December 2014
Volume 55, Issue 12
Free
Retina  |   December 2014
Variability and Repeatability of Quantitative, Fourier-Domain Optical Coherence Tomography Doppler Blood Flow in Young and Elderly Healthy Subjects
Author Affiliations & Notes
  • Faryan Tayyari
    Retina Research Group, School of Optometry and Vision Science, University of Waterloo, Waterloo, Ontario, Canada
  • Firdaus Yusof
    Retina Research Group, School of Optometry and Vision Science, University of Waterloo, Waterloo, Ontario, Canada
    Department of Optometry and Visual Science, International Islamic University of Malaysia, Bandar Indera Mahkota, Kuantan, Malaysia
  • Michal Vymyslicky
    Retina Research Group, School of Optometry and Vision Science, University of Waterloo, Waterloo, Ontario, Canada
  • Ou Tan
    Casey Eye Institute, Oregon Health & Science University, Portland, Oregon, United States
  • David Huang
    Casey Eye Institute, Oregon Health & Science University, Portland, Oregon, United States
  • John G. Flanagan
    Retina Research Group, School of Optometry and Vision Science, University of Waterloo, Waterloo, Ontario, Canada
    Department of Ophthalmology and Vision Sciences, University of Toronto, Toronto, Ontario, Canada
  • Christopher Hudson
    Retina Research Group, School of Optometry and Vision Science, University of Waterloo, Waterloo, Ontario, Canada
    Department of Ophthalmology and Vision Sciences, University of Toronto, Toronto, Ontario, Canada
  • Correspondence: Christopher Hudson, University of Waterloo, School of Optometry and Vision Science, 200 University Avenue West, Waterloo, ON, Canada N2L 3G1; chudson@uwaterloo.ca
Investigative Ophthalmology & Visual Science December 2014, Vol.55, 7716-7725. doi:10.1167/iovs.14-14430
  • Views
  • PDF
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Faryan Tayyari, Firdaus Yusof, Michal Vymyslicky, Ou Tan, David Huang, John G. Flanagan, Christopher Hudson; Variability and Repeatability of Quantitative, Fourier-Domain Optical Coherence Tomography Doppler Blood Flow in Young and Elderly Healthy Subjects. Invest. Ophthalmol. Vis. Sci. 2014;55(12):7716-7725. doi: 10.1167/iovs.14-14430.

      Download citation file:


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

      ×
  • Supplements
Abstract

Purpose.: The purpose of this study was to determine the within-session variability and between-session repeatability of spectral Fourier-domain optical coherence tomography (Doppler FD-OCT) Doppler retinal blood flow measurements in young and elderly subjects.

Methods.: Doppler FD-OCT blood flow was measured using the RTVue system. One eye of each of 20 healthy young (24.7 ± 2.7 years) and 16 healthy elderly (64.6 ± 5.1 years) subjects was randomly selected, and the pupil was dilated. The double circular scanning pattern of the RTVue was employed. Six Doppler FD-OCT measurements (i.e., each separate measurement comprising an upper and a lower nasal pupil scan) were acquired at each session. Measurements were repeated approximately 2 weeks later. Total retinal blood flow was calculated by summing flow from all detectable venules surrounding the optic nerve head. The coefficient of variation (COV) and coefficient of repeatability (COR) were calculated for each individual.

Results.: The individual COVs for retinal blood flow for young subjects ranged from 0.4% to 20.4% (median 7.5%) and for the elderly subjects ranged from 0.6% to 34.6% (median 9.2%). The group mean CORs for retinal blood flow for young participants were 6.4 μL/min (median 5.91 μL/min, relative to a mean effect 39.8 μL/min) and for elderly subjects were 10.5 μL/min (median 9.2 μL/min, relative to a mean effect 46.4 μL/min).

Conclusions.: Doppler FD-OCT gave consistent and repeatable blood flow measurements within retinal venules in normal subjects. Considering the individual variation in blood flow measurements, confidence limits for retinal hemodynamics need to be determined on an individual basis.

