Telemedicine for Diabetic Retinopathy

Current strategies, future directions, and the role of the retinal physician

Telemedicine for Diabetic Retinopathy

Current strategies, future directions, and the role of the retinal physician


Telemedicine allows healthcare providers to diagnose and manage diseases remotely via the use of information technology. This simple concept has transformed the practice of radiology, because teleradiologists can now diagnose disease even if they are thousands of miles away from the patient.

Radiology lends itself to a telemedicine approach in large part because it is such a visual specialty; diagnoses can be made based on images that are easily transmittable across great distances.

Like radiology, ophthalmology is another field in which the diagnosis can often be made based on visualization of the underlying pathology. This is especially true for diseases that affect the retina, such as diabetic retinopathy (Figure).

Figure. Clinical features of DR may include hemorrhages, hard exudates, cotton wool spots, and/or neovascularization.

In the case of DR, nearly all of the information required to make a decision regarding diagnosis, staging of the disease, and treatment can be gleaned via a thorough inspection of the retina. Now that it is possible to obtain high-quality images of the retina for later review by an expert reader (a “store and forward” method of telemedicine), DR has become a model ophthalmic disease for a telemedicine-based approach to reducing preventable blindness.

The burden of diabetes and DR in the United States is extraordinary. It is already a leading cause of vision loss in the Western world, and the number of people with DR is expected to triple by 2050.1

Maxwell S. Stem, MD, is a vitreoretinal fellow at Associated Retinal Consultants, PC, in Royal Oak, MI. Maria A. Woodward, MD, MS, is on the faculty of the Kellogg Eye Center of the University of Michigan in Ann Arbor. Dr. Woodward receives funding from the NEI. Dr. Stem reports no financial interest in products mentioned in this article. Dr. Stem can be reached via e-mail at

Given that DR progresses in a step-wise fashion from asymptomatic changes in the retinal vasculature to potentially blinding complications, such as retinal detachment and macular edema, it is extremely important that patients with diabetes are screened regularly for signs of DR.

Screening guidelines suggest that patients with type 1 or type 2 diabetes be examined annually for signs of DR.2 However, the way in which such screening should or can be accomplished is a matter of some debate.


Eyecare providers routinely screen for DR. In this model, patients with diabetes and coexisting eye diseases can have many of their ophthalmic needs addressed at a single appointment if DR has not developed. Active cases of DR, once discovered, are referred promptly to a retinal specialist for treatment.

For patients without comorbid eye conditions who simply attend ophthalmic visits to be screened for DR, the hassle of scheduling and attending an appointment represents time away from work that some patients cannot afford.

When the problem of screening patients with diabetes for DR is viewed through the lens of an expanding diabetic population and a relatively shrinking ophthalmic workforce, the current system of in-office eye examinations for all patients with diabetes does not appear sustainable.

The incidence of diabetes in the United States is estimated at 1.4 million cases per year.3 To put this number in perspective, 467 medical students matched into ophthalmology in 2016, and not all of these future graduates will screen patients for DR.4 In 2015, there were 1,557 graduates from optometry schools in the US, including Puerto Rico.5

Thus, each year, there are roughly 2,000 new eyecare professionals faced with screening 1.4 million patients newly diagnosed with diabetes, which equates to one eyecare provider per 700 patients. While some of the patients newly diagnosed with diabetes may already have eye doctors who can screen them, many will not.

Also, the predicted work shortages for ophthalmologists, combined with an increasing incidence of diabetes, will continue to shift the ratio of diabetic patients to eyecare practitioners to an increasingly large and unsustainable number.6


Some organizations in the US and abroad have adopted a telemedicine approach to DR monitoring. The American Telemedicine Association (ATA) categorizes screening programs into four categories, based on their ability to detect various stages of DR.7

In brief, category 1 systems can distinguish between patients who have no or minimal DR and those who have more than minimal DR. Category 2 systems can stratify patients into those with no or minimal DR and those with sight-threatening DR, such as diabetic macular edema, severe nonproliferative DR, or proliferative DR.

