PEER REVIEWED
Landmark Laser Trials Continue to Govern Standard of Care in Diabetic Retinopathy
ABDHISH R. BHAVSAR, MD

Vision loss is one of the many devastating complications of diabetes. People with diabetes are 25 times more likely to go blind than the average person, and each year ~25,000 new cases of diabetes-related blindness occur in the United States. These cases may involve vision loss caused by cataracts, glaucoma, optic neuropathy/papillopathy, or diabetic retinopathy, and most are preventable.

Diabetic retinopathy is the most serious eye condition associated with diabetes. It can be particularly severe in persons who have insulin-dependent diabetes mellitus and also occurs frequently with chronic noninsulin-dependent diabetes mellitus.

About 25% of persons with diabetes have at least some form of diabetic retinopathy, and the incidence increases with the duration of diabetes. At 10 years, the prevalence of retinopathy in patients with diabetes is 7%; after 25 years, it is >90%. Diabetic retinopathy can also accelerate during puberty and pregnancy.

Figure 1. Cumulative event rates of visual acuity less than 5/200 at 2 or more visits for patients in each treatment group. Reprinted with permission.3

NATURAL HISTORY OF DIABETIC RETINOPATHY

Diabetes causes microvascular abnormalities that lead to retinal vascular pericyte loss and endothelial damage, which cause microvascular occlusion and alterations in retinal vascular permeability. The result is often ischemia and retinal edema. The 2 types of diabetic retinopathy, nonproliferative and proliferative, encompass the spectrum of diabetes-related disease that affects the retina.

Nonproliferative retinopathy, also known as background diabetic retinopathy, involves microvascular abnormalities that are apparent during a clinical examination and include dot/blot retinal hemorrhages, cotton wool spots, dilation of retinal veins, venous beading, and attenuation of retinal arteries. Retinopathy also may involve diabetic macular edema in which the macula thickens as blood, fluid, lipids, or exudates leak into the retina as a result of abnormalities in microvascular permeability. Some patients may not notice this edema, while others may notice decreased vision. Diabetic macular edema increases with the duration of diabetes, and the prevalence is 5% within the first 5 years after diagnosis and 15% at 15 years following diagnosis.

Proliferative diabetic retinopathy, which occurs in about 5% of the diabetic population, involves neovascularization of the optic disc, retina, or iris. This harmful neovascularization occurs as a response to retinal ischemia that is caused by diabetes-induced retinal vascular compromise. During the early stages, affected people may not notice any symptoms. However, if the proliferating blood vessels remain untreated, vitreous hemorrhage, retinal detachment due to traction on the retina, or severe neovascular glaucoma can occur and cause severe vision loss and blindness.

The degree of retinopathy seems to be related to the duration of the diabetes and to the patient's blood glucose control. Proliferative retinopathy generally occurs 10 years or more after diagnosis of diabetes. For patients who have been diagnosed with diabetes prior to age 30, the incidence of proliferative diabetic retinopathy is 25% after 15 years and 55% after 20 years. For patients who have been diagnosed with diabetes after the age of 30, the incidence is 20% after 20 years.

	Figure 2. ETDRS treatment assignment schedule for patients with macular edema and mild to moderate diabetic retinopathy in one or both eyes. Reprinted with permission.5

TREATMENT OF DIABETIC RETINOPATHY

Many treatments for diabetic retinopathy are under investigation. However, at this time the only treatments that have been proven effective in large-scale, randomized, controlled clinical trials are several types of laser treatment. The clinical trials that demonstrated the benefits of laser treatment are landmark studies in the history of medicine as they are among the best-designed trials ever conducted. Recommendations from these trials remain clinically relevant today, and they form the basis for the current management of diabetic retinopathy.

THE DIABETIC RETINOPATHY STUDY

The Diabetic Retinopathy Study (DRS) was a randomized, controlled clinical trial that included 1758 patients at 15 clinical centers. Trial enrollment began in 1972 and was concluded in 1975.¹ Eligibility criteria included age <70 years, the presence of proliferative diabetic retinopathy in at least 1 eye or severe nonproliferative retinopathy in both eyes, visual acuity 20/100 or better in both eyes with both eyes suitable for photocoagulation, physician-assessed good survival outlook, and availability for 5 years of follow-up. Exclusion criteria included any prior photocoagulation or pituitary gland ablation.

