Comparison of Visual Outcomes in Coats’

Disease
A 20-Year Experience
Sally S. Ong, MD,1 Edward G. Buckley, MD,1 Brooks W. McCuen II, MD,1 Glenn J. Jaffe, MD,1
Eric A. Postel, MD,1 Tamer H. Mahmoud, MD, PhD,1 Sandra S. Stinnett, DrPH,1 Cynthia A. Toth, MD,1
Lejla Vajzovic, MD,1 Prithvi Mruthyunjaya, MD, MHS1,2

Purpose: To report differences in visual acuities among patients with Coats’ disease who sought treatment
at a tertiary care university-based practice.
Design: Single-center retrospective cohort study.
Participants: Patients with Coats’ disease diagnosed clinically, angiographically, or both from 1995
through 2015.
Methods: Patients were divided into 2 groups based on date of presentation: decade 1 (1995e2005) and
decade 2 (2006e2015).
Main Outcome Measures: Visual acuity (VA).
Results: Thirty-nine eyes of 39 patients were included with 19 eyes presenting in decade 1 and 20 eyes
presenting in decade 2. Three patients demonstrated bilateral disease, but only the worse eye was included for
analysis. Forty-seven percent of eyes in decade 1 demonstrated advanced stages of disease (stage 3B or worse)
compared with 20% of eyes in decade 2. There was a trend for the mean initial presenting VA (
standard
deviation) for decade 1 eyes to be worse (2.05
1.29 logarithm of the minimum angle of resolution [logMAR]) than
for decade 2 eyes (1.45
0.99 logMAR; P ¼ 0.1). From initial to final follow-up visit, mean VA also worsened for
decade 1 eyes (P ¼ 0.03), but remained stable for decade 2 eyes (P ¼ 1.0). At the end of follow-up, there was a
trend for mean VA for decade 1 eyes (2.28
1.17 logMAR) to be worse than for decade 2 eyes (1.60
1.15 log-
MAR; P ¼ 0.07). Eight eyes were observed initially in decade 1 compared with 1 eye in decade 2, and only 1 of the
observed eyes (in decade 2) developed painful glaucoma requiring enucleation. Decade 2 eyes had a higher
average number of procedures per eye (6.5
4.9) compared with decade 1 eyes (1.4
1.7; P < 0.001).
Conclusions: The earlier presentation of disease in decade 2 suggests improvements in disease detection
over time. Furthermore, there was a trend for eyes to have better final VA in this decade. This is due to
a combination of factors, including earlier presentation of disease, fewer eyes being observed without
treatment, and eyes, when treated, receiving a higher number of procedures. Ophthalmology 2017;124:
1368-1376 ª 2017 by the American Academy of Ophthalmology

Coats’ disease is a sporadic and nonhereditary condition both eyes.3e5 In cases where subretinal fluid prevents
characterized as idiopathic retinal telangiectasia that may adequate ablative therapy, Coats’ disease has been treated
progress to retinal exudation and exudative retinal detachment with intravitreal triamcinolone (IVT) or antievascular
in the absence of retinal or vitreous traction.1 Severe endothelial growth factor (VEGF) injections, or surgically
complications of untreated Coats’ disease can be devastating with vitrectomy, subretinal fluid drainage, or both, with or
and include neovascular glaucoma and phthisis bulbi.2 without scleral buckle.6e10
Traditionally, Coats’ disease has been diagnosed clini- Despite the introduction of new diagnostic methods and
cally with indirect ophthalmoscopy, and in eyes with early treatment strategies for Coats’ disease in recent years, few
stages of disease, has been treated with laser photocoagu- studies have been published to describe how these new ad-
lation or cryotherapy to the abnormal telangiectasic vessels.1 vances have affected patients’ anatomic and visual outcomes.
In recent years, the advent of wide-field imaging and fluo- Herein, we studied the population of Coats’ patients who
rescein angiography during examination under anesthesia were diagnosed and treated at a tertiary care academic vit-
has allowed earlier identification of peripheral vascular reoretinal practice over a 20-year period to compare treatment
anomalies and areas of abnormal retinal perfusion in one or approaches and outcomes over 2 distinct 10-year periods.

1368 ª 2017 by the American Academy of Ophthalmology http://dx.doi.org/10.1016/j.ophtha.2017.03.051

Published by Elsevier Inc. ISSN 0161-6420/17
 Ong et al 
Visual Outcomes in Coats’ Disease

