August 2015
Volume 56, Issue 9
Free
Retina  |   August 2015
Quantitative Noninvasive Angiography of the Fovea Centralis Using Speckle Variance Optical Coherence Tomography
Author Affiliations & Notes
  • Zaid Mammo
    Department of Ophthalmology and Visual Sciences University of British Columbia, Vancouver, Canada
  • Chandrakumar Balaratnasingam
    Department of Ophthalmology and Visual Sciences University of British Columbia, Vancouver, Canada
    Department of Physiology and Pharmacology, Centre for Ophthalmology and Visual Science, Lions Eye Institute, The University of Western Australia, Nedlands, Australia
    Vitreous Retina Macula Consultants of New York, New York, New York, United States
    Luesther T. Mertz Retinal Research Center, Manhattan, Eye, Ear and Throat Hospital, New York, New York, United States
  • Paula Yu
    Department of Physiology and Pharmacology, Centre for Ophthalmology and Visual Science, Lions Eye Institute, The University of Western Australia, Nedlands, Australia
  • Jing Xu
    School of Engineering Science, Simon Fraser University, Burnaby, British Columbia, Canada
  • Morgan Heisler
    School of Engineering Science, Simon Fraser University, Burnaby, British Columbia, Canada
  • Paul Mackenzie
    Department of Ophthalmology and Visual Sciences University of British Columbia, Vancouver, Canada
  • Andrew Merkur
    Department of Ophthalmology and Visual Sciences University of British Columbia, Vancouver, Canada
  • Andrew Kirker
    Department of Ophthalmology and Visual Sciences University of British Columbia, Vancouver, Canada
  • David Albiani
    Department of Ophthalmology and Visual Sciences University of British Columbia, Vancouver, Canada
  • K. Bailey Freund
    Vitreous Retina Macula Consultants of New York, New York, New York, United States
    Luesther T. Mertz Retinal Research Center, Manhattan, Eye, Ear and Throat Hospital, New York, New York, United States
  • Marinko V. Sarunic
    School of Engineering Science, Simon Fraser University, Burnaby, British Columbia, Canada
  • Dao-Yi Yu
    Department of Physiology and Pharmacology, Centre for Ophthalmology and Visual Science, Lions Eye Institute, The University of Western Australia, Nedlands, Australia
  • Correspondence: Chandrakumar Balaratnasingam, Centre for Ophthalmology and Visual Science, The University of Western Australia, Nedlands, WA 6009; balaratnasingam@gmail.com
Investigative Ophthalmology & Visual Science August 2015, Vol.56, 5074-5086. doi:10.1167/iovs.15-16773
  • Views
  • PDF
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Zaid Mammo, Chandrakumar Balaratnasingam, Paula Yu, Jing Xu, Morgan Heisler, Paul Mackenzie, Andrew Merkur, Andrew Kirker, David Albiani, K. Bailey Freund, Marinko V. Sarunic, Dao-Yi Yu; Quantitative Noninvasive Angiography of the Fovea Centralis Using Speckle Variance Optical Coherence Tomography. Invest. Ophthalmol. Vis. Sci. 2015;56(9):5074-5086. doi: 10.1167/iovs.15-16773.

      Download citation file:


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

      ×
  • Supplements
Abstract

Purpose: To demonstrate the utility of speckle variance optical coherence tomography (svOCT), a noninvasive angiographic technique, for evaluating the foveal vasculature.

Methods: Twelve normal human eyes were imaged with svOCT (1060-nm, 100-kHz custom-built system) and fluorescein angiography (FA; Topcon TRC-50DX with 5.0 megapixel resolution camera). Manual tracing techniques were used to quantify the foveal vasculature, including foveal avascular zone (FAZ) metrics (area, perimeter, greatest diameter, and lowest diameter). Reproducibility of these measurements was determined. The FAZ was imaged in 25 normal eyes using svOCT and 15 donor eyes using confocal scanning laser microscopy. Retinal capillary plexuses in donor eyes were perfusion-labeled with phalloidin conjugated to Alexa Fluor 546.

Results: Speckle variance OCT is able to stratify the foveal circulation into inner and deep capillary plexuses as well as reliably quantify and assess the morphometric dimensions of the human FAZ. Capillary density measurements were significantly greater in svOCT than FA (31.2 ± 1.6% vs. 19.3 ± 1.9% of total tissue area; P < 0.001). Measurements were highly reproducible (all P > 0.366). All FAZ metrics were significantly lower in histology than svOCT (all P < 0.001).

Conclusions: Speckle variance OCT permits precise, reproducible, and noninvasive visualization of the human foveal vasculature. Speckle variance OCT may become an important adjunct in evaluating patients with retinal vascular diseases.