Study Design of the DRS

One eye of each patient was assigned randomly to immediate photocoagulation, and the other eye was assigned to observation without treatment. The eye that was to receive photocoagulation was randomized to either argon laser treatment or xenon arc treatment.¹ Scatter photocoagulation treatment (panretinal or PRP) was typically completed in 1 or 2 sessions extending to or beyond the vortex vein ampullae. Argon photocoagulation involved the placement of 800 to 1600 500-micron burns or 500 to 1000 1000-micron burns at 0.1 seconds duration, to areas within one disc diameter of the optic disc, including over-neovascularization of the disc (NVD) or retina (NVE). Today, laser over NVD typically is not performed during PRP. Focal laser treatment was applied to microaneurysms causing macular edema (argon: 50- to 100-micron burns, 0.05 to 0.2 seconds; xenon: 1.5º to 3º burn size). Xenon arc photocoagulation was generally applied with a fewer number of burns, 400 to 800, 3º burn size or 200 to 400, 4.5º burn size. Additional treatments were performed at 4-month intervals.^{1, 2}

The primary endpoint, defined as the "event rate," was less than 5/200 visual acuity at 1 or more, 2 or more, and 3 or more consecutive 4-month follow-up visits. The protocol was changed in 1976 when it was discovered that eyes with high-risk characteristics had significant benefit from treatment compared with no treatment such that photocoagulation was to be considered for initially untreated eyes if they had high-risk characteristics.^{1, 3}

Results of the DRS

Of the 1758 patients initially enrolled, 16 were excluded from the study because they were assigned to a treatment group that had combined argon and xenon treatment but was later discontinued . Of the remaining 1742 patients, 867 were randomized to the argon group and 875 to the xenon group.¹ After 2 years of follow-up, the occurrence of visual acuity worse than 5/200 was 15.9% in untreated eyes and 6.4% in treated eyes. After 3 years of follow-up, this event rate was 26.4% in untreated eyes and 10.5% in treated eyes.³

	Figure 3. ETDRS photocoagulation treatment scheme for eyes without macular edema with moderate-to-severe nonproliferative or early proliferative retinopathy. Reprinted with permission.8
	Figure 4. ETDRS photocoagulation treatment scheme for eyes with macular edema and less severe retinopathy. Reprinted with permission.8
	Figure 5. ETDRS photocoagulation treatment scheme for eyes with macular edema and more severe retinopathy. Reprinted with permission.8

With respect to the different treatment groups, the treatment effect for xenon photocoagulation showed a decrease in the event rate of 11.1% after 2 years and 18.5% after 3 years of follow-up. The argon photocoagulation group showed a decrease in the event rate of 7.7% after 2 years and 13.3% after 3 years of follow-up (Figure 1).³

Early losses of visual acuity due to treatment were reported: a decrease of 1 line in visual acuity at 6 weeks after treatment in 15.7% of untreated eyes and 25.9% of treated eyes; a decrease of 1 to 4 lines in 48.4% of xenon group eyes, 35.2% of argon group eyes, and 21% of untreated eyes; a decrease of 2 to 4 lines in 22.8% of xenon group eyes, 9.8% of argon group eyes, and 5.6% of untreated eyes; and a decrease of 5 or more lines in 6.2% of xenon group eyes, 2.2% of argon group eyes, and 2.2% of untreated eyes.³

As a measure of persistent loss of visual acuity due to treatment, those patients who had a 1- to 4-line loss of visual acuity at 6 weeks and also at 1-year posttreatment included 9.3% of argon-treated eyes and 19.1% of xenon-treated eyes. The estimate of eyes with a decrease of 5 or more lines in visual acuity at 6 weeks and at 1 year after treatment was 2.3% in the xenon-treated eyes. No such effect was observed in the argon-treated eyes.³

Visual field changes were measured and showed small losses in visual field in argon-treated eyes and more substantial losses in visual field in xenon-treated eyes. At the 4-month follow-up visit, ~48% of xenon group eyes vs ~90% of eyes in all other groups had visual field scores of 500 degrees or more (visual field score = sum of 12 meridian scores minus the scotomas in each respective meridian).³