Introduction
The human eye is perfused mostly from the ophthalmic artery, the first branch of the internal carotid artery and the only branch of the internal carotid outside the cranium. There are two distinct vascular systems that support the mature neural retina: the inner retinal vessels and the choroidal vessels. The two beds vary in both their embryonic differentiation pattern and functionally in the adult human.1 The retinal vasculature is spread mainly within the inner retina and nourishes most inner retinal structures. The pigment epithelium and photoreceptors, however, are primarily nourished by the choroidal vasculature. The contraction of the arterioles determines the blood flow into the inner retinal capillary bed.2 Moreover, the downstream retinal capillaries are believed to be able to further fine-tune blood flow via the actions of the contractile pericytes in response to local tissue demands, such as the level of oxygenation.3 Taken together, these mechanisms sustain constant blood flow over an extensive range of perfusion pressures (i.e., autoregulation). 
The assessment of retinal blood flow (RBF) is important because its perturbation has been suggested to play a role in many ocular diseases such as diabetic retinopathy46 and glaucoma.714 Several techniques have been developed to quantify RBF in humans. Previous techniques for evaluating retinal perfusion have numerous limitations, such as being invasive (e.g., fluorescein angiography), or are subjective (e.g., blue field entoptic phenomenon), or are incapable of calculating blood flow (e.g., retinal vessel analyzer), since a surrogate parameter of flow is truly measured. In addition, a technique that truly measures volumetric RBF, that is, bidirectional laser Doppler velocimetry and simultaneous vessel densitometry, is currently limited to relatively large vessels and a single measurement site. To overcome these limitations, a new technique termed spectral Fourier-domain optical coherence tomography Doppler (Doppler FD-OCT) blood flow (e.g., the Optovue RTVue system) has been developed. The FD-OCT method uses noncontact imaging that mixes micrometer-scale resolution with millimeter image penetration depth.1517 It is similar to the ultrasound technique, with the exception of using light energy in the form of lasers instead of sound. The FD-OCT technique perceives the intensity of light, back-reflected from various features within the imaged object, resulting from spatial variations of the tissue refractive index. It is based on low-coherence interferometry, a classic optical measurement technique. The FD-OCT method makes high-resolution imaging possible. It is generally used in the diagnosis and management of retinal diseases.18,19 
Fourier-domain optical coherence tomography has the ability to detect motion within a sample from the back-reflected light, which delivers information on movement.20,21 Regarding terminology, Huang and coworkers15 and many other groups have generally used the term “Doppler shift” to describe the same phenomenon that other groups call a “phase shift” or “Doppler phase,” which represents a difference in the position of a given point within the waveform between two subsequent A-scans. From this point forward, we will use the term “Doppler phase shift” to describe this phenomenon. 
The variability and repeatability of this new RBF measurement technique needs to be established in order to characterize a significant change and to use it as a clinical technique to detect abnormal deviations in blood flow. The aim of this study is to investigate the within-session variability and between-session repeatability of the Doppler FD-OCT morphologic blood flow technology in young and elderly healthy subjects. 
Methods
Sample
This study received approval from the University of Waterloo Office of Research Ethics and the Research Ethics Board of the University Health Network, University of Toronto, Canada. Informed consent was obtained from each subject after explanation of the nature and possible consequences of the study according to the tenets of the Declaration of Helsinki. The sample consisted of 36 healthy volunteers in two groups of young and elderly subjects. One eye of each of 20 healthy young (mean age 24.7; SD 2.7 years) and 16 healthy elderly (mean age 64.6; SD 5.1 years) subjects was randomly selected for the study. All subjects had a corrected visual acuity of 20/40 or better. Subjects were excluded for any ocular or systemic disease, refractive error greater than ±6.00 Diopter sphere (DS) or ±2.50 Diopter cylinder (DC), glaucoma or diabetes in a first-degree relative, history of central nervous system disorders, or medications with known effects on blood flow (e.g., anticonvulsant or anti-inflammatory medications). Subject and appropriate summary group characteristics (age, sex, right or left eye, intraocular pressure, and blood pressure) are detailed in Table 1
Table 1
 
Individual Subject/Group Characteristics for the Young and Elderly Groups*
Table 1
 
Individual Subject/Group Characteristics for the Young and Elderly Groups*
Name Age M/F IOP, mm Hg Visit 1 Visit 2
Visit 1 Visit 2 Sys, mm Hg Dias, mm Hg Sys, mm Hg Dias, mm Hg
Young group
 1 27 F 11.00 10.33 116 68 121 71
 2 24 F 11.33 12.00 115 75 111 74
 3 29 F 12.33 11.33 125 80 121 82
 4 27 F 12.00 12.33 110 69 113 68
 5 28 F 11.00 11.66 97 63 110 64
 6 19 F 15.00 14.00 109 69 100 68
 7 24 F 13.00 12.66 113 67 124 71
 8 26 F 12.66 12.00 106 67 115 73
 9 28 F 11.00 12.00 113 78 102 81
 10 26 F 12.00 11.66 119 81 122 69
 11 22 M 15.00 14.00 128 78 119 69
 12 23 M 14.33 13.66 125 65 111 78
 13 25 M 9.00 10.33 128 71 120 81
 14 26 M 18.33 16.66 128 82 135 87
 15 24 M 13.66 12.33 131 76 129 75
 16 25 M 17.00 15.33 128 67 121 71
 17 20 M 12.00 12.66 129 69 125 73
 18 23 M 13.33 13.00 125 78 119 75
 19 26 M 17.66 16.33 127 81 125 78
 20 22 M 16.00 16.00 129 60 128 61
Mean 24.7 13.38 13.01 120.05 72.2 118.55 73.45
SD 2.7 2.48 1.86 9.59 6.69 8.77 6.4
Elderly group
 21 64 F 15.00 15.33 145 56 139 59
 22 63 F 13.33 14.00 105 69 111 70
 23 62 F 13.00 15.00 119 84 121 83
 24 62 F 14.00 13.00 136 83 129 86
 25 67 F 15.00 16.00 142 79 134 81
 26 61 F 14.00 13.66 138 83 145 84
 27 68 F 16.00 15.66 133 86 135 85
 28 70 F 11.00 12.00 150 79 145 87
 29 59 F 15.00 14.66 124 78 132 85
 30 78 F 12.66 12.00 124 78 129 75
 31 70 F 14.00 13.33 128 79 135 79
 32 60 M 15.00 14.33 131 72 122 76
 33 60 M 13.33 13.66 120 76 128 75
 34 67 M 14.00 15.33 122 74 124 78
 35 62 M 14.00 13.66 125 79 131 81
 36 60 M 14.00 14.33 128 78 112 73
Mean 64.6 13.64 13.51 125.58 74.36 123.49 75.72
SD 5.1 2.00 1.68 12.07 7.21 10.42 7.17
The Doppler FD-OCT Method
Doppler FD-OCT is a novel imaging method that provides in vivo noninvasive assessment of retinal structure and RBF using a physical phenomenon called Doppler phase shift.22 The principle of the Doppler phase shift has been integrated into the commercially available Doppler FD-OCT system (Optovue, Inc., Freemont, CA, USA). Doppler FD-OCT generates high-resolution cross-sectional images of the retina. This instrument utilizes a laser light source of 841 nm with a bandwidth of 49 nm and an incident power of 500 μW on the cornea. Theses parameters result in an axial resolution of 5.4 μm in tissue.23 System transverse resolution was 20 μm, as determined by the maximum aperture of the eye.24 
Unlike morphological FD-OCT systems that produce just structural images, the prototype Doppler FD-OCT analyzes the Doppler phase shift between two consecutive A-scans. Light reflected from moving particles undergoes Doppler phase shift. 
Flow velocity is determined by  where v is the flow velocity in an OCT voxel, ΔΦ is the Doppler phase shift (ΔΦ = Φ1 − Φ2, where Φ1 and Φ2 are the phase of voxels in the same position in consecutive OCT axial scans), λ0 is the source center wavelength, n is the refractive index of the medium, T is the time interval between consecutive scans, and θ is the Doppler angle defined by the OCT beam axis relative to the line perpendicular to blood vessel flow axis.  
The maximum detectable Doppler phase shift of 8.9 KHz is determined by the acquisition speed of the charged-coupled device (CCD) camera. Given this setup, the maximum measurable velocity in the retinal vessels was 2.8 mm/s.24 
Phase detection caused by RBF was incorporated into the prototype system by creating two circular scans centered on the optic nerve head. The RBF protocol consists of a double circular Doppler scan comprising two concentric rings of diameters 3.4 and 3.75 mm centered on the optic nerve head (Fig. 1).25 The double circular FD-OCT beam passes through the pupil nasally with two sets of scans (inferior and superior). Two sets of scans are taken to achieve optimum flow measurements from at least one of the two sets of scans. The circular scan is displayed as the sinusoidal variation in retinal height/morphology (Fig. 2). 
Figure 1
 