Category 3 systems are akin to an in-office examination by an eyecare professional; thus, treatment and follow-up recommendations can be made based on these images. Category 4 systems can match or exceed the gold standard of seven-field standard ETDRS photos for the diagnosis and management of DR (such a program does not exist at this time).

Current telemedicine DR screening programs were recently reviewed by Tozer and colleagues.8 A category 2 or 3 system would likely be preferred over a category 1 system because they would allow for direct referral to a retinal specialist if sight-threatening retinopathy is present.

One example of a category 2 system is the National Health Service Diabetic Eye Screening Program (NDESP) in the United Kingdom. Most centers in the NDESP obtain two-field mydriatic photos (one centered on the macula and the other centered on the optic disc), which are read in tiered fashion by trained graders who are not ophthalmologists — cases of retinopathy are reviewed by an additional grader before referral decisions are made. In a telling statistic, after implementation of the NDESP, DR lost its place as the leading cause of blindness in England and Wales for the first time in 50 years.9

In the United States, the Joslin Vision Network (JVN) provides an excellent example of a category 3 system. JVN protocols mandate fundus photos in three fields: (1) the optic disc and macula; (2) nasal to the disc; and (3) the superotemporal arcade. Unlike in the NDESP, a trained eye-specialist reviews the images and recommends follow-up based on disease severity. Using these three fields has good agreement (up to 95%) with seven-field ETDRS photos.10

The JVN protocol is in use at the Veterans’ Affairs Administration, in which a randomized trial found that 87% of patients enrolled in the telemedicine program had annual screenings for DR, compared to 77% in the traditional program.11

Potential benefits to incorporating a telemedicine program for DR screening include increased cost-effectiveness and adherence without sacrificing diagnostic accuracy. Telemedicine for DR is generally regarded as cost-effective, compared to traditional eyecare models with in-office screening.12-14

Adherence to screening programs is generally better among patients using a telemedicine approach, especially when the patients perceive the programs to be more convenient than in-office examinations.15

Studies comparing telemedicine diagnosis of DR to a fundus examination by a retinal specialist have found good agreement between the different strategies; in one study comparing a JVN protocol to a retinal specialist examination, there was 89% agreement within one ETDRS grade between the two modalities.16


Once the decision to incorporate a category 2 or 3 DR screening protocol into a healthcare system has been made, logistical considerations regarding the who, what, when, where, and how of obtaining and grading the images become extremely important. What follows is a potential framework for how such a screening system might be integrated into the US healthcare system.

Practical considerations for implementing a telemedicine approach to DR screening include the type of camera to be used, whether or not dilation will be used, the areas of the fundus that requires imaging, who will read the photos, and who will be responsible for referring patients to subspecialists.

There are many cameras that can be used to image the fundus, as evidenced by the extensive list at the NDESP Web site.17 The merits and drawbacks to each camera are beyond the scope of this article. Generally, higher-quality photos are obtained if the patient is dilated prior to obtaining the pictures, although dilation comes with a very rare (~1/20,000) risk of inducing angle-closure glaucoma in susceptible individuals.18 A good compromise would be to employ selective mydriasis, with mild dilating agents only used if adequate photographs cannot be obtained through a nondilated pupil.

Granular decisions about how to design the program include the method of acquiring photos, determining who will interpret the photographs, and the role of general ophthalmologists and retina specialists.

One approach would be to treat image acquisition as a laboratory test — primary care physicians could write prescriptions for the images, and the fundus photographs could then be obtained at local blood draw stations, along with the other routine blood tests that are used to monitor patients with diabetes.

Phlebotomists or other personnel could be trained to use fundus cameras, creating a “one-stop shop” for patients. In this model, an interpretation of the fundus images would be sent to the PCP, who would then refer the patient depending on the severity of the DR and the presence or absence of other ocular disease.

The interpretation of the fundus images could be performed by nonphysician graders (as in the NDESP) or by eyecare professionals (as in JVN). One advantage to having general ophthalmologists and optometrists grade the images is that they may be more likely to detect the coexistence of other ocular pathologies that require in-office examination.

