and superficial boundary of INL) and four (boundary of deep INL and OPL) were not significantly different between the patients with cardiovascular comorbidities and control patients, this may be due to the diseased patients being in the earliest stages of retinal vascular disease. Perhaps, given more time to allow for further retinal disease processes to occur, layers three and four of the retinal capillary network also may be significantly decreased. 
A clinical example of relevance once capillary density has decreased is cotton-wool spots. Cotton-wool spots are usually caused by ischemia in the NFL, with the pathogenesis being described as an interruption of axoplasmic transport.33 Ischemia may then lead to retinal vascular pathological changes, such as induced vessel dilatation. 
It was shown in our study that all capillary diameters were larger (nonsignificant) in the eyes of diseased patients compared with the control group. This finding was shown in a similar study,4 and represented similar reports to cerebrocortical models of hypoxia34 and hypercapnic hyperemia.3537 An increase in the capillary diameter allows a higher blood flow volume and therefore increased tissue oxygenation.38 A possible hypothesis for an increased capillary diameter in patients with cardiovascular comorbidities but no known ophthalmic disease, may be due to a relatively hypoxic retina in these eyes compared with normal eyes, leading to a compensatory increase in capillary diameter to provide nourishment to hypoxia-sensitive neuronal tissue. These findings allow us to further understand the relationship between cardiovascular comorbidities and retinal vascular disease processes. 
Even though all donors had no known presence of ocular disease, pathological vascular changes also were found in this study, with clinically similar appearances to microaneurysms and tortuous vasculature. Of these, microaneurysms serve as an important finding in the clinical setting, particularly with regard to diabetic retinopathy. Microaneurysms are often graded as being the first and earliest important clinical sign characteristic of diabetic retinopathy,39 and serve as a catalyst for the diagnosis of diabetic retinopathy. 
Stitt et al.40 have documented in detail four distinct groups of microaneurysms, and described the several stages in the formation of diabetic retinopathy microaneurysms. The microaneurysms we observed in our study were most likely Type I or Type II; however, given our novel technique of perfusion, we were unable to observe the polymorphonuclear or red blood cells in the lumen, and made our observations by the endothelial cell structures instead. 
Another important study written by Moore et al.41 classified microaneurysms into three categories corresponding to morphology: saccular, fusiform, and focal bulges. In our study, we predominantly saw saccular microaneurysms. 
Hypertension is another important risk factor in the development of microaneurysms. The exudative phase of retinal microvasculature anomalies usually shows classical features, such as microaneurysms and hemorrhages in hypertension.9 A recent classification on hypertensive retinopathy revealed that retinal signs, such as microaneurysm, cotton-wool spots, hemorrhages, and hard exudates, are associated with moderate hypertensive retinopathy.42 The three human donor eyes in which we observed vascular abnormalities had hypertension. This could potentially be a significant reason why these observations were noted in these eyes. Other risk factors in which microaneurysms would be important include smoking and pregnancy. 
Macula edema stemming from leaky diabetic microaneurysms is one of the leading causes of vision loss in working-aged people in the developed world43; thus, further research into this area is crucial. Knowledge in the past regarding microaneurysm structure and location have predominantly been histopathological studies with light microscopy or electron microscopy on trypsin-digested retinal flat mounts.40,44 More recently, confocal images,41 optical coherence tomography (OCT),43,45 and fluorescein angiograms46 also have been used. 
Our findings support previous data on the size and origin of the microaneurysms. Our random sampling of 30 microaneurysm diameters ranged from 29 μm to 116 μm, which was similar to the 14 μm to 136 μm range reported in another confocal study,41 and the 43 μm to 266 μm range reported in an OCT study.43 As shown in Figure 7, there were also microaneurysms seen originating in all four retinal layers, as well as occupying between one to four retinal layers. 
To our knowledge, there have been no studies that have observed the individual endothelial cells and nuclei that encapsulate the microaneurysms. Our data reveal there are multiple different-appearing nuclei framing the microaneurysm. Some nuclei were small and pyknotic-like, and other nuclei were larger, irregular in shape, and appeared to have less chromatin. Thus, the difference in the nuclei shape, size, and density could be due to the varying stages of cellular proliferation as the microaneurysm develops. Although f-actin staining was visible in the vasculature, most endothelial cell borders were indiscernible in the microaneurysm. However, this is to be expected with proliferating cells that have not completely differentiated. 
