You are on page 1of 7

412 Diabetes Care Volume 40, March 2017

Diabetic Retinopathy: A Position Sharon D. Solomon,1 Emily Chew,2


Elia J. Duh,1 Lucia Sobrin,3 Jennifer K. Sun,4

Statement by the American Brian L. VanderBeek,5 Charles C. Wykoff,6


and Thomas W. Gardner7

Diabetes Association
Diabetes Care 2017;40:412–418 | DOI: 10.2337/dc16-2641
POSITION STATEMENT

Diabetic retinopathy diagnostic assessment and treatment options have improved


dramatically since the 2002 American Diabetes Association Position Statement (1).
These improvements include the widespread adoption of optical coherence tomog-
raphy to assess retinal thickness and intraretinal pathology and wide-field fundus
photography to reveal clinically silent microvascular lesions. Treatment of diabetic
macular edema is now achieved by intravitreous injection of anti–vascular endo-
thelial growth factor agents, and the same drugs are now used for proliferative
diabetic retinopathy. Improvements in medications and devices for the systemic
therapy of diabetes have also improved the ability of patients to optimize their
metabolic control. This Position Statement incorporates these recent developments
1
for the use of physicians and patients. Wilmer Eye Institute, Johns Hopkins Medicine,
Diabetic retinopathy is a highly specific neurovascular complication of both type 1 and Baltimore, MD
2
National Eye Institute, National Institutes of
type 2 diabetes, the prevalence of which strongly correlates to both the duration of Health, Bethesda, MD
diabetes and level of glycemic control. A pooled meta-analysis involving 35 studies con- 3
Department of Ophthalmology, Massachusetts
ducted worldwide from 1980 to 2008 estimated global prevalence of any diabetic ret- Eye and Ear Infirmary, Boston, MA
4
inopathy and proliferative diabetic retinopathy (PDR) among patients to be 35.4% and Beetham Eye Institute, Joslin Diabetes Center,
Boston, MA
7.5%, respectively (2). Diabetic retinopathy is the most frequent cause of new cases of 5
Department of Ophthalmology, University of
blindness among adults aged 20–74 years in developed countries. Glaucoma, cataracts, Pennsylvania, Philadelphia, PA
and other disorders of the eye occur earlier and more frequently in people with diabetes. 6
Retina Consultants of Houston, Houston, TX
7
In addition to diabetes duration, factors that increase the risk of or are associated Department of Ophthalmology and Visual Sci-
with retinopathy include chronic hyperglycemia (3,4), nephropathy (5), hypertension ences, Kellogg Eye Center, University of Michi-
gan, Ann Arbor, MI
(6), and dyslipidemia (7). Intensive diabetes management with the goal of achieving
near-normoglycemia has been shown in large prospective randomized studies to Corresponding author: Thomas W. Gardner,
tomwgard@umich.edu.
prevent and/or delay the onset and progression of diabetic retinopathy (8,9).
This position statement was reviewed and ap-
Lowering blood pressure has been shown to decrease retinopathy progression in proved by the American Diabetes Association
people with type 2 diabetes, although tight targets (systolic blood pressure Professional Practice Committee in October
,120 mmHg) do not impart additional benefit over targets of ,140 mmHg (9,10). In 2016 and ratified by the American Diabetes As-
patients with dyslipidemia, retinopathy progression may be slowed by the addition of sociation Board of Directors in December 2016.
fenofibrate, particularly with very mild nonproliferative diabetic retinopathy (NPDR) at For further information on the ADA evidence-
baseline (7). Several case series and a controlled prospective study suggest that preg- grading system and the levels of evidence, please
see Table 1 in the American Diabetes Associa-
nancy in patients with type 1 diabetes may aggravate retinopathy and threaten vision,
tion’s Introduction section of the Standards of
especially when glycemic control is poor at the time of conception (11,12). Medical Care in Diabetesd2017. Diabetes Care
2017;40(Suppl. 1):S2.
NATURAL HISTORY © 2017 by the American Diabetes Association.
Readers may use this article as long as the work
Recommendations is properly cited, the use is educational and not
c Optimize glycemic control to reduce the risk or slow the progression of for profit, and the work is not altered. More infor-
diabetic retinopathy. A mation is available at http://www.diabetesjournals
.org/content/license.
care.diabetesjournals.org Solomon and Associates 413