The stage of retinopathy at which the benefits of photocoagulation outweighed the risks was identified and defined as high-risk proliferative diabetic retinopathy. High-risk characteristics were defined as:

1. Moderate or severe NVE with vitreous or preretinal hemorrhage or both;

2. Mild NVD with vitreous or preretinal hemorrhage or both;

3. Moderate or severe NVD without vitreous or preretinal hemorrhage or both;

4. Moderate or severe NVD with vitreous or preretinal hemorrhage or both.

The risk of severe visual loss at 2 years was 3.2% for eyes with nonproliferative diabetic retinopathy; 7.0% in eyes with PDR without high-risk characteristics; and 26.2% in eyes with high-risk characteristics.⁴ The group that had the greatest risk of severe visual loss (36.9%) included those eyes with vitreous or preretinal hemorrhage and NVD greater than that in Standard Photograph 10A (1/4 to 1/3 disc area).⁴ In addition, neovascularization seemed to follow shortly after the appearance or worsening of several features of intraretinal lesions termed NV-VH risk factors: extensive retinal hemorrhages, cotton-wool spots, intraretinal microvascular abnormalities (IRMA), venous beading, and arteriolar abnormalities. The risk of severe visual loss at 3 years increased from 11.1% in eyes without these NV-VH characteristics to 36.3% in eyes with 3 or 4 of these characteristics.⁴ Both argon and xenon arc photocoagulation reduced the risk of severe visual loss by >50%.³

Recommendations of the DRS

The study group concluded that photocoagulation reduced the risk of severe visual loss compared with observation and was effective in reducing the progression to more severe stages of proliferative retinopahy.^1,3 Scatter (panretinal) photocoagulation should be performed promptly in eyes that have high-risk characteristics as defined by the Diabetic Retinopathy Study.³

THE EARLY TREATMENT OF DIABETIC RETINOPATHY STUDY

The Early Treatment of Diabetic Retinopathy Study (ETDRS) was a multicenter, randomized clinical trial designed to evaluate photocoagulation and aspirin treatment in patients with early proliferative diabetic retinopathy or nonproliferative diabetic retinopathy.³

The study enrolled 3711 patients in 22 clinical centers between 1980 and 1985 who had early proliferative retinopathy, moderate to severe nonproliferative retinopathy, and/or diabetic macular edema in each eye and who met the following criteria: (1) no macular edema, with visual acuity of 20/40 or better and moderate or severe nonproliferative diabetic retinopathy or early proliferative diabetic retinopathy; or (2) macular edema, visual acuity of 20/200 or better, and mild, moderate, or severe nonproliferative diabetic retinopathy or early proliferative diabetic retinopathy. Patients were excluded if they had high-risk proliferative retinopathy, visual acuity worse than 20/200, or significant ocular disease.⁴

ETDRS Study Design

The study was designed to answer 3 questions: (1) When in the course of diabetic retinopathy is it most effective to initiate panretinal photocoagulation? (2) Is photocoagulation effective in the treatment of diabetic macular edema? (3) Is aspirin treatment effective in altering the course of diabetic retinopathy?

Patients were randomly assigned to either aspirin (650 mg/day) or placebo. One eye of each patient was randomly assigned to early argon laser photocoagulation or to deferral of photocoagulation.⁶ Both eyes were to be examined at least every 4 months.

The first report of the ETDRS was limited to the subgroup of eyes that had mild to moderate nonproliferative diabetic retinopathy and macular edema as determined by the baseline fundus photographs and flourescein angiograms graded at the ETDRS Fundus Photograph Reading Center, Madison, Wisconsin.⁵ Included were 1122 patients with bilateral macular edema (2244 eyes) and 754 patients with unilateral macular edema. Randomization 1 was to immediate photocoagulation of 1 eye and to deferral of photocoagulation in the other eye of patients with bilateral macular edema. For patients with unilateral macular edema, the eyes were divided about equally between immediate and deferral groups. Randomization 2 involved randomizing the immediate group to either combination initial panretinal photocoagulation with subsequent focal macular laser or initial focal laser with panretinal photocoagulation if there was progression to severe nonproliferative retinopathy or beyond (Figure 2).⁵

Macular edema was defined by the reading center as retinal thickening within 1 disc diameter of the center of the macula or definite hard exudates within this area. Clinically significant macular edema (CSME) was defined if at least 1 of the criteria listed in Table 1 was present.⁵