Another finding in one of the diseased eyes (donor G) showed a particularly tortuous vessel. Tortuosity of vasculature in the human body is common and has been linked with diabetes mellitus, atherosclerosis, hypertension, aging, and genetic defects.4750 It has been found to be generally asymptomatic when mild; however, when severe could lead to end-organ ischemia.51 
Specifically, retinal vessel abnormalities, such as retinal vasculature tortuosity, have been found to be in association with diabetes mellitus, hypertension, retinopathy, genetic disorders, lower high-density lipoprotein cholesterol level, ischemic heart disease, stroke, and cerebral vessel disease.5262 
Our study found the tortuous vessel to be draining into a main retinal vein. Many more studies research arterial tortuosity; however, there are few that examine venous tortuosity. In one population-based cross-sectional study, it was found that retinal venules were significantly more tortuous than retinal arterioles.60 As it is known that the venous system is highly complex, with diseases being 10 times more frequent in the venous than arterial system,63,64 it is thus important to study the venous system of the eye in more depth.65 
Last, there were multiple retinal vascular abnormalities seen in two diseased eyes, up to 30 per quadrant. As they had a “burst” appearance with minor leakage, they potentially may be representative of burst leaky microaneurysms. It is unclear whether this may have occurred as part of the perfusion and staining process from our novel technique, or whether these were already present. As we closely monitor the input perfusate pressure at the entry point to the eye and keep it in the physiological range, it is most likely that these leakage sites were already present in the diseased eye. 
The relationship between microaneurysms and leakage is not yet understood66; however, understanding its occurrences and the processes of leaky microaneurysms resulting in vision loss would be important. We have been able to demonstrate a three-dimensional reconstruction of retinal microaneurysms in Figure 6. Further research into areas such as microaneurysms, tortuous vasculature, and retinal lesions using three-dimensional techniques could be used to obtain more detailed information that we have otherwise been unable to obtain using two-dimensional studies. 
We acknowledge several limitations of this study. First, the sample size is too small to conduct detailed studies on specific cardiovascular comorbidities or risk factors. Second, due to the limitation with all postmortem histopathologic studies, we were unable to determine all the details of the patient's systemic past medical history. We specifically attempted to provide information on the latest eye examinations for each donor eye; however, due to confidentiality issues, we were unable to obtain this information, and thus is a potential limitation, particularly with regard to the nature of progression in retinal diseases. Confounding variables also may have been present, but all efforts were made to choose donor eyes carefully by evaluating all known medical history available, and including data from donor eyes with as few known variables as possible. In addition, most of the quantitative findings of this study were generally limited to the region 3 mm superior to the optic disc, and may be relevant only to this area. 
In conclusion, we found cardiovascular comorbidities were associated with a significant decrease in the capillary density of the human inner retina. Also, pathological vascular changes were described. These findings may suggest that retinal vascular diseases have the potential to commence earlier than clinically detectable retinal diseases. 
Acknowledgments
The authors thank the staff of the Lions Eye Bank of Western Australia, Lions Eye Institute, for providing human donor eyes; the staff of DonateWest, the Western Australian agency for organ and tissue donation, who facilitated the recruitment of donors into the study by referral and completion of consent processes; and Dean Darcey for his expert technical assistance. 
Supported by the National Health and Medical Research Council of Australia and the Australian Research Council Centre of Excellence in Vision Science 
Disclosure: P.E.Z. Tan, None; P.K. Yu, None; S.J. Cringle, None; D.-Y. Yu, None 
References
Wong TY Klein R Nieto FJ Retinal microvascular abnormalities and 10-year cardiovascular mortality: a population-based case-control study. Ophthalmology. 2003; 110: 933–940. [CrossRef] [PubMed]
Cheung N Wang JJ Klein R Diabetic retinopathy and the risk of coronary heart disease: the Atherosclerosis Risk in Communities Study. Diabetes Care. 2007; 30: 1742–1746. [CrossRef] [PubMed]
Wong TY Rosamond W Chang PP Retinopathy and risk of congestive heart failure. J Am Med Assoc. 2005; 293: 63–69. [CrossRef]
Chan G Balaratnasingam C Yu PK Quantitative changes in perifoveal capillary networks in patients with vascular comorbidities. Invest Ophthalmol Vis Sci. 2013; 54: 5175–5185. [CrossRef] [PubMed]
Tan PE Yu PK Balaratnasingam C Quantitative confocal imaging of the retinal microvasculature in the human retina. Invest Ophthalmol Vis Sci. 2012; 53: 5728–5736. [CrossRef] [PubMed]