The fovea centralis is a highly specialized region that represents the area of greatest visual acuity. The microanatomy of the fovea centralis is unique, and structures within it are highly dependent on layered capillary plexuses for the metabolic exchange of nutrients and toxic substrates.1,2 Histological correlation between capillary plexus compartmentalization and foveal topography has shown that capillary planes are located within, or adjacent to, regions of high energy demands.35 Understanding the topological characteristics of the foveal microcirculation is therefore critical for delineating vasculogenic mechanisms that govern foveal homeostasis. The physiological dependence of foveal neurons on vascular elements may also explain why the fovea centralis is highly vulnerable to insults that target regional capillary structures. 
Valuable pathobiological insights have been discerned from studies that have investigated the consequences of systemic and retinal diseases on the foveal circulation.68 Most of these studies have examined the macular circulation in the context of diabetic retinopathy, which remains a major cause of visual morbidity globally.9 Disruption of the terminal capillary ring, enlargement of the foveal avascular zone (FAZ), and reduction in foveal capillary density have all been negatively correlated with visual potential, suggesting that they may be useful surrogate markers of foveal health.68 Many of these previous studies used fluorescein angiography (FA) and, due to the morbidities associated with the administration of intravenous contrast, were largely limited to the examination of a single time point. A noninvasive and reproducible method for studying the macular circulation is likely to expand our understanding about retinal vascular diseases by permitting an in-depth documentation of the temporal evolution of various pathologic entities. This knowledge may in turn help refine treatment paradigms that are used for managing retinal vascular diseases. 
A number of detailed reports have demonstrated the utility of optical coherence tomography (OCT)-based angiographic techniques for visualizing the retinal circulation.1013 In addition to Doppler OCT, phase variance,14 optical micro-angiography,15 speckle variance,16 phase-resolved,17 and split spectrum amplitude decorrelation angiography,18 flow contrast imaging techniques have been shown to permit label-free, en face visualization of the retinal vasculature. Label-free visualization of the microcirculations has also been demonstrated recently using flow contrast and forward-scattering/offset pinhole techniques in combination with adaptive optics scanning laser ophthalmoscopy (AO-SLO).1922 
In this report, we use speckle variance OCT (svOCT), which was first reported by Mariampillai et al.16 as a novel technique for label-free retinal angiography, to perform a detailed, quantitative assessment of the human foveal circulation. Using quantifiable and reproducible techniques, we show that svOCT permits superior visualization of retinal capillary structures when compared with FA. We also compare svOCT images from healthy subjects with histology from a comparable group of donor eyes to demonstrate that it has the capability to precisely stratify the inner and deep foveal capillary plexuses, as well as attain useful quantitative information about the FAZ. The goal of this study was to clarify the advantages and limitations of svOCT angiography and thereby facilitate its possible integration into clinical practice. 
Methods
General
This study was approved by the human research ethics committees at the University of Western Australia and the University of British Columbia. All live-patient imaging was performed at the Eye Care Centre in Vancouver, British Columbia. Written informed consent was obtained from all subjects. All human tissue was handled according to the Tenets of the Declaration of Helsinki. 
Human Donor Eyes and Histological Labeling of Capillary Plexuses
Human donor eyes used in this report were obtained from the Lions Eye Bank of Western Australia or DonateLife (the Western Australian agency for organ donation). None of the donors had a known history of ocular disease. 
Our previously described method of perfusion staining of endothelia was used to label the retinal microvasculature in 15 human donor eyes.2 In brief, the central retinal artery was cannulated with a glass micropipette and residual blood was washed out with oxygenated Ringer's solution with 1% BSA. Following this, 4% paraformaldehyde in 0.1 M phosphate buffer was perfused for at least 20 minutes for fixation. An aldehyde-based detergent, 0.1% Triton X-100 in 0.1 M phosphate buffer solution, was then perfused for 5 to 7 minutes to aid in the permeation of endothelial cell membranes. The detergent was washed out by a further 30 minutes of 0.1 M phosphate buffer perfusion. The microfilament and cell nuclei were then labeled over the course of 2 hours using a mixture of phalloidin conjugated to Alexa Fluor 546 (30 U, A22283; Invitrogen, Carlsbad, CA, USA) and DNA label (bisbenzimide H 33258, 1.2 μg/mL; Sigma-Aldrich Corp., St. Louis, MO, USA). Residual label was cleared from the vasculature by further perfusion of 0.1 M phosphate buffer, after which the posterior chamber was immersion fixed in 4% paraformaldehyde overnight. 
Tissue Preparation and Confocal Scanning Laser Microscopy
The posterior globe was dissected at the equator to allow viewing of the posterior retina. The macular region and the fovea were identified by the anatomic orientation of the globe, yellow pigmentation around the fovea, and greater choroidal pigmentation. The retina was carefully dissected out without inclusion of the optic disc region. A few cuts were made to the peripheral retina to enable the retina to lie flat. 
One human donor eye was used to localize the position of capillary plexuses respective to different retinal layers. This eye did not undergo perfusion staining but instead the retina was cryosectioned at −25°C and the sections that involved the central fovea were used for whole-mount immunohistochemical labeling. The frozen sections were rehydrated and permeabilized using 3% H2O2 in methanol for 10 minutes. The sections were then blocked from nonspecific staining using 10% goat serum in buffer. After several buffer washes, the sections were incubated in a mixture of primary antibodies comprising mouse anti-Goα (Millipore MAB3073, 1:100; ON-bipolar cell marker; Millipore, Billerica, MA, USA) and a rabbit anti-parvalbumin (Swant PV 25, 1:100; Horizontal cell marker; Swant, Marly, Switzerland) for 1 hour at room temperature.23 Several buffer washes were applied before further incubation in secondary antibody mixture of goat anti-mouse antibody conjugated to Alexa Fluor 488 (Invitrogen A11008, 1:200), goat anti-rabbit antibody conjugated to Alexa Fluor 555 (Invitrogen A21429, 1:200) along with Lectins-tetramethylrhodamine isothiocyanate (TRITC; Sigma L5266, 1:50; Sigma-Aldrich Corp.) and bisbenzimide H33258 (Sigma B2883, 1:1000; Sigma-Aldrich Corp.) for 30 minutes. The sections were mounted in histomount after several buffer washes over 30 minutes and imaged using confocal scanning laser microscopy. 
A confocal system equipped with three lasers (405, 488, and 532 nm; C1; Nikon, Tokyo, Japan) was used in conjunction with a fluorescence microscope (E800; Nikon) to capture images. Imaging was performed with the EZ-C1 software (v3.20; Nikon). The image captured was centered at the foveola. The macula image was divided into different regions using the criteria previously defined by Hogan et al.24 The foveola was designated to extend out to a radius of 175 μm from its center. The fovea centralis extended out a further 750 μm, with the outer boundary delineated by a circle of 925-μm radius (Fig. 1). An optical stack was then collected from the region of interest using the confocal scanning laser microscope. 
Figure 1
 
Topology of the human macula and foveal microcirculation. Markings on multispectral color fundus image (A) (Heidelberg Spectralis; Heidelberg Engineering, Carlsbad, CA, USA) are used to illustrate the dimensions of the normal human macula ([1] yellow shade, 5.5-mm diameter) and, within it, the fovea centralis ([2] orange shade, 1.85-mm diameter). These regions also are indicated on a spectral-domain OCT image (B). A human donor eye that has been labeled with multiple antibodies demonstrates the microstructural anatomy of the fovea centralis (C). The antibodies that were used to label various retinal layers are provided. The foveola, which is the central depression of the fovea centralis, is demarcated by the sloped clivus. Inset (I) illustrates the distribution of capillary plexuses (arrows) in the inner layers of the fovea centralis. A cadaveric human eye (D) that has been perfused with endothelial markers reveals the asymmetric pattern of arterioles (Art) and venules (Vn) that supply the foveal centralis. It is evident that only a few arterioles supply the terminal capillary ring that surrounds the avascular foveola. The circumference of the fovea centralis is demarcated by a yellow dashed line. INL, inner nuclear layer; IPL, inner plexiform layer; NFL, nerve fiber layer; RGC, retinal ganglion cell layer. Scale bar: 100 μm.
Figure 1
 