Doppler FD-OCT image acquisition page. A double circular scan pattern with concentric radii is used for the measurement, which is typically centered at the optic nerve head.
Figure 1
 
Doppler FD-OCT image acquisition page. A double circular scan pattern with concentric radii is used for the measurement, which is typically centered at the optic nerve head.
Figure 2
 
Doppler FD-OCT image for beam passing through the superior nasal quadrant of the pupil. The peak of the sinusoidal should be positioned either within superior nasal or inferior nasal quadrants.
Figure 2
 
Doppler FD-OCT image for beam passing through the superior nasal quadrant of the pupil. The peak of the sinusoidal should be positioned either within superior nasal or inferior nasal quadrants.
The incident angle is estimated by vessel center depth difference within the two consecutive circular OCT scans from the Doppler FD-OCT images. The measured Doppler phase shift, the incident angle calculation, and the vessel area are used to compute absolute red blood cell velocity and RBF. 
Procedures
Refraction, logMAR visual acuity, Goldmann applanation tonometry, and resting blood pressure were assessed prior to dilation of the study eye. The pupil of the study eye was dilated using tropicamide 1% (Alcon Canada, Inc., Mississauga, ON, Canada) at the beginning of each visit to achieve an adequate view of the fundus for the RBF image acquisition. Visual fields were also tested in subjects using the Humphrey Field Analyzer II (Carl Zeiss Meditec, Inc., Dublin, CA, USA) with the 24–2 automated static threshold test. The clinical fundus view and the automated perimetry results were all confirmed to be normal. Subjects rested for 10 minutes before the start of each study visit to stabilize baseline cardiovascular and respiratory parameters. Each subject attended for two visits. The two visits were undertaken within 2 weeks, and the second visit was at the same time of day and under the same conditions as the initial visit. A minimum of six separate FD-OCT Doppler measurements (i.e., each separate measurement comprising an upper nasal pupil scan and a lower nasal pupil scan) was acquired at each session. 
Image Grading
Semiautomated Doppler OCT of Retinal Circulation (DOCTORC) Software.
Acquired Doppler FD-OCT scans and a scanning laser ophthalmoscopy (SLO) en face image were exported as raw data using RTVue Doppler transfer output software. The exported raw data were converted into DOCTORC (version 2.1.1.4) grading software–compatible data for image grading and RBF calculation.25 After loading the Doppler scan into DOCTORC, the initial automated assessment of the loaded scans underwent software processing that identified the vessel type based on the Doppler signal characteristics (i.e., arteriole or venule). Grading also required a color fundus image of the optic disc to confirm that the automated vessel identification undertaken by DOCTORC was correct by comparing the DOCTORC assignment (i.e., venule or arteriole) with that of a color fundus photograph. The Doppler scans with three-dimensional OCT image and SLO image were then registered to allow the grader to correlate the vessels that are identified on the en face image with those seen on the cross-sectional Doppler OCT B-scan. DOCTORC then computed the blood flow from the Doppler phase data after manual assessment of scan validity. The blood flow data were then automatically exported into the appropriate subject folder. 
The FD-OCT scans were manually graded based on their Doppler signal, size, and location between inner and outer circle, clarity of the vessel boundary, and finally the type of the vessel. The grader adjusted the dotted circle in order to make it the same size as the Doppler signal (Fig. 3). There was also a confidence score ranging from 0 to 5 with a confidence score guidance sheet (Doheney Eye Institute, Arcadia, CA, USA). The grader of each vessel on every scan assigned this score. After completion of all these procedures for all the vessels in the Doppler FD-OCT image, the software verified whether the flow calculation for each graded vessel was valid. Subsequently, the total retinal venous blood flow was automatically computed by summing all calculated flow values from all valid retinal venules and the estimated flow from venules25 using an Excel file. 
Figure 3
 
Cross-section of retinal blood vessels displaying Doppler signal. The grader adjusts the vessel circumference with a dotted circle and the diameter with a horizontal solid line.
Figure 3
 