Methods pictures or HOTV charts. Snellen VA equivalent of 20/125 was

selected as a cutoff to determine eyes that maintained functional
This was a retrospective review of all consecutive patients aged vision at the end of follow-up.
18 years or younger at presentation with Coats’ disease treated by
multiple vitreoretinal specialists (E.G.B., B.W.M., G.J.J., E.A.P., Statistical Analysis
T.H.M., C.A.T., L.V., P.M.) at the Duke University Eye Center
between January 1, 1995, and September 30, 2015, and who were Snellen VA equivalent was converted to logarithm of the mini-
followed up for at least 12 months. The study was designed in mum angle of resolution (logMAR) units for analysis. Snellen
accordance with the tenets of the Declaration of Helsinki, was acuities for counting fingers, hand movements, light perception,
approved by the Duke University Institutional Review Board, and and no light perception (NLP) measured at 2 feet were approx-
complied with the Health Insurance Portability and Accountability imated as described previously.12 Snellen acuities for
Act. Patients with a diagnosis code of exudative retinopathy (In- measurements made at other distances were extrapolated from
ternational Classification of Diseases, Ninth Revision, code those obtained at 2 feet. Only 1 eye from each patient was
362.12) were identified using the Duke Enterprise Data Unified included in statistical analysis. In patients with bilateral
Content Explorer database. Patients were excluded if they did not involvement, the worse eye was included in statistical analysis.
have clinical evidence of Coats’ disease, were treated at another For paired analysis (comparison of same eyes at different time
institution, or had missing or incomplete records. points), only eyes that had documented Snellen VA equivalent
Patients were diagnosed clinically with Coats’ disease based on at both of the visits being compared were included. The
the following criteria: presence of idiopathic retinal telangiectasia ManneWhitney U test and KruskaleWallis test were used to
with or without intraretinal or subretinal exudation and with or compare continuous variables between groups and among
without exudative retinal detachments. Demographic information, groups, whereas the Fisher exact test was used to compare
visual acuity (VA) at several time points (initial presentation, the categorical variables between groups. The Wilcoxon signed-
12- to 18-month follow-up, the intermediate period for decade 1, rank test was used to compare paired data. A P value of 0.05
and the final follow-up) when available were recorded, and details or less was considered statistically significant. Statistical anal-
of each examination were recorded. Pediatric patients usually were ysis was performed using SAS software version 9.3 (SAS
examined and treated under anesthesia. Fundus photography and Institute, Inc., Cary, NC).
fluorescein angiography were performed using a RetCam 2 (Clarity
Medical Systems, Pleasanton, CA), Zeiss FF450 (Carl Zeiss
Meditec, Okerkochen, Germany), or Optos 200Tx (Optos, Inc., Results
Dunfermline, United Kingdom) imaging device at the discretion of
the treating physician.
Baseline Characteristics of Study Participants
Eyes then were divided into 2 groups based on year of pre-
sentation: decade 1 (1995e2005) and decade 2 (2006e2015). A total of 55 patients with complete medical records who were
Furthermore, eyes were divided into 5 stages of disease at the time treated for Coats’ disease at our institution were identified. A total
of presentation as previously described by Shields et al.11 Briefly, of 16 patients were excluded: 14 had follow up less than 12 months
stages 1, 2A, 2B, 3A, 3B, 4, and 5 are defined as follows: stage 1
and 2 were adults older than age 18 years at the time of presen-
corresponds to eyes with retinal telangiectasia only; stage 2A refers
to eyes with retinal telangiectasia and extrafoveal exudation; stage tation. Thirty-nine patients met inclusion criteria. Three patients
2B denotes eyes with retinal telangiectasia and foveal exudation; demonstrated asymmetric bilateral disease at presentation (decade
stage 3A represents eyes with retinal telangiectasia, exudation, 1: stages 2B and 1 and stages 3B and 5; decade 2: stages 3A and 1).
and subtotal retinal detachment; stage 3B describes eyes with Only 1 eye from each patient was included in statistical analysis,
retinal telangiectasia, exudation, and total retinal detachment; and the better eye from patients with bilateral disease was
stage 4 eyes demonstrate retinal telangiectasia, exudation, total excluded.
retinal detachment, and secondary glaucoma; and stage 5 eyes Baseline demographics and ocular characteristics of the final
are defined as blind eyes with retinal telangiectasia, exudation, cohort of 39 eyes of 39 patients who met inclusion criteria are
total retinal detachment, and anterior chamber involvement or shown in Table 1 and are consistent with previously published
phthisis (advanced end-stage disease).
reports.1,13e19 There was no significant difference in baseline de-
Management methods included observation (defined as no
treatment for posterior manifestations of Coats’ disease; some pa- mographic characteristics between patients in the 2 decades. There
tients in this group still underwent cataract or strabismus surgery); was a trend for eyes in decade 1 to demonstrate more advanced
ablative therapies including cryotherapy, 532-nm laser, or both; stages of disease than those in decade 2: 47% of eyes in decade 1
vitreoretinal surgery, which included any combination of subretinal demonstrated stage 3B to 5 disease (9 of 19 eyes) compared with
fluid drainage, scleral buckle, pars plana vitrectomy, epiretinal 20% of eyes in decade 2 (4 of 20 eyes; P ¼ 0.1).
membrane removal, and intraocular gas or silicone oil tamponade; Given the small number of eyes with advanced disease in
IVT injection; or enucleation. After 2008, off-label intravitreal decade 2, eyes from both decades were combined to examine the
bevacizumab (IVB; 1.25 mg or 0.625 mg; Genentech, Inc., South relationship between stage of disease and presenting VA. Eyes
San Francisco, CA) was given to selected eyes based on treating with more advanced stages of disease presented with worse initial
physician preference. Anatomic disease resolution was defined as
VA
standard deviation (SD; stage 2A, 0.12
0.13 logMAR [n ¼
resolution of exudates, exudative subretinal fluid, or both. When
possible, best-corrected VA was obtained at each clinical visit. 4]; stage 2B, 0.87
0.48 logMAR [n ¼ 5]; stage 3A, 1.78
1.03
Visual acuity was measured using Snellen or Early Treatment logMAR [n ¼ 15]; stage 3B, 2.95
0.21 logMAR [n ¼ 7]; stage 4,
Diabetic Retinopathy Study charts. In young children who could 3.20 logMAR [n ¼ 1]; stage 5, 3.20 logMAR [n ¼ 1]; P ¼ 0.001).
not recognize the letters or numbers on Snellen or Early Treatment Correspondingly, there was a trend for eyes in decade 1 to
Diabetic Retinopathy Study charts, VA was measured using Allen demonstrate worse VA than eyes in decade 2 (P ¼ 0.1).

