Despite advances in eye care, diabetic macular edema remains the most common vision-impacting complication of diabetic retinopathy and represents one of the major causes of visual impairment in the developed world.1 With prolonged and poorly controlled blood glucose levels, the integrity of the microvasculature supplying the inner retina is compromised by inflammation and hyperglycemic-induced damage to pericytes, resulting in “leaky” vessels leading to retinal edema and thickening.2–4 This excess permeability can be treated with intravitreal anti-VEGF injections, which have become a mainstay of care for patients with visually significant DME. 


What’s ‘visually significant’ DME?

But, what is “visually significant” DME? The historic term “clinically significant macular edema” (CSME) was defined in the Early Treatment of Diabetic Retinopathy Study as a clinical diagnosis based on the exam findings, not optical coherence tomography or visual acuity results.5 Advances in imaging technology, particularly with the ubiquity of OCT, have allowed clinicians to detect more subtle DME, often before patients have appreciated a decline in visual acuity or visual function. In fact, one study showed more than 80 percent of patients with DME have a visual acuity of 20/40 or better.6 

High-quality, randomized, masked studies such as RISE and RIDE (leading to the approval of ranibizumab [Lucentis, Roche/Genentech]), or VIVID and
VISTA (aflibercept [Eylea, Regeneron Pharmaceuticals]), excluded patients with good vision, leaving retina specialists uncertain about how to manage patients with DME and good vision. Thankfully, the DRCR Retina Network has stepped in to help address this important question with Protocol V, assessing treatment options for patients with DME and “very” good vision. 7 




Protocol V design and findings

This was a multicenter, randomized trial that took place from 2013 to 2018. Patients with center-involving DME greater than 250 µm on time-domain OCT and an ETDRS best-corrected visual acuity of 20/25 or better were enrolled. Subjects were randomized to one of three arms: aflibercept injections; focal/grid laser photocoagulation; or observation. 

The primary outcome of the study was investigating the proportion of patients who lost at least one line (5 letters) of vision at two years. Patients were followed every eight to 16 weeks. Those randomized to the focal/grid laser photocoagulation and observation arms were allowed to be rescued if the visual acuity deteriorated by two lines or more during any visit or by one to two lines during two consecutive visits. 

Across all arms, the proportion of patients who lost at least 5 letters of visual acuity at two years was similar with 16-, 17- and 19-percent declines in the
aflibercept, laser/grid photocoagulation and observation arms, respectively (p=0.79). Also, 25 percent of patients in the photocoagulation group and 34 percent in the observation group received intravitreal aflibercept for significant visual acuity decline according to the study’s rescue parameters. The median number of injections in the laser and observation groups were seven and nine, respectively, for those who required rescue. After two years of follow-up, VA across all groups was comparable, with a mean of 20/20. 


Real-world implications

These results suggest that patients with CI-DME and good vision (20/25 or better) can be managed initially with observation and close follow-up. This is important information for retina specialists and patients alike, and can provide the opportunity for clinicians to spend the time educating their patients on diabetic retinopathy and the importance of blood-sugar and blood-pressure control. 

However, as with any good study that seeks to answer a particular question, Protocol V raises several new questions as we try to apply these trial results to the clinical setting. They include:

• Clinic vision vs. ETDRS vision. In a busy clinical setting, many factors can affect the measured VA, including the exam lane used, the technician performing the assessment, whether the patient brought his or her glasses that day, whether pinhole measurement is used, the patient’s motivation, the severity of his or her dry-eye symptoms that day, etc. 

How should we interpret the normal fluctuations in vision that can occur in patients, when the clinical exam and/or imaging remain unchanged? In addition, for those of us who participate in clinical trials, we know that 20/25 ETDRS vision is very different than 20/25 clinic vision. Typically, the ETDRS vision is better than the vision measured in the clinic. Should we therefore extrapolate Protocol V findings to patients with 20/30 or 20/40 vision? 

• Other factors that impact vision. What should we do for patients with CI-DME and a cataract? Or ocular surface disease? Sometimes it can be difficult to tell which factors are contributing to the visual decline. 

