Treatment and Prevention of Diabetic Retinopathy

Diabetic retinopathy is the most frequent cause of new cases of blindness among adults aged 20-74 years in industrialized countries, causing blindness in more than 12,000 to 24,000 people each year in the US alone. The prevalence of retinopathy is strongly related to the duration of diabetes. During the first 20 years of disease, nearly all patients with type 1 and more than 60% of patients with type 2 diabetes develop retinopathy.[1] Up to 21% of patients with type 2 diabetes have retinopathy when first diagnosed.[1] The most common factors contributing to the development of retinopathy include duration of disease,[1] hyperglycemia, hypertension, and possibly dyslipidemia. Medical and surgical treatments are currently available to help the clinician treat retinopathy, and more are in development.

Intensive metabolic control

The first step in the treatment and prevention of diabetic retinopathy is to achieve and maintain optimal metabolic control. Clear-cut evidence supports the benefits of glycemic and blood pressure control in the primary prevention of diabetic retinopathy.[2] The benefits of lipid-lowering therapy in primary prevention and treatment of diabetic retinopathy are less certain, but may be important if hard exudates are present.[3]

Intensive glycemic control in patients with type 1 diabetes reduced the mean risk of retinopathy by 76%, and the risk of progression by 54% in the landmark DCCT trial.[1] The protective effect of intensive glycemic control has also been confirmed for patients with type 2 diabetes. In the UKPDS, the risk of developing retinopathy was 1.4% for those with A1C of 6% to 2-7.4%, but 2.5% for those with A1C ≥7.5%.[4] The risk of progression was even higher than the risk for initially developing retinopathy in those with greater degrees of hyperglycemia, 4.1% for those with A1C of 6.2% to 7.4%, but 8.1% for those with A1C ≥7.5%.[4]

Systolic blood pressure was significantly associated with retinopathy incidence, but not progression.[4] The relative risk of developing retinopathy in patients with systolic readings of 125-139 mm Hg was 1.5, but 2.8 for those with pressures ≥140 mm Hg.[4] Some evidence suggests ACE inhibitors or ARBs may be especially beneficial in the treatment of retinopathy beyond the benefits obtained by lowering blood pressure.[5],[6] The ongoing DIRECT trial will attempt to differentiate the specific benefits of angiotensin-II receptor blockade with candesartan apart from the general benefits of blood pressure reduction.[6]

Multiple observational studies support a relationship between elevated triglycerides and/or elevated LDL-C with the presence or number of hard exudates in diabetic macular edema (DME).[3] The severity of retinopathy was positively associated with triglycerides and negatively associated with HDL cholesterol in a population of male and female patients with type 1 diabetes from the DCCT/EDIC cohort.[7] Small studies and case reports support aggressive lipid control to reverse hard exudates.[8],[9] The ongoing ASPEN study will examine the effects of atorvastatin therapy on ocular endpoints.[10]

Other clinical trials describing the benefits of intensive glycemic and blood pressure control in diabetic retinopathy can be found in the review articles by Porta and Allione,[11] and Sjolie and Moller.[10] Transient worsening of retinopathy may be noted with rapid improvement of glycemic control.[11]

Additional information regarding current guidelines for control of glycemia, blood pressure, and lipids can be found at the following web pages:

Treatment of Type 1 Diabetes
Treatment of Type 2 Diabetes
Treatment and Prevention of Diabetic Neuropathy
Treatment and Prevention of Diabetic Nephropathy

Laser photocoagulation therapy

Laser photocoagulation techniques can be classified as scatter, focal, or grid.[12] Scatter photocoagulation is also called panretinal photocoagulation, and is used to treat proliferative diabetic retinopathy (Table 1).[12] Scatter photocoagulation indirectly treats neovascularization on the retina, optic nerve, or anterior chamber angle by placing laser burns throughout the peripheral fundus.[12] Focal and grid photocoagulation use smaller laser burns than scatter photocoagulation, and are used to treat diabetic macular edema (DME).[12] Focal photocoagulation is used to treat leaking microaneurysms; grid photocoagulation is applied in a pattern similar to scatter photocoagulation and is used to treat areas of edema caused by diffuse capillary leakage or nonperfusion.[12]