producing severe and often irreversible vi- studies performed in the modern era of
c Optimize blood pressure and se-
sion loss. Third, the new blood vessels may expanded options for glucose, lipid, and
rum lipid control to reduce the
bleed, adding the further complication of blood pressure control.
risk or slow the progression of di-
preretinal or vitreous hemorrhage. These The WESDR found a relationship be-
abetic retinopathy. A
clinically evident vascular changes are ac- tween onset of retinopathy and du-
companied by damage to retinal neurons ration of diabetes. It established that
In general, retinopathy advances from (15), the final common pathway for vision progression of retinopathy was a function
mild nonproliferative abnormalities, char- loss. of baseline retinopathy. More severe base-
acterized by increased numbers of micro- Several epidemiological studies have line retinopathy led to a greater frequency
aneurysms that may wax and wane. With described the progression rates for di- of progression to vision-threatening re-
increasing severity, there is increased vas- abetic retinopathy. The cohort with the tinopathy. Among patients with type 2
cular permeability and occlusion and pro- longest follow-up is the Wisconsin Epide- diabetes whose baseline photographs
gression from moderate and severe NPDR miologic Study of Diabetic Retinopathy showed no retinopathy, there was 54%
to PDR, characterized by the growth of (WESDR), which reported the 25-year less progression to PDR over 10 years com-
new blood vessels on the retina and pos- progression of diabetic retinopathy in pared with those with severe NPDR at
terior surface of the vitreous (see Table 1 patients with type 1 diabetes (16). How- baseline (17). The WESDR epidemiologi-
for definitions of diabetic retinopathy ever, the WESDR started recruitment in cal data were limited primarily to white
stages). Pregnancy and puberty can accel- 1979 when options for glycemic, blood Northern European extraction popula-
erate these changes (12,13). Cataract sur- pressure, and lipid control were mark- tions and may not be applicable to Afri-
gery has not been definitely demonstrated edly limited compared with the options can American, Hispanic American, or
by recent studies to accelerate the pro- available today. The risk factors identi- Asian American populations or to others
gression of diabetic retinopathy, espe- fied in WESDRdlonger duration of dia- with a high prevalence of diabetes and
cially in the more recent era of treating betes, greater hyperglycemia, increased retinopathy.
both diabetic macular edema (DME) blood pressure, and dyslipidemiad After duration of diabetes, hypergly-
and PDR with the use of anti–vascular remain relevant while the progression cemia has been the most consistently
endothelial growth factor (anti-VEGF) rates in more recent studies may differ associated risk factor for retinopathy.
agents (14). markedly. For example, the WESDR pro- A large and consistent set of observa-
Vision loss due to diabetic retinopathy gression data predicted a progression tional studies and clinical trials document
results from several mechanisms. First, rate near 40% over 4 years for the Ac- the association of poor glucose control
central vision may be impaired by macular tion to Control Cardiovascular Risk in Di- and retinopathy. The Diabetes Control
edema as the result of increased vascular abetes (ACCORD) trial, but the actual and Complications Trial (DCCT), a ran-
permeability and/or capillary nonperfu- progression rate in people with type 2 domized controlled clinical trial of inten-
sion. Second, the new blood vessels of PDR diabetes at the study’s conclusion was sive glycemic control versus conventional
and contraction of the accompanying fi- only 10%. Table 2 shows the odds ratios glycemic control in people with type 1
brous tissue can distort the retina and associated with the most consistently as- diabetes, demonstrated that intensive
lead to tractional retinal detachment, sociated retinopathy risk factors in therapy reduced the development or
progression of diabetic retinopathy
by 34–76% (51). In addition, the DCCT
demonstrated a definitive relationship
Table 1—Diabetic retinopathy stages* between hyperglycemia and diabetic
Diabetic retinopathy microvascular complications, including
stage Description
retinopathy (18). Early treatment with
Mild NPDR Small areas of balloon-like swelling in the retina’s tiny blood vessels, called intensive therapy was most effective. In
microaneurysms, occur at this earliest stage of the disease. These addition, intensive therapy had a sub-
microaneurysms may leak fluid into the retina.
stantial beneficial effect over the entire
Moderate NPDR As the disease progresses, blood vessels that nourish the retina may swell range of retinopathy. A 10% reduction in
and distort. They may also lose their ability to transport blood. Both
HbA1c, for example from 10 to 9% or
conditions cause characteristic changes to the appearance of the retina
and may contribute to DME. from 8 to 7.2%, reduces the risk of reti-
Severe NPDR Many more blood vessels are blocked, depriving blood supply to areas of nopathy progression by 43% (52).
the retina. These areas secrete growth factors that signal the retina to The UK Prospective Diabetes Study
grow new blood vessels. (UKPDS) of patients newly diagnosed
PDR At this advanced stage, growth factors secreted by the retina trigger the with type 2 diabetes conclusively dem-
proliferation of new blood vessels, which grow along the inside surface onstrated that improved blood glucose
of the retina and into the vitreous gel, the fluid that fills the eye. The new control in those patients reduced the
blood vessels are fragile, which makes them more likely to leak and risk of developing retinopathy and ne-
bleed. Accompanying scar tissue can contract and cause retinal
phropathy and possibly reduced the
detachmentdthe pulling away of the retina from underlying tissue,
like wallpaper peeling away from a wall. Retinal detachment can lead risk for neuropathy (8). The overall mi-
to permanent vision loss. crovascular complication rate was de-
creased by 25% in patients receiving
*Adapted from https://nei.nih.gov/health/diabetic/retinopathy.
intensive therapy versus conventional
414 Position Statement Diabetes Care Volume 40, March 2017