Cross-section of retinal blood vessels displaying Doppler signal. The grader adjusts the vessel circumference with a dotted circle and the diameter with a horizontal solid line.
The coefficient of variation (COV) is defined as the ratio of the standard deviation σ to the mean μ, such that Display FormulaImage not available . The Bland-Altman plot26 is a graphical approach to compare two measurement methodologies of the same object. The differences between the two methods are mapped as the function of the averages of the two. This plot may be used to assess the repeatability of a method. The coefficient of repeatability (COR) can be calculated as 1.96 times the standard deviation of the differences between the two measurements (d2 and d1), such that Display FormulaImage not available . Repeatability or test-retest reliability is defined as the variation in measurements taken by a single operator and instrument under the consistent conditions.  
Results
Typically, it takes between 2 and 5 minutes to analyze each vessel. Depending on a given patient's vascular tree characteristics, the total time required to undertake the analysis of a data set acquired at a single visit varied between 20 and 60 minutes. Box plots of the COVs for blood flow in both young and elderly participant groups are displayed in Figure 4. The individual COVs for blood flow in the young ranged from 0.4% to 20.4% (median 7.5%) and in the elderly subjects ranged from 0.6% to 34.6% (median 9.2%). The individual COR values for flow ranged from 0.2 to 13.8 μL/min (median 4.5 μL/min) for young participants and from 0.4 to 38.8 μL/min (median 9.2 μL/min) for elderly participants. 
Figure 4
 
Box plot of individual COVs for blood flow in young and elderly participants. The error bars show the nonoutlier range (±1.5 times the height of the box).
Figure 4
 
Box plot of individual COVs for blood flow in young and elderly participants. The error bars show the nonoutlier range (±1.5 times the height of the box).
The group mean CORs for RBF for young participants were 6.4 μL/min (median 5.9 μL/min, relative to a mean effect 39.8 μL/min) and for elderly subjects were 10.5 μL/min (median 9.2 μL/min, relative to a mean effect 46.4 μL/min). Difference versus mean plots of RBF (Figs. 5 and 6) revealed two clear outliers in the elderly group data (see Bland and Altman26). Removal of these outliers reduced the elderly group mean COR for blood flow to 9.88 μL/min (median 8.3 μL/min relative to a mean effect 42.9 μL/min). 
Figure 5
 
Bland-Altman plot showing the difference in total blood flow as a function of mean blood flow between sessions for young subjects. MoD, mean of differences; 95% Conf. Limit, 1.96 × SD of the MoD (also known as limits of agreement).
Figure 5
 
Bland-Altman plot showing the difference in total blood flow as a function of mean blood flow between sessions for young subjects. MoD, mean of differences; 95% Conf. Limit, 1.96 × SD of the MoD (also known as limits of agreement).
Figure 6
 
Bland and Altman plot showing the difference in total blood flow as a function of mean blood flow between sessions for elderly subjects. MoD, mean of differences; 95% Conf. Limit, 1.96 × SD of the MoD (also known as limits of agreement).
Figure 6
 
Bland and Altman plot showing the difference in total blood flow as a function of mean blood flow between sessions for elderly subjects. MoD, mean of differences; 95% Conf. Limit, 1.96 × SD of the MoD (also known as limits of agreement).
A scatter plot was used to illustrate test-retest characteristics of the Doppler FD-OCT measurement of total RBF (in micrometers per minute) (Fig. 7). 
Figure 7
 
Scatter plot illustrating test-retest characteristics of the Doppler FD-OCT for measuring total retinal blood flow in microliters per minute.
Figure 7
 
Scatter plot illustrating test-retest characteristics of the Doppler FD-OCT for measuring total retinal blood flow in microliters per minute.
The COV values for venous area, velocity, and flow can be found in Table 2. The COR values for both young and elderly subjects have been reported in Table 3 and 4, respectively. 
Table 2
 
Coefficient of Variation (%) for Venous Area, Velocity, and Blood Flow
Table 2
 
Coefficient of Variation (%) for Venous Area, Velocity, and Blood Flow
COV, %
Venous Area Velocity Flow
Median Range Median Range Median Range
Young 4.7 0.8–16.8 10.4 0.1–28.9 7.5 0.4–20.4
Elderly 4.8 0.4–20.4 10.8 1.1–43.8 9.2 0.6–34.6
Table 3
 
Coefficient of Repeatability for Young Individuals as a Group
Table 3
 
Coefficient of Repeatability for Young Individuals as a Group
COR in Young Subjects
Individual COR Overall COR Mean Effect Size
Median Range
Venous area 0.003 0.0003–0.017 0.005 0.033
Velocity 4.25 1.11–17.53 6.12 21.90
Flow 5.91 0.23–13.76 6.43 39.76
Table 4
 