• Progressive disease. The retreatment criteria in Protocol V were visual acuity-based. Given the challenges listed earlier with clinic visual acuities, how should we respond to the patient whose DME is worsening on OCT? Or if his or her peripheral diabetic retinopathy is worsening? Sometimes, even if we don’t routinely treat nonproliferative diabetic retinopathy with anti-VEGF injections, we may be more prone to do so when DME is present, even with good vision.

• Follow-up intervals. Clinical trials have predefined schedules for evaluation that often are more frequent than what’s practical in the real world. In addition, patients who participate in clinical trials understand the commitment they are making over a one- or two-year period. Many patients choose not to participate in trials because the commitment is too burdensome. For these patients who cannot commit to Protocol V’s close follow-up standards, how should we extrapolate the safety of observation, particularly when one-third of observed patients ultimately required some treatment?

• Improved to 20/25. Protocol V enrolled patients who had treatment-naïve DME. But, what about the huge cohort of patients who are treated for DME and improve their vision to Protocol V standards (i.e., 20/25 or better)? Can we then safely observe them until their vision declines again? Or should we continue with regular treatment based on the algorithms of prior studies like RISE, RIDE, VIVID, VISTA or DRCR Protocols I and T?8,9 

Bottom line

The DRCR Retina Network plays a critical role in helping to address important clinical questions in the management of diabetic retinopathy. Often, these are studies that drug manufacturers may not be motivated to support directly. Protocol V aimed to answer one such question: How should we manage patients with good vision despite center-involving  DME on exam and OCT? 

Based on the results of Protocol V, clinicians should feel more confident if they decide to observe these patients while following them closely. Like most great studies, and perhaps it's always the case, understanding one thing better opens up a whole new line of questions, some of which may never be fully answered. We will then have to apply these scientific results to the individual patient sitting in the exam chair in front of us. In this way, the art of medicine lives on. RS



1. Soloman SD, Chew E, Duh EJ, et al. Diabetic retinopathy: A position statement by the American Diabetes Association. Diabetes Care. 2017;40:412-418. 

2. Romero-Aroca P, Baget-Bernaldiz M, Pareja-Rios A, Lopez-Galvez M, Navarro-Gil R, Verges R. Diabetic macular edema pathophysiology: vasogenic versus inflammatory. J Diabetes Res. 2016;2016:2156273. Published online September 28, 2016.

3. Satirtav G, Oltulu R, Kerimoglu H. Current evidence of pathophysiology of diabetic macular edema: A review. World J Ophthalmol. 2014;4:147-151.

4. Musat O, Cernat C, Labib M, et al. Diabetic macular edema. Rom J Ophthalmal. 2015;59:133-136.

5. Early Treatment Diabetic Retinopathy Study Research Group. Photocoagulation for diabetic macular edema: Early Treatment Diabetic Retinopathy Study report number 1. Arch Ophthalmol. 1985;103:1796-1806.

6. Bressler NM, Varma R, Doan QV, et al. Underuse of healthcare system by persons with diabetes mellitus and diabetic macular edema in the United States. JAMA Ophthalmol. 2014;132:168-173.

7. Baker CW, Glassman AR, Beaulieu WT, et al., for the DRCR Retina Network. Effect of initial management with aflibercept vs laser photocoagulation vs. observation on vision loss among patients with diabetic macular edema involving the center of the macula and good visual acuity. A randomized clinical trial. JAMA. 2019;321:1880-1894.

8. Elman MJ, Aiello LP, Beck RW, et al., for the Diabetic Retinopathy Clinical Research Network. Randomized trial evaluating ranibizumab plus prompt or deferred laser or triamcinolone plus prompt laser for diabetic macular edema. Ophthalmology. 2010;117:1064-1077.

9. Wells J, Glassman A, Ayala A, et al., for the he Diabetic Retinopathy Clinical Research Network. Aflibercept, bevacizumab or ranibizumab for diabetic macular edema. N Engl J Med 2015;372:1193-1203.