Table 2—Recent estimates of the association between major risk factors and fenofibrate was not evident after the
diabetic retinopathy drug was stopped in the clinical trial of
Risk factor Reference Strength of association, odds ratio (95% CI) ACCORD. This suggests that the treat-
ment with fenofibrate therapy may in-
Duration of diabetes Xu et al. (48) 1.16 (1.10–1.22) per year increase
deed be real.
Kajiwara et al. (49) 1.13 (1.09–1.17) per year increase
The results of these two large ran-
HbA1c Xu et al. (48) 1.73 (1.35–2.21) per 1% increase
domized trials, ACCORD Eye Study and
Kajiwara et al. (49) 1.21 (1.08–1.36) per 1% increase
Jin et al. (50) 1.12 (1.01–1.24) per 1% increase
FIELD, suggest that fenofibrate may be a
potential therapy for people with dia-
Blood pressure Kajiwara et al. (49) 1.02 (1.01–1.03) per mmHg increase
in systolic blood pressure betic retinopathy. These results were
Jin et al. (50) 1.80 (1.14–2.86) if systolic blood not subgroup analyses, and these bene-
pressure .140 mmHg and/or ficial effects were supported by two
diastolic blood pressure .90 mmHg large randomized controlled clinical tri-
als. Because of the lack of beneficial ef-
fects on cardiovascular disease, medical
physicians have been reluctant to pre-
therapy. Epidemiological analysis of the UKPDS showed a 37% reduction in mi- scribe fenofibrate for people with dia-
UKPDS data showed a continuous rela- crovascular abnormalities, including di- betic retinopathy. There are sufficient
tionship between the risk of microvas- abetic retinopathy and specifically DME, data to suggest developing collaboration
cular complications and glycemia, such with lowering of systolic blood pressure between the ophthalmologists (eye care
that every percentage point decrease in from a mean of 154 mmHg to 144 mmHg providers) and the medical physician to
HbA1c (e.g., 9% to 8%) was associated (20). However, the more recent ACCORD consider this treatment for people af-
with a 35% reduction in the risk of mi- Eye Study did not show either a harmful fected with diabetic retinopathy.
crovascular complications. or a beneficial effect when comparing sys-
More recently, the ACCORD trial of tolic pressure of 120 mmHg vs. 140 mmHg SCREENING
medical therapies demonstrated that in- in a similar cohort of patients (9). Recommendations
tensive glycemic control reduced the Several observational studies have sug-
c Adults with type 1 diabetes should
risk of progression of diabetic retinopa- gested that dyslipidemia may play a role
have an initial dilated and compre-
thy in people with type 2 diabetes of in the progression of diabetic retinopa-
hensive eye examination by an oph-
10 years duration (9). This study in- thy. Dyslipidemia is associated with reti-
thalmologist or optometrist within
cluded 2,856 ACCORD participants who nal hard exudate and visual loss. Two
5 years after the onset of diabetes. B
were enrolled into the ACCORD Eye trials of fenofibrate have been conducted
c Patients with type 2 diabetes should
Study and followed for 4 years. to reduce the levels of serum triglycerides
have an initial dilated and compre-
The results of the DCCT, UKPDS, and in an effort to reduce cardiovascular risk
hensive eye examination by an oph-
ACCORD Eye Study showed that while (9,21). Although fenofibrate does not
thalmologist or optometrist at the
intensive therapy does not prevent ret- have an effect on cardiovascular risk,
time of the diabetes diagnosis. B
inopathy completely, it reduces the risk both studies showed an effect on the pro-
c If there is no evidence of retinopa-
of the development and progression of gression of diabetic retinopathy. The
thy for one or more annual eye ex-
diabetic retinopathy. This can be trans- Fenofibrate Intervention and Event
ams, then exams every 2 years may
lated clinically to a higher likelihood of Lowering in Diabetes (FIELD) study
be considered. If any level of dia-
preserving sight and to a reduced need demonstrated the beneficial effects of
betic retinopathy is present, subse-
for treatment. Furthermore, all three fenofibrate (200 mg daily) versus placebo
quent dilated retinal examinations
studies demonstrated that years after in reducing the need for laser photocoag-
for patients with type 1 or type 2
the initial clinical trial ended, the treat- ulation (hazard ratio 0.69, 95% CI 0.56–
diabetes should be repeated at
ment effect of intensive glycemic control 0.84, P = 0.00002) (21). A substudy of the
least annually by an ophthalmolo-
persisted, despite the fact that both treat- FIELD participants with fundus photo-
gist or optometrist. If retinopathy
ment groups had similar levels of HbA1c. In graphs showed the beneficial effect on
is progressing or sight-threatening,
fact, 25 years after the cessation of the the Early Treatment Diabetic Retinopathy
then examinations will be required
DCCT, ocular surgery rates were reduced Study (ETDRS) scale, especially in those
more frequently. B
in those who had been assigned to inten- with retinopathy at baseline and also on
c Women with preexisting type 1 or
sive glycemic control (19). In the DCCT, at the development of macular edema (haz-
type 2 diabetes who are planning
varying intervals, the beneficial effects of ard ratio 0.66, 95% CI 0.47–0.94, P = 0.02).
pregnancy or who have become
intensive glycemic control persisted but The ACCORD Study also compared feno-
pregnant should be counseled on
declined over time. This persistent ben- fibrate 160 mg daily with simvastatin
the risk of development and/or pro-
eficial effect beyond the clinical trial was versus placebo with simvastatin and
gression of diabetic retinopathy. B
true for people with type 1 and type 2 found that the risk of progression of
c Eye examinations should occur be-
diabetes. diabetic retinopathy was reduced by
fore pregnancy or in the first trimes-
Blood pressure control has also been one-third. The effect was particularly
ter in patients with preexisting
studied in several observational and demonstrated in those with preexisting
type 1 or type 2 diabetes, and then
clinical trials, including the UKPDS. The diabetic retinopathy. The effect of
care.diabetesjournals.org Solomon and Associates 415