Full scatter photocoagulation treatment was performed with 500-micron spot size, 0.1 seconds duration, moderate intensity, 1200 to 1600 burns, 1/2 burn-width apart >2 disc areas form the fovea out to the equator. Mild scatter photocoagulation was performed with 500-micron spot size, 0.1 seconds duration, moderate intensity, 1 burn-width apart >2 disc areas from the fovea out to the equator.⁷

Microaneurysms between 500 to 3000 microns from the foveal center were treated with 50- to 100-micron burns of 0.05 to 0.1 seconds duration with argon blue-green or green only to achieve whitening around the microaneurysm or whitening or darkening of the microaneurysm itself if >40 microns in diameter. Treatment of lesions closer than 500 microns from the fovea was not required initially, but treatment no closer than 300 microns from the fovea was recommended if retinal edema and leakage persisted. Areas of diffuse leakage or nonperfusion were treated within 2 disc areas of the fovea in a grid pattern with 50- to 200-micron spot size, with 1-burn-width spacing. Since the results were the same in eyes that received aspirin vs no aspirin, the results were presented without regard to aspirin treatment.² More details regarding the concepts of clinically significant macular edema, treatable lesions, and focal treatment parameters are contained in ETDRS Report Number 2.⁷

Patients were divided into 3 different categories with several schemes of photocoagulation within each category. Category 1 included eyes without macular edema. Category 2 included eyes with macular edema and less severe retinopathy. Category 3 included eyes with macular edema and more severe retinopathy (Figures 3, 4, 5).⁸

The primary endpoint for comparing early photocoagulation with deferral of photocoagulation was severe visual loss defined as visual acuity <5/200 at 2 consecutive follow-up visits. The primary endpoint for determining the effects of photocoagulation on macular edema was moderate visual loss defined as loss of 15 or more letters (3 or more lines) of visual acuity from baseline to follow-up.⁸

The protocol was changed after 5 years since focal photocoagulation was effective at reducing moderate visual loss. All eyes initially assigned to the deferral group were then to be treated with focal photocoagulation whenever CSME developed.⁸

Results of the ETDRS

ETDRS Report Number 1 was based on data from 80% of enrolled patients who completed 12 months of follow-up since the initial treatment, and 35% of patients who completed 36 months of follow-up. Eyes that were in the immediate focal photocoagulation group were about half as likely to lose >= 15 ETDRS letters compared with eyes in the deferral group: 5% vs 8% at 1 year, 7% vs 16% at 2 years, and 12% vs 24% at 3 years.³ More data from ETDRS Report Number 9 are presented in Table 2.⁸

With regard to visual acuity improvement, in eyes with baseline 20/40 or worse acuity, improvement in visual acuity of 6 or more letters was more frequent in treated eyes compared with eyes in the deferral group.⁵ With respect to clinically significant macular edema, a clear benefit was observed in eyes that received immediate focal treatment compared with eyes in the deferral group, with statistical significance at the 8-month visit and beyond.⁵ With respect to retinal thickening, 74% of eyes with CSME had retinal thickening involving the fovea as graded by the reading center. At 12 months, 35% of such eyes in the treatment group had foveal thickening compared with 63% of such eyes in the deferral group.⁵

At study closeout, of the 3711 patients enrolled, 706 patient deaths occurred, and 2807 patients (93%) completed a closeout examination. The 5-year rates for severe visual loss were 3.7% in deferral group eyes and 2.6% in early photocoagulation eyes.⁸

For eyes without macular edema, no benefit was noted in preventing moderate or severe visual loss with early photocoagulation (immediate full or mild scatter, with delayed focal if needed for CSME). Eyes that received early full scatter photocoagulation were more likely to have moderate visual loss during the first 2 years than eyes in the deferral group.⁸ For eyes with macular edema and less severe retinopathy, each of the schemes for early photocoagulation gave a reduction of severe visual loss compared with deferral.⁸ For eyes with macular edema and more severe retinopathy, the 5-year risk of severe visual loss in the deferral eyes was 6.5%. This risk was reduced to between 3.8 and 4.7% in the eyes assigned to early photocoagulation.⁸