1369
 Ophthalmology Volume 124, Number 9, September 2017

Table 1. Patient Demographics and Baseline Ocular whereas 8 eyes (50%) had maintained or progressed to NLP VA.
Characteristics In contrast, mean initial and final VA for decade 2 eyes remained
stable (P ¼ 1.0). Specifically, VA in this decade remained stable
Variable Decade 1 Decade 2 P Value for 1 eye, improved for 8 eyes, and worsened for 5 eyes. Five of 14
Total no. of patients 19 20 eyes (36%) maintained VA of 20/125 or better, whereas 1 eye (7%)
Age (mos) had progressed to NLP VA at the end of follow-up. Overall, there
Mean
SD 78.1
63.7 95.2
73.3 0.5 was a trend of better mean final VA in decade 2 (1.60
1.15 log-
Range 1e221 5e224 MAR) than in decade 1 (2.28
1.17 logMAR; P ¼ 0.07).
Gender, no. (%)
Male 16 (84.2) 17 (85.0) 1.0
To determine if the difference in length of follow-up between
Female 3 (15.8) 3 (15.0) 1.0 decades 1 and 2 influenced the final visual outcomes, VA at an
Race, no (%) intermediate period of follow-up for decade 1 also was examined.
White 13 (68.4) 10 (50.0) 0.3 The mean intermediate follow-up period was 38.2
10.9 months
Black 5 (26.3) 4 (20.0) 0.7 (range, 14e61 months). Mean VA in decade 1 again worsened
Asian 0 2 (10.0) 0.5 from initial to intermediate follow-up (2.27
1.15 logMAR; P ¼
Hispanic 0 1 (5.0) 1.0
0.02), and there was no difference in mean VA between interme-
Not reported 1 (5.3) 3 (15.0) 0.6
Eye laterality, no. (%) diate and final follow-up (P ¼ 0.9). Visual acuity for eyes in
Right eye 10 (52.6) 12 (60.0) 1.0 decade 1 at the final visit was stable for 13 eyes, improved for 1
Left eye 9 (47.4) 8 (40.0) 1.0 eye, and worsened for 2 eyes compared with the intermediate
Stage of disease at follow-up period. Two of the 16 eyes (13%) maintained VA of 20/
presentation, no. (%) 125 or better and 8 eyes (50%) maintained or progressed to NLP
1 0 0 N/A
VA by this intermediate period of follow-up. Given the comparable
2A 2 (10.5) 2 (10.0) 1.0
2B 3 (15.8) 4 (20.0) 1.0 follow-up periods between intermediate follow-up in decade 1 and
3A 5 (26.3) 10 (50.0) 0.2 final follow-up in decade 2, mean VA for these 2 groups were then
3B 7 (36.8) 4 (20.0) 0.3 compared. Overall, there was a trend of better mean final VA in
4 1 (5.3) 0 0.5 decade 2 (1.60
1.15 logMAR) than mean intermediate VA in
5 1 (5.3) 0 0.5 decade 1 (2.27
1.15 logMAR; P ¼ 0.07).
Combined stages of
Next, VA at a common time point (12e18 months of follow-
disease, no. (%)
Stages 1 to 3A 10 (52.6) 16 (80.0) 0.1 up) in both decades was examined. Visual acuity during this
Stages 3B to 5 9 (47.4) 4 (20.0) 0.1 window of follow-up was available only for 13 eyes in decade 1
Best-corrected visual and 10 eyes in decade 2. There was no significant difference in
acuity (logMAR) mean VA in decade 1 from initial visit (1.84
1.24 logMAR) to the
at presentation by stage, 12- to 18-month follow-up (2.12
1.17 logMAR; P ¼ 0.3). Simi-
mean
SD (no.)
2A 0.05
0.07 (2) 0.20
0.14 (2) 0.4 larly, there was no significant difference in mean VA in decade 2
2B 0.95
0.38 (3) 0.75
0.78 (2) 1.0 from presentation (1.68
0.95 logMAR) to the 12- to 18-month
3A 2.00
1.45 (5) 1.69
0.83 (10) 0.7 follow-up (1.55
0.93 logMAR; P ¼ 0.6). Additionally, mean
3B 2.95
0.23 (6) 2.90 (1) N/A 12- to 18-month VA in decade 2 was not statistically different from
4 3.20 (1) 0 N/A mean 12- to 18-month VA in decade 1 (P ¼ 0.2).
5 3.20 (1) 0 N/A Globe Salvage and Anatomic Outcomes. Although more
P value 0.05* 0.07
All stages 2.05
1.29 (18) 1.45
0.99 (15) 0.1
eyes were managed with observation alone in decade 1 (n ¼ 8)
compared with decade 2 (n ¼ 1), the globe salvage rates did not
differ in the 2 decades. Only 1 observed eye was enucleated
logMAR ¼ logarithm of the minimum angle of resolution; N/A ¼ not
because of uncontrolled pain after progression to angle-closure
applicable; SD ¼ standard deviation.
*Statistically significant. glaucoma, and this eye had stage 3B disease and had presented
in decade 2. Before enucleation, this eye had been treated with
transscleral diode cyclophotocoagulation with inadequate intraoc-
Outcomes by Decade of Treatment ular pressure relief. None of the observed eyes in decade 1
demonstrated glaucoma and none had to be enucleated.
Nineteen eyes presented in decade 1 and 20 eyes presented in In addition to the 1 stage 3B eye in decade 2 that was enucleated
decade 2. As expected, the mean
SD follow-up was longer for after progression to uncontrolled painful glaucoma, 12 other eyes
decade 1 eyes (110.9
74.2 months) than for decade 2 eyes across both decades demonstrated stage 3B or worse disease. Of
(46.2
28.9 months; P ¼ 0.01). these, 2 eyes (with stages 3B and 5 disease) in decade 1 and 1 eye
Visual Acuity Outcomes. Average VA at presentation and (with stage 3B disease) in decade 2 were enucleated at presentation
end of follow-up were compared for each of the 2 groups, as shown because of concern about retinoblastoma, although pathologic ex-
in Table 2. This analysis examined only eyes that had VA amination eventually showed Coats’ disease. Despite poor visual
documented at both initial and final follow-up visits (16 eyes in outcomes, the globe was salvaged in the other 9 eyes with stage 3B
decade 1 and 14 eyes in decade 2). Mean VA worsened from initial disease or worse. The 1 stage 4 eye was observed without treatment
to final visit for decade 1 eyes (P ¼ 0.03). As shown in Figure 1, for posterior segment manifestations of Coat’s disease, but under-
VA in decade 1 stayed the same for 3 eyes, improved for 2 eyes, went cataract extraction without intraocular lens placement for
and worsened for 11 eyes. Only 2 of the 16 eyes (13%) angle-closure glaucoma in the setting of total retrolental retinal
maintained VA of 20/125 or better by the final follow-up, detachment. The glaucoma resolved after cataract surgery.

1370
 Ong et al 
Visual Outcomes in Coats’ Disease

Table 2. Visual Acuity Measured at Presentation and Final Follow-up for All Eyes, Eyes in Decades 1 and 2, and by Stage of Disease

Overall Decade 1 Decade 2

Initial Visual Final Visual Initial Visual Final Visual Initial Visual Final Visual
Stage No. Acuity Acuity P Value No. Acuity Acuity P Value No. Acuity Acuity P Value
All stages 30 1.73
1.15 1.96
1.19 0.2 16 1.90
1.29 2.28
1.17 0.03* 14 1.53
0.97 1.60
1.15 1.0
2A 3 0.06
0.06 0.93
1.45 0.8 2 0.05
0.07 1.39
1.71 0.5 1 0.10 0 N/A
2B 5 0.87
0.48 0.90
0.72 1.0 3 0.95
0.38 0.90
0.46 0.9 2 0.75
0.78 0.91
1.29 1.0
3A 15 1.78
1.03 1.98
1.08 0.5 5 1.96
1.45 2.41
1.11 0.1 10 1.69
0.83 1.77
1.06 1.0
3B 6 2.90
0.19 3.10
0.16 0.1 5 2.90
0.21 3.14
0.13 0.1 1 2.90 2.90 N/A
4 1 3.20 3.20 N/A 1 3.20 3.20 N/A 0 N/A N/A N/A
5 0 N/A N/A N/A 0 N/A N/A N/A 0 N/A N/A N/A

N/A ¼ not applicable.

Visual acuity (VA) measured in logarithm of the minimum angle of resolution (logMAR) units. Data are mean
standard deviation unless otherwise
indicated. P values from Wilcoxon signed-rank test for comparison of mean VA at initial and final follow-up visits. Only eyes that had documented VA at
both the initial and final visits were included in this analysis.
*Statistically significant.

