Decision Making in Diabetic Retinopathy Treatment

A variety of tools enable individualized care.


With 1.5 million new cases of diabetes diagnosed in the United States each year,1 retina specialists will likely see a higher incidence of diabetic retinopathy (DR) in their patient populations. Given the steady rise of cases and renewed public interest in preventive care, it is incumbent upon us to continuously seek modalities to minimize progression and reduce long-term vision loss. These concepts, along with the compendium of evidence we have for treating diabetic eye disease, guide our decision making for DR treatment.


We are fortunate in the field that there have been well-designed, large, randomized, controlled trials evaluating different approaches to the treatment of diabetic eye disease. Until recently, established, evidence-based paradigms adequately delineated first-line treatment of various manifestations of DR: panretinal laser photocoagulation (PRP) for proliferative disease; focal laser for non-center-involving, clinically significant macular edema, and intravitreal anti-vascular endothelial growth factor (VEGF) injections for center-involving macular edema with vision loss. Now, the landscape for treating DR is changing again, as we are learning more about the risks and benefits of pharmacotherapy.


In the setting of proliferative diabetic retinopathy (PDR), the Diabetic Retinopathy Study (DRS) defined cases that benefitted from prompt panretinal laser. Laser treatment of this “high-risk PDR” reduced the incidence of severe vision loss by approximately 50%, but not without adverse effects, such as a decrease in acuity, an increase in macular edema, and decreases in visual field and scotopic vision.2 These data have shaped our treatment of PDR for the last 2 generations.


Similarly, parameters for clinically significant macular edema were defined in the Early Treatment for Diabetic Retinopathy Study (ETDRS), with anatomic improvement and a reduction of progressive central acuity loss by approximately 50% in patients treated with macular laser. The ETDRS examined patients with both good vision and non-center-involved diabetic macular edema (DME), as well as patients with decreased vision and center-involving DME. Although there was a reduction in the proportion of patients with moderate vision loss, the average improvement in vision for patients with decreased vision and center-involving macular edema was small in the ETDRS.3 Visual acuity (VA) data in subsequent trials over the past decade have prompted a transition from laser to intravitreal anti-VEGF therapy as the “gold standard” first-line treatment for center-involving DME.

For center-involved DME with vision loss, the RISE and RIDE ranibizumab (Lucentis; Genentech) studies demonstrated improvement in vision and central subfield thickness with ranibizumab vs sham groups with DME.4 Subsequently, VIVID and VISTA demonstrated similar results for intravitreal aflibercept (Eylea; Regeneron).5 In concert with the pivotal industry-sponsored ranibizumab and aflibercept trials, the Diabetic Retinopathy Clinical Research Network’s ( ) Protocol I prospectively compared prompt laser to deferred laser in combination with intravitreal ranibizumab, and it can be argued that the results of Protocol I, demonstrating better VA results with deferred laser for center-involving visually significant DME, were the evidence that moved retina solidly into the pharmacotherapeutic era for center-involving DME.6’s Protocol T suggested that aflibercept outperformed ranibizumab and bevacizumab (Avastin; Genentech) in DME patients with VA of 20/50 or worse over the first year of that trial, but ranibizumab and aflibercept demonstrated equivalency by the end of the second year, with bevacizumab performing statistically significantly worse.7 Interestingly, a significant minority of the population in Protocol T failed to have “dry” maculas throughout the trial despite frequent anti-VEGF treatment. In clinical practice, patients representative of this group are candidates for other available treatment modalities, including intravitreal steroids.


The benefits of periocular and intraocular steroid in DME within the context of the anti-VEGF landscape are less well elucidated in randomized, controlled trials than the data for intravitreal anti-VEGF agents, but there are historic and emerging data as well as practice patterns supporting their utilization. In Protocol I, intravitreal triamcinolone acetonide performed essentially as well as ranibizumab anti-VEGF therapy with regard to visual gains in pseudophakic patients over the first 2 years of the trial.6 The MEAD study showed somewhat improved vision in groups receiving monotherapy with an intravitreal dexamethasone implant (Ozurdex; Allergan) compared to sham with no rescue treatment, and subanalysis of patients who underwent cataract surgery during the trial showed even better results in those patients.8 This study was followed recently by’s Protocol U, which investigated the utility of dexamethasone implant used in combination with ranibizumab vs ranibizumab monotherapy for persistent DME.9

While the combination-therapy arm in Protocol U yielded anatomical improvement compared to monotherapy, it came without significant average visual gains in the overall population. However, pseudophakic subjects again showed a numerical advantage over phakic subjects, even in this short 6-month study. Steroids carry well-known risks of increased cataract and steroid-responsive intraocular pressure in a proportion of patients. A compelling theme throughout the modern randomized, controlled trials employing steroid therapy is that pseudophakic subjects — free of the confounder of cataract advancement — repeatedly showed a trend toward better acuity when treated with intravitreal steroid or combination therapy.

