Diabetic Retinopathy

Numerous hypotheses have been proposed regarding the precise mechanisms of action responsible for laser’s benefits in both PRP and focal/grid photocoagulation. Our current understanding of the mechanisms leading to retinal neovascularization and macular edema is that hypoxia of the inner retina leads to increased production of proangiogenic cytokines such as VEGF that also lead to increased vascular permeability. Budzynski and associates showed that average PO ₂ across the inner 50% of the retina was higher in photocoagulated feline retinas than in untreated eyes. Increased oxygenation of the inner retina after photocoagulation may decrease the stimulus for VEGF production, thus decreasing neovascularization and vascular permeability that leads to macular edema. Alternatively, laser may directly damage or destroy the cells that produce proangiogenic cytokines. In cases with areas of focal leakage from microaneurysms, direct closure also may be important. Finally, effects on retinal pigment epithelial function also may be significant. Laser likely changes the microenvironment in important ways, such as cytokine levels.

Thermal laser first achieved widespread popularity and validity in ophthalmology in 1976 when the Diabetic Retinopathy Study published conclusive evidence that peripheral scatter or panretinal photocoagulation (PRP) reduced the risk of severe visual acuity loss in eyes with PDR by 50% or more. Side effects included decreased peripheral visual field and dark adaptation, plus occasional loss of a few lines of Snellen visual acuity, presumed to be from exacerbation of macular edema. The Diabetic Retinopathy Study Group recommended treatment for eyes with certain high-risk characteristics. Today, argon green laser PRP is still widely used and remains a common practice for treatment of not only PDR, but also anterior or posterior segment neovascularization resulting from other causes, such as central retinal vein occlusion (CRVO), branch retinal vein occlusion (BRVO), sickle cell retinopathy, and the ocular ischemic syndrome. However, it is plausible that pharmacotherapies could supplant laser in the future, because PRP does destroy functioning retinal neurons.

In 1985, another therapeutic indication for thermal laser emerged when the Early Treatment Diabetic Retinopathy Study showed that focal/grid photocoagulation of DME reduced the 3-year risk of losing 3 lines or more of vision by approximately 50%, and for eyes with edema involving the center of the macula and visual acuity loss, this treatment also increased the chance of visual improvement. The technique of focal/grid photocoagulation involves 2 strategies within areas of retinal thickening: (1) direct treatment to microaneurysms within 2 disc diameters of the center of the fovea, and (2) scatter or grid treatment approximately 2 burn widths apart to other areas of thickening. Over the years, a modified version of the Early Treatment Diabetic Retinopathy Study laser protocol has been used more commonly, employing less intense burns, which seems to result in outcomes similar to those seen in the Early Treatment Diabetic Retinopathy Study, whereas grid treatment alone does not seem to be as effective and certainly has not been shown to be superior to focal/grid treatment.

Focal/grid laser therapy continued as the primary treatment method for DME with vision impairment involving the center of the macula until very recently. Potential challengers such as preservative-free intravitreal triamcinolone (IVTA) attempted to displace photocoagulation. It was presumed that by inhibiting expression of VEGF as well as other proinflammatory cytokines, corticosteroids could reduce capillary permeability, thus decreasing macular edema, and preliminary studies with short follow-up suggested that this treatment was superior to laser. However, in 2008, the Diabetic Retinopathy Clinical Research Network ( DRCR.net ) published data from a multicenter, randomized, clinical trial of 840 eyes that compared laser with 2 doses of IVTA. Despite early results favoring 4 mg IVTA at 4 months, IVTA was not shown to be superior to focal/grid photocoagulation, and laser seemed to yield better mean visual acuity results at 2 years than 1- and 4-mg doses of IVTA. Optical coherence tomography (OCT) results mirrored the visual acuity results. As expected, rates of intraocular pressure elevation (IOP) and cataract formation were higher in the triamcinolone groups than in the laser groups. Focal/grid photocoagulation thus retained its position as the usual practice for treatment of DME.

Although one third of eyes with DME and visual impairment treated with focal/grid photocoagulation see improvement in visual acuity of at least 2 lines, approximately 20% of eyes worsen by at least 2 lines. In addition, photocoagulation does come with inherent risks and side effects. Burns close to the fovea may damage vision acutely or may enlarge over time, leading to atrophic retina and retinal pigment epithelium encroaching on fixation with visual acuity loss. Occasionally, choroidal neovascularization may develop and may create permanent central scotomas with substantial vision loss.

Recently, researchers have reported more efficacious and potentially safer treatments for DME. In June 2010, the DRCR.net published convincing evidence that the anti-VEGF agent ranibizumab (Lucentis; Genentech, South San Francisco, California, USA) with or without prompt laser, leads to superior visual acuity outcomes compared with laser alone. The rationale for exploring anti-VEGF agents as a treatment method for DME stems from the knowledge that VEGF increases vascular permeability. Thus, anti-VEGF agents theoretically could block VEGF from activating its receptors, prevent increased vessel permeability, thus decreasing DME.