Coefficient of Repeatability for Elderly Individuals as a Group
Table 4
 
Coefficient of Repeatability for Elderly Individuals as a Group
COR in Elderly Subjects
Individual COR Overall COR Mean Effect Size
Median Range
Venous area 0.005 00.1–0.02 0.006 0.035
Velocity 7.19 2.22–18.12 8.20 21.77
Flow 9.16 0.37–38.83 10.53 46.93
Discussion
Several studies have reported reduced ocular blood flow and poor blood flow regulation in glaucomatous eyes.714 There is also evidence that primary open-angle glaucoma is linked to factors related to ocular blood flow regulation and a breakdown of autoregulation.7 Our group has demonstrated a reduction in the magnitude of retinal arteriolar vascular reactivity in both untreated primary open-angle glaucoma and progressive primary open-angle glaucoma.27 There are also studies that show diabetic retinopathy is preceded by subclinical disturbances in the retinal vasculature. Hence, many studies have considered the quantification of inner RBF as a promising investigation approach for the early detection and improved monitoring of various diseases. However, the results of inner RBF disturbance in diabetic retinopathy have been contradictory,2837 although virtually all indicate some aspect of disturbance. 
This study shows that the Doppler FD-OCT can reliably and consistently measure RBF in healthy participants, despite the subjectivity of the vessel area determination. Doppler FD-OCT offers a quantifiable and repeatable method of assessing RBF. Blue field entoptic and fluorescein angiographic techniques deliver mostly subjective or semiquantitative measurements of RBF. Laser Doppler flowmetry offers flow values only in arbitrary units, and readings between subjects are dependent on interpretation. Pulsatile ocular blood flow assessments are dependent on controversial assumptions of ocular physiology and measure total ocular blood flow, a majority of which arises from the choroidal circulation. 
Wang and coworkers23 used a single-beam FD-OCT to measure the total RBF of all vessels around the optic nerve head. Their results showed values of 45.6 ± 3.8 μL/min for total RBF (TRBF) with the coefficient of variation of 10.5%. Our results showed a range of variability in individual blood flow, with a COV ranging from 0.4% in young subjects (median 7.5%) to 34.6% in elderly subjects (median 9.2%). Our data are in agreement with the data of Wang and coworker.23 There are several sources that might contribute to the individual variability, including the impact of eye motion, tear film quality, or normal biological variations. The individual variability may also reflect the quality of measurement and possibly the analysis method. Subjects with higher COV values could be excluded from study based upon their outlier status. 
The differences in COV between the two groups (median 7.5% for the young group versus 9.2% for the elderly group) are negligible, and the differences in COR between groups are relatively small (median 5.91 μL/min relative to a mean effect of 39.76 μL/min for the young group versus median 9.16 μL/min relative to a mean effect of 46.39 μL/min for the elderly group). This represents a difference in RBF calculations between sessions/days of approximately 4 μL/min in real terms, or in terms of the known range of TRBF values, a difference of approximately 8%. 
The coefficient of variation relates the group mean SD of a given parameter as a function of the group mean effect (on a percentage scale). Therefore, comparison of the COV of venous area and venous velocity for young and elderly participants can be used to determine which of the two parameters contributes the greatest variability to the RBF value. We found that the COV for venous area to be approximately half that of the COV for venous velocity (i.e., 4.7%–4.8% for area versus 10.4%–10.8% for velocity). We conclude that the velocity measurement is the greater contributor to blood flow variability, a finding that is in agreement with that of the established Canon Laser Blood Flowmeter (CLBF) methodology.38,39 A mathematical perspective of this issue would be to argue that a summary velocity value is derived by dividing the measured area by the measured flow value. Consequently, the fundamental error components are Doppler angle error caused by motion error, vessel boundary segmentation error, Doppler phase error due to phase wrapping, residual bulk motion error (after compensation), and system phase noise (Huang D, personal communication, 2014). 
One of the advantages of OCT is the exceptional level of resolution used to discern the retinal vasculature. Another, more established method to assess RBF is laser Doppler velocimetry (LDV), which has been applied to investigate retinal blood velocities in the major arterioles and venules of the retina.40 The most established instrument to utilize the LDV methodology is the CLBF, which simultaneously measures vessel diameter and centerline blood velocity in order to derive blood flow values in absolute units.41 
Using the CLBF, the individual COVs for flow were reported to be from 4.8% to 37.3% (median 19.3%). The group mean CORs for flow were found to be 2.6 μL/min (relative to a mean effect of 8.8 μL/min). These data are from CLBF measurements of blood flow within retinal arterioles in normal subjects.38 Our group studied the retinal arteriolar and venular blood flow in healthy subjects and concluded that the level of variability between the two visits was equivalent before and during hyperoxic provocation of vascular reactivity. This study showed the COV for venular flow was 9.9% during the baseline period.39 These findings can be found in the Table 5
Table 5
 
Comparison of Coefficient of Variation Using Different Doppler-Based Technologies
Table 5
 
Comparison of Coefficient of Variation Using Different Doppler-Based Technologies
COV, %
CLBF Diameter or SD-OCT Doppler Area Velocity Flow
Median Range Median Range Median Range
CLBF39 2.0 0.5–6.5 19.9 4.8–39.7 19.3 4.8–37.3
CLBF/arteriole40 1.8 0.6–3.3 10.7 0.8–22.1 11.8 1.0–23.2
CLBF/venule40 1.4 0.1–7.1 9.8 1.2–20.8 9.96 1.6–20.8
SD-OCT Doppler/young 4.7 0.8–16.8 10.4 0.1–28.9 7.5 0.4–20.4
SD-OCT Doppler/elderly 4.8 0.4–20.4 10.8 1.1–43.8 9.2 0.6–34.6
Doppler FD-OCT repeatability for young and elderly subjects, in terms of the COR, was found to be 6.43 μL/min and 10.53 μL/min, respectively; the acceptable level of between-session variability needs to be interpreted based upon the magnitude of the effect size. The difference in repeatability between sessions was also greater in subjects with crowded vascular beds, which resulted in difficulty distinguishing the arteriolar from the venular Doppler phase-shift signals (Fig. 8). Other factors, such as tortuous vessels or curved and tapering vessels, can also play a role in the variability through the generation of irregular or non-Poiseuille flow conditions. The within-session variability of blood flow measurements could be a result of eye motion, tear film, the anatomy of the vessels, or intrinsic variation of RBF. The tear breakup can alter the laser intensity projected onto the retina and may result in displacement of the laser from the center of the vessel due to optical blurring effects. We attempted to keep all parameters (such as diet) in both sessions the same. Other factors such as consuming red meat or caffeine could play a role in the difference as well.42,43 These factors can contribute to the individual blood flow variation measured by Doppler FD-OCT. Another limitation of this study is that the maximum detectable volume of the Doppler FD-OCT system is slower than the maximum flow velocity, especially in retinal arterioles, assuming equal flow in arteriole and venule; as a result, all reported values were based upon venular measurements only. 
Figure 8
 
Fundus image of an elderly subject with tortuous vessels and high COR value.
Figure 8
 