The FAME registration study of the long-acting fluocinolone implant (Iluvien; Alimera Sciences) demonstrates the risks of steroids, with almost all phakic patients developing cataracts in the trial, and a small but significant percentage of subjects requiring incisional glaucoma surgery.10 In a telling counterpoint to FAME, more recent real-world fluocinolone implant studies have highlighted potential practical benefits of steroid therapy. The USER and PALADIN studies demonstrated a significant reduction in treatment burden with maintenance of vision after intravitreal implant placement. Real-world data also show a much smaller number of patients requiring glaucoma procedures compared to the FAME registration trial, an indication that retina specialists are selecting good candidates for lower risk in real-world fluocinolone use than in the earlier registration trial.11,12 Research remains aimed toward integrating currently available steroids into our treatment plans, as well as developing ocular steroid delivery systems that will mitigate these adverse effects.


The landscape of DR treatment has been further advanced by data supporting the regression of retinopathy with serial anti-VEGF injections in both non-PDR and PDR, with randomized controlled data showing noninferiority of anti-VEGF injection to PRP in PDR diabetic retinopathy. The expansion of the FDA indications for ranibizumab injection to include DR without DME was largely supported by’s Protocol S data studying PDR13 and secondarily from RISE and RIDE, which studied mostly non-PDR.

The CLARITY study also demonstrated noninferiority of aflibercept vs panretinal laser in PDR.14 These data sets suggest that serial anti-VEGF adequately prevents worsening of DR and supports regression of the level of retinopathy, without the disadvantage of tissue destruction and adverse visual events from laser. Some secondary endpoints in both PDR trials suggest superior results with anti-VEGF therapy for proliferative disease, including an improvement in VA, even in subjects coming into the trials with decent vision and no DME. Improvements in nonacuity parameters of visual function and less need for vitrectomy in the arm randomized to anti-VEGF primary therapy for PDR were shown in both Protocol S and CLARITY. Thus, anti-VEGF is now a viable treatment for both non-PDR and PDR, and has rapidly gained popularity in place of, or in combination with, PRP for some PDR patients, per the ASRS Preferences and Trends survey.

The disadvantages of anti-VEGF therapy include a limited half-life necessitating a relatively high treatment burden, increased costs, and infrequent yet serious risks of endophthalmitis, iatrogenic cataract, and retinal tear or detachment. There is uncertainty regarding the real-world endpoint for anti-VEGF therapy in PDR, and there is less clarity regarding the benefit of anti-VEGF therapy in non-PDR with good vision and no DME, despite the probability that serial anti-VEGF therapy could regress retinopathy in a number of non-PDR patients. Furthermore, anti-VEGF monotherapy may not be appropriate for extremely severe presentations of PDR with macular traction, as these cases were excluded from Protocol S and CLARITY. These tractional cases are typically reserved for individualized combination approaches, including selective use of anti-VEGF injections with PRP and/or vitrectomy.


So how does one approach a typical diabetic patient? It begins with a conversation about lifestyle habits and an assessment of blood glucose control, as well as overall medical fitness. Often, the diabetic patients who make it to our clinics tend to struggle in these areas. We always emphasize the importance of controlling systemic risk factors to achieve a maximal response to any intervention. All diabetic patients in our clinics undergo macular optical coherence tomography (OCT), and most of those with significant retinopathy on exam undergo an initial fluorescein angiography to assess for posterior or peripheral nonperfusion, subclinical edema, and/or neovascularization. Patients who are fortunate enough to have mild nonproliferative disease without edema are seen annually for dilated fundus exams, or some may return to their referring comprehensive ophthalmologist. We follow edema-free moderate nonproliferative cases as often as every 6-9 months and severe nonproliferative cases every 6 months or more. For classification purposes, we use the practical DRS 4:2:1 definition of severe non-PDR: intraretinal hemorrhages in 4 quadrants, venous beading in 2, or intraretinal microvascular abnormality in 1.