minimal to no retinopathy. Exams every 78 diopter), indirect ophthalmoscopy,


these patients should be monitored
2 years may be cost-effective after one and testing as appropriate that may in-
every trimester and for 1 year post-
or more normal eye exams and, in a pop- clude optical coherence tomography
partum as indicated by the degree of
ulation with well-controlled type 2 dia- and fluorescein angiography.
retinopathy. B
betes, there was essentially no risk of Retinal photography, with remote
c While retinal photography may
development of significant retinopathy reading by experts, has great potential
serve as a screening tool for retinop-
with a 3-year interval after a normal exam- to provide screening services in areas
athy, it is not a substitute for a com-
ination (26). Examinations will be required where qualified eye care professionals
prehensive eye exam, which should
more frequently by the ophthalmologist if are not readily available (31). High-quality
be performed at least initially and
retinopathy is progressing. fundus photographs can detect most
at intervals thereafter as recom-
Pregnancy can be associated with rapid clinically significant diabetic retinopathy.
mended by an eye care profes-
progression of diabetic retinopathy in the Interpretation of the images should be
sional. E
setting of type 1 and type 2 diabetes (27). performed by a trained eye care pro-
Women who develop gestational diabetes vider. Retinal photography can also en-
Screening strategies depend on the mellitus do not require an eye examination hance efficiency and reduce costs when
rates of appearance and progression of during pregnancy and do not appear to be the expertise of ophthalmologists can
diabetic retinopathy and on risk factors at increased risk of developing diabetic ret- be used for more complex examinations
that alter these rates. While population- inopathy during pregnancy (28). Women and for therapy (32). In-person exams
based studies often are the best source with preexisting type 1 or type 2 diabetes are still necessary when the retinal pho-
for evaluating the rates of progression, who plan to become pregnant should have tos are unacceptable and for follow-up if
data from other studies, including ob- an ophthalmic examination prior to preg- abnormalities are detected. Retinal pho-
servational studies and clinical trials, nancy and receive counseling about the tos are not a substitute for a compre-
have provided important information risk of development and progression of di- hensive eye exam, which should be
as well. A summary of screening recom- abetic retinopathy. When pregnant, an performed at least initially and at inter-
mendations is in Table 3. eye examination should be performed dur- vals thereafter as recommended by an
With regard to retinopathy onset, ing the first trimester with follow-up visits eye care professional. Results of eye ex-
vision-threatening retinopathy rarely scheduled depending on retinopathy se- aminations should be documented and
appears in type 1 diabetes patients in verity (12,29). Rapid implementation of transmitted to the referring health care
the first 3–5 years of diabetes or before tight glycemic control in the setting of ret- professional.
puberty (22,23). Because retinopathy inopathy can be associated with worsen-
takes at least 5 years to develop after ing of retinopathy (12). TREATMENT
the onset of hyperglycemia, adults with For patients with diabetes, regular follow- Recommendations
type 1 diabetes should have an initial up with early detection and treatment
c Promptly refer patients with any
dilated and comprehensive eye exami- of vision-threatening retinopathy enables
level of macular edema, severe
nation by an ophthalmologist or optom- the prevention of up to 98% of visual loss
nonproliferative diabetic retinop-
etrist within 5 years after the diagnosis due to diabetic retinopathy (30). The
athy (a precursor of proliferative
of diabetes. preventive effects of therapy and the
diabetic retinopathy), or any pro-
Up to one-fifth of patients with type 2 fact that patients with PDR or macular
liferative diabetic retinopathy to
diabetes have retinopathy at the time of edema may be asymptomatic provide
an ophthalmologist who is knowl-
first diagnosis of diabetes (24,25). Pa- strong support for screening to detect
edgeable and experienced in the
tients with type 2 diabetes who may diabetic retinopathy.
management and treatment of di-
have had years of undiagnosed diabetes An ophthalmologist or optometrist
abetic retinopathy. A
and have a significant risk of diabetic who is knowledgeable and experienced
c Laser photocoagulation therapy
retinopathy at the time of diagnosis in diagnosing diabetic retinopathy
reduces the risk of vision loss in
should have an initial dilated and com- should perform the examinations. If di-
patients with high-risk prolifera-
prehensive eye examination by an oph- abetic retinopathy is present, prompt
tive diabetic retinopathy and, in
thalmologist or optometrist at the time referral to an ophthalmologist is recom-
some cases, severe nonprolifera-
of diagnosis. mended. Comprehensive evaluation by
tive diabetic retinopathy. A
Subsequent examinations for patients an ophthalmologist will include dilated
c Intravitreous injections of anti–
with type 1 or type 2 diabetes are gener- slit-lamp examination including biomi-
vascular endothelial growth factor
ally repeated annually for patients with croscopy with a hand-held lens (90 or
are indicated for central-involved
diabetic macular edema, which oc-
curs beneath the foveal center and
Table 3—Screening recommendations for patients with diabetes
may threaten reading vision. A
Classification Examination by ophthalmologist or optometrist
c The presence of retinopathy is
Type 1 diabetes Within 5 years after onset of diabetes not a contraindication to aspirin
Type 2 diabetes At time of diabetes diagnosis therapy for cardioprotection, as
Women with preexisting diabetes planning Before pregnancy or in first trimester aspirin does not increase the risk
pregnancy or who have become pregnant of retinal hemorrhage. A
416 Position Statement Diabetes Care Volume 40, March 2017