An increased risk of moderate visual loss was observed at 6 weeks posttreatment in eyes with macular edema and more severe retinopathy compared with deferral eyes. However, after the first year, all the treatment arms for early photocoagulation had a lower risk of moderate visual loss. The least visual loss was found in the immediate mild scatter combined with the immediate focal photocoagulation group.⁸

Visual fields and color vision were measured. In the eyes assigned to deferral, there was no difference in the visual field between baseline and the 4-month visit. In all eyes assigned to immediate full scatter photocoagulation, there was a greater loss of visual field compared with eyes in the deferral groups and an intermediate loss of visual field in eyes assigned to mild scatter photocoagulation. No significant visual field loss was observed in eyes with macular edema and less severe retinopathy that received immediate focal and delayed scatter photocoagulation.There was less loss of color vision by FM-100 testing at the 4-year visit in eyes with macular edema and less severe retinopathy assigned to immediate focal with delayed scatter photocoagulation compared with deferral.⁸ Photocoagulation strategies that involved immediate full scatter photocoagulation reduced the rate of developing high-risk proliferative retinopathy by ~50%, whereas those strategies that involved mild scatter reduced the rate by ~25%.⁸

Recommendation of the ETDRS

Early photocoagulation reduces the risk of severe visual loss and progression of retinopathy and the need for vitrectomy. However, the side effects of decreased visual field and central visual acuity should be considered in determining when to initiate panretinal photocoagulation. The ETDRS concluded: "Focal photocoagulation should be considered for all eyes with clinically significant macular edema. Focal photocoagulation reduced the risk of moderate visual loss, increased the chance of visual improvement, lessened loss of color vision, and was associated with only minor losses of visual field."⁸

CONCLUSIONS

The DRS demonstrated that scatter (panretinal) photocoagulation reduced the risk of severe visual loss by >50% in eyes with high-risk proliferative diabetic retinopathy. The ETDRS demonstrated that immediate focal photocoagulation for clinically significant macular edema reduced the risk of visual loss and reduced macular thickening by ~50%.

Although many potentially promising treatments for diabetic retinopathy are under investigation, including sub-Tenon's and intravitreal steroids, other steroid-like substances, anti-VEGF aptamers and antibodies, hyaluronidase and other substances to dissolve vitreous hemorrhage, and pars plana vitrectomy and ILM dissection, for now we are going back to the future for laser treatment.

From the Department of Ophthalmology, Phillips Eye Institute, and the Retina Center, P.A., Minneapolis, Minn. Dr. Bhavsar serves as state chair of the Minnesota Diabetes Eye Exam Initiative.

Table 2. Occurrence of moderate visual loss by visit. Reprinted with permission.8

REFERENCES

1. The Diabetic Retinopathy Study Research Group. Preliminary report on the effect of photocoagulation therapy. Am J Ophthalmol. 1976;81:383-396.

2. The Diabetic Retinopathy Study Research Group. Photocoagulation treatment of proliferative diabetic retinopathy. Clinical application of Diabetic Retinopathy Study (DRS) Findings. DRS Report Number 8. Ophthalmol. 1981;88:583-600.

3. The Diabetic Retinopathy Study Research Group. Photocoagulation treatment of proliferative diabetic retinopathy. DRS Report Number 2.Ophthalmol. 1978; 85:82-106.

4. The Diabetic Retinopathy Study Research Group. Indications for photocoagulation treatment of diabetic retinopathy. DRS Report Number 14. Int Ophthalmol Clin. 1987;27:239-253.

5. Early Treatment Diabetic Retinopathy Study Research Group. Photocoagulation for Diabetic Macular Edema: ETDRS Report Number 1. Arch Ophthalmol. 1985; 103;1796 -1806.

6. Early Treatment Diabetic Retinopathy Study Research Group. Early treatment diabetic retinopathy study design and baseline patient characteristics. EDTRS Report Number 7. Ophthalmol. 1991;98:741-756.

7. Early Treatment Diabetic Retinopathy Study Research Group. Treatment techniques and clinical guidelines for photocoagulation of diabetic macular edema: ETDRS Report Number 2. Ophthalmol. 1987;94;761-774.

8. Early Treatment Diabetic Retinopathy Study Research Group. Early photocoagulation for diabetic retinopathy: ETDRS Report Number 9. Ophthalmol. 1991; 98:766-785.