Yu PK Tan PE Morgan WH Age-related changes in venous endothelial phenotype at human retinal artery-vein crossing points. Invest Ophthalmol Vis Sci. 2012; 53: 1108–1116. [CrossRef] [PubMed]
Yu PK Balaratnasingam C Cringle SJ Microstructure and network organization of the microvasculature in the human macula. Invest Ophthalmol Vis Sci. 2010; 51: 6735–6743. [CrossRef] [PubMed]
Chan G Balaratnasingam C Yu PK Quantitative morphometry of perifoveal capillary networks in the human retina. Invest Ophthalmol Vis Sci. 2012; 53: 5502–5514. [CrossRef] [PubMed]
Wong TY Klein R Klein BE Retinal microvascular abnormalities and their relationship with hypertension, cardiovascular disease, and mortality. Surv Ophthalmol. 2001; 46: 59–80. [CrossRef] [PubMed]
Yu PK Balaratnasingam C Morgan WH The structural relationship between the microvasculature, neurons, and glia in the human retina. Invest Ophthalmol Vis Sci. 2010; 51: 447–458. [CrossRef] [PubMed]
R Development Core Team. R: A Language and Environment for Statistical Computing. Vienna, Austria: R Foundation for Statistical Computing; 2011.
Attwell D Buchan AM Charpak S Glial and neuronal control of brain blood flow. Nature. 2010; 468: 232–243. [CrossRef] [PubMed]
Drake CT Iadecola C. The role of neuronal signaling in controlling cerebral blood flow. Brain and Language. 2007; 102: 141–152. [CrossRef] [PubMed]
Yu DY Cringle SJ Yu PK Su EN. Intraretinal oxygen distribution and consumption during retinal artery occlusion and graded hyperoxic ventilation in the rat. Invest Ophthalmol Vis Sci. 2007; 48: 2290–2296. [CrossRef] [PubMed]
Yu D-Y Cringle SJ. Oxygen distribution and consumption within the retina in vascularised and avascular retinas and in animal models of retinal disease. Prog Retina Eye Res. 2001; 20: 175–208. [CrossRef]
Jespersen SN Ostergaard L. The roles of cerebral blood flow, capillary transit time heterogeneity, and oxygen tension in brain oxygenation and metabolism. J Cereb Blood Flow Metab. 2012; 32: 264–277. [CrossRef] [PubMed]
Pournaras CJ Rungger-Brandle E Riva CE Hardarson SH Stefansson E. Regulation of retinal blood flow in health and disease. Prog Retin Eye Res. 2008; 27: 284–330. [CrossRef] [PubMed]
Klaassen I Van Noorden CJ Schlingemann RO. Molecular basis of the inner blood-retinal barrier and its breakdown in diabetic macular edema and other pathological conditions. Prog Retin Eye Res. 2013; 34: 19–48. [CrossRef] [PubMed]
Cunha-Vaz J Bernardes R Lobo C. Blood-retinal barrier. Eur J Ophthalmol. 2011; 21: S3–S9.
Cunha-Vaz JG. The blood-retinal barriers. Doc Ophthalmol. 1976; 41: 287–327. [CrossRef] [PubMed]
Shaver SW Pang JJ Wall KM Sposito NM Gross PM. Subregional topography of capillaries in the dorsal vagal complex of rats: I. Morphometric properties. J Comp Neurol. 1991; 306: 73–82. [CrossRef] [PubMed]
Shaver SW Sposito NM Gross PM. Quantitative fine structure of capillaries in subregions of the rat subfornical organ. J Comp Neurol. 1990; 294: 145–152. [CrossRef] [PubMed]
Cheung CY Ikram MK Sabanayagam C Wong TY. Retinal microvasculature as a model to study the manifestations of hypertension. Hypertension. 2012; 60: 1094–1103. [CrossRef] [PubMed]
Wong TY Hubbard LD Klein R Retinal microvascular abnormalities and blood pressure in older people: the Cardiovascular Health Study. Br J Ophthalmol. 2002; 86: 1007–1013. [CrossRef] [PubMed]
Leung H Wang JJ Rochtchina E Impact of current and past blood pressure on retinal arteriolar diameter in an older population. J Hypertens. 2004; 22: 1543–1549. [CrossRef] [PubMed]
De Silva DA Woon FP Manzano JJ The relationship between aortic stiffness and changes in retinal microvessels among Asian ischemic stroke patients. J Hum Hypertens. 2012; 26: 716–722. [CrossRef] [PubMed]
Crosby-Nwaobi R Heng LZ Sivaprasad S. Retinal vascular calibre, geometry and progression of diabetic retinopathy in type 2 diabetes mellitus. Ophthalmologica. 2012; 228: 84–92. [CrossRef] [PubMed]
Cheung CY Thomas GN Tay W Retinal vascular fractal dimension and its relationship with cardiovascular and ocular risk factors. Am J Ophthalmol. 2012; 154: 663–674. [CrossRef] [PubMed]
Wang JJ Liew G Wong TY Retinal vascular calibre and the risk of coronary heart disease-related death. Heart. 2006; 92: 1583–1587. [CrossRef] [PubMed]
Sluimer JC Kolodgie FD Bijnens AP Thin-walled microvessels in human coronary atherosclerotic plaques show incomplete endothelial junctions relevance of compromised structural integrity for intraplaque microvascular leakage. J Am Coll Cardiol. 2009; 53: 1517–1527. [CrossRef] [PubMed]
Akagi Y Kador PF Kuwabara T Kinoshita JH. Aldose reductase localization in human retinal mural cells. Invest Ophthalmol Vis Sci. 1983; 24: 1516–1519. [PubMed]
Wolf S Arend O Schulte K Ittel TH Reim M. Quantification of retinal capillary density and flow velocity in patients with essential hypertension. Hypertension. 1994; 23: 464–467. [CrossRef] [PubMed]
Leulescu C. The pathogenesis of cotton-wool spots [in Romanian]. Oftalmologia. 1997; 41: 296–299. [PubMed]
Levy MN Sauceda G. Diffusion of oxygen from arterial to venous segments of renal capillaries. Am J Physiol. 1959; 196: 1336–1339. [PubMed]