The DRCR.net study was a multicenter, randomized clinical trial of 854 eyes with DME involving the center of the macula with vision impairment. At 1 year, the groups assigned to ranibizumab with prompt or deferred (> 24 weeks) focal/grid laser had nearly identical outcomes, gaining an average of 2 lines of vision, with approximately 50% gaining 2 lines or more of visual acuity and less than 5% losing 2 lines or more. Prompt focal/grid photocoagulation yielded no additional visual acuity benefit when compared with the deferred laser group, in which approximately 60% of the study participants did not receive a laser treatment within the 2 years. Results in a triamcinolone plus prompt laser group were not superior to those of the laser alone group. Also noteworthy was the observation that eyes assigned to either ranibizumab group or the triamcinolone group were less likely than the laser control group to have progression of diabetic retinopathy or vitreous hemorrhage or to require PRP for PDR, suggesting an additional benefit for anti-VEGF agents and triamcinolone (other than a decrease in macular edema).

These recently published results have changed many retina specialists’ management approach for DME. Anti-VEGF injections not only seem to provide superior visual acuity results, but also are faster acting and potentially safer than focal/grid photocoagulation (except for the risk of rare endophthalmitis). Also, whereas fluorescein angiography was necessary frequently to facilitate or guide placement of focal/grid laser treatment, intravitreal anti-VEGF therapy often can be used based on clinical examination and OCT findings, obviating the need for fluorescein angiograms. The often rapid decrease in retinal thickness and accompanying vision improvement seen with anti-VEGF injections provide an incentive for patients to return for treatment, potentially improving compliance and efficacy of the therapy. Also, anti-VEGF agents do not pose the dangers of cataract formation and elevation of IOP seen with intravitreal steroids. The era of solitary focal/grid photocoagulation being the most effective treatment for DME has come to a close.

Although the treatment approach for applying anti-VEGF therapy in the DRCR.net study may seem complex, the underlying rationale is fairly straightforward. After initiating treatment, it seems reasonable to continue to treat at monthly intervals until one is relatively certain that macular thickness is no longer improving (for example, 2 consecutive OCT measurements showing no further thinning of the macula). After retreatment is not given, injections would be resumed if edema recurs or worsens. If it does not recur or worsen after withholding treatment for at least 1 month, follow-up injections might be extended to 2 months and then 4 months, unless edema recurs or worsens to warrant resumption of monthly intravitreal injections, which are used again until improvement no longer occurs. Focal/grid laser as often as every 4 months can be added if edema persists and no longer improves despite multiple monthly injections of intravitreal anti-VEGF drugs.

The DRCR.net study showed that such an approach can lead to improvement in visual acuity, on average, without loss if or when injections are withheld, as long as injections are resumed when edema recurs or worsens, with a median of 6 injections within the first 6 months, followed by a median of 2 to 3 injections in the second 6 months and a median of another 2 to 3 injections in the second year. This potential avoidance of monthly injections indefinitely to maintain improvements noted through at least 6 months is different from the use of anti-VEGF drugs for neovascular AMD (see below), where withholding treatment when thickening no longer is improving and resuming treatment when thickening recurs or worsens does not maintain the maximum visual acuity improvements obtained after the first few months of monthly treatments.

For some patients with DME, however, focal laser may still be a good treatment option. Patients who cannot tolerate or comply with the potential need for monthly visits and injections may be better suited to laser treatments, rather than no treatment, given their potentially longer-lasting effect and extended interval between visits of up to 4 months. At present, there is little evidence to suggest that laser therapy in combination with anti-VEGF injections will decrease the injection treatment burden substantially, but further follow-up in a DRCR.net study is addressing this. Because it remains unclear if there is a small increased risk of arterial thromboembolic events with intravitreal anti-VEGF agents (current studies confidently rule out a moderate or large risk), some practitioners are hesitant to use these agents in patients with, or at risk for, cardiovascular or cerebrovascular disease. In addition, intravitreal injections come with their own risks, including rare but devastating endophthalmitis.

Although some investigators have reported contraction of fibrovascular tissue after anti-VEGF injection in eyes with proliferative disease about to undergo vitrectomy, presumably increasing the risk of tractional retinal detachment, rhegmatogenous detachment, hemorrhage, or a combination thereof, there is no evidence to suggest that an eye with PDR and DME or an eye that has been treated for PDR and in which DME develops has an increased risk of tractional retinal detachment when undergoing intravitreal ranibizumab for DME. At this time, neither IVTA nor ranibizumab are approved by the United States Food and Drug Administration approved for use for DME, and thus are used in an off-label fashion in the United States, although ranibizumab for this indication has been approved in Europe by the European Medical Evaluation Agency. Some retina practices also have used bevacizumab (Avastin; Genentech), rather than ranibizumab, as their anti-VEGF treatment of choice for DME because of its lower price, and evidence, albeit less definitive, of its effectiveness.