While optimization of blood glucose, data from the Diabetic Retinopathy Clin- severe vision loss in eyes affected with
blood pressure, and serum lipid levels ical Research Network (DRCRN) suggest PDR (39). The benefit was greatest
in conjunction with appropriately sched- that for eyes with CIDME and good levels among patients whose baseline evalua-
uled dilated eye examinations can sub- of acuity, 20/40 or better, each agent tion revealed high-risk characteristics
stantially decrease the risk of vision loss effectively and similarly improves visual (HRCs) consisting of disc neovasculari-
from complications of diabetic retinopa- acuity. However, in eyes with CIDME and zation greater than or equal to one-
thy, a significant proportion of those af- lower levels of acuity, 20/50 or worse, quarter of a disc area in size, any disc
fected with diabetes develop DME or aflibercept appears to be most effective neovascularization with vitreous hem-
proliferative changes that require inter- at improving visual acuity (38). Most orrhage, or vitreous hemorrhage with
vention (Table 4). patients require near-monthly adminis- retinal neovascularization greater than
tration of intravitreous therapy with anti- or equal to one-half of a disc area in
CENTRAL-INVOLVED DME VEGF agents during the first 12 months of size. Although some eyes, especially
Historically, focal laser photocoagula- treatment, with fewer injections needed in those of patients with type 2 diabetes,
tion has been the standard treatment subsequent years to maintain remission benefit from early PRP prior to the de-
for eyes with clinically significant macu- from CIDME. velopment of HRCs, given the risk of a
lar edema (CSME), defined as either ret- Multiple emerging therapies for reti- modest loss of visual acuity and of con-
inal edema located at or within 500 mm nopathy that target alternative pathways, traction of visual field from PRP, laser
of the center of the macula or edema provide sustained intravitreous delivery of therapy has been primarily recom-
of a disc area or more within a disc di- pharmacological agents, or allow oral or mended for eyes approaching or reach-
ameter of the foveal center. The ETDRS topical noninvasive delivery systems are ing HRCs.
(33) showed that treated eyes with currently under investigation for the treat- PRP is still commonly used to manage
CSME had a significantly reduced risk ment of CIDME. Intravitreous steroid ther- eyes with PDR. However, widespread
of further visual loss. apy for CIDME has been evaluated in observations that rapid regression of
Current treatment thresholds are multiple phase 3 studies, and the steroid retinal neovascularization occurs in
based on the presence of central- agents dexamethasone and fluocinolone eyes receiving intravitreous anti-VEGF
involved DME (CIDME), or edema affect- acetonide are approved by the U.S. Food therapy for CIDME has made these
ing the 1 mm in diameter retinal central and Drug Administration for the indication agents a potentially viable alternative
subfield, rather than the presence of of CIDME. Nonetheless, given the inferior treatment for PDR. In a randomized trial
CSME. Intravitreous therapy with agents visual acuity outcomes to anti-VEGF seen by the DRCRN comparing intravitreous
that neutralize VEGF is currently the with intravitreous steroid therapy in a ranibizumab to PRP for visual acuity out-
standard of care in the management of large DRCRN trial, as well as the increased comes in patients with PDR, there was
eyes with CIDME, following numerous adverse events of cataract and glaucoma no statistically significant visual acuity
well-designed randomized phase 3 clinical associated with steroid use, these agents difference between the ranibizumab
trials that have shown benefit compared are rarely used as first-line therapy in eyes and PRP groups at 2 years (40). How-
with monotherapy or even combination with CIDME. ever, average visual acuity outcomes
therapy with laser (34–37). There are over the course of 2 years favored the
currently three anti-VEGF agents com- PDR ranibizumab-treated group. Further-
monly used to treat eyes with CIDMEd The Diabetic Retinopathy Study (DRS) more, significantly more eyes in the
bevacizumab, ranibizumab, and afliber- showed that panretinal laser photoco- PRP group experienced peripheral visual
cept. Of these anti-VEGF agents, recent agulation (PRP) reduced the risk of field loss and underwent vitrectomy for

Table 4—Recommended follow-up


Referral to
Indication ophthalmologist Follow-up Recommended intraocular treatment*
No diabetic retinopathy Within 1 year Every 1–2 years None
Mild NPDR Within 1 year Every year None
Moderate NPDR Within 3–6 months Every 6–9 months None
Severe NPDR Immediate Every 3–6 months Can consider early PRP for patients with type 2 diabetes
PDR Immediate Every 3 months PRP or intravitreous anti-VEGF therapy, especially if
HRCs are present
No DME Within 1 year Every 1–2 years None
Non-CIDME Within 3–6 months Every 6 months None, but observe carefully for progression to CIDME
CIDME Immediate Every 1–4 months Anti-VEGF as first-line therapy for most eyes. Consider
macular laser as an adjunctive therapy in eyes with persistent
CIDME despite anti-VEGF therapy. Intravitreous steroid
treatment can be used as an alternative in selected cases.
*In addition to optimizing systemic control of blood glucose, cholesterol, and hypertension, as well as educating the patient about importance of
routine follow-up regardless of whether visual symptoms are present or absent.
care.diabetesjournals.org Solomon and Associates 417