Duelli R Kuschinsky W. Changes in brain capillary diameter during hypocapnia and hypercapnia. J Cereb Blood Flow Metab. 1993; 13: 1025–1028. [CrossRef] [PubMed]
Stefanovic B Hutchinson E Yakovleva V Functional reactivity of cerebral capillaries. J Cereb Blood Flow Metab. 2008; 28: 961–972. [CrossRef] [PubMed]
Atkinson JL Anderson RE Sundt TM Jr. The effect of carbon dioxide on the diameter of brain capillaries. Brain Res. 1990; 517: 333–340. [CrossRef] [PubMed]
van der Linden P De HS Belisle S Comparative effects of red blood cell transfusion and increasing blood flow on tissue oxygenation in oxygen supply-dependent conditions. Am J Respir Crit Care Med. 2001; 163: 1605–1608. [CrossRef] [PubMed]
Massin P Erginay A Gaudric A. Rétinopathie diabétique. Paris: Elsevier, 2000.
Stitt AW Gardiner TA Archer DB. Histological and ultrastructural investigation of retinal microaneurysm development in diabetic patients. Br J Ophthalmol. 1995; 79: 362–367. [CrossRef] [PubMed]
Moore J Bagley S Ireland G McLeod D Boulton ME. Three dimensional analysis of microaneurysms in the human diabetic retina. J Anat. 1999; 194: 89–100. [CrossRef] [PubMed]
Wong TY Mitchell P. Hypertensive retinopathy. N Engl J Med. 2004; 351: 2310–2317. [CrossRef] [PubMed]
Wang H Chhablani J Freeman WR Characterization of diabetic microaneurysms by simultaneous fluorescein angiography and spectral-domain optical coherence tomography. Am J Ophthalmol. 2012; 153: 861–867. [CrossRef] [PubMed]
Fryczkowski AW Chambers RB Craig EJ Walker J Davidorf FH. Scanning electron microscopic study of microaneurysms in the diabetic retina. Ann Ophthalmol. 1991; 23: 130–136. [PubMed]
Horii T Murakami T Nishijima K Optical coherence tomographic characteristics of microaneurysms in diabetic retinopathy. Am J Ophthalmol. 2010; 150: 840–848. [CrossRef] [PubMed]
Spencer T Phillips RP Sharp PF Forrester JV. Automated detection and quantification of microaneurysms in fluorescein angiograms. Graefes Arch Clin Exp Ophthalmol. 1992; 230: 36–41. [CrossRef] [PubMed]
Owen CG Newsom RS Rudnicka AR Diabetes and the tortuosity of vessels of the bulbar conjunctiva. Ophthalmology. 2008; 115: e27–e32. [CrossRef] [PubMed]
Pancera P Ribul M Presciuttini B Lechi A. Prevalence of carotid artery kinking in 590 consecutive subjects evaluated by Echocolordoppler. Is there a correlation with arterial hypertension? J Intern Med. 2000; 248: 7–12.
Hiroki M Miyashita K Oda M. Tortuosity of the white matter medullary arterioles is related to the severity of hypertension. Cerebrovasc Dis. 2002; 13: 242–250. [CrossRef] [PubMed]
Callewaert BL Willaert A Kerstjens-Frederikse WS Arterial tortuosity syndrome: clinical and molecular findings in 12 newly identified families. Hum Mutat. 2008; 29: 150–158. [CrossRef] [PubMed]
Han HC. Twisted blood vessels: symptoms, etiology and biomechanical mechanisms. J Vasc Res. 2012; 49: 185–197. [CrossRef] [PubMed]
Taarnhoj NC Munch IC Sander B Straight versus tortuous retinal arteries in relation to blood pressure and genetics. Br J Ophthalmol. 2008; 92: 1055–1060. [CrossRef] [PubMed]
Hayreh SS Servais GE Virdi PS. Retinal arteriolar changes in malignant arterial hypertension. Ophthalmologica. 1989; 198: 178–196. [CrossRef] [PubMed]
Hughes AD Martinez-Perez E Jabbar AS Quantification of topological changes in retinal vascular architecture in essential and malignant hypertension. J Hypertens. 2006; 24: 889–894. [CrossRef] [PubMed]
Amemiya T Bhutto IA. Retinal vascular changes and systemic diseases: corrosion cast demonstration. Ital J Anat Embryol. 2001; 106: 237–244. [PubMed]
Sutter FK Helbig H. Familial retinal arteriolar tortuosity: a review. Surv Ophthalmol. 2003; 48: 245–255. [CrossRef] [PubMed]
Chaudhuri S Chatterjee S Katz N Nelson M Goldbaum M. Detection of blood vessels in retinal images using two-dimensional matched filters. IEEE Trans Med Imaging. 1989; 8: 263–269. [CrossRef] [PubMed]
Streicher T Spirkova J. Tortuous retinal vessels [in Slovak]. Cesk Oftalmol. 1993; 49: 246–251. [PubMed]
Vahedi K Massin P Guichard JP Hereditary infantile hemiparesis, retinal arteriolar tortuosity, and leukoencephalopathy. Neurology. 2003; 60: 57–63. [CrossRef] [PubMed]
Cheung CY Zheng Y Hsu W Retinal vascular tortuosity, blood pressure, and cardiovascular risk factors. Ophthalmology. 2011; 118: 812–818. [CrossRef] [PubMed]
Kwa VI van der Sande JJ Stam J Tijmes N Vrooland JL. Retinal arterial changes correlate with cerebral small-vessel disease. Neurology. 2002; 59: 1536–1540. [CrossRef] [PubMed]
Witt N Wong TY Hughes AD Abnormalities of retinal microvascular structure and risk of mortality from ischemic heart disease and stroke. Hypertension. 2006; 47: 975–981. [CrossRef] [PubMed]
Meissner MH Moneta G Burnand K The hemodynamics and diagnosis of venous disease. J Vasc Surg. 2007; 46: 4S–24S. [CrossRef] [PubMed]
Monos E Berczi V Nadasy G. Local control of veins: biomechanical, metabolic, and humoral aspects. Physiol Rev. 1995; 75: 611–666. [PubMed]
Tan PE Yu PK Cringle SJ Morgan WH Yu DY. Regional heterogeneity of endothelial cells in the porcine vortex vein system. Microvasc Res. 2013; 89: 70–79. [CrossRef] [PubMed]
Ezra E Keinan E Mandel Y Boulton ME Nahmias Y. Non-dimensional analysis of retinal microaneurysms: critical threshold for treatment. Integr Biol (Camb). 2013; 5: 474–480. [CrossRef] [PubMed]
Figure 1
 
