QUALITATIVE AND QUANTITATIVE

FOLLOW-UP USING OPTICAL

COHERENCE TOMOGRAPHY
ANGIOGRAPHY OF RETINAL VEIN
OCCLUSION TREATED WITH ANTI-VEGF
Optical Coherence Tomography Angiography
Follow-up of Retinal Vein Occlusion
ALEXANDRE SELLAM, MD, AGNES GLACET-BERNARD, MD, FLORENCE COSCAS, MD,
ALEXANDRA MIERE, MD, GABRIEL COSCAS, MD, ERIC H. SOUIED, MD, PHD

Purpose: To evaluate changes of vascular flow of patients treated with intravitreal

injections of anti-vascular endothelial growth factor for macular edema secondary to retinal
vein occlusion (RVO) with optical coherence tomography angiography (OCTA).
Methods: Patients with RVO with macular edema and treated with intravitreal injections of
anti-vascular endothelial growth factors were retrospectively evaluated. The following
examinations were performed before and after treatment: best-corrected visual acuity, spectral
domain optical coherence tomography, fluorescein angiography, and OCTA (Optovue, Inc).
Automatic measurement of vascular density of the superficial and deep capillary plexus was
also performed and compared with age- and sex-matched healthy subjects.
Results: Twenty-eight eyes of 28 patients (mean age 66.2 years; males 19%) were
evaluated, including 13 central RVO, 11 branch RVO, and 4 hemicentral RVO. After
treatment, mean central macular thickness significantly decreased from 644 mm to 326 mm
and best-corrected visual acuity increased from 20/125 to 20/63 (P , 0.01 for both results).
On OCTA, perifoveal capillary disruption (P = 0.029) and the number of cysts in the super-
ficial capillary plexus and deep capillary plexus (P , 0.002) significantly decreased after
treatment. The mean vascular density in the superficial capillary plexus slightly decreased
during follow-up from 46.44% to 45.01% (not significantly). These densities were signifi-
cantly less than those observed in healthy controls (P , 0.001).
Conclusion: Optical coherence tomography angiography showed regression of macular
edema, reduced capillary disruption and cysts, and slight decrease in mean macular
vascular density with time and despite treatment. Thus, OCTA enables qualitative and
quantitative evaluation during follow-up of patients treated for RVO.
RETINA 0:1–9, 2016

R etinal vein occlusion (RVO) is the second most

common retinal vascular disease affecting the ret-
ina and is an important cause of visual loss.1,2 Retinal
vein occlusions are divided into central (CRVO),
hemi, and branch retinal vein occlusions (BRVO).3,4
Visual acuity loss after RVO commonly occurs as
a result of macular edema but may also result from
macular ischemia or neovascular complications such
as vitreous hemorrhage and neovascular glaucoma.5
The features of RVO have been extensively described
using fundus photography, spectral domain optical
coherence tomography (SD-OCT), and fluorescein
angiography (FA).6–8 The diagnosis of RVO is based
on fundus examination although FA is often per-
formed to confirm the diagnosis (delayed filling of
the occluded retinal vein) and to identify macular
and peripheral nonperfused areas and other compli-
cations such as macular edema, neovascularization,

Copyright ª by Ophthalmic Communications Society, Inc. Unauthorized reproduction of this article is prohibited.
2 RETINA, THE JOURNAL OF RETINAL AND VITREOUS DISEASES  2016  VOLUME 0  NUMBER 0

Table 1. Baseline Characteristics of Patients With Retinal Venous Occlusion

Total Group, n = 28 CRVO, n = 13 BRVO-HCVO, n = 15
Age, mean ± SD (years) 66.2 ± 12.3 64.1 ± 14.1 68.1 ± 10.5
Male, n (%) 18 (64.3) 10 (77) 8 (53)
HTA, n (%) 11 (39) 4 (31) 7 (47)
Diabetes 8 (29) 2 (15) 6 (40)
Naive patient, n (%) 20 (71) 9 (69) 11 (73)
Duration of RVO at baseline, 12.2 ± 18.3 15.6 ± 19.5 9.3 ± 17.3
mean ± SD (months)
Duration of follow-up, mean ± SD 19.5 ± 12.9 19.5 ± 13.4 19.5 ± 13.1
(weeks)
Number of previous intravitreal 1.54 ± 2.82 5.7 ± 4.1 3.5 ± 1.9
injection in nonnaive patients,
mean ± SD
Previous peripheral laser, n (%) 14 (50) 7 (54) 7 (47)
Previous focal or grid laser, n (%) 2 (7) 0 2 (13)
BCVA, mean ± SD (logMAR) 0.78 ± 0.41 0.84 ± 0.43 0.73 ± 0.38
Mean BCVA (Snellen) 20/125 20/125 20/100
Medication rnb: 18, afb: 10 rnb: 8, afb: 5 rnb: 10, afb: 5
Number of intravitreal injection 3 ± 1.7 3.15 ± 2.1 2.8 ± 1.3
during follow-up, mean ± SD
afb, aflibercept; HCVO, hemicentral retinal vein occlusion; HTA, hypertension; rnb, ranibizumab.