secondary complications of PDR than in looking at the latest advancement in ret- 5. Estacio RO, McFarling E, Biggerstaff S, Jeffers
the ranibizumab group. In addition, inopathy treatment, anti-VEGF therapy. BW, Johnson D, Schrier RW. Overt albuminuria
predicts diabetic retinopathy in Hispanics with
whereas 28% of eyes receiving PRP de- These eye injections have been shown NIDDM. Am J Kidney Dis 1998;31:947–953
veloped DME over the course of 2 years, in numerous studies to be more cost- 6. Leske MC, Wu S-Y, Hennis A, et al.; Barbados Eye
only 9% of ranibizumab-treated eyes did effective than laser monotherapy for DME Study Group. Hyperglycemia, blood pressure, and
so. Only 6% of eyes in the ranibizumab (45–47). Future studies will be needed to the 9-year incidence of diabetic retinopathy: the
group received PRP during the course of determine the cost-effectiveness of the Barbados Eye Studies. Ophthalmology 2005;112:
799–805
the study. Systemic safety outcomes ap- anti-VEGF medications as a first-line 7. Chew EY, Davis MD, Danis RP, et al.; Action to
peared equivalent between the groups, treatment for PDR. Control Cardiovascular Risk in Diabetes Eye
and injection-related endophthalmitis Study Research Group. The effects of medical
occurred in only one eye (0.5%) in the management on the progression of diabetic ret-
ranibizumab group. Funding and Duality of Interest. S.D.S. re- inopathy in persons with type 2 diabetes: the
ceives academic support through the Katharine Action to Control Cardiovascular Risk in Diabe-
These results suggest that intravitre- tes (ACCORD) Eye Study. Ophthalmology 2014;
M. Graham Professorship at Wilmer Eye Insti-
ous anti-VEGF may be a viable alterna- 121:2443–2451
tute, Johns Hopkins School of Medicine. E.C. is a
tive or adjunct to PRP for treatment of government employee. L.S. did the work without 8. UK Prospective Diabetes Study (UKPDS)
eyes with PDR through at least 2 years. receiving any financial support from any third Group. Intensive blood-glucose control with sul-
However, in applying these findings to party and over the past 36 months has served as a phonylureas or insulin compared with conven-
consultant for Santen and Genentech. J.K.S. tional treatment and risk of complications in
clinical practice, factors such as fre-
performed the work without any financial sup- patients with type 2 diabetes (UKPDS 33). Lan-
quency of follow-up, treatment cost, port from any third party; her relevant financial cet 1998;352:837–853
and patient preference must be consid- activities outside the submitted work over the 9. ACCORD Study Group, ACCORD Eye Study
ered in addition to these safety and ef- last 36 months include research support from Group, Chew EY, et al. Effects of medical thera-
ficacy outcomes. Complete application Genentech, Optovue, Boston Micromachines, pies on retinopathy progression in type 2 diabe-
Adaptive Sensory Technology, Optos, KalVista, tes. N Engl J Med 2010;363:233–244
of PRP can sometimes be accomplished 10. Do DV, Wang X, Vedula SS, et al. Blood pres-
and Roche, and she has received fees for con-
in as little as one visit, whereas intravit- sulting or invited talks from Allergan, Bayer, sure control for diabetic retinopathy. Cochrane
reous ranibizumab may be required Eisai, Eleven Biotherapeutics, Kowa, Merck, Database Syst Rev 2015;1:CD006127
chronically, over numerous visits, to ad- Novartis, and Regeneron Pharmaceuticals. B.L.V. 11. Aiello LP, Gardner TW, King GL, et al. Diabetic
equately maintain regression of PDR. receives financial support from National Institutes retinopathy. Diabetes Care 1998;21:143–156
of Health K23 Award (1K23EY025729-01). Addi- 12. Diabetes Control and Complications Trial Re-
PRP costs less than a ranibizumab injec- search Group. Effect of pregnancy on microvascular
tional funding was provided by Research
tion and carries no risk of endophthal- to Prevent Blindness and the Paul MacKall & complications in the Diabetes Control and Compli-
mitis. However, if CIDME is present in an Evanina MacKall Foundation Trust as block cations Trial. Diabetes Care 2000;23:1084–1091