Topology of the human macula and foveal microcirculation. Markings on multispectral color fundus image (A) (Heidelberg Spectralis; Heidelberg Engineering, Carlsbad, CA, USA) are used to illustrate the dimensions of the normal human macula ([1] yellow shade, 5.5-mm diameter) and, within it, the fovea centralis ([2] orange shade, 1.85-mm diameter). These regions also are indicated on a spectral-domain OCT image (B). A human donor eye that has been labeled with multiple antibodies demonstrates the microstructural anatomy of the fovea centralis (C). The antibodies that were used to label various retinal layers are provided. The foveola, which is the central depression of the fovea centralis, is demarcated by the sloped clivus. Inset (I) illustrates the distribution of capillary plexuses (arrows) in the inner layers of the fovea centralis. A cadaveric human eye (D) that has been perfused with endothelial markers reveals the asymmetric pattern of arterioles (Art) and venules (Vn) that supply the foveal centralis. It is evident that only a few arterioles supply the terminal capillary ring that surrounds the avascular foveola. The circumference of the fovea centralis is demarcated by a yellow dashed line. INL, inner nuclear layer; IPL, inner plexiform layer; NFL, nerve fiber layer; RGC, retinal ganglion cell layer. Scale bar: 100 μm.
Confocal imaging of the whole-mount retina was performed sequentially for the different wavelengths with emission signals separated into different channels. A z series was taken using a ×4 plan apo objective lens to include the thickness of the retina containing retinal vasculature and at z-intervals of 6.15 μm. The confocal stack was separated into inner and deep layers based on the anatomical appearance and the retinal location of the vessels.3,4 The deep capillary plexus was the easiest to recognize, as it was very flat and located at the border of the inner nuclear layer and outer plexiform layer. The inner macular vasculatures were projected as one inner layer, as the configuration varies between donor eyes. 
For the cryosections, confocal images were collected using a ×10 plan apo lens. Sequential images were collected for the different wavelengths and fluorescent probes. Image sequences from the different channels were merged using Image Pro Plus software (Media Cybernetics, Herndon, VA, USA) to produce a color composite sequence. Images from the composite sequence were then used to form a projected image. 
Optical Coherence Tomography Instrumentation
A custom-built 1060-nm swept-source OCT system was used in the acquisition of the svOCT data.2528 The system operated at an A-scan rate of 100 kHz with a total of 1024 points per A-scan. The swept source had a full width half maximum (FWHM) spectral bandwidth of approximately 61.5 nm centered at approximately 1060 nm, corresponding to an axial coherence length of approximately 6 μm in tissue. The sample arm optics delivered a beam of approximately 1.5-mm diameter to the subject's pupil, with a vertical fast axis scan. The size of the focal waist on the retina was estimated to be ωo = approximately 7.3 μm (calculated using Gaussian optics), corresponding to a lateral FWHM of approximately 8.6 μm (calculated as FWHM = Display FormulaImage not available ωo).  
Speckle Variance Method
Speckle variance OCT calculates the variation in intensity on a pixel-to-pixel basis between consecutive B-scans acquired at the same position of the sample. Static tissues have a low variance, whereas the speckle pattern corresponding to locations of movement, such as blood cells flowing through vessels, generate a large variance. This method allowed for both a simple acquisition and processing algorithm, while providing similar vasculature network maps as other techniques. For the speckle variance calculation, three repeat acquisitions at each B-scan location were acquired. The total acquisition time for each svOCT volume, consisting of 1024 × 300 × 300 × 3, was approximately 3.15 seconds. Scan dimensions were calibrated based on the eye length of each participant, measured using the IOL Master 500 (Carl Zeiss Meditec, Inc., Dublin, CA, USA). 
Processing of svOCT Images
The svOCT images were acquired and processed using an open-source graphics processing unit–accelerated platform that permitted visualization of the retinal vasculature in real time during imaging.25,29 The same processing platform was used for postprocessing generation of the intensity-based OCT volume, as well as the svOCT-processed volume. The retinal vascular layers were semiautomatically segmented using the OCT intensity volume data as the inputs to a 3D Graph Cut–based segmentation tool implemented in MATLAB (Natick, MA, USA).30 The 3D Graph Cut algorithm segmented the inner limiting membrane (ILM) and the retinal pigment epithelium/Bruch's membrane complex for each B-scan location so as to capture the retinal shape in three dimensions. The intraretinal regions corresponding to the depth locations of the vascular layers were manually determined by applying an offset to the segmented surfaces, and qualitatively examining the appearance of each en face svOCT image. 
Qualitative Comparisons Between Histology and svOCT
A perfused human donor eye from a 27-year-old male with no known history of ocular disease was used for this part of the study. A ×4 plan apo lens was used to collect an optical stack at 6.15-μm intervals that was centered at the foveola. All sections collected along the z-axis were then projected to provide an en face image of the entire capillary circulation (Fig. 2). The stack was then divided into inner and deep capillary plexuses as described by Zhang et al.31 The inner plexus was constructed by projecting optical slices extending from the ILM to the deep margin of the retinal ganglion cell layer. The deep capillary plexus was constructed by projecting optical slices extending from the inner plexiform layer to the outer nuclear layer. 
Figure 2
 
Separation of the foveal vasculature into inner and deep plexuses. A human donor eye (A) that has been perfusion-labeled with endothelial markers is compared with an svOCT image (B). The regions of interest (ROI) are demarcated by yellow dashed lines. Stratifications of ROIs into inner and deep plexuses are provided in the panels below. Scale bar: 100 μm.
Figure 2
 
Separation of the foveal vasculature into inner and deep plexuses. A human donor eye (A) that has been perfusion-labeled with endothelial markers is compared with an svOCT image (B). The regions of interest (ROI) are demarcated by yellow dashed lines. Stratifications of ROIs into inner and deep plexuses are provided in the panels below. Scale bar: 100 μm.
Speckle variance OCT images were captured from the left eye of a 26-year-old-male with no ocular disease. The microcirculation was divided into inner and deep capillary plexuses by segmenting the same retinal layers as the histological specimen. En face visualizations of the retinal microvasculature were generated by summing the svOCT data along the depth direction within each segmented retinal region. The morphology of the inner and deep capillary plexuses serving the fovea centralis was then compared between histology and svOCT images. 
Quantitative Comparisons Between svOCT and FA
Thirty-eight eyes from 19 healthy subjects underwent FA. Inclusion criteria included clear ocular media, normal ocular examination, stable fixation, best corrected visual acuity greater than or equal to 20/25, and normal FA. 
Fundus FA images were captured with a fundus camera (TRC-50DX; Topcon, Itabashi-ku, Tokyo, Japan) with 5.0-megapixel resolution. Before FA imaging, pupil dilation was performed with tropicamide 1.0% and phenylephrine 2.5% eye drops in all subjects. Sodium fluorescein 10% in water was then injected intravenously through the cubital fossa followed by a flush of normal saline. The images of the macula were captured during the transit phase of the angiogram. The entire angiogram sequence was then evaluated to determine if the best available image of the macula allowed reliable quantification of capillary plexuses. Our previous work,32 together with that of Weinhaus et al.,33 has shown that capillary density does not vary significantly at different time points after fluorescein injection. Therefore, only the best single frame was used for quantitative analysis. Our previous study established a grading scheme that identified FA capillary images from which reliable quantitative information could be derived.32 The same criteria were used in this study. Specifically, the ratio of mean vascular to background fluorescence was determined for the best FA macula image in each eye. Manual histogram equalization, to improve the contrast of finer capillary structures was applied to each image using Adobe Photoshop (CS6, version 13.0; Adobe Systems, Inc., San Jose, CA, USA) before grading. Those subjects with a mean vascular fluorescence to background fluorescence ratio greater than 4:1 (defined as grade 1 images in our previous study)32 also underwent svOCT imaging of the macula, after which quantitative comparisons were performed (Table; Fig. 3). The remaining subjects were not used for quantitative analysis. Eight eyes from four subjects with a ratio of mean vascular to background fluorescence between 4:1 and 2:1 (defined as grade 2 images in our previous study)32 also underwent svOCT imaging to assess the quality of capillary detail that could be captured with the latter device (Fig. 4). These eyes, however, were not used for quantitative analysis. 
Table
 
Capillary Density Comparisons Between FA and svOCT Images
Table
 
Capillary Density Comparisons Between FA and svOCT Images
Figure 3
 
Quantification of capillary density. The morphological characteristics of capillary structures in the fovea centralis are compared between an FA (A) and svOCT image (B) from the same patient. Clearly defined capillary structures were manually traced in both images (C, D). Tracings were then used to determine the area of retinal tissue that was composed of vascular structures (E, F). Results were expressed as a percentage of tissue area occupied by vessels. The density of capillary information was consistently greater in svOCT images. Scale bar: 100 μm.
Figure 3
 
Quantification of capillary density. The morphological characteristics of capillary structures in the fovea centralis are compared between an FA (A) and svOCT image (B) from the same patient. Clearly defined capillary structures were manually traced in both images (C, D). Tracings were then used to determine the area of retinal tissue that was composed of vascular structures (E, F). Results were expressed as a percentage of tissue area occupied by vessels. The density of capillary information was consistently greater in svOCT images. Scale bar: 100 μm.
Figure 4
 