Despite the new data supporting regression with intravitreal injections, most of our asymptomatic, nonproliferative patient population has elected not to treat asymptomatic, nonproliferative disease with serial anti-VEGF injections. We have opened the conversation about the option of therapy with a number of these patients given the newest evidence of regression. Some severe non-PDR patients have elected to undergo injections for their asymptomatic disease. Most them have visual problems related to diabetes in their contralateral eye, and the thought of “turning back the clock” on the better eye is appealing in this clinical setting. Some of the hesitation in treating asymptomatic edema-free non-PDR patients is the absence of data regarding timing of treatment initiation, dosing, or endpoints. Subgroup analyses of existing clinical trials15 and ongoing randomized, controlled trials, such as DRCR’s Protocol W and PANORAMA, are under way to provide prospective evidence and to help formulate clear recommendations for patients who fall into this category.

Decisions in patients with NPDR and diabetic macular edema are more clear cut because we have more evidence and experience with this common manifestation of DR. Perhaps the best “golden rule” when it comes to decision making in DR is to treat most symptomatic center-involving macular edema with anti-VEGF therapy as the first line. In our hands, the risk of vision loss with observation and the good results with injections encourage us to almost always initiate treatment. A trial of anti-VEGF injections has a very high benefit-to-risk ratio. We will start with a series of approximately 3-6 injections while evaluating for improvement in vision and OCT anatomy. If there is measurable, consistent improvement, progression to a dry macula, and patient compliance with the treatment plan, we will continue regular injections until response plateaus and switch the patient to treat-and-extend or PRN dosing. If the response is inadequate, with persistent intraretinal fluid or recurrent fluid with a short extension, we consider switching to another anti-VEGF or adding intravitreal corticosteroid. We will also consider adding intravitreal corticosteroid in a patient who is performing well on anti-VEGF injections, but who requires ongoing high-frequency treatment, resulting in frustration or fatigue with injections. The decision to add intravitreal steroid earlier is easier in pseudophakic patients or older patents with some pre-existing cataract. There are some patients who we consider for intravitreal steroid first-line therapy, such as patients with DME who have recently had cataract surgery and suffer an inflammatory, postoperative exacerbation in their edema.

For non-center-involving DME, there are a variety of good strategies. If the patient is symptomatic and/or meets the ETDRS criteria for clinical significance, focal laser monotherapy is reasonable, but we can also offer anti-VEGF and defer focal laser. The benefit of a series of injections, especially in a motivated patient, is that the injections address both the underlying retinopathy and the presenting noncentral edema. Because macular edema, center involving or otherwise, is more common with more severe retinopathy, injection therapy is a reasonable thing to consider for its diffuse effect.


For patients with PDR, excluding tractional detachment and dense vitreous hemorrhage, we often offer a series of anti-VEGF injections and monitor for regression of neovascularization. If there is high suspicion for ongoing noncompliance on presentation, there is a developing pattern of noncompliance with follow-up or intolerance of injection, or we are concerned that the PDR is worsening or responding poorly after 4-6 injections, we will often perform PRP, usually over 2 sessions, placing at least 1,200 spots in a similar fashion to the DRS laser protocol. We sometimes avoid anti-VEGF if there are significant tractional membranes for fear of exacerbating a detachment. We may perform sectoral PRP in these patients, being careful to avoid the areas of traction, often in preparation for vitrectomy surgery. For PDR patients with macular edema, we almost always start with a series of anti-VEGF injections, as PRP monotherapy carries a substantial risk of worsening the edema. As mentioned before, early vitrectomy is reserved for the more aggressive proliferative cases or cases in which there is dense vitreous hemorrhage.


It is gratifying to treat a patient with severe DR, as the disease incorporates angiogenic, inflammatory, vascular, and mechanical pathologies. We have a variety of approaches available, including 2 distinct classes of intravitreal medical therapy with several commercially available agents in each class, as well as laser therapy and vitrectomy surgery.

Our preferences notwithstanding, one must maintain an individualized approach to the patient with this relatively diverse toolbox. Outside the eye, these patients are not homogenous and may have circumstances dictating a particular treatment algorithm. Patients who are at risk of noncompliance or have an aversion to injections may benefit from the somewhat finite nature of focal and PRP laser. Others may be uncooperative and not conducive to laser delivery, or they might understand and prefer the regression of disease afforded by injection therapy; CLARITY even demonstrated a surprising patient preference for injection therapy over PRP for PDR.