and associated artery occlusions.9 Monitoring of pa-
tients with RVO treated for macular edema is based
on visual acuity evaluation and SD-OCT with mea-
surement of the central macular thickness (CMT).
Fluorescein angiography enables detection of the
presence or extension of nonperfused areas.
Recently, the introduction of optical coherence
tomography angiography (OCTA) has provided
a novel method for imaging the capillary network
in a noninvasive manner.10 Optical coherence
tomography angiography uses the split-spectrum
amplitude–decorrelation angiography algorithm to
detect erythrocyte movement. Optical coherence
tomography angiography has previously been used
to describe the retinal vasculature in normal eyes
and diseases such as macular telangiectasia type 2,
RVO, and diabetes.11–13
Kuehlewein et al14 reported a patient with branch
RVO who was evaluated with FA and swept-source
OCT microangiography. Areas of nonperfusion after
BRVO could be precisely delineated at the deep cap-
From the Department of Ophthalmology, Centre Hospitalier Inter-
communal de Creteil, University Paris Est Creteil, Creteil, France.
None of the authors have any financial/conflicting interests to
disclose.
Study concept and design: F. Coscas, A. Glacet-Bernard, and
A. Sellam; Acquisition of data: F. Coscas and A. Sellam; Analysis
and interpretation of data: A. Sellam, A. Glacet-Bernard, and F. Coscas;
Drafting the manuscript: A. Sellam, F. Coscas, and A. Glacet-Bernard;
Critical revision of the manuscript for important intellectual content:
E. H. Souied, A. Glacet-Bernard, and G. Coscas.
Reprint requests: Eric H. Souied, MD, PhD, Centre Hospitalier
Intercommunal de Créteil, Université Paris Est-Créteil, 40 Avenue
Verdun, 94010 Créteil, France; e-mail: esouied@hotmail.com
illary plexus (DCP) level using this imaging modality.
De Carlo et al15 also described an area of diffuse
capillary nonperfusion, continuous with the foveal
avascular zone and telangiectatic vessels, in a patient
with CRVO. More recently, our team described the
main RVO features on OCTA and correlated the peri-
foveal capillary arcade disruption observed using OC-
TA with the peripheral ischemia found on FA.12
Suzuki et al16 found that OCTA could visualize, as
well as FA, the microvascular abnormalities in mac-
ular edema associated with BRVO.
A recent version of the software released by the
AngioVue system enables quantification of vascular
density around the macula. This tool might be useful
in evaluating retinal vascular disorders such as RVO
or diabetes.17 We are not aware of any studies that
described the follow-up of patients with RVO using
OCTA. Therefore, the aim of our study was to follow
up, using OCTA, patients with RVO treated with anti-
vascular endothelial growth factor (anti-VEGF) and to
compare these results with those of SD-OCT and FA.
Our follow-up was both qualitative and quantitative
using vascular density measurement.
Materials and Methods
This retrospective, observational, and consecutive
case series was conducted at the University Eye Clinic
of Creteil from November 1, 2014, through October
31, 2015. The study design was approved by the
Institutional Review Board of Creteil University. All
patients provided written informed consent for

Copyright ª by Ophthalmic Communications Society, Inc. Unauthorized reproduction of this article is prohibited.
 OCTA Follow up of RVO  SELLAM ET AL 3

Table 2. Optical Coherence Tomography and Angiographic Data of Patients With RVO at Baseline and Final Follow-up
Initial Visit, n = 28 Final Visit, n = 28 P
OCT features
CMT, mean ± SD (mm) 634 ± 203 326 ± 135 ,0.001*
Hyperreflective dots, n (%) 27 (96) 3 (11) ,0.001*
Ellipsoid zone, n (%)
Interrupted 15 (54) 9 (32)
Fragmented 9 (32) 2 (7)
Normal 2 (7) 15 (54) 0.009*
Serous retinal detachment, n (%) 11 (39) 11 (39) 0.196
Epimacular membrane or 4 (14) 7 (25) 0.125
vitreomacular traction, n (%)
Initial Visit, n = 20 Final Visit, n = 20 P
FA, n (%)
Disruption the perifoveolar 14 (78) 14 (78) 0.375
capillary arcade
Late macular leakage 16 (88) 5 (28) ,0.001*
Macular ischemia 12 (63) 12 (63) 0.5
Peripheral ischemia 13 (65) 14 (70) 0.5
*Significant value.
OCT, optical coherence tomography.