Fluorescein angiogram images that could not be used for quantification. Comparisons between FA and svOCT images from two different patients (A, B) and (C, D) illustrate the significant disparity in vascular detail between the two modalities despite the presence of clear ocular media. Fluorescein angiogram images were defined as grade 2 images according to the established criteria from our previous study.32 These eyes were not used for quantitative comparisons. Scale bar: 100 μm.
Figure 4
 
Fluorescein angiogram images that could not be used for quantification. Comparisons between FA and svOCT images from two different patients (A, B) and (C, D) illustrate the significant disparity in vascular detail between the two modalities despite the presence of clear ocular media. Fluorescein angiogram images were defined as grade 2 images according to the established criteria from our previous study.32 These eyes were not used for quantitative comparisons. Scale bar: 100 μm.
Because it is not possible to stratify FA images of the foveal microcirculation into inner and deep capillary plexuses,32 only en face svOCT images of the entire retinal microvasculature were used for quantitative comparisons. En face svOCT images were generated by summing svOCT data from the nerve fiber layer to the outer nuclear layer. Fluorescein angiography and svOCT images were cropped to an equivalent size of 2 × 2 mm before analysis. Large vessels and FAZ margins were aligned and superimposed for each pair of FA and svOCT images to ensure comparisons of identical areas. 
Quantification of all images was performed using image analysis software (Image Pro Plus, version 7.0). Using our previously described technique32 and that of Weinhaus et al.,33 retinal vessels were manually traced in FA and svOCT images (Fig. 3). Only capillaries that were distinguishable from background without difficulty were traced. The proportion of the image occupied by retinal vessels was expressed as a percentage, and the unit of measurement was calculated as the percentage retinal area occupied by capillary plexus. To determine the reproducibility of these measurements, five FA and five svOCT images were manually traced and quantified on two separate occasions, each at least 1 month apart, by the same observer. 
Quantitative Study of FAZ Morphology
Twenty-five eyes from 13 healthy human subjects, with no history of ocular disease, were imaged with svOCT (Supplementary Table S1). The human subjects used for this part of the study were different from those who underwent FA. Best corrected visual acuity of all eyes was 20/25 or greater. All images were centered on the FAZ. En face svOCT images were generated by summing svOCT data from the nerve fiber layer to the outer nuclear layer. 
Fifteen eyes from 13 human donors with no known history of ocular disease were used for comparison (Supplementary Table S1). All eyes had been perfusion stained, as described above, to label the retinal endothelium. The visual acuity of these eyes was unknown. 
The following quantitative FAZ metrics were acquired from the svOCT and histology images using manual tracing techniques and Image Pro Plus software (Fig. 5): 
Figure 5
 
The human FAZ. Quantitative measurements of the FAZ included perimeter and area ([A] red outline and blue shade, respectively), and major axis and minor axis ([B] green arrow and yellow arrow, respectively). Human donor and svOCT images (CH) illustrate the wide variation in FAZ morphology between healthy human subjects. The variation in the arteriolar network that feeds into the foveal capillary ring is also illustrated in these images. Histology images appear on the left, and svOCT images on the right. Scale bars: 100 μm.
Figure 5
 
The human FAZ. Quantitative measurements of the FAZ included perimeter and area ([A] red outline and blue shade, respectively), and major axis and minor axis ([B] green arrow and yellow arrow, respectively). Human donor and svOCT images (CH) illustrate the wide variation in FAZ morphology between healthy human subjects. The variation in the arteriolar network that feeds into the foveal capillary ring is also illustrated in these images. Histology images appear on the left, and svOCT images on the right. Scale bars: 100 μm.
  1.  
    FAZ perimeter: This was determined by manually delineating the boundary of the FAZ;
  2.  
    FAZ area: The area enclosed within the contour-delineating FAZ perimeter was calculated using image analysis software;
  3.  
    Major axis: Largest diameter of the FAZ; and
  4.  
    Minor axis: Smallest diameter of the FAZ.
Statistical Analysis
All statistical testing was performed using commercial software (SigmaStat, version 3.1; SPSS, Chicago, IL, USA). Kolmogorov-Smirnov testing was performed on all data before analysis to determine whether data were normally distributed. Normally distributed data were analyzed using ANOVA with post hoc factor comparison performed using a paired Student's t-test with Bonferroni correction. Nonnormally distributed data were analyzed using ANOVA on ranks, with the Tukey test used for post hoc paired analysis. Pearson's correlation coefficient was used to determine the strength of relationships between variables that were used to quantify the FAZ. Results are expressed as mean ± SE. 
Results
Microcirculatory Topology of the Fovea Centralis
The organization of the foveal microcirculation and the colocalization of inner and deep capillary plexuses respective to various retinal layers are presented in Figure 1. Arterioles and venules in the macula were organized as paired vessels and formed a radial pattern around the fovea centralis.3 A capillary plexus was seen to separate arterioles and veins and there was a notable absence of direct anastomotic shunt vessels between them.3 Unlike the peripheral retinal eccentricities, arterioles in the macula were relatively devoid of physiological capillary-free zones. Not all macular arterioles entered the fovea centralis and only a few contributed to the formation of the terminal capillary ring.3 The dimensions of the foveola approximate that of the FAZ and is surrounded by the terminal capillaries that are located in the clivus of the foveal pit.4,5 On transverse retinal section, capillary lumens comprising the inner plexus were evident within the retinal ganglion cell and nerve fiber layers (Fig. 1). Capillary lumens of the deep plexus were seen at the outer boundary of the inner nuclear layer and outer plexiform layer. 
En face images of the inner and deep capillary plexuses following confocal microscopy and svOCT imaging appeared comparable and are presented in Figure 2. The terminal capillaries bordering the FAZ were seen to arise from the inner plexus in histological and svOCT images. Projection of optical slices comprising the inner plexus demonstrated that the arterioles and venules supplying this microcirculation were situated in a similar plane to the inner capillaries. 
Capillaries in the deep plexus did not approach the central fovea as closely as the inner plexus. Arterioles and venules were not seen in the plane of the deep plexus in the svOCT image. In the histological specimen, an arteriole was seen within the projection of slices that comprised the deep capillary plexus. Capillaries with an oblique trajectory, linking capillaries in the deep plexus to large vessels in the inner plexus, were more evident in the histological specimen. In both svOCT and histological images, the deep capillary plexus demonstrated a laminar configuration. 
Comparisons Between FA and svOCT
Of the 19 subjects who underwent FA, the images from 6 subjects (31.6%) satisfied grade 1 criteria and thus were used for quantitative analysis. Both eyes from each of these subjects were suitable for quantification. Four subjects who satisfied grade 2 criteria on FA and were deemed unsuitable for quantitative analysis demonstrated accurate visualization of the FAZ and foveal capillary beds on svOCT imaging (representative images of two grade 2 eyes are presented in Fig. 4). 
The demographic details of the subjects who underwent quantitative analysis are provided in the Table. The mean age of subjects was 34.8 ± 3.5 years (age range, 25–46 years) and consisted of one male and five females. The mean capillary density in FA images was 19.3% ± 1.9% of total tissue area, and in svOCT images was 31.2% ± 1.6% of total tissue area. Capillary density values were significantly greater in svOCT images compared with FA images (P < 0.001). 
There was no significant difference between day of measurement and capillary density for FA images (P = 0.366) and svOCT images (P = 0.427) that underwent repeat tracing at different time points. 
Morphometric Features of the FAZ
Demographic details of the subjects and human donors who were used for FAZ analysis are provided in Supplementary Table S1. The mean age of subjects who underwent svOCT imaging was 27.6 ± 1.3 years (age range, 21–35 years) and consisted of five males and eight females. The mean age of human donors was 45.0 ± 4.8 years (age range, 22–78 years) and consisted of 13 males. The mean age of the svOCT and histology groups was significantly different (P = 0.003). 
The shape of the FAZ was clearly delineated in histological and svOCT images (Fig. 5). There were great variations in the geometry of the FAZ between healthy subjects in both groups. Some eyes demonstrated a large and nearly circular foveal avascular zone, whereas other eyes demonstrated a small-diameter FAZ with a bridging capillary that crossed the foveola. There was relatively strong concordance between left and right eyes from the same subject with respect to FAZ geometry. There was great variation in the topography of arterioles and venules that supplied the fovea centralis in both svOCT and histology images. In both groups, the intensity of arteriolar vessels was high but the intensity of venules appeared greater in svOCT images than histological images. 
Quantitative FAZ measurements for each subject and human donor eye are presented in Supplementary Table S1. The mean FAZ perimeter was measured at 2049.5 ± 80.1 μm for svOCT images and 1579.7 ± 173.1 μm for histology images; the mean value of svOCT images was greater than histology (P = 0.001). The mean FAZ area was measured at 261,729.9 ± 20,458.7 μm2 for svOCT images and 140,897.1 ± 28,406.2 μm2 for histology images; the mean value of svOCT images was greater than histology (P = 0.008). The mean value for the major axis on svOCT images was measured at 619.1 ± 24.9 μm and on histology was 488.9 ± 51.4 μm; the mean value of svOCT images was greater than histology (P = 0.015). The mean value for minor axis on svOCT images was measured at 529.4 ± 25.2 μm and on histology was 331.8 ± 41.5 μm; the mean value of svOCT images was greater than histology (P < 0.001). Comparison between right and left eyes using svOCT data did not reveal a significant difference for FAZ perimeter (P = 0.504), FAZ area (P = 0.466), or major axis (P = 0.771); however, the minor axis did reveal a significant difference (P = 0.022). 
The relationships between variables that were used to quantify the FAZ are graphically presented in Figure 6. Calculation of Pearson's correlation coefficient demonstrated a strong linear relationship between area and perimeter (r = 0.945), area and major axis (r = 0.897), area and minor axis (r = 0.953), perimeter and major axis (r = 0.955), and perimeter and minor axis (r = 0.924) for histology images. Strong relationships also were identified between area and perimeter (r = 0.952), area and major axis (r = 0.978), area and minor axis (r = 0.972), perimeter and major axis (r = 0.951), and perimeter and minor axis (r = 0.945) for svOCT images. 
Figure 6
 