As expected, none of the observed-only eyes showed anatomic (with stage 3A disease) developed phthisis despite treatment with
resolution of disease. Of the eyes that received therapeutic in- vitreoretinal surgery and ablation.
terventions, exudation with or without subretinal fluid resolved in 5
of 10 eyes (50%) in decade 1 and in 11 of 18 eyes (61%) in decade Treatment Differences between the First and
2. Two of the eyes in decade 2 that did not show anatomic reso-
Second Decade of the Study
lution of disease were still being treated at the end of the study
(1 eye had follow-up of 19 months, whereas the other had follow- We also analyzed if there was a difference in the number of pro-
up of 12 months by the end of the study). One eye in decade 2 cedures per eye in both groups that could explain the difference in

Figure 1. Scatterplot showing initial and final visual acuity (VA) measurements in logarithm of the minimum angle of resolution (logMAR) units in
each eye by decade of presentation. The VA by Snellen equivalent is marked on the right axis. Eyes that presented in decade 1 are listed on the left, and
eyes that presented in decade 2 are listed on the right. Visual acuity on presentation is plotted as a blue cross, whereas VA at the end of follow-up is
plotted as a red diamond. Reasons for final VA worse than 0.80 logMAR or 20/125 despite treatment are denoted as arrows. Persistent exudates,
subretinal fluid, or both are denoted as orange arrows, macular scarring is denoted as purple arrows, tractional retinal detachment (TRD) is denoted as
gray arrows, and glaucoma is denoted as light blue arrows. The 1 eye that developed phthisis is marked with an asterisk. Two eyes in decade 2 were still
undergoing treatment by the end of follow-up because of persistent exudates. Three eyes were enucleated because of concern for retinoblastoma (RB).
HM ¼ hand movements; IVB ¼ intravitreal bevacizumab; IVT ¼ intravitreal triamcinolone; LP ¼ light perception; NLP ¼ no light perception;
VR ¼ vitreoretinal.

1371
 Ophthalmology Volume 124, Number 9, September 2017

VAs between groups. Eyes that presented in decade 2 underwent as in the prior analysis (Fig 1). We found no difference in TRD
a higher average number of procedures per patient (mean
SD, development among eyes that underwent primary therapy with
6.5
4.9) compared with eyes that presented in decade 1 (mean ablation only (3 of 11 eyes) versus ablation with IVB (0 of 7

SD, 1.4
1.7; P < 0.001). After removing eyes that were eyes) or among eyes that underwent primary vitreoretinal surgery
observed only or had been enucleated, the difference between with ablation (3 of 4 eyes) versus primary vitreoretinal surgery
the number of procedures between the 2 groups remained with ablation and IVB (2 of 3 eyes). There was also no
significant (7.1
4.8 in decade 2 vs. 2.4
1.8 in decade 1; difference between TRD development in primary ablation-only
P ¼ 0.003). groups treated in decade 1 (0 of 4 eyes) versus in decade 2 (3 of
7 eyes) or in primary vitreoretinal surgery with ablation groups
Comparison of Eyes Treated with and without treated in decade 1 (2 of 3 eyes) versus in decade 2 (1 of 1 eye).
Intravitreal Bevacizumab over Both Decades Eyes treated with IVB all sought treatment in decade 2, so a
comparison across decades was not performed.
Intravitreal bevacizumab was introduced after 2008 and was used
We then analyzed the difference in rate of macular scarring be-
in eyes with stage 2B disease and greater. As shown in Figure 1, of
tween eyes that had received IVB therapy versus those that had not.
the stage 2B eyes, 2 of 5 eyes treated with ablation only had final
We found no difference among eyes that underwent primary therapy
VA of 20/125 or better. Reasons for VA worse than 20/125
of ablation only (6 of 11 eyes) versus ablation with IVB (2 of 7 eyes)
included failure to achieve anatomic resolution and macular
or among eyes that underwent primary vitreoretinal surgery with
scarring. One of 2 stage 2B eyes treated with ablation and IVB
ablation (1 of 1 eye) versus primary vitreoretinal surgery with
also achieved VA of 20/125 or better. The remainder stage 2B
ablation and IVB (3 of 3 eyes). There was only 1 eye in the primary
eye treated with ablation and IVB was still undergoing treatment
vitreoretinal surgery with ablation group because the other eyes in
at the end of the study (length of follow-up, 12 months) because
this group had severe pathologic features precluding assessment for
of unresolved exudation and telangiectasia.
macular scarring. There was also no difference between macular
Only 3 of the 15 stage 3A eyes (20%) had final VA of 20/125 or
scarring development in primary ablation-only groups treated in
better, and all 3 eyes had undergone ablation and IVB therapy.
decade 1 (3 of 4 eyes) versus those treated in decade 2 (3 of 7 eyes)
Another stage 3A eye that had been treated primarily with ablation
or in primary vitreoretinal surgery and ablation groups treated in
and IVB therapy later demonstrated nonclearing vitreous hemor-
decade 1 (1 of 1 eye) versus those treated in decade 2 (0 eyes).
rhage requiring secondary vitreoretinal surgery, and this eye’s final
Secondary glaucoma was found at presentation for the stage 4
VA was worse than 20/125 in the setting of macular scarring
eye in decade 1. After lensectomy, the angle-closure glaucoma
(Table 3). The rest of the stage 3A eyes were observed (n ¼ 2),
resolved in this eye. Glaucoma developed after presentation in 2
underwent ablation only (n ¼ 3), underwent ablation only
eyes in decade 1 and 3 eyes in decade 2. Only 1 of the 5 eyes (1
initially followed by secondary vitreoretinal surgery for tractional
observed stage 3B eye in decade 2) required enucleation because of
retinal detachment (TRD; n ¼ 1), underwent ablation with IVT
painful glaucoma that remained uncontrolled despite transscleral
(n ¼ 1), and underwent primary vitreoretinal surgery with
diode photocoagulation. Glaucoma in the other 4 eyes was
ablation (n ¼ 2) or primary vitreoretinal surgery with ablation
controlled with ocular hypotensive drops (decade 1, 1 stage 3A eye
and IVB (n ¼ 2). Reasons for VA worse than 20/125 in these
treated with ablation and IVT and 1 stage 3B eye treated with
eyes included failure of anatomic improvement, macular scarring,
vitreoretinal surgery and ablation; decade 2, 1 stage 3A eye treated
TRD, and glaucoma. Phthisis developed in 1 eye. The decision
with vitreoretinal surgery and ablation) or ocular hypotensive drops
to perform vitreoretinal surgery as primary therapy in stage 3A
with IVB (1 stage 3B eye treated with vitreoretinal surgery, abla-
eyes was dependent on the extent of retinal detachment. Eyes
tion, and IVB in decade 2).
with 2 quadrants or fewer of retinal detachment were treated
with ablation only or ablation with IVB without vitreoretinal
surgery, whereas those with more extensive retinal detachment
Discussion
also underwent vitreoretinal surgery. See Table 3 for details of
individual eyes that underwent vitreoretinal surgery.
Coats’ disease presents numerous challenges in the man-
Stage 3B eyes had poor final visual outcomes ranging from
agement and preservation of VA across the disease spec-
light perception to NLP regardless of treatment method. Treated
trum. Our study demonstrated differences in ocular
eyes received vitreoretinal surgery with ablation (n ¼ 2), ablation
characteristics, treatment approach, and outcomes spanning
with IVB (n ¼ 1), or vitreoretinal surgery with ablation and IVB
2 different decades in a university-based Coats’ disease
(n ¼ 1). Reasons for poor VA included failure to achieve anatomic
population. In this study, eyes that presented in decade 1
improvement, macular scarring, TRD, and glaucoma. Five of 11
were more likely to show advanced stages of disease
stage 3B eyes (45%) were observed, and 1 of the observed eyes
compared with eyes that presented in decade 2, which
demonstrated glaucoma and eye pain requiring enucleation.
suggests earlier recognition and detection of disease over
Development of Tractional Retinal Detachment, time. This could be related to improvements in visual
Macular Scarring, and Glaucoma screenings in schools and the general pediatrics office. The
American Academy of Pediatrics and the United States
Because of previous reports that IVB could induce TRD and Preventative Services Task Force have advocated for
macular fibrosis,20 we examined differences in rates of TRD and screenings for visual impairment in children at a young
macular scarring among all eyes that were treated with and age,21,22 and automated devices that take less time per child
without IVB. This analysis of TRD and macular scarring and allow the screening of an uncooperative child have
development did not exclude eyes with VA of 20/125 or better become more available and easier to use over time.23,24

