There are two facets of diabetes: diabetic macular edema (DME) and proliferative diabetic retinopathy (PDR).

DME can be treated medically with steroids (triamcinolone, dexamethasone, fluocinolone, etc.) and anti-VEGFs (both FDA and non-FDA approved), including ranibizumab (Lucentis, Genentech), bevacizumab (Avastin, Genentech), and aflibercept (Eylea, Regeneron). It can also be treated with focal laser.

For PDR, we have medical treatment, which consists mostly of anti-VEGFs, and laser treatment, consisting of panretinal photocoagulation (PRP).

DRCR PROTOCOLS

Let’s take a quick look at the Diabetic Retinopathy Clinical Research (DRCR) network protocols.

• Protocol I compared ranibizumab and focal laser (either immediate or delayed) treatment versus focal laser, and steroids versus just focal laser alone for DME. Ranibizumab maintained the letter gain over the other groups.
• Protocol S was slightly more controversial, but it pitted ranibizumab versus PRP for PDR. Ranibizumab was noninferior, so it at least gave us another weapon to use for PDR other than laser, perhaps with less visual field loss.
• Protocol T was a comparison of all the anti-VEGFs—ranibizumab, bevacizumab, and aflibercept. At 1 year, aflibercept was better for vision gain, if the vision was 20/50 or worse, but that difference disappeared at 2 years between aflibercept and ranibizumab. Bevacizumab never caught up.
• Protocol U had dexamethasone and ranibizumab versus ranibizumab alone, with the results showing no difference in the visual acuity, but a better central macular thickness with the dexamethasone and ranibizumab group.
• Protocol V examined cases of DME with 20/25 vision or better. It showed no difference between aflibercept versus observation versus focal laser. However, if the patient began to deteriorate with only observation or laser, they were rescued with a course of aflibercept. This is a protocol I use regularly if the patient’s vision is good and they have very minimal swelling. I use the scare tactic that, if they don't get their diabetes under control within the next 3 months, they will require injections or laser.
• Protocol AB compared aflibercept versus pars plana vitrectomy (PPV) for vitreous hemorrhage. Visual outcomes were similar, with PPV having a quicker visual recovery. However, about a third of each group required the opposite therapy.

CASE 1: THINK PROTOCOL S

A 67-year-old Black male with diabetes for 15 years, and an A1C between 7 and 9 over the years, presented with 20/50 OD and 20/40 OS, with clinically significant swelling and neovascularization. He had received approximately 14 to 16 ranibizumab injections in both eyes previously. I gave him 1 ranibizumab injection and then did PRP. Nine months later, he is doing great—ocularly and logistically—with no more injections.

CASE 2: MODIFIED PROTOCOL U

A 43-year-old Caucasian male with diabetes for 4 years, and an A1C of 7.5, had acuities of 20/50 OD and 20/20 OS (Figure 1). He was given 1 dose of bevacizumab (I usually give 3 doses). But his vision got worse, so he was given dexamethasone. Things got better and then worse again, so we alternated between aflibercept and dexamethasone at 6-week intervals.

He has since received a fluocinolone implant in the hope that it will keep the inflammatory component at bay, while we treat with anti-VEGF, extending the intervals between injections.