Correlations between variables that were used to quantify the FAZ. Significant correlation between FAZ area and perimeter (A, B), FAZ perimeter and major axis measurement (C, D), and FAZ area and major axis measurement (E, F) were identified in histological and svOCT images. Graphs from histological images are presented on the left and svOCT graphs are presented on the right.
Figure 6
 
Correlations between variables that were used to quantify the FAZ. Significant correlation between FAZ area and perimeter (A, B), FAZ perimeter and major axis measurement (C, D), and FAZ area and major axis measurement (E, F) were identified in histological and svOCT images. Graphs from histological images are presented on the left and svOCT graphs are presented on the right.
Discussion
The major findings in this study are as follows: (1) svOCT permits reproducible assessment of the foveal microcirculation and allows accurate delineation of capillary structures in a greater proportion of subjects than FA, (2) svOCT has the capacity to stratify the foveal circulation into inner and deep capillary plexuses, and (3) svOCT can reliably quantify and assess the morphometric dimensions of the human FAZ. 
Speckle variance OCT is a noninvasive device that uses the change in speckle pattern, for example, due to red blood cell movement, and the corresponding intensity variance of structural images to image the retinal microvasculature.34 A major advantage with svOCT, similar to other OCT-based angiographic techniques, is that it is noninvasive and is capable of producing information about fine capillary structures without the administration of dye. Speckle variance OCT imaging therefore avoids the wide range of complications that are commonly encountered with FA, including nausea, vomiting, dye infiltration, and pruritus.35 It also does not carry with it the risk of anaphylaxis and death, which are rare consequences of contrast administration.36 
With the exception of the foveola, which is exclusively nourished by the choroidal circulation, the inner retinal layers of the fovea centralis are dependent on retinal capillary beds for nutrient delivery and waste exchange.1 In vivo techniques that permit evaluation of retinal capillary plexuses may therefore be a valuable means for monitoring the health of this region. Fluorescein angiography has the capacity to image retinal capillaries, but clinical-histologic correlation studies performed in our laboratory have shown that the morphologic information presented on FA images is predominantly that of the inner plexuses.32 Additionally, the density of capillary structures on FA is approximately 60% of what is measured histologically.32 Pinhas et al.37 and Scoles et al.38 incorporated AO-SLO with FA to acquire high-resolution images of foveal and peripapillary capillaries. Although this technique permits exquisite visualization of retinal capillaries, it requires the administration of contrast and therefore does not circumvent the morbidities that are associated with fluorescein administration. 
In this study, we quantified foveal capillary density and performed direct comparisons between svOCT and FA. Performing comparisons in the same eye allowed us to account for the wide variation in capillary plexus morphometry between healthy individuals. The influence of variables, such as patient cooperation and media opacities, which have the potential to confuse the interpretation of results, also was minimized by enrolling only healthy subjects with clear ocular media and stable fixation. Despite stringent inclusion criteria, we found that only 30% of eyes that were imaged with FA allowed accurate and quantifiable delineation of capillary structures. This finding is consistent with our previous study.32 Excessive choroidal fluorescence, with subsequent reduction in contrast between vasculature and background, was the most common reason why capillary structures could not be quantified on FA. Of the subjects who underwent svOCT imaging, we were able to reliably quantify capillary detail in 100% of them. Furthermore, it was possible to acquire quantifiable svOCT images from subjects whose fluorescein angiograms were deemed to be of poor quality. 
Manual tracing techniques have been used previously to quantify and measure the morphometric properties of human retinal capillary plexuses. Previous studies have demonstrated that this method of quantification has high reproducibility when applied to FA and histological images.32,39 Macular capillary density in FA images was measured to be approximately 19% of total tissue areas, whereas it was 31% on svOCT images. This discrepancy is largely because FA has limited capability to image the deep capillary plexus. To our knowledge, direct quantitative comparisons between FA and svOCT have not been performed before. The results in this study, therefore, provide evidence that svOCT provides retinal capillary detail superior to that of FA. Our method of quantification in this report was also shown to be highly reproducible. 
The FAZ is an anatomical site of great research and clinical interest. From a developmental perspective, the FAZ is critically linked to the formation of the foveal pit.40 Although there is significant interindividual variation in the metrics of the FAZ, time-dependent changes in the dimensions of an individual's FAZ may be an important indicator of disease onset and progression. Previous work has shown that enlargement in the diameter and area of the FAZ signifies a deleterious progression of retinal microvascular disease.6,7 Disruption of the foveal capillary ring also has been shown to be associated with an abnormal-pattern ERG.41 Clinical techniques that permit quantifiable visualization of the FAZ, therefore, appear to be a valuable means of detecting disturbances in macula vascular hemodynamics. 
In this report, we were able to demonstrate that svOCT can be used to accurately delineate the dimensions of the human FAZ. Expectedly, there was great variation in the metrics of the FAZ in healthy subjects and also in human donor eyes that were studied histologically. In both groups, the morphology of the FAZ varied from large, near-circular structures to small, irregular shapes with a bridging vessel that crossed the foveola. These findings are consistent with previous anatomical studies.3,42 The quantitative measurements of the FAZ attained in this study were also comparable to previous reports that have examined this region using fluorescein angiography, adaptive optics, phase variance OCT angiography, and SLO.6,19,4346 Our perfusion-based histological technique allows exact visualization of capillary structures in the human retina and therefore provides a gold standard technique for studying the FAZ. In comparison to histology, we found that FAZ metrics were significantly higher in images collected using svOCT. The lack of perfect overlap between histology and svOCT data may be resolved by the following postulations: (1) The influence of postmortem processing techniques, including fixation,47 may have influenced the dimensions of the FAZ on histological tissue. (2) As there is significant interindividual variation in FAZ metrics between normal eyes, it is possible that the difference between histology and svOCT was within the normal range; however, appeared significant due to the limited sample size of the two groups. With a larger sample size it is possible that the differences between histology and svOCT will not reach significance. (3) The mean age of histology and svOCT groups was significantly different. Previous studies have demonstrated an age-dependent change in the size of the FAZ.48 