Comparison of the capillary network in control and diseased groups. (AD) Representative images of retinal vasculature from control group ([A] donor J; [B] donor M; [C] and [D] donor K). (EH) Representative images of retinal vasculature from diseased group (donor D). Whole-mount confocal microscopy images capture both vascular and nuclei information at 3 mm superior to the optic disc. Phalloidin labeled the endothelial f-actin, and Hoechst or YO-PRO-1 labeled the nuclei. (A) The vasculature of the NFL from the control group demonstrates a higher density of vasculature compared with the vasculature from the diseased group at the same region (E). This is also shown in the comparison of the vasculature in the RGC layer from the control group (B) and the diseased group (F) in the same region. In the RGC layer, venules and arterioles also are observed in addition to capillaries. (C) The control capillary network that lies between the IPL and the superficial INL; (G) the capillary network in this region from the diseased group. (D) and (H) display the deepest capillary network of the control and diseased groups respectively, and lie at the border of the deep INL and OPL. Scale bar: 100 μm.
Figure 1
 
Comparison of the capillary network in control and diseased groups. (AD) Representative images of retinal vasculature from control group ([A] donor J; [B] donor M; [C] and [D] donor K). (EH) Representative images of retinal vasculature from diseased group (donor D). Whole-mount confocal microscopy images capture both vascular and nuclei information at 3 mm superior to the optic disc. Phalloidin labeled the endothelial f-actin, and Hoechst or YO-PRO-1 labeled the nuclei. (A) The vasculature of the NFL from the control group demonstrates a higher density of vasculature compared with the vasculature from the diseased group at the same region (E). This is also shown in the comparison of the vasculature in the RGC layer from the control group (B) and the diseased group (F) in the same region. In the RGC layer, venules and arterioles also are observed in addition to capillaries. (C) The control capillary network that lies between the IPL and the superficial INL; (G) the capillary network in this region from the diseased group. (D) and (H) display the deepest capillary network of the control and diseased groups respectively, and lie at the border of the deep INL and OPL. Scale bar: 100 μm.
Figure 2
 