participation in the study. The described research
methods and analysis adhered to the tenets of the
Declaration of Helsinki.
We retrospectively evaluated the case records of 28
consecutive patients with macular edema secondary to
RVO. Inclusion criteria were the presence of macular
edema secondary to RVO and central (CRVO), hemi-
central, or branch RVO (BRVO). Macular edema was
treated by intravitreal injections of anti-VEGF (either
ranibizumab or aflibercept). Exclusion criteria were
patients with poor-quality images on OCTA (signal
strength index lower than 50) due to eye movements or
media opacities, previous retinal surgery, pathologic
myopia, ocular trauma, and uncontrolled diabetes.
All patients underwent complete eye examination
including best-corrected visual acuity (BCVA) mea-
surement with the Early Treatment Diabetic Retinopathy
Study (ETDRS) charts, anterior segment examination,
intraocular pressure measurement, color fundus photog-
raphy, FA, SD-OCT (Spectralis Heidelberg Engineer-
ing, Heidelberg, Germany), and OCTA using the
split-spectrum amplitude–decorrelation angiography
algorithm on the AngioVue OCTA system version
2015.100.0.35 (Optovue RTVue XR 100 AVANTI,
Inc, Fremont, CA) intended for retinal imaging
(AngioRetina mode, software AngioAnalytics) as
previously described.14,15 The calculated amplitude–
decorrelation signal from consecutive B-scans

Table 3. Optical Coherence Tomography Angiography Data of Patients With RVO at Baseline and Final Follow-up
Initial Visit, n = 24 Final Visit, n = 24 P
Disruption the perifoveolar capillary 4 (18) 10 (45) 0.029*
arcade of less than the half, n (%)
Capillary loss at the SCP in 3 6 (27) 5 (23) 0.287
quadrants or more, n (%)
Capillary loss at the DCP in 3 10 (44) 5 (22) —
quadrants or more, n (%)
Venous dilation at the SCP, n (%) 21 (87) 12 (50) 0.002*
Venous dilation at the DCP, n (%) 19 (79) 13 (54) 0.031*
Vascular ectasia at the SCP, n (%) 10 (42) 12 (50) 0.219
Vascular ectasia at the DCP, n (%) 16 (64) 17 (72) 0.167
Numerous cysts located outside the 12 (50) 1 (4) ,0.001*
central area at the SCP (grading $
2), n (%)
Numerous cysts located outside the 21 (88) 5 (21) 0.002*
central area at the DCP (grading
$ 2), n (%)
*Significant value.

Copyright ª by Ophthalmic Communications Society, Inc. Unauthorized reproduction of this article is prohibited.
4 RETINA, THE JOURNAL OF RETINAL AND VITREOUS DISEASES  2016  VOLUME 0  NUMBER 0