Fundus image of an elderly subject with tortuous vessels and high COR value.
Factors that will cause Doppler FD-OCT measurements to be invalid fall into three main causes. Objects that are close to perpendicular to the incident OCT beam will result in a relatively weak Doppler signal because the cosine of 90° is zero. Also, errors in the measurement of the vessel lumen area will greatly influence the estimation of flow; this effect is compounded by the fact that the area estimation requires the subjective determination of the Doppler signal boundary and then the subjective assignment of confidence in the area estimation. In addition, the impact of eye movements will introduce further error into the estimation of vessel lumen area. 
To summarize, the degree of variability and repeatability can differ significantly between individuals, but overall, Doppler FD-OCT gave consistent and repeatable blood flow measurements within retinal venules in normal subjects. 
Acknowledgments
We thank Optovue, Inc., for the loan of two RTVue FD-OCT systems (CH and JGF). 
Supported by the Ontario Research Fund for Research Excellence (Grant RE-04-034) and by an anonymous donor. 
Disclosure: F. Tayyari, None; F. Yusof, None; M. Vymyslicky, None; O. Tan, Optovue, Inc. (F), P; D. Huang, Optovue, Inc. (F, I), P; J.G. Flanagan, Optovue, Inc. (F, R), Carl Zeiss Meditec (C); C. Hudson, Optovue, Inc. (F, R) 
References
Saint-Geniez M D'Amore PA. Development and pathology of the hyaloid, choroidal and retinal vasculature. Int J Dev Biol. 2004; 48: 1045–1058. [CrossRef] [PubMed]
Haefliger IO Flammer J Bény JL Lüscher TF. Endothelium-dependent vasoactive modulation in the ophthalmic circulation. Prog Retin Eye Res. 2001; 20: 209–225. [CrossRef] [PubMed]
Alm A. Ocular circulation. In: Hart WM ed. Adler's Physiology of the Eye. 9th ed. St. Louis, MO: Mosby-Year Book, Inc.; 1992: 198–227.
Schmetterer L Wolzt M. Ocular blood flow and associated functional deviations in diabetic retinopathy. Diabetologia. 1999; 42: 387–405. [CrossRef] [PubMed]
Archer DB. Bowman Lecture 1998. Diabetic retinopathy: some cellular, molecular and therapeutic considerations. Eye (Lond). 1999; 13: 497–523. [CrossRef] [PubMed]
Hayreh SS. The blood supply of the optic nerve head and the evaluation of it—myth and reality. Prog Retin Eye Res. 2001; 20: 563–593. [CrossRef] [PubMed]
Flammer J Orgül S Costa VP The impact of ocular blood flow in glaucoma. Prog Retin Eye Res. 2002; 21: 359–393. [CrossRef] [PubMed]
Grieshaber MC Mozaffarieh M Flammer J. What is the link between vascular dysregulation and glaucoma? Surv Ophthalmol. 2007; 52: S144–S154. [CrossRef] [PubMed]
Nicolela MT Walman BE Buckley AR Drance SM. Ocular hypertension and primary open-angle glaucoma: a comparative study of their retrobulbar blood flow velocity. J Glaucoma. 1996; 5: 308–310. [CrossRef] [PubMed]
Fuchsjäger-Mayrl G Wally B Georgopoulos M Ocular blood flow and systemic blood pressure in patients with primary open-angle glaucoma and ocular hypertension. Invest Ophthalmol Vis Sci. 2004; 45: 834–839. [CrossRef] [PubMed]
Grieshaber MC Flammer J. Blood flow in glaucoma. Curr Opin Ophthalmol. 2005; 16: 79–83. [CrossRef] [PubMed]
Harris A Zarfati D Zalish M Reduced cerebrovascular blood flow velocities and vasoreactivity in open-angle glaucoma. Am J Ophthalmol. 2003; 135: 144–147. [CrossRef] [PubMed]
Hafez AS Bizzarro RL Lesk MR. Evaluation of optic nerve head and peripapillary retinal blood flow in glaucoma patients, ocular hypertensives, and normal subjects. Am J Ophthalmol. 2003; 136: 1022–1031. [CrossRef] [PubMed]
Emre M Orgül S Gugleta K Flammer J. Ocular blood flow alteration in glaucoma is related to systemic vascular dysregulation. British J Ophthal. 2004; 88: 662–666. [CrossRef]
Huang D Swanson EA Lin CP Optical coherence tomography. Science. 1991; 254: 1178–1181. [CrossRef] [PubMed]
Fujimoto JG Brezinski ME Tearney GJ Optical biopsy and imaging using optical coherence tomography. Nat Med. 1995; 1: 970–972. [CrossRef] [PubMed]
Fercher AF. Optical coherence tomography. J Biomed Opt. 1996; 1: 157–173. [CrossRef] [PubMed]
Hee MR Puliafito CA Wong C Optical coherence tomography of macular holes. Ophthalmology. 1995; 102: 748–756. [CrossRef] [PubMed]
Schuman JS Hee MR Arya AV Optical coherence tomography: a new tool for glaucoma diagnosis. Curr Opin Ophthalmol. 1995; 6: 89–95. [CrossRef] [PubMed]
Wang XJ Milner TE Nelson JS. Characterization of fluid flow velocity by optical Doppler tomography. Opt Lett. 1995; 20: 1337–1339. [CrossRef] [PubMed]
Zhao Y Chen Z Saxer C Doppler standard deviation imaging for clinical monitoring of in vivo human skin blood flow. Opt Lett. 2000; 25: 1358–1360. [CrossRef] [PubMed]
Wang Y Bower BA Izatt JA Tan O Huang D. In vivo total retinal blood flow measurement by Fourier domain Doppler optical coherence tomography. J Biomed Opt. 2007; 12: 041215. [CrossRef] [PubMed]
Wang Y Lu A Gil-Flamer J Tan O Izatt JA Huang D. Measurement of total blood flow in the normal human retina using Doppler Fourier-domain optical coherence tomography. Br J Ophthalmol. 2009; 93: 634–637. [CrossRef] [PubMed]
Wang Y Bower BA Izatt JA Tan O Huang D. Retinal blood flow measurement by circumpapillary Fourier domain Doppler optical coherence tomography. J Biomed Opt. 2008; 13; 064003.
Konduru RK Tan O Nittala MG Huang D Sadda SR. Reproducibility of retinal blood flow measurements derived from semi-automated Doppler OCT analysis. Ophthalmic Surg Lasers Imaging. 2012; 43: 25–31. [CrossRef] [PubMed]
Bland JM Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet. 1986; 1: 307–310. [CrossRef] [PubMed]
Venkataraman ST Hudson C Rachmiel R Retinal arteriolar vascular reactivity in untreated and progressive primary open-angle glaucoma. Invest Ophthalmol Vis Sci. 2010; 51: 2043–2050. [CrossRef] [PubMed]
Bertram B Wolf S Arend O Schulte K Reim M. Retinal circulation and current blood glucose value in diabetic retinopathy [in German]. Klin Monbl Augenheilkd. 1992; 200: 654–657. [CrossRef] [PubMed]
Bursell SE Clermont AC Kinsley BT Retinal blood flow changes in patients with insulin-dependent diabetes mellitus and no diabetic retinopathy. Invest Ophthalmol Vis Sci. 1996; 37: 886–897. [PubMed]
Feke GT Buzney SM Ogasawara H Retinal circulatory abnormalities in type 1 diabetes. Invest Ophthalmol Vis Sci. 1994; 35: 2968–2975. [PubMed]
Ferris FL III Patz A. Macular edema. A complication of diabetic retinopathy. Surv Ophthalmol. 1984; 28 (suppl): 452–461. [CrossRef] [PubMed]
Gilmore ED Hudson C Nrusimhadevara RK Retinal arteriolar diameter, blood velocity, and blood flow response to an isocapnic hyperoxic provocation in early sight-threatening diabetic retinopathy. Invest Ophthalmol Vis Sci. 2007; 48: 1744–1750. [CrossRef] [PubMed]
Gilmore ED Hudson C Nrusimhadevara RK Retinal arteriolar hemodynamic response to an acute hyperglycemic provocation in early and sight-threatening diabetic retinopathy. Microvasc Res. 2007; 73: 191–197. [CrossRef] [PubMed]
Grunwald JE Riva CE Baine J Brucker AJ. Total retinal volumetric blood flow rate in diabetic patients with poor glycemic control. Invest Ophthalmol Vis Sci. 1992; 33: 356–363. [PubMed]
Kohner EM Patel V Rassam SM. Role of blood flow and impaired autoregulation in the pathogenesis of diabetic retinopathy. Diabetes. 1995; 44: 603–607. [CrossRef] [PubMed]
Nyengaard JR Ido Y Kilo C Williamson JR. Interactions between hyperglycemia and hypoxia: implications for diabetic retinopathy. Diabetes. 2004; 53: 2931–2938. [CrossRef] [PubMed]
Zimmet P Alberti KG Global Shaw J. and societal implications of the diabetes epidemic. Nature. 2001; 414: 782–787. [CrossRef] [PubMed]
Guan K Hudson C Flanagan JG. Variability and repeatability of retinal blood flow measurements using the Canon Laser Blood Flowmeter. Microvasc Res. 2003; 65; 145–151. [CrossRef]
Rose PA Hudson C. Comparison of retinal arteriolar and venular variability in healthy subjects. Microvasc Res. 2007; 73: 35–38. [CrossRef]
Riva C Ross B Benedek GB. Laser Doppler measurements of blood flow in capillary tubes and retinal arteries. Invest Ophthalmol. 1972; 11: 936–944.
Feke GT Goger DG Tagawa H Delori FC. Laser Doppler technique for absolute measurement of blood speed in retinal vessels. IEEE Trans Biomed Eng. 1987; 34; 673–680. [CrossRef]
Field AS Laurienti PJ Yen YF Burdette JH Moody DM. Dietary caffeine consumption and withdrawal: confounding variables in quantitative cerebral perfusion studies? Radiology. 2003; 227: 129–135. [CrossRef]
Ozkan B Yüksel N Anik Y The effect of caffeine on retrobulbar hemodynamics. Curr Eye Res. 2008; 33: 804–809. [CrossRef]
Figure 1
 