Graphs showing the averaged data of capillary diameter and density in the control and diseased groups in each of the capillary networks of the retina at 3 mm superior to the optic disc. Layer 1 = NFL, layer 2 = RGC layer, layer 3 = border of IPL and superficial INL, layer 4 = border of deep INL and OPL. The diameter bar graph shows there were no significant differences in each of the capillary networks (P > 0.05). Density box plots show the mean value and the SEs. Significant differences (P < 0.05) were noted between the densities of the control versus diseased eyes in layer 1 and layer 2, as indicated by the asterisk.
Figure 2
 
Graphs showing the averaged data of capillary diameter and density in the control and diseased groups in each of the capillary networks of the retina at 3 mm superior to the optic disc. Layer 1 = NFL, layer 2 = RGC layer, layer 3 = border of IPL and superficial INL, layer 4 = border of deep INL and OPL. The diameter bar graph shows there were no significant differences in each of the capillary networks (P > 0.05). Density box plots show the mean value and the SEs. Significant differences (P < 0.05) were noted between the densities of the control versus diseased eyes in layer 1 and layer 2, as indicated by the asterisk.
Figure 3
 
Graphs showing the average capillary diameter and density in the young (<60 years old) and elderly (>60 years old) groups in each of the different capillary networks. Layer 1 = NFL, layer 2 = RGC layer, layer 3 = border of IPL and superficial INL, layer 4 = border of deep INL and OPL. The diameter bar graph shows there were no significant differences in each of the capillary networks (P > 0.05) in those older than or younger than 60 years old. The density box plot shows the mean value and the SEs. There were no significant differences (P > 0.05) between all capillary layers.
Figure 3
 
Graphs showing the average capillary diameter and density in the young (<60 years old) and elderly (>60 years old) groups in each of the different capillary networks. Layer 1 = NFL, layer 2 = RGC layer, layer 3 = border of IPL and superficial INL, layer 4 = border of deep INL and OPL. The diameter bar graph shows there were no significant differences in each of the capillary networks (P > 0.05) in those older than or younger than 60 years old. The density box plot shows the mean value and the SEs. There were no significant differences (P > 0.05) between all capillary layers.
Figure 4
 
(A) Confocal projected image of 24 optical sections that show capillary microaneurysms within donor H's retinal vasculature from the diseased group using the ×20 objective lens. Saccular microaneurysms (arrowheads) were predominantly noted. (B) Confocal projected image of 17 optical sections with the ×60 objective lens shows a microaneurysm that is seen as saccular capillary outpouching containing proliferated endothelial cells. (C) A z-series of 110 optical sections with the ×60 objective lens captures a round-shaped saccular microaneurysm among a background of numerous nuclei of neural cells in the INL. (D) A magnified image of (C) with fewer stacked images of 72 optical sections, to allow identification of the proliferated endothelial cells on the wall. The f-actin staining on the wall of this microaneurysm shows remarkably uneven intensity. The cell boundary of most endothelial cells cannot be clearly defined except an endothelial cell (outlined in yellow). The difference in nuclei size and density staining of the endothelial cells also is found in this microaneurysm. (E) and (F) are schematic diagrams of images (C) and (D), respectively. Both (E) and (F) show the outline of microaneurysms and highlight the difference in nuclei size and density. Some nuclei are relatively larger with a speckled chromatin-filled appearance shown in purple (F), and some nuclei show relatively dark staining (pyknotic-like appearance) shown in blue (F). These changes of endothelial cells may indicate varying stages of endothelial proliferation. Scale bar: (A) 100 μm, (B) 17.5 μm, (C) 33.3 μm, (D) 15.9 μm.
Figure 4
 
(A) Confocal projected image of 24 optical sections that show capillary microaneurysms within donor H's retinal vasculature from the diseased group using the ×20 objective lens. Saccular microaneurysms (arrowheads) were predominantly noted. (B) Confocal projected image of 17 optical sections with the ×60 objective lens shows a microaneurysm that is seen as saccular capillary outpouching containing proliferated endothelial cells. (C) A z-series of 110 optical sections with the ×60 objective lens captures a round-shaped saccular microaneurysm among a background of numerous nuclei of neural cells in the INL. (D) A magnified image of (C) with fewer stacked images of 72 optical sections, to allow identification of the proliferated endothelial cells on the wall. The f-actin staining on the wall of this microaneurysm shows remarkably uneven intensity. The cell boundary of most endothelial cells cannot be clearly defined except an endothelial cell (outlined in yellow). The difference in nuclei size and density staining of the endothelial cells also is found in this microaneurysm. (E) and (F) are schematic diagrams of images (C) and (D), respectively. Both (E) and (F) show the outline of microaneurysms and highlight the difference in nuclei size and density. Some nuclei are relatively larger with a speckled chromatin-filled appearance shown in purple (F), and some nuclei show relatively dark staining (pyknotic-like appearance) shown in blue (F). These changes of endothelial cells may indicate varying stages of endothelial proliferation. Scale bar: (A) 100 μm, (B) 17.5 μm, (C) 33.3 μm, (D) 15.9 μm.
Figure 5
 