eye for which intravitreous anti-VEGF grants to the Scheie Eye Institute. C.C.W. is a 13. Goldstein DE, Blinder KJ, Ide CH, et al. Gly-
therapy is planned, concomitant treat- consultant for Alimera Sciences, Allergan, Alnylam, cemic control and development of retinopathy
Bayer, Clearside Biomedical, DORC, Genentech, in youth-onset insulin-dependent diabetes mel-
ment with PRP may not be necessary, as litus. Results of a 12-year longitudinal study.
ONL Therapeutics, Regeneron Pharmaceuticals,
the anti-VEGF agent will likely effec- Thrombogenics, and Valeant; has minor equity in Ophthalmology 1993;100:1125–1131; discus-
tively manage both the CIDME and the ONL Therapeutics; receives research support from sion 1131–1132
PDR. Allergan, Apellis Pharmaceuticals, Clearside Bio- 14. Rashid S, Young LH. Progression of diabetic
medical, Iconic Therapeutics, Genentech, Regeneron retinopathy and maculopathy after phacoemul-
COST-EFFECTIVENESS OF Pharmaceuticals, Diabetic Retinopathy Clinical Re- sification surgery. Int Ophthalmol Clin 2010;50:
search Network, Ophthotech, ThromboGenics, and 155–166
SCREENING AND TREATMENT FOR Tyrogenex; and has been a speaker for Allergan 15. Sun JK, Radwan SH, Soliman AZ, et al. Neu-
DIABETIC RETINOPATHY and Regeneron Pharmaceuticals. T.W.G. is a ral retinal disorganization as a robust marker of
The cost-effectiveness of both screening consultant for Novo Nordisk, KalVista, Janssen visual acuity in current and resolved diabetic
and traditional laser treatment for dia- Research, PureTech Health, Johnson & Johnson, macular edema. Diabetes 2015;64:2560–2570
and BetaStem, which have no relationship to this 16. Klein R, Knudtson MD, Lee KE, Gangnon R,
betic retinopathy has been established manuscript. No other potential conflicts of inter- Klein BEK. The Wisconsin Epidemiologic Study of
long ago and is no longer in dis- est relevant to this article were reported. Diabetic Retinopathy: XXII the twenty-five-year
pute. More recent literature on cost- progression of retinopathy in persons with type 1
effectiveness has now focused on the References diabetes. Ophthalmology 2008;115:1859–1868
impact of telemedicine on the detection 1. American Diabetes Association. Diabetic reti- 17. Klein R, Klein BE, Moss SE, Cruickshanks
nopathy. Diabetes Care 2002;25(Suppl. 1):S90–S93 KJ. The Wisconsin Epidemiologic Study of Dia-
and eventual management of diabetic betic Retinopathy. XIV. Ten-year incidence and
2. Yau JWY, Rogers SL, Kawasaki R, et al.; Meta-
retinopathy (41). Multiple studies have Analysis for Eye Disease (META-EYE) Study progression of diabetic retinopathy. Arch Oph-
argued both in support of and against Group. Global prevalence and major risk factors thalmol 1994;112:1217–1228
whether telemedicine represents an im- of diabetic retinopathy. Diabetes Care 2012;35: 18. The Diabetes Control and Complications
provement over eye care provider–based Trial Research Group. The effect of intensive
556–564
treatment of diabetes on the development
screening (42,43). Although a consensus 3. Harris Nwanyanwu K, Talwar N, Gardner TW,
and progression of long-term complications in
has yet to be reached, telemedicine Wrobel JS, Herman WH, Stein JD. Predicting de-
insulin-dependent diabetes mellitus. N Engl J
appears to be most effective when the velopment of proliferative diabetic retinopathy. Med 1993;329:977–986
Diabetes Care 2013;36:1562–1568 19. DCCT/EDIC Research Group, Aiello LP, Sun
ratio of providers to patients is low, the
4. Klein R, Lee KE, Gangnon RE, Klein BEK. The W, et al. Intensive diabetes therapy and ocular
distance to reach a provider is prohibi- 25-year incidence of visual impairment in type 1 surgery in type 1 diabetes. N Engl J Med 2015;
tive, or the alternative is no patient diabetes mellitus the Wisconsin Epidemiologic 372:1722–1733
screening (44). In terms of treatment, Study of Diabetic Retinopathy. Ophthalmology 20. UK Prospective Diabetes Study Group. Tight
cost-effectiveness literature has begun 2010;117:63–70 blood pressure control and risk of macrovascular
418 Position Statement Diabetes Care Volume 40, March 2017