Several laser photocoagulation procedures have proven benefits for reducing the progression of diabetic retinopathy, and in some cases, improving visual acuity. Strong evidence of benefit has been found for macular photocoagulation in clinically significant macular edema (CSME), peripheral retinal laser photocoagulation in moderate-severe nonproliferative retinopathy and DME, and peripheral retinal laser photocoagulation in proliferative diabetic retinopathy (PDR).[2] Grid photocoagulation to zones of retinal thickening in DME is likely to be beneficial, but the efficacy of laser treatment for non-CSME and nonproliferative retinopathy without DME is not known.[2]

Evidence from randomized controlled trials[2] and practical experience[13] suggests that stabilization of retinopathy progression is a more typical outcome of successful laser photocoagulation therapy than improvement in visual acuity. Patients may require multiple treatments to resolve DME or PDR.[13] Vision may continue to worsen in the long term despite successful laser treatment.[13]

The Early Treatment of Diabetic Retinopathy Study (ETDRS), a randomized controlled trial of early photocoagulation in 3711 patients with type 1 and type 2 diabetes, found that patients with type 2 diabetes were more likely to benefit from early scatter photocoagulation than patients with type 1 diabetes.[14] The patients most likely to benefit from early photocoagulation had severe NPDR or early PDR. If patients can be closely monitored for the development of high-risk retinopathy, the ETDRS data do not demonstrate that early photocoagulation will reduce the risk of severe vision loss in patients with type 1 diabetes.[14] The need for scatter photocoagulation in long-term followup studies of the ETDRS cohort has been high.[15] Mortality in this cohort has also been high: 61% of patients were deceased at 13-19.5 years after their initial laser photocoagulation.[15]

Side effects and complications of scatter photocoagulation usually include peripheral vision loss with poor night vision, and occasionally include some central vision loss.[12],[13] Occasionally, vitreous hemorrhage may occur during treatment.[12] Side effects of focal photocoagulation may cause a transient decrease in central vision; focal photocoagulation may also alter color vision or create blind spots just outside the fovea.[12],[13] Rarely, subretinal fibrosis with choroidal neovascularization may occur, causing permanent central vision loss.[12]

Table 1. Indications for laser photocoagulation therapy[12]

Vitrectomy

Despite preventive regimens such as intensive metabolic control and treatments such as laser photocoagulation, a substantial number of patients go on to develop complications of progressive PDR and may become candidates for vitrectomy (Table 2 ).[16], [17] The objectives of vitrectomy for severe PDR are the following:[16]

• Remove axial opacities

• Relieve retinal traction

• Segment or peel epiretinal membrane

• Effect hemostasis

• Treat all retinal breaks

• Deliver laser treatment

Vitrectomy can safely clear medial opacities and relieve retinal traction in most cases.[17] Clinical evidence suggests vitrectomy is likely to be beneficial in severe vitreous hemorrhage and proliferative retinopathy, if performed early.[2] The efficacy of vitrectomy for DME is unknown.[2] Outcomes for this complex surgical procedure vary by complication,[16] but retrospective studies suggest roughly half of the procedures result in improvements in functional vision.[18],[19],[20] Retinal ischemia appears to be the chief obstacle preventing better functional results, but no current surgical approach effectively reperfuses occluded retinal capillaries.[17]

While improvements in technique over the past several decades have lowered the threshold for performing surgery,[17] vitrectomy has potentially serious complications, including retinal detachment, recurring vitreous hemorrhage, rubeosis iridis, severe visual loss and eye pain.[12]

Table 2. Indications for pars plana vitrectomy[12]

Emerging medical treatments

Given the limitations of current medical treatments and the drawbacks associated with laser and other surgical treatments, there is a great deal of interest in increasing the number of effective medical treatments for diabetic retinopathy. Current avenues of investigation include intravitreal treatments, new indications for currently marketed drugs, and investigational new drugs.