This study provides a stepwise assessment of the utility of svOCT angiography and, by extension, other label-free angiographic techniques in the clinical assessment of the human foveal circulation. We have shown that svOCT can provide quantitative and reproducible information about capillary plexuses and the FAZ. Although svOCT can stratify the circulation into separate plexuses, shadow artifacts can influence the quantitative information presented in the deep plexus. For this reason, we purposefully did not quantify capillary characteristics within individual plexuses; however, this will be the focus of further research. The presented work shows the ability of svOCT to delineate and segment the different capillary layers within healthy subjects. Pathologic conditions with foveal thickening and macular edema can distort spatial relationships between retinal layers and thereby influence the speckle patterns in the OCT data. The major limitation posed by such anatomical distortion concerns image analysis and segmentation. Conditions such as macular edema may not allow clear delineation of the retinal layers as compared with subjects with normal macular thickness. This could mean a more difficult segmentation process of the retinal layers due to lower contrast in the OCT intensity data. However, areas of edema have relatively little flow and therefore will not affect the contrast of the svOCT signal from regional blood vessels. Results from preliminary studies by our group have demonstrated that it is possible to acquire good-quality speckle variance flow contrast images delineating the foveal terminal ring in patients with diabetic macular edema (Mammo Z, Balaratnasingam C, Yu P, et al., unpublished data, 2015). 
Although we have demonstrated the utility of svOCT in clinical practice, we acknowledge several limitations of this work. This study assessed only relatively young subjects with clear ocular media. The presence of media opacities and poor fixation in older subjects may limit the amount of capillary information that can be attained from images. Additionally, manual tracing techniques were used in this report so as to accurately delineate capillary structures and thus make reliable comparisons among svOCT, FA, and histology. Such labor-intensive techniques are, however, not feasible in clinical practice, and automated methods for segmenting the microanatomy of the foveal circulation are required. Previous studies have described automated measurement techniques for quantifying the FAZ45 and the strong correlations demonstrated in this study between FAZ area and other parameters suggest that the measurement of mean axis or minor axis may be adequate surrogate markers for identifying changes in FAZ metrics. Addressing some of these limitations may expand the already valuable role of OCT angiography in clinical practice. Future directions will focus on evaluating the utility of svOCT imaging in subjects with retinal vascular disease and its potential in patient care. 
Acknowledgments
The authors thank the staff of the Lions Eye Bank of Western Australia for providing human donor eyes; the staff of DonateLife, the Western Australian agency for organ and tissue donation, for facilitating the recruitment of donors into the study; and Dean Darcey for his expert technical assistance. The authors thank the ophthalmic imaging staff at the Eye Care Centre and Laura Hall for their technical assistance. 
Supported by Michael Smith Foundation for Health Research, Vancouver, Canada; Natural Sciences and Engineering Research Council of Canada, Ottawa, Canada; Canadian Institutes of Health Research and National Health, Ottawa, Canada; Brain Canada, Montreal, Canada; Medical Research Council of Australia, Canberra, Australia; LuEsther T. Mertz Retinal Research Center, Manhattan Eye, Ear and Throat Hospital, New York, New York, United States; and The Macula Foundation, Inc., New York, New York, United States. 
Disclosure: Z. Mammo, None; C. Balaratnasingam, None; P. Yu, None; J. Xu, None; M. Heisler, None; P. Mackenzie, None; A. Merkur, None; A. Kirker, None; D. Albiani, None; K.B. Freund, Genentech (C), Regeneron (C), ThromboGenics (C), Ohr Pharmaceutical (C), Bayer HealthCare (C), Heidelberg Engineering (C), Alexion (C): M.V. Sarunic, None; D.-Y. Yu, None 
References
Yu DY, Cringle SJ. Oxygen distribution and consumption within the retina in vascularised and avascular retinas and in animal models of retinal disease. Prog Retin Eye Res. 2001; 20: 175–208.
Yu PK, Balaratnasingam C, Morgan WH, Cringle SJ, McAllister IL, Yu DY. The structural relationship between the microvasculature, neurons, and glia in the human retina. Invest Ophthalmol Vis Sci. 2010; 51: 447–458.
Yu PK, Balaratnasingam C, Cringle SJ, McAllister IL, Provis J, Yu DY. Microstructure and network organization of the microvasculature in the human macula. Invest Ophthalmol Vis Sci. 2010; 51: 6735–6743.
Snodderly DM, Weinhaus RS. Retinal vasculature of the fovea of the squirrel monkey, Saimiri sciureus: three-dimensional architecture, visual screening, and relationships to the neuronal layers. J Comp Neurol. 1990; 297: 145–163.
Iwasaki M, Inomata H. Relation between superficial capillaries and foveal structures in the human retina. Invest Ophthalmol Vis Sci. 1986; 27: 1698–1705.
Conrath J, Giorgi R, Raccah D, Ridings B. Foveal avascular zone in diabetic retinopathy: quantitative vs qualitative assessment. Eye (Lond). 2005; 19: 322–326.
Arend O, Wolf S, Jung F, et al. Retinal microcirculation in patients with diabetes mellitus: dynamic and morphological analysis of perifoveal capillary network. Br J Ophthalmol. 1991; 75: 514–518.
Bresnick GH, Condit R, Syrjala S, Palta M, Groo A, Korth K. Abnormalities of the foveal avascular zone in diabetic retinopathy. Arch Ophthalmol. 1984; 102: 1286–1293.
Varma R, Bressler NM, Doan QV, et al. Prevalence of and risk factors for diabetic macular edema in the United States. JAMA Ophthalmol. 2014; 132: 1334–1340.