Doppler FD-OCT image acquisition page. A double circular scan pattern with concentric radii is used for the measurement, which is typically centered at the optic nerve head.
Figure 1
 
Doppler FD-OCT image acquisition page. A double circular scan pattern with concentric radii is used for the measurement, which is typically centered at the optic nerve head.
Figure 2
 
Doppler FD-OCT image for beam passing through the superior nasal quadrant of the pupil. The peak of the sinusoidal should be positioned either within superior nasal or inferior nasal quadrants.
Figure 2
 
Doppler FD-OCT image for beam passing through the superior nasal quadrant of the pupil. The peak of the sinusoidal should be positioned either within superior nasal or inferior nasal quadrants.
Figure 3
 
Cross-section of retinal blood vessels displaying Doppler signal. The grader adjusts the vessel circumference with a dotted circle and the diameter with a horizontal solid line.
Figure 3
 
Cross-section of retinal blood vessels displaying Doppler signal. The grader adjusts the vessel circumference with a dotted circle and the diameter with a horizontal solid line.
Figure 4
 
Box plot of individual COVs for blood flow in young and elderly participants. The error bars show the nonoutlier range (±1.5 times the height of the box).
Figure 4
 
Box plot of individual COVs for blood flow in young and elderly participants. The error bars show the nonoutlier range (±1.5 times the height of the box).
Figure 5
 
Bland-Altman plot showing the difference in total blood flow as a function of mean blood flow between sessions for young subjects. MoD, mean of differences; 95% Conf. Limit, 1.96 × SD of the MoD (also known as limits of agreement).
Figure 5
 
Bland-Altman plot showing the difference in total blood flow as a function of mean blood flow between sessions for young subjects. MoD, mean of differences; 95% Conf. Limit, 1.96 × SD of the MoD (also known as limits of agreement).
Figure 6
 