1372
 Ong et al 
Visual Outcomes in Coats’ Disease

Table 3. Vitreoretinal Surgical Methods by Decade of Presentation and Whether Surgery Was Conducted to Address Primary or
Secondary Ocular Manifestations of Coats’ Disease

Surgery for Primary or

Decade of Initial Visual Final Visual Secondary Manifestations
Presentation Stage Other Treatments Acuity Acuity Vitreoretinal Surgery of Coats’ Disease
1 2A Ablation and IVT 20/20 HM PPV with MP Primary
1 3A Ablation 20/80 6/200 SRF drainage Primary
2 3A Ablation 20/400 NLP PPV with MP and gas, Primary
then PPV with MP and SO
2 3A Ablation 20/200 20/400 PPV with MP Secondary TRD
2 3A Ablation and IVB LP 20/800 PPV with MP Primary
2 3A Ablation and IVB 20/800 LP SB, PPV with MP, retinectomy Primary
and SO, then PPV with
SO removal
2 3A Ablation and IVB HM CF @ 2 ft PPV Secondary Nonclearing vitreous
hemorrhage
1 3B Ablation LP NLP SRF drainage Primary
1 3B Ablation LP NLP SRF drainage Primary
2 3B Ablation and IVB LP LP SRF drainage Primary

CF ¼ counting fingers; HM ¼ hand movements; IVB ¼ intravitreal bevacizumab; IVT ¼ intravitreal triamcinolone; LP ¼ light perception; NLP ¼ no light
perception; MP ¼ membrane peel; PPV ¼ pars plana vitrectomy; SB ¼ scleral buckle; SO ¼ silicone oil; SRF ¼ subretinal fluid; TRD ¼ tractional retinal
detachment; VA ¼ visual acuity.
Only patients who received treatment at Duke were included in the study.

Furthermore, ultrawide-field imaging, which was designed months). Given the trends toward more advanced stage of
originally for use in pediatric patients, has made examina- disease at presentation and worse presenting VA, as well as
tion of the peripheral retina in children more accessible in worsening of VA over the length of follow-up in decade 1, it
pediatric ophthalmology clinics.25 is not surprising that there was also a trend toward worse final
Correspondingly, the initial VA for eyes that presented visual outcomes in eyes in this decade compared with those
in decade 1 trended toward being worse than that of eyes that that had presented in decade 2.
presented in decade 2. These data are consistent with those of More eyes were observed without treatment in the first
previous reports11,13,17,19,26,27 showing that eyes with early- decade (n ¼ 8) compared with the second decade (n ¼ 1),
stage Coats’ disease have better presenting VA compared which may reflect either the practice pattern of some in our
with eyes with more advanced disease stages. Presenting VA group at that time or the more aggressive approach to treat-
was worse in eyes with stage 2B disease, coincident with ment as imaging and surgical technology improved in
foveal exudates, and is especially poor in eyes with total decade 2. Without treatment, Coats’ disease, with rare
exudative retinal detachment for which the average VA was exception,28e32 progresses and causes VA loss.11,19 In our
light perception for stage 3B or NLP for stage 4 and 5 disease study, 1 stage 2A eye in decade 1 that was observed initially
and which did not improve despite treatment. showed worsening of disease. Despite treatment after disease
Over time, mean VA in decade 1 worsened from initial to progression, the eye had poor visual outcome. For the other 8
final follow-up, whereas mean VA in decade 2 remained observed eyes, 2 had stage 3A disease and 6 had stage 3B or
stable. Moreover, from presentation to final follow-up, VA in worse disease at presentation, and as previously reported,
decade 1 worsened for most eyes, whereas VA in decade 2 observation was thought to be a reasonable option as long as
improved for most eyes. Because Coats’ disease is a pro- the patient was comfortable, because treatment for advanced
gressive disease and the difference in length of follow-up disease often is challenging and may not lead to improved
could have influenced final VA, mean VA at different time visual outcomes.11,33 However, other authors have advocated
points also were compared. Mean VA in decade 1 at an in- treatment of eyes with advanced disease to prevent painful
termediate period of follow-up (intended to mimic the glaucoma requiring enucleation.11,27 In the present report, of
average time to final follow-up for decade 2) also was worse the 6 observed eyes with stage 3B or worse disease, 1 (17%)
when compared with that at presentation. However, at a required enucleation because of uncontrolled glaucoma
common time point (12e18 months of follow-up), mean VA causing a painful eye. This rate is similar to that of the study
in both decades did not differ significantly from that at pre- by Shields et al,11 in which 3 of 16 eyes (19%) with stages 3B
sentation. Taken together, this suggests that mean VA did not and 5 disease that were observed demonstrated painful
change significantly from presentation to the 12- to 18-month glaucoma and required enucleation. However, Silodor
follow-up for either decade, but became worse for decade 1 at et al27 reported that 4 of the 6 eyes (67%) with total retinal
an average of 38 months of follow-up, and this difference detachment that were observed progressed to painful
persisted until final follow-up (mean, 111 months). In neovascular glaucoma requiring enucleation.
contrast, mean VA for decade 2 did not differ significantly at Among eyes that were treated, eyes in decade 2 were
different time points, including final follow-up (mean, 46 treated with a higher number of procedures than eyes in