and microvascular complications in type 2 diabetes: 31. Bragge P, Gruen RL, Chau M, Forbes A, retinopathy in type I diabetics. Ophthalmology
UKPDS 38. BMJ 1998;317:703–713 Taylor HR. Screening for presence or absence 1989;96:255–264
21. Keech AC, Mitchell P, Summanen PA, et al.; of diabetic retinopathy: a meta-analysis. Arch 42. Phan A-DT, Koczman JJ, Yung C-W, Pernic
FIELD study investigators. Effect of fenofibrate Ophthalmol 2011;129:435–444 AA, Doerr ED, Kaehr MM. Cost analysis of tele-
on the need for laser treatment for diabetic 32. Ahmed J, Ward TP, Bursell S-E, Aiello LM, retinal screening for diabetic retinopathy in a
retinopathy (FIELD study): a randomised con- Cavallerano JD, Vigersky RA. The sensitivity and county hospital population. Diabetes Care
trolled trial. Lancet 2007;370:1687–1697 specificity of nonmydriatic digital stereoscopic 2014;37:e252–e253
22. Canadian Ophthalmological Society Dia- retinal imaging in detecting diabetic retinopa- 43. Kirkizlar E, Serban N, Sisson JA, Swann JL,
betic Retinopathy Clinical Practice Guideline Ex- thy. Diabetes Care 2006;29:2205–2209 Barnes CS, Williams MD. Evaluation of teleme-
pert Committee, Hooper P, Boucher MC, et al. 33. Early Treatment Diabetic Retinopathy dicine for screening of diabetic retinopathy in
Canadian Ophthalmological Society evidence- Study research group. Photocoagulation for di- the Veterans Health Administration. Ophthal-
based clinical practice guidelines for the manage- abetic macular edema. Early Treatment Diabetic mology 2013;120:2604–2610
ment of diabetic retinopathy. Can J Ophthalmol Retinopathy Study report number 1. Arch Oph- 44. Pasquel FJ, Hendrick AM, Ryan M, Cason E,
2012;47:91–96 thalmol 1985;103:1796–1806 Ali MK, Narayan KMV. Cost-effectiveness of dif-
23. Klein R, Klein BE, Moss SE, Davis MD, 34. Diabetic Retinopathy Clinical Research ferent diabetic retinopathy screening modali-
DeMets DL. The Wisconsin Epidemiologic Study Network, Elman MJ, Aiello LP, et al. Randomized ties. J Diabetes Sci Technol 2015;10:301–307
of Diabetic Retinopathy. II. Prevalence and risk trial evaluating ranibizumab plus prompt or de- 45. Haig J, Barbeau M, Ferreira A. Cost-effectiveness
of diabetic retinopathy when age at diagnosis is ferred laser or triamcinolone plus prompt laser of ranibizumab in the treatment of visual im-
less than 30 years. Arch Ophthalmol 1984;102: for diabetic macular edema. Ophthalmology pairment due to diabetic macular edema. J Med
520–526 2010;117:1064–1077.e35 Econ 2016;19:663–671
24. Klein R, Klein BE, Moss SE, Davis MD, 35. Mitchell P, Bandello F, Schmidt-Erfurth U, 46. Brown GC, Brown MM, Turpcu A, Rajput Y.
DeMets DL. The Wisconsin Epidemiologic Study et al.; RESTORE study group. The RESTORE study: The cost-effectiveness of ranibizumab for the
of Diabetic Retinopathy. III. Prevalence and risk ranibizumab monotherapy or combined with laser treatment of diabetic macular edema. Ophthal-
of diabetic retinopathy when age at diagnosis is versus laser monotherapy for diabetic macular mology 2015;122:1416–1425
30 or more years. Arch Ophthalmol 1984;102: edema. Ophthalmology 2011;118:615–625 47. Pershing S, Enns EA, Matesic B, Owens DK,
527–532 36. Elman MJ, Bressler NM, Qin H, et al.; Dia- Goldhaber-Fiebert JD. Cost-effectiveness of
25. Klein R, Klein BE, Moss SE, Davis MD, betic Retinopathy Clinical Research Network. treatment of diabetic macular edema. Ann In-
DeMets DL. The Wisconsin Epidemiologic Study Expanded 2-year follow-up of ranibizumab tern Med 2014;160:18–29
of Diabetic Retinopathy. IX. Four-year incidence plus prompt or deferred laser or triamcinolone 48. Xu J, Xu L, Wang YX, You QS, Jonas JB, Wei
and progression of diabetic retinopathy when plus prompt laser for diabetic macular edema. WB. Ten-year cumulative incidence of diabetic
age at diagnosis is 30 years or more. Arch Oph- Ophthalmology 2011;118:609–614 retinopathy. The Beijing Eye Study 2001/2011.
thalmol 1989;107:237–243 37. Nguyen QD, Brown DM, Marcus DM, et al.; PLoS One 2014;9:e111320
26. Agardh E, Tababat-Khani P. Adopting 3-year RISE and RIDE Research Group. Ranibizumab for 49. Kajiwara A, Miyagawa H, Saruwatari J, et al.
screening intervals for sight-threatening retinal diabetic macular edema: results from 2 phase III Gender differences in the incidence and pro-
vascular lesions in type 2 diabetic subjects without randomized trials: RISE and RIDE. Ophthalmol- gression of diabetic retinopathy among Japa-
retinopathy. Diabetes Care 2011;34:1318–1319 ogy 2012;119:789–801 nese patients with type 2 diabetes mellitus: a
27. Axer-Siegel R, Hod M, Fink-Cohen S, et al. 38. Diabetic Retinopathy Clinical Research clinic-based retrospective longitudinal study.
Diabetic retinopathy during pregnancy. Oph- Network, Wells JA, Glassman AR, et al. Afliber- Diabetes Res Clin Pract 2014;103:e7–e10
thalmology 1996;103:1815–1819 cept, bevacizumab, or ranibizumab for diabetic 50. Jin P, Peng J, Zou H, et al. The 5-year onset
28. Gunderson EP, Lewis CE, Tsai A-L, et al. A macular edema. N Engl J Med 2015;372:1193– and regression of diabetic retinopathy in Chi-
20-year prospective study of childbearing and 1203 nese type 2 diabetes patients. PLoS One 2014;
incidence of diabetes in young women, control- 39. The Diabetic Retinopathy Study Research 9:e113359
ling for glycemia before conception: the Coro- Group. Preliminary report on effects of photo- 51. Diabetes Control and Complications Trial
nary Artery Risk Development in Young Adults coagulation therapy. Am J Ophthalmol 1976;81: Research Group. Progression of retinopathy
(CARDIA) Study. Diabetes 2007;56:2990–2996 383–396 with intensive versus conventional treatment
29. Chew EY, Mills JL, Metzger BE, et al. Metabolic 40. Writing Committee for the Diabetic Reti- in the Diabetes Control and Complications Trial.
control and progression of retinopathy. The Diabe- nopathy Clinical Research Network, Gross JG, Ophthalmology 1995;102:647–661
tes in Early Pregnancy Study. National Institute of Glassman AR, et al. Panretinal photocoagulation 52. The Diabetes Control and Complications
Child Health and Human Development Diabetes in vs intravitreous ranibizumab for proliferative di- Trial Research Group. The relationship of glyce-
Early Pregnancy Study. Diabetes Care 1995;18: abetic retinopathy: a randomized clinical trial. mic exposure (HbA1c) to the risk of development
631–637 JAMA 2015;314:2137–2146 and progression of retinopathy in the Diabetes
30. Ferris FL 3rd. How effective are treatments for 41. Javitt JC, Canner JK, Sommer A. Cost effec- Control and Complications Trial. Diabetes 1995;
diabetic retinopathy? JAMA 1993;269:1290–1291 tiveness of current approaches to the control of 44:968–983