A holistic, patient-centered approach ultimately provides the best outcome. Keep communication lines open while setting expectations early and often, and be willing to change your approach as both the patient and the retinopathy change. Clinical decision making requires a particular blend of anecdotal experience shaped by evidence-based medicine, which is probably the definition of the art of clinical practice. We encourage you to be flexible and evolve with your retina colleagues, as we refine our management of this all too familiar, yet formidable, disease. RP


  1. Statistics About Diabetes. American Diabetes Association. Available at: . Accessed July 29, 2018.
  2. Photocoagulation treatment of proliferative diabetic retinopathy: clinical application of diabetic retinopathy study (DRS) findings. DRS Report 8. The Diabetic Retinopathy Study Research Group. Ophthalmology. 1981;88(7):583-600.
  3. Photocoagulation for diabetic macular edema. Early Treatment Diabetic Retinopathy Study report number 1. Early Treatment Diabetic Retinopathy Study research group. Arch Ophthalmol. 1985;103:1796-1806.
  4. Nguyen QD, Brown DM, Marcus DM, et al; RISE and RIDE Research Group. Ranibizumab for DME: results from 2 phase III randomized trials: RISE and RIDE. Ophthalmology. 2012;119(4):789-801.
  5. Brown DM, Schmidt-Erfurth U, Do CV, et al. Intravitreal aflibercept for diabetic macular edema: 100-week results from the VISTA and VIVID studies. Ophthalmology. 2015;122(10):2044-2052.
  6. Bressler SB, Qin H, Melia M, et al. Exploratory analysis of the effect of intravitreal ranibizumab or triamcinolone on worsening of diabetic retinopathy in a randomized clinical trial. JAMA Ophthalmol. 2013;131(8):1033-1040.
  7. Wells JA, Glassman AR, Ayala AR, et al. Aflibercept, bevacizumab, or ranibizumab for diabetic macular edema: two-year results from a comparative effectiveness randomized clinical trial. Ophthalmology. 2016;123(6):1351-1359.
  8. Boyer DS, Yoon YH, Belfort R Jr., et al; OZURDEX® MEAD Study Group. Three-year, randomized, sham-controlled trial of dexamethasone intravitreal implant in patients with diabetic macular edema. Ophthalmology. 2014 121(10):1904-1914.
  9. Maturi RK, Glassman AR, Liu D, et al; Diabetic Retinopathy Clinical Research Network. Effect of adding dexamethasone to continued ranibizumab treatment in patients with persistent diabetic macular edema: A DRCR Network Phase 2 randomized clinical trial. JAMA Ophthalmol. 2018;136:29-38.
  10. Campochiaro PA, Brown DM, Pearson A, et al; FAME Study Group. Sustained delivery fluocinolone acetonide vitreous inserts provide benefit for at least 3 years in patients with diabetic macular edema. Ophthalmology. 2012;119(10):2125-2132.
  11. Singer M. Treatment burden associated with intravitreal injections in the real world: PALADIN Phase 4 trial with fluocinolone acetonide 0.2 µg/day. Paper presented at: Annual Meeting of the Association of Research in Vision and Ophthalmology (ARVO); April 29-May 3, 2018; Honolulu, HI.
  12. Lai J. Prior steroid response as a predictor of real-world IOP safety with 0.2 µg/day fluocinolone acetonide (FAc) in diabetic macular edema (DME) therapy. Paper presented at: Annual Meeting of the Association of Research in Vision and Ophthalmology (ARVO); April 29-May 3, 2018; Honolulu, HI.
  13. Gross JG, Glassman AR, Jampol LM, et al. Panretinal photocoagulation vs intravitreous ranibizumab for proliferative diabetic retinopathy: a randomized clinical trial. JAMA. 2015;314:2137-2146.
  14. Sivaprasad S, Prevost AT, Vasconcelos JC, et al. Clinical efficacy of intravitreal aflibercept versus panretinal photocoagulation for best corrected visual acuity in patients with proliferative diabetic retinopathy at 52 weeks (CLARITY): a multicentre, single-blinded, randomised, controlled, phase 2b, non-inferiority trial. Lancet. 2017;389:2193-2203.
  15. Eichenbaum D, Quezada-Ruiz C, Hill L, Ghanekar A, Haskova Z. Regression of diabetic retinopathy with ranibizumab in patients with diabetic macular edema and highest-risk nonproliferative diabetic retinopathy. Poster presented at: Annual meeting of the American Academy of Ophthalmology; October 2016; Chicago, IL.