Bland and Altman plot showing the difference in total blood flow as a function of mean blood flow between sessions for elderly subjects. MoD, mean of differences; 95% Conf. Limit, 1.96 × SD of the MoD (also known as limits of agreement).
Figure 6
 
Bland and Altman plot showing the difference in total blood flow as a function of mean blood flow between sessions for elderly subjects. MoD, mean of differences; 95% Conf. Limit, 1.96 × SD of the MoD (also known as limits of agreement).
Figure 7
 
Scatter plot illustrating test-retest characteristics of the Doppler FD-OCT for measuring total retinal blood flow in microliters per minute.
Figure 7
 
Scatter plot illustrating test-retest characteristics of the Doppler FD-OCT for measuring total retinal blood flow in microliters per minute.
Figure 8
 
Fundus image of an elderly subject with tortuous vessels and high COR value.
Figure 8
 
Fundus image of an elderly subject with tortuous vessels and high COR value.
Table 1
 
Individual Subject/Group Characteristics for the Young and Elderly Groups*
Table 1
 
Individual Subject/Group Characteristics for the Young and Elderly Groups*
Name Age M/F IOP, mm Hg Visit 1 Visit 2
Visit 1 Visit 2 Sys, mm Hg Dias, mm Hg Sys, mm Hg Dias, mm Hg
Young group
 1 27 F 11.00 10.33 116 68 121 71
 2 24 F 11.33 12.00 115 75 111 74
 3 29 F 12.33 11.33 125 80 121 82
 4 27 F 12.00 12.33 110 69 113 68
 5 28 F 11.00 11.66 97 63 110 64
 6 19 F 15.00 14.00 109 69 100 68
 7 24 F 13.00 12.66 113 67 124 71
 8 26 F 12.66 12.00 106 67 115 73
 9 28 F 11.00 12.00 113 78 102 81
 10 26 F 12.00 11.66 119 81 122 69
 11 22 M 15.00 14.00 128 78 119 69
 12 23 M 14.33 13.66 125 65 111 78
 13 25 M 9.00 10.33 128 71 120 81
 14 26 M 18.33 16.66 128 82 135 87
 15 24 M 13.66 12.33 131 76 129 75
 16 25 M 17.00 15.33 128 67 121 71
 17 20 M 12.00 12.66 129 69 125 73
 18 23 M 13.33 13.00 125 78 119 75
 19 26 M 17.66 16.33 127 81 125 78
 20 22 M 16.00 16.00 129 60 128 61
Mean 24.7 13.38 13.01 120.05 72.2 118.55 73.45
SD 2.7 2.48 1.86 9.59 6.69 8.77 6.4
Elderly group
 21 64 F 15.00 15.33 145 56 139 59
 22 63 F 13.33 14.00 105 69 111 70
 23 62 F 13.00 15.00 119 84 121 83
 24 62 F 14.00 13.00 136 83 129 86
 25 67 F 15.00 16.00 142 79 134 81
 26 61 F 14.00 13.66 138 83 145 84
 27 68 F 16.00 15.66 133 86 135 85
 28 70 F 11.00 12.00 150 79 145 87
 29 59 F 15.00 14.66 124 78 132 85
 30 78 F 12.66 12.00 124 78 129 75
 31 70 F 14.00 13.33 128 79 135 79
 32 60 M 15.00 14.33 131 72 122 76
 33 60 M 13.33 13.66 120 76 128 75
 34 67 M 14.00 15.33 122 74 124 78
 35 62 M 14.00 13.66 125 79 131 81
 36 60 M 14.00 14.33 128 78 112 73
Mean 64.6 13.64 13.51 125.58 74.36 123.49 75.72
SD 5.1 2.00 1.68 12.07 7.21 10.42 7.17
Table 2
 
Coefficient of Variation (%) for Venous Area, Velocity, and Blood Flow
Table 2
 
Coefficient of Variation (%) for Venous Area, Velocity, and Blood Flow
COV, %
Venous Area Velocity Flow
Median Range Median Range Median Range
Young 4.7 0.8–16.8 10.4 0.1–28.9 7.5 0.4–20.4
Elderly 4.8 0.4–20.4 10.8 1.1–43.8 9.2 0.6–34.6
Table 3
 
Coefficient of Repeatability for Young Individuals as a Group
Table 3
 
Coefficient of Repeatability for Young Individuals as a Group
COR in Young Subjects
Individual COR Overall COR Mean Effect Size
Median Range
Venous area 0.003 0.0003–0.017 0.005 0.033
Velocity 4.25 1.11–17.53 6.12 21.90
Flow 5.91 0.23–13.76 6.43 39.76
Table 4
 
Coefficient of Repeatability for Elderly Individuals as a Group
Table 4
 
Coefficient of Repeatability for Elderly Individuals as a Group
COR in Elderly Subjects
Individual COR Overall COR Mean Effect Size
Median Range
Venous area 0.005 00.1–0.02 0.006 0.035
Velocity 7.19 2.22–18.12 8.20 21.77
Flow 9.16 0.37–38.83 10.53 46.93
Table 5
 
Comparison of Coefficient of Variation Using Different Doppler-Based Technologies
Table 5
 
Comparison of Coefficient of Variation Using Different Doppler-Based Technologies
COV, %
CLBF Diameter or SD-OCT Doppler Area Velocity Flow
Median Range Median Range Median Range
CLBF39 2.0 0.5–6.5 19.9 4.8–39.7 19.3 4.8–37.3
CLBF/arteriole40 1.8 0.6–3.3 10.7 0.8–22.1 11.8 1.0–23.2
CLBF/venule40 1.4 0.1–7.1 9.8 1.2–20.8 9.96 1.6–20.8
SD-OCT Doppler/young 4.7 0.8–16.8 10.4 0.1–28.9 7.5 0.4–20.4
SD-OCT Doppler/elderly 4.8 0.4–20.4 10.8 1.1–43.8 9.2 0.6–34.6
×
×

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.

×