1373
 Ophthalmology Volume 124, Number 9, September 2017

decade 1. We suspect the shift away from observation and In the current report, only 2 stage 3B eyes were treated
toward increasing numbers of treatments in the decade 2 with ablation and anti-VEGF therapy with or without vit-
group reflects increasing evidence showing the benefits of reoretinal surgery. Both eyes showed poor anatomic and
aggressive repetitive treatments to achieve anatomic reso- visual outcomes. Given the small number of eyes in this
lution of disease.11,34 group, it was not possible to conclude if anti-VEGF was
Although most stage 2B eyes treated with ablation only beneficial in the treatment of stage 3B eyes. However, other
(4 of 5 eyes) showed good anatomic outcomes, only 40% studies have shown that anatomic resolution was possible in
(2 of 5) showed good visual outcomes (20/125 VA). stage 3B eyes treated with anti-VEGF and other therapies.
Macular scarring accounted for poor visual outcomes in the For example, in another retrospective case series, although
3 eyes treated with ablation only. Meanwhile, the 1 stage 2B treatment failed in 2 eyes (both stage 3B) of 10 patients
eye that completed treatment with ablation and IVB ach- treated with ablation only, all 10 patients matched by
ieved anatomic resolution and maintained final VA better macular appearance, quadrants of subretinal fluid, and
than 20/125. It has been reported previously that prolonged quadrants of telangiectasia (stages ranged from 2B to 3B)
macular exudation can develop into macular fibrosis leading treated with ablation and IVB achieved anatomic resolu-
to visual loss.17 Shields et al11 similarly noted that visual tion.9 Meanwhile, Zheng and Jiang39 reported that although
prognosis usually is limited when a dense yellow-gray all eyes in their study responded to treatment with complete
nodule is centered within the foveal exudation or when or partial resolution of telangiectasia, subretinal fluid, and
there is late subretinal fibrosis at the fovea. Given the small exudation, average VA improved for the 14 pediatric
sample size, it is not possible to determine if IVB had a role patients (stages 2 to 3B), but not for the 5 adult patients
in preventing prolonged macular exudation and conse- (stages 3A and 3B) treated with IVB with or without
quently scarring in stage 2B eyes. other therapies. Furthermore, in a retrospective review of
There are higher levels of VEGF in eyes with Coats 24 children with exudative retinal detachment (not
disease.35,36 Accordingly, one might hypothesize that anti- differentiated between stage 3A or 3B) associated with
VEGF treatment would be effective to manage the VEGF- Coats’ disease who were treated with large-spot diode
induced exudation and sequelae in this condition. In our laser and IVB, Villegas et al40 showed that all 24 patients
study, IVB was introduced in decade 2 and seemed to have achieved resolution of exudative retinal detachment,
had a positive impact in some stage 3A eyes. Intravitreal vascular telangiectasia, and anatomic improvement. Like
bevacizumab in conjunction with ablation with or without our study, most of these reports are limited by small
vitreoretinal surgery was associated with anatomic resolu- sample sizes, lack of adequate control groups, and
tion in 5 of the 6 stage 3A eyes and with final VA of 20/125 nonstandardization of the number of treatment sessions,
or better in 3 of the 6 stage 3A eyes. The eyes with final VA but taken together, there is evidence to suggest treatment
of 20/125 or better were also 3 of the 4 eyes treated with benefit of adding IVB for eyes with all stages of disease,
ablation and IVB without primary vitreoretinal surgery. but particularly for eyes with stage 3A or worse disease.
Because stage 3A eyes that were treated with ablation and Although anti-VEGF therapy has been shown to have
IVB without primary vitreoretinal surgery usually had 2 or favorable results in the previously mentioned studies, the
fewer quadrants of retinal detachment, this observation may long-term effects of anti-VEGF therapy in children remain
suggest that the advantage of using IVB may be most unknown. Ramasubramanian and Shields20 first reported the
apparent in stage 3A eyes with limited quadrants of retinal risk of incident TRD with adjunctive IVB. Of the 8 patients
detachment because VA potential is good after treatment. In with Coats disease who were treated with ablation and IVB,
contrast, only 2 of the 6 stage 3A eyes treated with ablation 4 demonstrated vitreous fibrosis, whereas 3 demonstrated
with or without vitreoretinal surgery and without IVB TRD. Other studies also have cautioned that patients
showed anatomic resolution, 1 eye demonstrated phthisis, treated with IVB in conjunction with other treatments may
and all 6 eyes showed unfavorable visual outcomes. be at increased risk of vitreoretinal fibrosis and
Similar to our report, other studies have demonstrated TRD.8,39,41 In our study, we compared eyes that under-
that adjunctive anti-VEGF therapy may be beneficial in eyes went ablation with or without vitreoretinal surgery with IVB
with stage 3A disease or better. Sein et al37 showed that as compared with similar treatment groups without IVB and
100% globe salvage and reasonable visual outcomes were did not find a difference in TRD or macular scarring rates
achieved in the 26 patients with stages 2A and 3A disease among the groups. However, it is important to note that the
treated with laser and IVB (presenting VA was not small sample size limits the power of detecting differences
reported, but final VA was 20/20 for both stage 2A eyes across groups. Similarly, Ferrone42 showed that in the
and ranged from 20/20 to counting fingers at less than 1 absence of anti-VEGF use, TRD developed after ablative
foot for the 24 stage 3A eyes). Additionally, in a therapy in 5 Coats’ disease eyes with total exudative
retrospective consecutive interventional study, Lin et al38 detachment. Interestingly, Daruich et al43 recently reported
reported that 6 stage 2B eyes (treated with ablation only, that TRD and macular fibrosis developed at a higher rate
n ¼ 2; treated with ablation and anti-VEGF, n ¼ 3; or in patients with extramacular fibrosis and questioned
treated with ablation, anti-VEGF, and IVT, n ¼ 1), 1 stage whether the higher rate of TRD found in
3A eye (treated with ablation and anti-VEGF), and 1 stage Ramasubramanian and Shields’ report could be because
3B eye (treated with ablation only) showed anatomic the eyes studied had advanced (stage 3B) disease.
improvement, but 1 stage 3B eye (treated with ablation and The strengths of this study include available anatomic
anti-VEGF) did not show anatomic improvement. and visual outcomes over a long follow-up. Limitations of