allowed visualization of blood flow and, therefore, also matched to a healthy subject for age and sex using the
of the capillary network. Image analysis was per- normative database previously published.18
formed using automated retinal segmentation derived
from the machine’s software. Motion correction was
performed using registration of 2 orthogonally cap- Statistical Analysis
tured imaging volumes.
On SD-OCT scans, we evaluated CMT, the presence The BCVA was converted to the logarithm of the
of subretinal fluid, hyperreflective dots (presence when minimum angle resolution (logMAR) for statistical
they were at least 5, or otherwise absence), epimacular evaluation. Fischer’s exact test was used for qualitative
membrane, or vitreomacular traction, and the appearance variables and the Student’s t test for quantitative data
of the ellipsoid zone (intact, fragmented, or nonobserv- analysis (IBM SPSS Statistics 19, Chicago, IL).
able). On FA, we evaluated perifoveal capillary arcade
disruption, the delay between arterial and venous filling,
macular and peripheral nonperfused areas, and the late-
stage leakage of fluorescein occasionally resulting in
petaloid macular edema. When the nonperfused area was
more than 10 disk areas (CRVO) or 5 disk areas
(BRVO), RVO was considered as ischemic.
Two retinal specialists (A.S. and F.C.) indepen-
dently read OCT-angiographic images to determine
the disruption of the perifoveal capillary arcade at the
superficial capillary plexus (SCP), in quadrants,
grading 0 to 4 (0, none; 1–4, only one to all 4
ETDRS quadrants, respectively) capillary dropout
in different quadrants of the macula, dilation of ves-
sels, vascular ectasia, and macular edema cysts (SCP
and DCP).12 “En face” OCT enabled macular edema
cysts to be graded from 0 to 3 (none, central, central,
and paracentral, diffuse to all the macular field). All
these parameters were evaluated before and after
treatment.
Quantitative follow-up was reported using the
measurement of vascular density. Vascular density
was defined as the percentage of the sample area
occupied by vessel lumens after binary reconstruc-
tion of images. The percentage of vessels was
defined in sectors based on a simplified ETDRS
chart (superior, temporal, inferior, and nasal) in
which the foveal center was automatically deter-
mined from the structure SD-OCT data. The inner
and outer rings, with a diameter of 1 mm and
2.5 mm, respectively, around the fovea, were
analyzed. The foveal vascular density corresponds
to the 1-mm ring around the fovea. The parafoveal
vascular density corresponds to the mean vascular
density of all 4 quadrants of the ETDRS chart
(superior, temporal, inferior, and nasal).
The flow density map software AngioAnalytics is the
first OCTA automatic quantification tool that enables
measurement of flow, nonflow, and flow area density.
AngioAnalytics assesses and compares the relative
density of flow as a percentage of the total area. These Fig. 1. Fluorescein angiography of a 45-year-old male with CRVO
(BCVA: 20/80). Early-phase (left) and late-phase (right) FA shows the
measurements were performed before and after treat- absence of nonperfused areas in the posterior pole and periphery, and
ment and then statistically analyzed. Each patient was dye staining in the macular cysts in the late phase.

Copyright ª by Ophthalmic Communications Society, Inc. Unauthorized reproduction of this article is prohibited.
 OCTA Follow up of RVO  SELLAM ET AL 5

McNemar’s test was used to compare qualitative paired
data. P , 0.05 was considered statistically significant.
Results
The case records of 28 eyes of 28 patients with
RVO treated with anti-VEGF injections were ana-
lyzed; there were 13 CRVO, 11 BRVO, and 4 patients
with hemicentral RVO. The mean age was 66.2 years
and 18 (64%) were males. The baseline characteristics
of the study patients are shown in Table 1. The median
period between the onset of RVO and inclusion in our
study was 3 months (range, 0–61 months). Twenty
eyes (71%) were treatment naive. Among treated pa-
tients, the mean number of previous injections was 5
(range, 1–10). Eighteen patients received ranibizumab
injections and the other 10 received aflibercept. The
mean number of anti-VEGF injections was 3 (range,
1–7) during the mean follow-up period of 19 weeks
(range, 4–51). At baseline, 14 patients had undergone
peripheral photocoagulation of the nonperfused retina
and 2 patients with BRVO had undergone partial grid
macular photocoagulation.
At the final evaluation, BCVA had increased
statistically significantly from a mean of 20/125 to
20/63 (in logMAR from 0.78 to 0.5, P , 0.01),
whereas mean CMT had significantly decreased from
644 mm to 326 mm (P , 0.001). On SD-OCT, 11 eyes
(39%) had serous detachment of the fovea, 27 (96%)
had hyperreflective dots, and 4 (14%) had an epiretinal
membrane. The ellipsoid zone was intact in only 2
eyes (7%), fragmented in 9 (32%), and nonobservable
in 17 eyes (61%). After treatment, the ellipsoid zone
appeared to significantly recover with an intact appear-
ance in 15 (54%) eyes (P = 0.009). Similarly, SD-OCT
showed a significant decrease in the number of hyper-
reflective dots after treatment (P , 0.001).
On FA (n = 20 patients), perifoveal capillary arcade
disruption was observed in 14 (78%) eyes, macular
ischemia in 12 (63%), peripheral ischemia in 13
(65%), and late-stage leakage in 16 (88%) eyes. The
mean arteriovenous transit time was 9 seconds. All
initial and final data obtained from SD-OCT and FA
are shown in Table 2.
On OCTA, perifoveal capillary arcade was intact in
only 1 eye (4%), involved 1 to 2 quadrants in 8 (35%)
eyes, and involved 3 to 4 quadrants in 14 (61%) eyes.
After treatment, there was a statistically significant
decrease in perifoveal capillary arcade disruption (P =
0.029). Table 3 shows the OCTA of the 24 eyes with
analyzable results in both initial and final examinations
(before and after treatment). After treatment, there was
a significant decrease in the number of macular cysts