Schwartz DM, Fingler J, Kim DY, et al. Phase-variance optical coherence tomography: a technique for noninvasive angiography. Ophthalmology. 2014; 121: 180–187.
Fingler J, Readhead C, Schwartz DM, Fraser SE. Phase-contrast OCT imaging of transverse flows in the mouse retina and choroid. Invest Ophthalmol Vis Sci. 2008; 49: 5055–5059.
Makita S, Jaillon F, Yamanari M, Miura M, Yasuno Y. Comprehensive in vivo micro-vascular imaging of the human eye by dual-beam-scan Doppler optical coherence angiography. Opt Express. 2011; 19: 1271–1283.
Spaide RF, Klancnik JM,Jr, Cooney MJ. Retinal vascular layers imaged by fluorescein angiography and optical coherence tomography angiography. JAMA Ophthalmol. 2015; 133: 45–50.
Fingler J, Schwartz D, Yang C, Fraser SE. Mobility and transverse flow visualization using phase variance contrast with spectral domain optical coherence tomography. Opt Express. 2007; 15: 12636–12653.
Wang RK, Jacques SL, Ma Z, Hurst S, Hanson SR, Gruber A. Three dimensional optical angiography. Opt Express. 2007; 15: 4083–4097.
Mariampillai A, Standish BA, Moriyama EH, et al. Speckle variance detection of microvasculature using swept-source optical coherence tomography. Opt Lett. 2008; 33: 1530–1532.
Braaf B, Vienola KV, Sheehy CK, et al. Real-time eye motion correction in phase-resolved OCT angiography with tracking SLO. Biomed Opt Express. 2013; 4: 51–65.
Jia Y, Tan O, Tokayer J, et al. Split-spectrum amplitude-decorrelation angiography with optical coherence tomography. Opt Express. 2012; 20: 4710–4725.
Tam J, Martin JA, Roorda A. Noninvasive visualization and analysis of parafoveal capillaries in humans. Invest Ophthalmol Vis Sci. 2010; 51: 1691–1698.
Chui TY, Vannasdale DA, Burns SA. The use of forward scatter to improve retinal vascular imaging with an adaptive optics scanning laser ophthalmoscope. Biomed Opt Express. 2012; 3: 2537–2549.
Chui TY, Dubow M, Pinhas A, et al. Comparison of adaptive optics scanning light ophthalmoscopic fluorescein angiography and offset pinhole imaging. Biomed Opt Express. 2014; 5: 1173–1189.
Bedggood P, Metha A. Direct visualization and characterization of erythrocyte flow in human retinal capillaries. Biomed Opt Express. 2012; 3: 3264–3277.
Eliasieh K, Liets LC, Chalupa LM. Cellular reorganization in the human retina during normal aging. Invest Ophthalmol Vis Sci. 2007; 48: 2824–2830.
Hogan MJ, Alvarado JA, Weddell JE. Histology of the Human Eye; An Atlas and Textbook. Philadelphia: Saunders; 1971.
Xu J, Wong K, Jian Y, Sarunic MV. Real-time acquisition and display of flow contrast using speckle variance optical coherence tomography in a graphics processing unit. J Biomed Opt. 2014; 19: 026001.
Xu J, Han S, Balaratnasingam C, et al. Retinal angiography with real-time speckle variance optical coherence tomography [published online ahead of print March 2, 2015]. Br J Ophthalmol. doi:10.1136/bjophthalmol-2014-306010.
Chan G, Balaratnasingam C, Xu J, et al. In vivo optical imaging of human retinal capillary networks using speckle variance optical coherence tomography with quantitative clinico-histological correlation. Microvasc Res. 2015; 100: 32–39.
Tan P, Balaratnasingam C, Xu J, et al. Quantitative comparison of retinal capillary images derived by speckle variance optical coherence tomography with histology. Invest Ophthalmol Vis Sci. 2015; 56: 3989–3996.
Jian Y, Wong K, Sarunic MV. Graphics processing unit accelerated optical coherence tomography processing at megahertz axial scan rate and high resolution video rate volumetric rendering. J Biomed Opt. 2013; 18: 26002.
Lee KK, Mariampillai A, Yu JX, et al. Real-time speckle variance swept-source optical coherence tomography using a graphics processing unit. Biomed Opt Express. 2012; 3: 1557–1564.
Zhang H. Scanning electron-microscopic study of corrosion casts on retinal and choroidal angioarchitecture in man and animals. Prog Retin Eye Res. 1994; 13: 243–269.
Mendis KR, Balaratnasingam C, Yu P, et al. Correlation of histologic and clinical images to determine the diagnostic value of fluorescein angiography for studying retinal capillary detail. Invest Ophthalmol Vis Sci. 2010; 51: 5864–5869.
Weinhaus RS, Burke JM, Delori FC, Snodderly DM. Comparison of fluorescein angiography with microvascular anatomy of macaque retinas. Exp Eye Res. 1995; 61: 1–16.
Mahmud MS, Cadotte DW, Vuong B, et al. Review of speckle and phase variance optical coherence tomography to visualize microvascular networks. J Biomed Opt. 2013; 18: 50901.
Lipson BK, Yannuzzi LA. Complications of intravenous fluorescein injections. Int Ophthalmol Clin. 1989; 29: 200–205.
Bearelly S, Rao S, Fekrat S. Anaphylaxis following intravenous fluorescein angiography in a vitreoretinal clinic: report of 4 cases. Can J Ophthalmol. 2009; 44: 444–445.
Pinhas A, Razeen M, Dubow M, et al. Assessment of perfused foveal microvascular density and identification of nonperfused capillaries in healthy and vasculopathic eyes. Invest Ophthalmol Vis Sci. 2014; 55: 8056–8066.
Scoles D, Gray DC, Hunter JJ, et al. In-vivo imaging of retinal nerve fiber layer vasculature: imaging histology comparison. BMC Ophthalmol. 2009; 9: 9.
Chan G, Balaratnasingam C, Yu PK, et al. Quantitative morphometry of perifoveal capillary networks in the human retina. Invest Ophthalmol Vis Sci. 2012; 53: 5502–5514.
Dubis AM, Hansen BR, Cooper RF, Beringer J, Dubra A, Carroll J. Relationship between the foveal avascular zone and foveal pit morphology. Invest Ophthalmol Vis Sci. 2012; 53: 1628–1636.
Holder GE, Acheson JF, Griffiths MFP, Green W, Shilling JS. Pattern electroretinography in patients with branch retinal vein occlusion. J Physiol. 1991; 438: 295P.
Bird AC, Weale RA. On the retinal vasculature of the human fovea. Exp Eye Res. 1974; 19: 409–417.
Chui TY, VanNasdale DA, Elsner AE, Burns SA. The association between the foveal avascular zone and retinal thickness. Invest Ophthalmol Vis Sci. 2014; 55: 6870–6877.
Popovic Z, Knutsson P, Thaung J, Owner-Petersen M, Sjostrand J. Noninvasive imaging of human foveal capillary network using dual-conjugate adaptive optics. Invest Ophthalmol Vis Sci. 2011; 52: 2649–2655.
Zheng Y, Gandhi JS, Stangos AN, Campa C, Broadbent DM, Harding SP. Automated segmentation of foveal avascular zone in fundus fluorescein angiography. Invest Ophthalmol Vis Sci. 2010; 51: 3653–3659.
Kim DY, Fingler J, Zawadzki RJ, et al. Noninvasive imaging of the foveal avascular zone with high-speed, phase-variance optical coherence tomography. Invest Ophthalmol Vis Sci. 2012; 53: 85–92.
Kerns MJ, Darst MA, Olsen TG, Fenster M, Hall P, Grevey S. Shrinkage of cutaneous specimens: formalin or other factors involved? J Cutan Pathol. 2008; 35: 1093–1096.
Laatikainen L, Larinkari J. Capillary-free area of the fovea with advancing age. Invest Ophthalmol Vis Sci. 1977; 16: 1154–1157.
Figure 1
 