1374
 Ong et al 
Visual Outcomes in Coats’ Disease

this study include its retrospective nature, diverse group of 6. Bergstrom CS, Hubbard 3rd GB. Combination intravitreal
treating specialists, small number of participants because of triamcinolone injection and cryotherapy for exudative retinal
the rarity of the disease, and variable length of follow-up. detachments in severe Coats disease. Retina. 2008;28(3
Moreover, given the retrospective nature of the study, Suppl):S33-S37.
best-corrected VA measurements in children by Snellen 7. Suesskind D, Altpeter E, Schrader M, et al. Pars plana vitrec-
tomy for treatment of advanced Coats’ diseasedpresentation of
charts, Early Treatment Diabetic Retinopathy Study charts, a modified surgical technique and long term follow up. Graefes
Allen pictures, or HOTV charts were not standardized Arch Clin Exp Ophthalmology. 2014;252(6):873-879.
across eyes or visits. Additionally, patients were managed 8. Gaillard MC, Mataftsi A, Balmer A, et al. Ranibizumab in the
with a wide variety of methods, thus further decreasing the management of advanced Coats disease stages 3B and 4: long-
number of participants in each study group. The limited term outcomes. Retina. 2014;34(11):2275-2281.
study power in individual treatment groups precluded sta- 9. Ray R, Baranano DE, Hubbard GB. Treatment of Coats’ dis-
tistically meaningful comparisons among treatment groups. ease with intravitreal bevacizumab. Br J Ophthalmol.
Based on the results from this study and a review of the 2013;97(3):272-277.
literature, eyes with stage 2A or more advanced disease 10. Lin CJ, Hwang JF, Chen YT, Chen SN. The effect of intra-
should be treated promptly to prevent disease progression vitreal bevacizumab in the treatment of Coats disease in chil-
dren. Retina. 2010;30(4):617-622.
and worsening of VA. Stage 2A and 2B eyes that were 11. Shields JA, Shields CL, Honavar SG, et al. Classification and
treated with ablation to full resolution of exudates showed management of Coats disease: the 2000 Proctor Lecture. Am J
good visual outcomes when macular scarring did not Ophthalmol. 2001;131(5):572-583.
develop. In this study, some stage 3A eyes with 2 or fewer 12. The Ischemic Optic Neuropathy Decompression Trial
quadrants of retinal detachment also responded to ablation (IONDT): design and methods. Control Clin Trials.
with IVB with anatomic resolution and good visual out- 1998;19(3):276-296.
comes, but given the small sample size, a prospective study 13. Char DH. Coats’ syndrome: long term follow up. Br J Oph-
with larger sample size is needed to determine treatment thalmol. 2000;84(1):37-39.
efficacy definitively. Advanced-stage eyes should still 14. Gomez Morales A. Coats’ disease. Natural history and results
receive treatment to prevent progression to painful uncon- of treatment. Am J Ophthalmol. 1965;60(5):855-865.
15. Haik BG. Advanced Coats’ disease. Trans Am Ophthalmol
trolled glaucoma requiring enucleation. Soc. 1991;89(371e476):371-476.
In conclusion, this study demonstrated the evolution of 16. Harris GS. Coats’ disease, diagnosis and treatment. Can J
practice patterns for management of Coats’ disease at a Ophthalmol. 1970;5(4):311-320.
single institution over the past 2 decades. In the second 17. Tarkkanen A, Laatikainen L. Coats’ disease: clinical, angio-
decade of the study, fewer eyes presented with advanced graphic, histopathological findings and clinical management.
stages of disease, fewer eyes were observed without treat- Br J Ophthalmol. 1983;67:766-776.
ment, and eyes were treated with a higher mean number of 18. Theodossiadis GP. Some clinical, fluorescein-angiographic
procedures. Additionally, fewer eyes had NLP VA at the and therapeutic aspects of Coats disease. J Pediatr Oph-
final follow-up in decade 2 compared with both the inter- thalmol Strabismus. 1974;16:257-262.
mediate and final follow-ups in decade 1. The results of this 19. Shields JA, Shields CL, Honavar SG, Demirci H. Clinical
variations and complications of Coats disease in 150 cases: the
study also suggest that adding IVB therapy may benefit 2000 Sanford Gifford Memorial Lecture. Am J Ophthalmol.
vision in stage 3A eyes with 2 or fewer quadrants of retinal 2001;131(5):561-571.
detachment; further prospective studies to answer this spe- 20. Ramasubramanian A, Shields CL. Bevacizumab for Coats’
cific question are needed. Although the globe salvage rate disease with exudative retinal detachment and risk of vitreor-
was high, the data from this report and those described in the etinal traction. Br J Ophthalmol. 2012;96(3):356-359.
literature also highlight the need for better treatments for 21. American Academy of Pediatrics, Committee on Practice and
advanced-stage Coats’ disease. Ambulatory Medicine. Recommendations for preventive pe-
diatric health care. Pediatrics. 1995;96(2 pt 1):373-374.
22. Force UPST. Vision screening for children 1 to 5 years of age:
References US Preventive Services Task Force Recommendation state-
ment. Pediatrics. 2011;127:340-346.
23. Donahue SP, Arthur B, Neely DE, et al. Guidelines for auto-
1. Shields J, Shields C. Review: Coats disease, the 2001 mated preschool vision screening: a 10-year, evidence-based
LuEsther T. Mertz Lecture. Retina. 2001;22(1):80-91. update. J AAPOS. 2013;17(1):4-8.
2. Morris B, Foot B, Mulvihill A. A population-based study of 24. Bregman J, Donahue SP. Validation of photoscreening tech-
Coats’ disease in the United Kingdom I: epidemiology and nology in the general pediatrics office: a prospective study.
clinical features at diagnosis. Eye (Lond). 2010;24:1797-1801. J AAPOS. 2016;20(2):153-158.
3. Suzani M, Moore AT. Intraoperative fluorescein angiography- 25. Tsui I, Franco-Cardenas V, Hubschman JP, Schwartz SD.
guided treatment in children with early Coats’ disease. Pediatric retinal conditions imaged by ultra wide field fluo-
Ophthalmology. 2015;122(6):1195-1202. rescein angiography. Ophthalmic Surg Lasers Imaging Retina.
4. Blair MP, Ulrich JN, Elizabeth Hartnett M, Shapiro MJ. Pe- 2013;44(1):59-67.
ripheral retinal nonperfusion in fellow eyes in coats disease. 26. Budning AS, Heon E, Gallie BL. Visual prognosis of Coats’
Retina. 2013;33(8):1694-1699. disease. J AAPOS. 1998;2(6):356-359.
5. Shane TS, Berrocal AM, Hess DJ. Bilateral fluorescein 27. Silodor SW, Augsburger JJ, Shields JA, Tasman W. Natural
angiographic findings in unilateral Coats’ disease. Ophthalmic history and management of advanced Coats’ disease.
Surg Lasers Imaging. 2011;42(online):e15-e17. Ophthalmic Surg. 1988;19(2):89-93.