Fig. 2. Optical coherence tomography angiography of the superficial and DCPs at baseline (same patient as in Figure 1). At the SCP level, ETDRS grid on
10° central macula with foveal ring (1 mm), 4 quadrants (superior, inferior, temporal, and nasal) and parafoveal ring 2.5 mm: focal foveal capillary arcade
disruption in the inferonasal part (top left); en face optical coherence tomography shows dark central cysts (top middle left); macular vessel density map
shows predominant lack of vessels in the inferonasal quadrant with whole en face density of 50.63%. SD-OCT shows the place of segmentation at the SCP
level (top right); at the DCP, there is dilation of superior macular capillaries with hypersignal (white dashed arrow) (bottom left); en face OCT shows
numerous central and paracentral dark cysts (bottom middle left); on the macular vessel density map, the low perfusion area appears to be larger and the
whole en face density is low (51.64%) (bottom middle right); SD-OCT shows the place of segmentation at the DCP (bottom right).

Copyright ª by Ophthalmic Communications Society, Inc. Unauthorized reproduction of this article is prohibited.
6 RETINA, THE JOURNAL OF RETINAL AND VITREOUS DISEASES  2016  VOLUME 0  NUMBER 0
in the SCP and DCP (P , 0.001 and P = 0.002,
respectively). Vascular dilation was also significantly
decreased after treatment in both plexuses (P = 0.002
for the SCP and P = 0.031 for the DCP) (Figures 1–5).
Five patients had OCTA that was not interpretable
because of high artifacts.
During the study follow-up period, mean vascular
density slightly decreased in all quadrants except the
foveal zone, particularly in the SCP (Table 4 shows the
values of vascular densities in the SCP and DCP in
Fig. 3. Fluorescein angiography of the patient in Figures 1 and 2 after
treatment with aflibercept. Three months after aflibercept intravitreal
injection, FA in the early-phase (left) and late-phase (right) shows
significant reduction in macular edema with recovery of the CRVO
appearance (BCVA: 20/20).
Copyright ª by Ophthalmic Communications Society, Inc. Unauthorized reproduction of this article is prohibited.
both study patients and healthy subjects). The mean
whole en face vascular density in the SCP slightly
decreased during follow-up from 46.44% to 45.01%
and the density in the parafoveal zone decreased from
48.02% to 46.34%; both differences were not statisti-
cally significant. Decreased vascular density was also
observed in the DCP, but it was tiny compared with
the SCP; and this difference was also not statistically
significant. There was no statistically significant dif-
ference between the SCP and the DCP in the change in
vascular density.
Nine patients (36%) had increased whole en face
vascular density in the SCP, 5 (20%) had the same
vascular density, and 11 (44%) had decreased vascular
density; the mean change in vascular density in these
subgroups was +3.53, 0, and 26.10, respectively.
Compared with patients with decreased vascular den-
sity, patients who experienced an increase had lower
vascular density at baseline (43.31% vs. 47.95%, P =
0.048), were more frequently men (P = 0.058), and
had hypertension (P = 0.002). There was no difference
between both subgroups in age, initial and final
BCVA, CMT, the number of injections, or the time
period before treatment. Improvement in vascular den-
sity did not correlate with improvement in visual acu-
ity, but initial whole en face density in the SCP and
DCP correlated with initial BCVA (P = 0.013 and P ,
0.001, respectively) and final BCVA (P = 0.030 and
P , 0.001, respectively).
No difference was observed between BRVO and
CRVO eyes in terms of vascular density. The
comparison between patients with RVO (before and
after treatment) and healthy subjects for vascular
density on OCTA demonstrated a significant differ-
ence in all measurements—all quadrants in the SCP
and the DCP, except the foveal zone (Table 4).
Discussion
Anti-VEGF injections are commonly used to effec-
tively treat cystoid macular edema secondary to
RVO.19 In our study, as expected, anti-VEGF injec-
tions significantly improved BCVA and decreased
CMT on SD-OCT. The follow-up of patients with
RVO using OCTA before and after treatment of mac-
ular edema with anti-VEGF injections showed quali-
tative improvement in the capillary plexus with
a significant decrease in vascular dilation (P = 0.002
for the SCP and P = 0.031 for the DCP) and perifoveal
capillary arcade recovery (P = 0.029).
However, this qualitative improvement is somewhat
contradicted by a slight quantitative decrease in
vascular density during the follow-up period,
 OCTA Follow up of RVO  SELLAM ET AL 7