Topology of the human macula and foveal microcirculation. Markings on multispectral color fundus image (A) (Heidelberg Spectralis; Heidelberg Engineering, Carlsbad, CA, USA) are used to illustrate the dimensions of the normal human macula ([1] yellow shade, 5.5-mm diameter) and, within it, the fovea centralis ([2] orange shade, 1.85-mm diameter). These regions also are indicated on a spectral-domain OCT image (B). A human donor eye that has been labeled with multiple antibodies demonstrates the microstructural anatomy of the fovea centralis (C). The antibodies that were used to label various retinal layers are provided. The foveola, which is the central depression of the fovea centralis, is demarcated by the sloped clivus. Inset (I) illustrates the distribution of capillary plexuses (arrows) in the inner layers of the fovea centralis. A cadaveric human eye (D) that has been perfused with endothelial markers reveals the asymmetric pattern of arterioles (Art) and venules (Vn) that supply the foveal centralis. It is evident that only a few arterioles supply the terminal capillary ring that surrounds the avascular foveola. The circumference of the fovea centralis is demarcated by a yellow dashed line. INL, inner nuclear layer; IPL, inner plexiform layer; NFL, nerve fiber layer; RGC, retinal ganglion cell layer. Scale bar: 100 μm.
Figure 1
 
Topology of the human macula and foveal microcirculation. Markings on multispectral color fundus image (A) (Heidelberg Spectralis; Heidelberg Engineering, Carlsbad, CA, USA) are used to illustrate the dimensions of the normal human macula ([1] yellow shade, 5.5-mm diameter) and, within it, the fovea centralis ([2] orange shade, 1.85-mm diameter). These regions also are indicated on a spectral-domain OCT image (B). A human donor eye that has been labeled with multiple antibodies demonstrates the microstructural anatomy of the fovea centralis (C). The antibodies that were used to label various retinal layers are provided. The foveola, which is the central depression of the fovea centralis, is demarcated by the sloped clivus. Inset (I) illustrates the distribution of capillary plexuses (arrows) in the inner layers of the fovea centralis. A cadaveric human eye (D) that has been perfused with endothelial markers reveals the asymmetric pattern of arterioles (Art) and venules (Vn) that supply the foveal centralis. It is evident that only a few arterioles supply the terminal capillary ring that surrounds the avascular foveola. The circumference of the fovea centralis is demarcated by a yellow dashed line. INL, inner nuclear layer; IPL, inner plexiform layer; NFL, nerve fiber layer; RGC, retinal ganglion cell layer. Scale bar: 100 μm.
Figure 2
 
Separation of the foveal vasculature into inner and deep plexuses. A human donor eye (A) that has been perfusion-labeled with endothelial markers is compared with an svOCT image (B). The regions of interest (ROI) are demarcated by yellow dashed lines. Stratifications of ROIs into inner and deep plexuses are provided in the panels below. Scale bar: 100 μm.
Figure 2
 
Separation of the foveal vasculature into inner and deep plexuses. A human donor eye (A) that has been perfusion-labeled with endothelial markers is compared with an svOCT image (B). The regions of interest (ROI) are demarcated by yellow dashed lines. Stratifications of ROIs into inner and deep plexuses are provided in the panels below. Scale bar: 100 μm.
Figure 3
 
Quantification of capillary density. The morphological characteristics of capillary structures in the fovea centralis are compared between an FA (A) and svOCT image (B) from the same patient. Clearly defined capillary structures were manually traced in both images (C, D). Tracings were then used to determine the area of retinal tissue that was composed of vascular structures (E, F). Results were expressed as a percentage of tissue area occupied by vessels. The density of capillary information was consistently greater in svOCT images. Scale bar: 100 μm.
Figure 3
 
Quantification of capillary density. The morphological characteristics of capillary structures in the fovea centralis are compared between an FA (A) and svOCT image (B) from the same patient. Clearly defined capillary structures were manually traced in both images (C, D). Tracings were then used to determine the area of retinal tissue that was composed of vascular structures (E, F). Results were expressed as a percentage of tissue area occupied by vessels. The density of capillary information was consistently greater in svOCT images. Scale bar: 100 μm.
Figure 4
 
Fluorescein angiogram images that could not be used for quantification. Comparisons between FA and svOCT images from two different patients (A, B) and (C, D) illustrate the significant disparity in vascular detail between the two modalities despite the presence of clear ocular media. Fluorescein angiogram images were defined as grade 2 images according to the established criteria from our previous study.32 These eyes were not used for quantitative comparisons. Scale bar: 100 μm.
Figure 4
 
Fluorescein angiogram images that could not be used for quantification. Comparisons between FA and svOCT images from two different patients (A, B) and (C, D) illustrate the significant disparity in vascular detail between the two modalities despite the presence of clear ocular media. Fluorescein angiogram images were defined as grade 2 images according to the established criteria from our previous study.32 These eyes were not used for quantitative comparisons. Scale bar: 100 μm.
Figure 5
 
The human FAZ. Quantitative measurements of the FAZ included perimeter and area ([A] red outline and blue shade, respectively), and major axis and minor axis ([B] green arrow and yellow arrow, respectively). Human donor and svOCT images (CH) illustrate the wide variation in FAZ morphology between healthy human subjects. The variation in the arteriolar network that feeds into the foveal capillary ring is also illustrated in these images. Histology images appear on the left, and svOCT images on the right. Scale bars: 100 μm.
Figure 5
 
The human FAZ. Quantitative measurements of the FAZ included perimeter and area ([A] red outline and blue shade, respectively), and major axis and minor axis ([B] green arrow and yellow arrow, respectively). Human donor and svOCT images (CH) illustrate the wide variation in FAZ morphology between healthy human subjects. The variation in the arteriolar network that feeds into the foveal capillary ring is also illustrated in these images. Histology images appear on the left, and svOCT images on the right. Scale bars: 100 μm.
Figure 6
 
Correlations between variables that were used to quantify the FAZ. Significant correlation between FAZ area and perimeter (A, B), FAZ perimeter and major axis measurement (C, D), and FAZ area and major axis measurement (E, F) were identified in histological and svOCT images. Graphs from histological images are presented on the left and svOCT graphs are presented on the right.
Figure 6
 
Correlations between variables that were used to quantify the FAZ. Significant correlation between FAZ area and perimeter (A, B), FAZ perimeter and major axis measurement (C, D), and FAZ area and major axis measurement (E, F) were identified in histological and svOCT images. Graphs from histological images are presented on the left and svOCT graphs are presented on the right.
Table
 
Capillary Density Comparisons Between FA and svOCT Images
Table
 
Capillary Density Comparisons Between FA and svOCT Images
×
×

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.

×