We suspect that the role of the retinal physician in managing patients with DR would remain largely unchanged if a telemedicine approach to screening were adopted.


Given the increasing prevalence of diabetes in the United States and the relative paucity of eyecare professionals able to perform in-office examinations for DR screening, it seems to be only a matter of time before more streamlined approaches to DR screening become the standard of care.

A telemedicine-based approach to DR screening has been successfully adopted in the United Kingdom and United States at facilities such as the VA and the Joslin Diabetes Center. The implementation of a category 2 or 3 screening program for DR would likely reduce healthcare costs and improve screening adherence among eligible patients.

While such programs would reduce the number of routine diabetic eye exams performed by optometrists and general ophthalmologists, retinal specialists would be unlikely to experience a huge shift in their clinical practice.

While they may receive more referrals to treat sight-threatening retinopathy (as adherence to screening improves), retinal specialists will likely continue to be primarily responsible for the management, and not the screening, of DR. RP


1. Saadine JB, Fong DS, Yao J. Factors associated with follow-up eye examinations among persons with diabetes. Retina. 2008;28:195-200.

2. Standards of medical care in diabetes--2015: summary of revisions. Diabetes Care. 2015;38 (Suppl):S4.

3. American Diabetes Association. Statistics about diabetes. Available at: Accessed September 3, 2016.

4. SF Match. Ophthalmology residency. Available at: Accessed September 3, 2016.

5. Association of Schools and Colleges of Optometry. Applicant/student profile and prerequisites. Available at: Accessed September 3, 2016.

6. U.S. Department of Health and Human Services. Physician Supply and Demand: Projections to 2020. Available at: Accessed September 3, 2016.

7. American Telemedicine Association. Telehealth practice recommendations for diabetic retinopathy. Available at: Accessed September 3, 2016.

8. Tozer K, Woodward MA, Newman-Casey PA. Telemedicine and diabetic retinopathy: Review of published screening programs. J Endocrinol Diabetes. 2015;2(4).

9. Liew G, Michaelides M, Bunce C. A comparison of the causes of blindness certifications in England and Wales in working age adults (16-64 years), 1999-2000 with 2009-2010. BMJ Open. 2014;4:e004015.

10. Moss SE, Meuer SM, Klein R, Hubbard LD, Brothers RJ, Klein BE. Are seven standard photographic fields necessary for classification of diabetic retinopathy? Invest Ophthalmol Vis Sci. 1989;30:823-828.

11. Conlin PR, Fisch BM, Orcutt JC, Hetrick BJ, Darkins AW. Framework for a national teleretinal imaging program to screen for diabetic retinopathy in Veterans Health Administration patients. J Rehabil Res Dev. 2006;43:741-748.

12. Li Z, Wu C, Olayiwola JN, Hilaire DS, Huang JJ. Telemedicine-based digital retinal imaging vs standard ophthalmologic evaluation for the assessment of diabetic retinopathy. Conn Med. 2012;76:85-90.

13. Kirkizlar E, Serban N, Sisson JA, Swann JL, Barnes CS, Williams MD. Evaluation of telemedicine for screening of diabetic retinopathy in the Veterans Health Administration. Ophthalmology. 2013;120:2604-2610.

14. Bashshur RL, Shannon GW, Smith BR, Woodward MA. The empirical evidence for the telemedicine intervention in diabetes management. Telemed J E Health. 2015;21:321-354.

15. Valikodath NG, Leveque TK, Wang SY, et al. Patient Attitudes Toward Telemedicine for Diabetic Retinopathy. Telemed J E Health. 2016.

16. Cavallerano AA, Cavallerano JD, Katalinic P, et al. Use of Joslin Vision Network digital-video nonmydriatic retinal imaging to assess diabetic retinopathy in a clinical program. Retina. 2003;23:215-223.

17. Diabetic eye screening: approved cameras and settings. Available at: Accessed September 3, 2016.

18. Pandit RJ, Taylor R. Mydriasis and glaucoma: exploding the myth. A systematic review. Diabet Med. 2000;17:693-699.