Age-Related Macular Degeneration

Although diabetic eye disease is the leading cause of vision loss among young to middle-aged individuals, neovascular AMD, left untreated, is the most frequent cause of irreversible severe vision loss among older people in the developed world. Before the Anti-VEGF Antibody for the Treatment of Predominantly Classic Choroidal Neovascularization in Age-Related Macular Degeneration and Minimally Classic/Occult Trial of the Anti-VEGF Antibody Ranibizumab in the Treatment of Neovascular Age-Related Macular Degeneration trials, exudative AMD was treated with thermal laser or PDT. Thermal laser photocoagulation works by destroying abnormal blood vessels. The Macular Photocoagulation Study reported that laser photocoagulation for extrafoveal, juxtafoveal, and even subfoveal CNV resulting from AMD decreased the risk of severe visual loss. Drawbacks to treatment of subfoveal CNV (the most common group) were substantial and included immediate loss of visual acuity as a result of collateral damage to overlying neurosensory retina and the retinal pigment epithelium, the relatively small number of eligible lesions (small with substantial vision loss), the absence of visual acuity improvement, as well as frequent recurrences.

In 2000, cold laser treatment in combination with verteporfin gained acceptance as a therapy for neovascular AMD. PDT is believed selectively to target the photosensitizing compound only within choroidal neovascular vessels, hence leading to much less collateral damage to overlying retina and RPE. The Treatment of Age-related Macular Degeneration with Photodynamic Therapy Study Group showed that for classic CNV lesions, a higher percentage of verteporfin-treated eyes than placebo lost fewer than 15 letters of visual acuity at 12 months, although few eyes had substantial improvement. Fluorescein angiographic studies also showed that PDT with verteporfin reduced classic lesion growth, leakage, and progression. The Verteporfin in Photodynamic Therapy Study Group shortly thereafter investigated verteporfin therapy for subfoveal lesions with occult but no classic CNV and evidence of recent disease progression. Although the rate of 15-letter visual acuity loss was similar among treatment and placebo groups through 12 months, by 24 months, 55% of verteporfin-treated eyes versus 68% percent of placebo eyes had lost at least 15 letters. The benefits seemed greatest for those with smaller lesions and lower initial visual acuity levels. In the United States, PDT with verteporfin was approved by the Food and Drug Administration only for predominantly classic lesions.

Although PDT of subfoveal CNV showed better visual acuity results than thermal laser and slowed the progression of vision loss, most patients still continued to lose vision, and very few actually gained vision. Hence, the enthusiasm in 2006 when the Minimally Classic/Occult Trial of the Anti-VEGF Antibody Ranibizumab in the Treatment of Neovascular Age-Related Macular Degeneration and Anti-VEGF Antibody for the Treatment of Predominantly Classic Choroidal Neovascularization in Age-Related Macular Degeneration trials showed that almost 95% of patients treated monthly for 2 years with intravitreal ranibizumab avoided vision loss of 15 letters and more than one third of the patients in these 2 studies gained at least 3 lines of vision. These 2 studies were groundbreaking in proving for the first time that a treatment for neovascular AMD could result in visual acuity gain for at least one third of treated patients.

Since 2006, the use of anti-VEGF agents has increased markedly, and neovascular AMD patients have benefited, maintaining and often recovering vision in a disease that previously often led to inevitable severe visual loss. In contrast, thermal laser is almost obsolete in the treatment of AMD, and PDT’s use has dwindled dramatically, with some practices not even owning the laser. That being said, thermal laser and PDT may still be beneficial in particular AMD scenarios. Patients with CNV far from the fovea (a rare event) should respond well to treatment with thermal laser without the risk, expense, or time investment of repeated injections.

In addition, we have all treated patients whose disease process responds minimally if at all to anti-VEGF therapy. PDT may be beneficial in some of these cases. Indocyanine green angiography may reveal the presence of polyps under the retina, leading to the diagnosis of a pattern of CNV termed idiopathic polypoidal choroidal vasculopathy . Authors have shown that idiopathic polypoidal choroidal vasculopathy often responds poorly to anti-VEGF monotherapy. Despite decreasing exudation and retinal thickness, anti-VEGF therapy largely is ineffective in actually eliminating the polyps. PDT, however, seems to eradicate the actual root of the disease more successfully. Experience in an Asian population has suggested that this pattern of CNV may respond with an average visual acuity gain that is similar with either anti-VEGF therapy or PDT. The use of reduced-fluence PDT, combined with the fact that the polyps often are peripapillary or extrafoveal, rather than subfoveal, has decreased the apprehension of many retina specialists to treat these lesions with PDT. Trials are ongoing to investigate whether various combinations of anti-VEGF agents, PDT, and intravitreal steroids can yield visual acuity outcomes that are the same or better than those seen with anti-VEGF monotherapy, and with fewer treatments. So for now, hold onto those PDT lasers!