1375
 Ophthalmology Volume 124, Number 9, September 2017

28. Friedenwald H, Friedenwald JS. Terminal stage in a case of role of bevacizumab. Graefes Arch Clin Exp Ophthalmol.
retinitis with massive exudation. Trans Am Ophthalmol Soc. 2010;248(10):1519-1521.
1929;27:188-194. 37. Sein J, Tzu JH, Murray TG, Berrocal AM. Treatment of Coats’
29. Campbell FP. Coats’ disease and congenital vascular retinop- disease with combination therapy of intravitreal bevacizumab,
athy. Trans Am Ophthalmol Soc. 1976;74:365-424. laser photocoagulation, and sub-Tenon corticosteroids.
30. Deutsch TA, Rabb MF, Jampol LM. Spontaneous regression Ophthalmic Surg Lasers Imaging Retina. 2016;47:443-449.
of retinal lesions in Coats’ disease. Can J Ophthalmol. 38. Lin CJ, Chen SN, Hwang JF, Yang CM. Combination treat-
1982;17(4):169-172. ment of pediatric Coats’ disease: a bicenter study in Taiwan.
31. Gass JDM. Stereoscopic Atlas of Macular Diseases: Diagnosis J Pediatr Ophthalmol Strabismus. 2013;50(6):356-362.
and Treatment. 4th ed. St. Louis, MO: Mosby; 1997. 39. Zheng X, Jiang Y. The effect of intravitreal bevacizumab in-
32. Wolfe JD, Hubbard 3rd GB. Spontaneous regression of sub- jection as the initial treatment for Coats’ disease. Graefes Arch
retinal exudate in Coats disease. Arch Ophthalmol. Clin Exp Ophthalmol. 2014;252:35-42.
2006;124(8):1208-1209. 40. Villegas VM, Gold AS, Berrocal AM, Murray TG. Advanced
33. Ghazi NG, Al Shamsi H, Larsson J, Abboud E. Intravitreal Coats’ disease treated with intravitreal bevacizumab combined
triamcinolone in Coats’ disease. Ophthalmology. 2012;119(3): with laser vascular ablation. Clin Ophthalmol. 2014;8:973-976.
648-649. 41. Bhat V, D’Souza P, Shah PK, Narendran V. Risk of tractional
34. Schefler AC, Berrocal AM, Murray TG. Advanced Coats’ retinal detachment following intravitreal bevacizumab along
disease. Management with repetitive aggressive laser ablation with subretinal fluid drainage and cryotherapy for stage 3B
therapy. Retina. 2008;28(3 Suppl):S38-S41. Coats’ disease. Middle East Afr J Ophthalmol. 2016;23(2):
35. Sun Y, Jain A, Moshfeghi DM. Elevated vascular endothelial 208-211.
growth factor levels in Coats disease: rapid response to 42. Ferrone PJ. Long term follow-up of treated severe Coats dis-
pegaptanib sodium. Graefes Arch Clin Exp Ophthalmol. ease complicated by secondary traction retinal detachments.
2007;245(9):1387-1388. Bordeaux, France: Club Jules Gonin; 2016.
36. He YG, Wang H, Zhao B, et al. Elevated vascular endothelial 43. Daruich A, Matet A, Tran HV, et al. Extramacular fibrosis in
growth factor level in Coats’ disease and possible therapeutic Coats’ disease. Retina. 2016;36(10):2022-2028.

Footnotes and Financial Disclosures

Originally received: January 11, 2017. Author Contributions:
Final revision: March 23, 2017. Conception and design: Ong, Toth, Vajzovic, Mruthyunjaya
Accepted: March 28, 2017. Analysis and interpretation: Ong, Stinnett, Toth, Vajzovic, Mruthyunjaya
Available online: April 28, 2017. Manuscript no. 2016-1123.
1 Data collection: Ong, Buckley, McCuen, Jaffe, Postel, Mahmoud, Toth,
Department of Ophthalmology, Duke University Medical Center,
Vajzovic, Mruthyunjaya
Durham, North Carolina.
2 Obtained funding: none
Department of Ophthalmology, Stanford University Medical Center, Palo
Overall responsibility: Ong, Buckley, McCuen, Jaffe, Postel, Mahmoud,
Alto, California.
Stinnett, Toth, Vajzovic, Mruthyunjaya
Presented as a poster at: American Academy of Ophthalmology Annual
Meeting, Chicago, Illinois, October, 2016. Abbreviations and Acronyms:
Financial Disclosure(s): HIPAA ¼ Health Insurance Portability and Accountability Act;
The author(s) have made the following disclosure(s): T.H.M.: Consultant e IVB ¼ intravitreal bevacizumab; IVT ¼ intravitreal triamcinolone;
logMAR ¼ logarithm of the minimum angle of resolution; NLP ¼ no light
Alimera, DORC, Spark Therapeutics; Financial support  Genentech.
perception; SD ¼ standard deviation; TRD ¼ tractional retinal detachment;
C.A.T.: Financial support  Genentech. VA ¼ visual acuity; VEGF ¼ vascular endothelial growth factor.
L.V.: Financial support  Alcon, Roche, DORC, Janssen.
Correspondence:
P.M.: Consultant e Castle Biosciences, Spark Therapeutics.
Prithvi Mruthyunjaya, MD, MHS, Department of Ophthalmology, Byers
Supported by the American Society of Cataract and Refractive Surgery Eye Institute, 2462 Watson Court, Palo Alto, CA 94306. E-mail: prithvi9@
(Resident Excellence Award [S.S.O.]); the Knights Templar Foundation stanford.edu.
(L.V.); Second Sight (L.V.).

1376