Fig. 4. Optical coherence tomography angiography after aflibercept injection (same patient as in Figures 1–3) at the superficial (top) and DCP (bottom)
levels. At the SCP, ETDRS grid on 10° central macula, shows the complete absence of dark cysts and an intact perifoveal capillary arcade (top left); en
face optical coherence tomography shows the complete disappearance of central cysts (top middle left); on macular vessel density map, all perfusion
areas appear normal and whole en face density (51.47%) was increased (top middle right); optical coherence tomography shows the place of seg-
mentation at the SCP (top right); at the DCP, there is an absence of dark cysts and persistence of focal capillary dropout in the inferior part of the fovea;
en face optical coherence tomography shows the absence of central cystoid edema (bottom middle left); macular vessel density map shows the dis-
appearance of almost the whole area of capillary dropout observed before treatment (except in the inferior part) with increased whole en face density of
55.63% (bottom middle right); optical coherence tomography shows the place of segmentation at the DCP (bottom right).

particularly in the SCP (the mean whole en face
vascular density slightly decreased from 46.44% to
45.01%). This probably reflects the extension of retinal
nonperfusion over time in some patients with CRVO
and BRVO, a situation that has been recognized for
many years as characterizing the conversion from
nonischemic to ischemic RVO.3 Campochiaro et al,20
in 2014, demonstrated that an aggressive blockade of
VEGF by monthly anti-VEGF injections may reduce
the progression of retinal nonperfusion at the posterior
pole but cannot totally prevent it. However, retinal non-
perfusion could be present at the initial stage of the
RVO but was revealed only after treatment with the
regression of the cysts. The detection of capillary non-
perfusion areas on OCTA could also be due to very
slow blood flow in these areas, below the detection limit
of 0.3 mm/second on OCTA.10 In addition, before and
after treatment, the comparison of vascular density on
OCTA between patients treated for RVO and healthy
subjects matched for age and sex showed a significant
difference in both plexuses and in all quadrants. This
confirms the ischemic potential of RVO.
In this study, we used the prototype software
(AngioAnalytics) that achieved excellent interexamin-
er and intraexaminer repeatability and interexaminer
reproducibility in our previous study.18 Another soft-
ware (ImageJ) has also been used to evaluate vascular
density, but when images are skeletonized, the diam-
eter of each vessel is the same and so it could not
consider vessel dilations.17 The vascular density data
obtained with AngioAnalytics could facilitate differen-
tiating healthy subjects and patients with progressive
stages of various retinal vascular diseases. However,
AngioAnalytics software considered the foveal vascu-
lar density in an area of 1-mm ring around the center,
including the foveal avascular zone. This could
explain the absence of a significant difference in foveal
vascular density between patients with RVO and
healthy subjects.
After treatment with anti-VEGF injections, on
OCTA, retinal cysts disappeared in both plexuses,
but vascular density as measured by AngioAnalytics
was improved in only 36% of patients. In the case of
diabetic macular edema, Couturier et al proposed two
hypotheses to explain the fact that no capillaries were
detected in the areas of the cysts in both the superficial
and DCPs: the capillaries could be displaced at the
periphery of the cysts or the cysts had preferential
development in nonperfused areas.13 The possibility of
increased vascular density in both plexuses after treat-
ment in some patients of the current study and the fact
that in naive patients the capillary bed is normal just

Copyright ª by Ophthalmic Communications Society, Inc. Unauthorized reproduction of this article is prohibited.
8 RETINA, THE JOURNAL OF RETINAL AND VITREOUS DISEASES  2016  VOLUME 0  NUMBER 0

Fig. 5. Optical coherence tomography angiography of a patient with an important retinal ischemia persistent after treatment by intravitreal injections for
a superotemporal BRVO, at the superficial (top) and DCP (bottom) levels. At the SCP, ETDRS grid on 10° central macula, shows the complete absence
of the capillary on the superior part (top left); en face optical coherence tomography shows the absence of central cysts (top middle left); on macular
vessel density map, the perfusion appears very low at the superior part and whole en face density (39.48%) was decreased (top middle right); optical
coherence tomography shows the place of segmentation at the SCP (top right); at the DCP, there is also a capillary dropout in the superior part. En face
optical coherence tomography shows the absence of central cystoid edema (bottom middle left); on macular vessel density map, the perfusion appears
also very low at the superior part and whole en face density (46.90%) was decreased compared with healthy subject (bottom middle right); optical
coherence tomography shows the place of segmentation at the DCP (bottom right).