(A) A large sausage-shaped, dilated, and tortuous retinal venule with irregularities in f-actin staining and shape in the vessel wall among sparse vasculature from donor G in the diseased group. It comprises 19 optical sections to complete a confocal projected image using the ×20 objective lens. Phalloidin labels the endothelial f-actin, and Hoechst labels the nuclei. (B) The lower portion of this vascular abnormality magnified and captured using the ×60 objective lens. (C) and (D) are schematic diagrams of the vasculature in images (A) and (B), respectively. These schematics highlight with clarity the tortuosity of the vasculature. Scale bar: (A) 100 μm, (B) 33.3 μm.
Figure 5
 
(A) A large sausage-shaped, dilated, and tortuous retinal venule with irregularities in f-actin staining and shape in the vessel wall among sparse vasculature from donor G in the diseased group. It comprises 19 optical sections to complete a confocal projected image using the ×20 objective lens. Phalloidin labels the endothelial f-actin, and Hoechst labels the nuclei. (B) The lower portion of this vascular abnormality magnified and captured using the ×60 objective lens. (C) and (D) are schematic diagrams of the vasculature in images (A) and (B), respectively. These schematics highlight with clarity the tortuosity of the vasculature. Scale bar: (A) 100 μm, (B) 33.3 μm.
Figure 6
 
(A) Confocal projected image of the beginnings of a small retinal vascular abnormality comprising 61 optical sections using the ×20 objective lens from donor H. Claudin-5 labels the transmembrane tight junction protein, Hoechst labels the nuclei, glial fibrillary acidic protein labels the astrocytes, and cellular retinaldehyde-binding protein labels the Muller cells. (B) Three-dimensional reconstruction of (A) showing the diseased vasculature from the scleral surface. Vasculature is labeled in red, the Muller cell processes in blue, and the astrocytes and nuclei are labeled in green. The white arrow points toward microaneurysms in (A) and correlates with the three-dimensional reconstructions in (B), and the yellow arrows point toward a larger retinal vascular abnormality. The retinal vascular changes are seen to be in closer association with the Muller cells, but a limited association with the astrocytes. Scale bar: (A) 100 μm, (B) 200 μm.
Figure 6
 
(A) Confocal projected image of the beginnings of a small retinal vascular abnormality comprising 61 optical sections using the ×20 objective lens from donor H. Claudin-5 labels the transmembrane tight junction protein, Hoechst labels the nuclei, glial fibrillary acidic protein labels the astrocytes, and cellular retinaldehyde-binding protein labels the Muller cells. (B) Three-dimensional reconstruction of (A) showing the diseased vasculature from the scleral surface. Vasculature is labeled in red, the Muller cell processes in blue, and the astrocytes and nuclei are labeled in green. The white arrow points toward microaneurysms in (A) and correlates with the three-dimensional reconstructions in (B), and the yellow arrows point toward a larger retinal vascular abnormality. The retinal vascular changes are seen to be in closer association with the Muller cells, but a limited association with the astrocytes. Scale bar: (A) 100 μm, (B) 200 μm.
Figure 7
 
Whole-mount confocal images capture the four different vascular layers in a diseased eye (donor H) at 6 mm nasal to the optic disc. (A) to (D) show Claudin-5 labeling the transmembrane tight junction proteins and particularly highlight the microaneurysms. (E) to (H) demonstrate labeling of Claudin-5 (green), glial fibrillary acidic protein (blue), and cellular retinaldehyde-binding protein (red). After merging the images, the microaneurysms appear yellow, demonstrating the colocalization of the Claudin-5 and cellular retinaldehyde-binding protein. Microaneurysms are seen to be present in all layers of the retinal vasculature. As they are relatively large compared with the capillaries, some of the microaneurysms are seen to traverse several retinal vasculature layers. (A) and (E) NFL (layer 1); (B) and (F) RGC layer (layer 2); (C) and (G) border of IPL and superficial INL (layer 3); (D) and (H) border of deep INL and OPL (layer 4). Scale bar: 100 μm.
Figure 7
 