Intravitreal injections or implants may permit local treatment of the vitreous and/or retina, and are listed in Table 3. Inflammation appears to be one of the physiological processes underlying tissue damage associated with diabetes. The corticosteroids are anti-inflammatory treatments, but long-term systemic treatment with this drug class is associated with serious side effects. Local corticosteroid injections may provide ocular benefits while reducing the risk of systemic side effects. Several clinical studies have demonstrated that ocular vascular endothelial growth factor (VEGF) concentration elevations lead to retinal neovascularization, and that these levels decrease following adequate laser photocoagulation.[21] Treatment with pegaptanib may slow or halt neovascularization by blocking the action of VEGF without laser photocoagulation. Abnormal connective tissue can develop in the eye following vitreous hemorrhage. Treatment with hyaluronidase may dissolve proteoglycans, a component of connective tissue. Hyaluronidase may also clear vitreous hemorrhage.[22] In May, 2004, hyaluronidase for injection was approved as an adjunct to other injected drugs to increase their absorption and dispersion.[23]

Table 3. Investigational intravitreal treatments for diabetic retinopathy in Phase II or III clinical trials[24]

Notes: VEGF = vascular endothelial growth factor

Certain drugs currently marketed for other indications are being tested for their effects on DME and PDR, as shown in Table 4. The beneficial effects of lowering blood pressure and cholesterol levels in preventing DME have been mentioned previously. Current clinical trials of antihypertensive medications and statins in DR and DME are attempting to determine if there are unique ocular benefits of these medications independent of their established cardiovascular benefits. It has been observed that cyclooxygenase-2 (COX-2) is upregulated in early diabetic retinopathy, and that COX-2 inhibition may indirectly block VEGF production in experimental models.[28] A trial of celecoxib will prospectively assess the effects of a COX-2 inhibitor on DME. Prior to the availability of laser photocoagulation, hypophysectomy was used to treat PDR.[29] While morbidity and mortality associated with this surgery were high, it was effective for halting neovascularization.[29] The current multicenter trial of octreotide, a somatostatin analog, will investigate whether this agent can prevent NPDR from progressing to PDR.[29]

Table 4. Systemic treatments for diabetic retinopathy in Phase II or III clinical trials–new indications of currently marketed drugs

Notes: ACE = angiotensin converting enzyme, ARB = angiotensin-II receptor blocker, COX-2 = cyclooxygenase-2

A somatostatin receptor analog, BIM 23190, is being tested for its effects on PDR (Table 5). There are also several treatments in development targeting various metabolic pathways associated with the development of multiple diabetic microvascular complications. Among these are the aldose reductase pathway and the protein kinase C pathway.[31] Epalrestat, an aldose reductase inhibitor (ARI) is marketed as a treatment for diabetic peripheral neuropathy in Japan, but not in the US or Europe. Clinical trials for epalrestat in diabetic retinopathy are underway but have not been reported. Ruboxistaurin mesylate, a protein kinase C (PKC) β inhibitor, shows significant reduction in the progression of DME for patients with A1C ≤10% (~25% on 32 mg ruboxistaurin vs ~40% on placebo after 3 yrs).[32] A nonsignificant trend toward reduced progression of PDR or photocoagulation (~43% on 32 mg ruboxistaurin vs ~47% on placebo at 3 yrs) has also been observed.[33]

Additional information regarding ARIs and PKC β inhibitors for the treatment of diabetic peripheral neuropathy can be found at the following web page:

Treatment and Prevention of Diabetic Neuropathy

Table 5. Systemic treatments for diabetic retinopathy in Phase II or III clinical trials–investigational new drugs[24]

Notes: ARI = aldose reductase inhibitor, PKC = protein kinase C

References