before the onset of RVO and macular edema seem to
be more consistent with the first hypothesis as regard
to RVO. Automatic segmentation provided by the An-
gioVue OCTA system could not validate this hypoth-
esis. Indeed, plexus autosegmentation showed the
image of the SCP variably superimposed on the
DCP and failed to clearly separate both plexuses.21
Therefore, a change in capillary density at the level
of the SCP would have an effect on DCP vessel den-
sity as assessed by AngioAnalytics. Only thinner
cross-sections at the level of the cysts would be able
to confirm this hypothesis. It is also important to note
that cysts were observed only when they were en-
circled by vessels.

Table 4. Vascular Density Measurements in Patients With RVO Using Optical Coherence Tomography-Angiography at
Baseline and Final Follow-up, and Comparison With Healthy Controls
RVO–Initial Visit RVO–Final Visit Controls, n = 37
(Mean ± SD) (Mean ± SD) Initial/Final, P (Mean ± SD) Initial/Controls, P
SCP
Whole 46.44 ± 4.98 45.01 ± 5.32 0.195 50.91 ± 3.53 ,0.001*
Foveal 33.95 ± 9.03 30.03 ± 9.91 0.064 31.11 ± 4.31 0.244
Parafoveal 48.02 ± 4.89 46.35 ± 5.83 0.17 52.74 ± 3.70 ,0.001*
Temporal 47.01 ± 4.81 45.43 ± 6.61 0.232 51.39 ± 4.60 ,0.001*
Superior 47.09 ± 7.42 47.43 ± 6.85 0.828 53.11 ± 5.65 ,0.001*
Nasal 47.99 ± 5.75 45.73 ± 6.71 0.138 51.83 ± 3.58 0.003*
Inferior 48.91 ± 5.98 46.82 ± 7.23 0.127 54.48 ± 4.02 ,0.001*
DCP
Whole 48.14 ± 4.71 48.14 ± 6.01 0.729 56.64 ± 3.08 ,0.001*
Foveal 30.77 ± 12.48 30.05 ± 10.07 0.768 31.06 ± 5.52 0.719
Parafoveal 50.29 ± 4.67 50.28 ± 5.92 0.995 58.85 ± 3.79 ,0.001*
Temporal 48.46 ± 5.28 47.84 ± 6.89 0.658 57.32 ± 4.97 ,0.001*
Superior 50.38 ± 5.86 50.28 ± 9.30 0.957 59.45 ± 5.23 ,0.001*
Nasal 50.06 ± 7.32 49.91 ± 7.20 0.884 57.33 ± 3.80 ,0.001*
Inferior 52.26 ± 8.56 51.81 ± 7.08 0.638 60.94 ± 3.32 ,0.001*
*Significant value.