Whole-mount confocal images capture the four different vascular layers in a diseased eye (donor H) at 6 mm nasal to the optic disc. (A) to (D) show Claudin-5 labeling the transmembrane tight junction proteins and particularly highlight the microaneurysms. (E) to (H) demonstrate labeling of Claudin-5 (green), glial fibrillary acidic protein (blue), and cellular retinaldehyde-binding protein (red). After merging the images, the microaneurysms appear yellow, demonstrating the colocalization of the Claudin-5 and cellular retinaldehyde-binding protein. Microaneurysms are seen to be present in all layers of the retinal vasculature. As they are relatively large compared with the capillaries, some of the microaneurysms are seen to traverse several retinal vasculature layers. (A) and (E) NFL (layer 1); (B) and (F) RGC layer (layer 2); (C) and (G) border of IPL and superficial INL (layer 3); (D) and (H) border of deep INL and OPL (layer 4). Scale bar: 100 μm.
Table 1
 
Donor Demographic Details: Age, Sex, Eye, Vascular Comorbidities, Cause of Death, and Time to Cannulation for Donor Eyes
Table 1
 
Donor Demographic Details: Age, Sex, Eye, Vascular Comorbidities, Cause of Death, and Time to Cannulation for Donor Eyes
Donor ID Age, y Sex Eye Cardiovascular Comorbidities Cause of Death Time to Cannulation, h
A 68 M L Ischemic heart disease, hypertension Cerebral vascular accident 11.5
B 41 M R Atherosclerosis Intracranial hemorrhage 17.0
C 51 M L + R Ischemic heart disease Myocardial infarction 19.0
D 65 M L + R Cerebral vascular accident Intracranial hemorrhage 14.0
E 64 M L Ischemic heart disease Cardiogenic shock 14.0
F 65 M R Cardiomyopathy Heart failure 6.5
G 75 F L* Hypertension, AF, vascular disease Cerebral vascular accident 18.0
H 72 M L* + R* Hypertension, hypercholesterolaemia Myocardial infarction 12.0
I 22 M L MVA 15.0
J 32 M L MVA 20.0
K 49 F L + R Cancer 5.5
L 67 M L Cancer 9.5
M 23 M L Suicide 22.0
N 66 M L + R Cancer 15.0
O 22 M L Suicide 15.0
P 53 M L MVA 14.0
Q 60 M L Collapse 18.0
R 72 M R Drowning 15.0
S 72 F L Sepsis 3.0
T 60 M R Liver disease 3.5
U 59 M R Melanoma 12.0
V 39 M L Sepsis 20.0
Table 2
 
Quantitative Capillary Diameter and Density Data for Each Network of the Diseased Human Eyes
Table 2
 
Quantitative Capillary Diameter and Density Data for Each Network of the Diseased Human Eyes
Capillary Network NFL RGC IPL/Superficial INL Deep INL/OPL
Capillary diameter, μm 8.86 ± 0.09 (495) 8.93 ± 0.09 (495) 8.93 ± 0.10 (495) 8.96 ± 0.10 (495)
Total capillary density, % 8.15 ± 0.01 (11) 21.42 ± 0.02 (11) 9.69 ± 0.01 (11) 14.19 ± 0.01 (11)
Table 3
 
Morphometric Comparisons Between Control and Diseased Groups
Table 3
 
Morphometric Comparisons Between Control and Diseased Groups
Capillary Network NFL RGC IPL/Superficial INL Deep INL/OPL
Control Diseased Control Diseased Control Diseased Control Diseased
Capillary diameter, μm 8.47 ± 0.05 (540) 8.86 ± 0.09 (495) 8.01 ± 0.05 (540) 8.93 ± 0.09 (495) 7.99 ± 0.06 (540) 8.93 ± 0.10 (495) 8.56 ± 0.06 (540) 8.96 ± 0.10 (495)
Total capillary density, % 13.69 ± 0.01 (12) 8.15 ± 0.01 (11) 26.74 ± 0.01 (12) 21.42 ± 0.02 (11) 11.28 ± 0.01 (12) 9.69 ± 0.01 (11) 16.12 ± 0.01 (12) 14.19 ± 0.01 (11)
<60yo >60yo <60yo >60yo <60yo >60yo <60yo >60yo
Capillary diameter, μm 8.48 ± 0.05 (540) 8.85 ± 0.09 (495) 8.27 ± 0.06 (540) 8.65 ± 0.09 (495) 8.25 ± 0.06 (540) 8.65 ± 0.10 (495) 8.58 ± 0.06 (540) 8.94 ± 0.10 (495)
Total capillary density, % 12.02 ± 0.01 (12) 9.96 ± 0.02 (11) 26.43 ± 0.01 (12) 21.76 ± 0.02 (11) 10.50 ± 0.01 (12) 10.54 ± 0.01 (11) 15.51 ± 0.01 (12) 14.86 ± 0.01 (11)
×
×

This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.

×