Copyright ª by Ophthalmic Communications Society, Inc. Unauthorized reproduction of this article is prohibited.
 OCTA Follow up of RVO  SELLAM ET AL
This study suffers from limitations such as the small
sample size and its retrospective nature. Furthermore,
OCTA using AngioVue requires steady patient fixa-
tion for approximately 4 seconds without any blinking,
whereas useful FA images can be obtained in a fraction
of a second. Fully automatic segmentation of the
superficial and DCPs is not always reliable in healthy
subjects and especially in the presence of cystoid
macular edema. This imperfection allows variable
superimposition of different layers.21 The index of
vessel density may be erroneous in the DCP because
of the presence of vessels from the SCP.22 The algo-
rithm of delimitation for the segmentation of the dif-
ferent layers is not yet well adapted, particularly for
the outer limits. Unfortunately, manual correction of
such errors will exclude automatic quantitative meas-
urements. There is no clear distinction in quantitative
evaluation, in virtual colors, between the low perfusion
areas and cystic spaces. This limitation could be par-
tially compensated with analysis of en face images that
enables clear definition of the cystic spaces.
In conclusion, the follow-up of patients with RVO
with macular edema treated with anti-VEGF injections,
using OCTA, showed qualitative improvement in the
capillary plexuses and significant recovery of the peri-
foveal capillary arcade. However, quantitatively, OCTA
showed a slight decrease in macular vascular density with
time and despite anti-VEGF therapy. This study suggests
the potential contribution of OCTA as a new noninvasive
imaging technology that enables qualitative and quanti-
tative monitoring of macular edema in RVO. Multimodal
imaging, including not only FA and OCT but also
OCTA, could provide the optimal approach for clinicians
to monitor patients with RVO.
Key words: optical coherence tomography angiog-
raphy, retinal vein occlusion, vascular density, anti-
VEGF.
References
1. Rogers S, McIntosh RL, Cheung N, et al. The prevalence of
retinal vein occlusion: pooled data from population studies
from the United States, Europe, Asia, and Australia. Ophthal-
mology 2010;117:313–319.
2. Jaulim A, Ahmed B, Khanam T, Chatziralli IP. Branch retinal
vein occlusion: epidemiology, pathogenesis, risk factors, clin-
ical features, diagnosis, and complications. An update of the
literature. Retina 2013;33:901–910.
3. Hayreh SS, Zimmerman MB, Podhajsky P. Incidence of vari-
ous types of retinal vein occlusion and their recurrence and
demographic characteristics. Am J Ophthalmol 1994;117:
429–441.
Copyright ª by Ophthalmic Communications Society, Inc. Unauthorized reproduction of this article is prohibited.
9
4. Jonas J, Paques M, Mones J, Glacet-Bernard A. Retinal vein
occlusions. Dev Ophthalmol 2010;47:111–135.
5. Central Vein, Occlusion, Study, Group. Natural history and
clinical management of central retinal vein occlusion. Arch
Ophthalmol 1997;115:486–491.
6. Hayreh SS, Zimmerman MB. Fundus changes in central retinal
vein occlusion. Retina 2015;35:29–42.
7. Coscas G, Cunha-Vaz J, Soubrane G. Macular edema: defini-
tion and basic concepts. Dev Ophthalmol 2010;47:1–9.
8. Patel M, Kiss S. Ultra-wide-field fluorescein angiography in
retinal disease. Curr Opin Ophthalmol 2014;25:213–220.
9. 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.
10. Jia Y, Tan O, Tokayer J, et al. Split-spectrum amplitude-
decorrelation angiography with optical coherence tomography.
Opt Express 2012;20:4710–4725.
11. 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.
12. Coscas F, Glacet-Bernard A, Miere A, et al. OCT-angiography
in retinal vein occlusion: evaluation of superficial and deep
capillary plexa. Am J Ophthalmol 2016;161:160–171.
13. Couturier A, Mané V, Bonnin S, et al. Capillary plexus anom-
alies in diabetic retinopathy on optical coherence tomography
angiography. Retina 2015;35:2384–2391.
14. Kuehlewein L, An L, Durbin MK, Sadda SR. Imaging areas of
retinal nonperfusion in ischemic branch retinal vein occlusion
with swept-source OCT microangiography. Ophthalmic Surg
Lasers Imaging Retina 2015;46:249–252.
15. De Carlo TE, Salz DA, Waheed NK, et al. Visualization of the
retinal vasculature using wide-field montage optical coherence
tomography angiography. Ophthalmic Surg Lasers Imaging
Retina 2015;46:611–616.
16. Suzuki N, Hirano Y, Yoshida M, et al. Microvascular abnor-
malities on optical coherence tomography angiography in mac-
ular edema associated with branch retinal vein occlusion. Am J
Ophthalmol 2016;161:126–132.
17. Agemy SA, Scripsema NK, Shah CM, et al. Retinal vascular
perfusion density mapping using optical coherence tomogra-
phy angiography in normals and diabetic retinopathy patients.
Retina 2015;35:2353–2363.
18. Coscas F, Sellam A, Glacet-Bernard A, et al. Normative data for
vascular density in superficial and deep capillary plexuses of
healthy adults assessed by optical coherence tomography angi-
ography. Invest Ophthalmol Vis Sci 2016;57:OCT211–223.
19. Yeh S, Kim SJ, Ho AC, et al. Therapies for macular edema
associated with central retinal vein occlusion: a report by the
American Academy of Ophthalmology. Ophthalmology 2015;
122:769–778.
20. Campochiaro PA, Bhisitkul RB, Shapiro H, Rubio RG.
Vascular endothelial growth factor promotes progressive
retinal nonperfusion in patients with retinal vein occlusion.
Ophthalmology 2013;120:795–802.
21. Bonnin S, Mané V, Couturier A, et al. New insight into the
macular deep vascular plexus imaged by optical coherence
tomography angiography. Retina 2015;35:2347–2352.
22. Spaide RF, Fujimoto JG, Waheed NK. Image artifacts in optical
coherence tomography angiography. Retina 2015;35:2163–2180.