Clinical Trials in Retina

KAREN GEHRS AND SHARAM DANESH

Choroidal neovascularization (CNV) is a significat cause of severe visual loss in patients with age-related macular degeneration (AMD) and ocular histoplasmosis syndrome (OHS). The Macular Photocoagulation Study (MPS) was a randomized, controlled clinical trial designed to determine whether laser photocoagulation of CNV prevents severe visual loss (SVL). Several clinical trials were conducted. These included (1) extrafoveal CNV in patients with AMD, OHS and idiopathic CNV (ICNV); (2) juxtafoveal CNV in patients with AMD, OHS, and ICNV; and (3) new and recurrent subfoveal CNV in patients with AMD only.

Three trials investigated the role of laser treatment to extrafoveal CNV in patients with AMD, OHS, and ICNV. To be eligible for these trials, patients were required to have angiographic evidence of a choroidal neovascular membrane (CNV) with the foveal edge 200 to 2500 from the center of the foveal avascular zone (FAZ). The visual acuity was 20/100 or better in the study eye. Eyes with peripapillary membrane were included, if the laser treatment spared at least 1.5 clock hours of peripapillary nerve fiber layer adjacent to the temporal half of the disc. Patients enrolled in the AMD trial had evidence of drusen. Patients in the OHS trial had at least one atrophic scar (histo spot). Patients in the ICNV trial had no drusen, histo spots, or other clinical findings that could account for the CNV.

Eligible eyes were randomly assigned to argonlaser photocoagulation or no treatment. Argon blue-green laser was used to obliterate the entire neovascular complex, including any areas of blood or blocked hyperfluorescence. Treatment resulted in uniform whitening of the overlying retina. Treated patients were seen at 3 to 9 weeks to assess the adequacy of treatment. Treated patients were encouraged to monitor their vision with an Amsler grid and report any new distortion or enlargement of scotoma. Additional treatment was promptly applied when persistence or recurrence of a treated membrane was found.

The primary outcome measure for the MPS was occurrence of SVL, defined as a loss of six or more lines of visual acuity (equivalent to a quadrupling of the visual angle).

Three trials investigated the role of krypton laser treatment to juxtafoveal CNV in patients with AMD, OHS, and ICNV. Krypton-red laser was used for treatment within the FAZ because the red wavelength is not absorbed by the xanthophyll in the inner retina. To be eligible for these trials, patients were required to have angiographic evidence of neovascularization with foveal edges between 1 and 199 μ from the center of the FAZ or between 200 and 2500 μ from the center if associated blood or blocked fluorescence extended within 200 μ of the FAZ center. These lesions could have blood or blocked fluorescence due to pigment that extended through the entire FAZ. The visual acuity was 20/400 or better in the study eye. Eyes with peripapillary membranes were included if the laser treatment spared at least 1.5 clock hours of peripapillary nerve-fiber layer adjacent to the temporal half of the disc. Patients enrolled in the AMD trial had evidence of drusen. Patients in the OHS trial had at least one atrophic scar (histo spot). Patients in the ICNV trial had no drusen, histo spots, or other clinical findings that could account for the CNVM.

Eligible eyes were randomly assigned to krypton-laser photocoagulation or no treatment. Treatment required a fluorescein angiogram taken within 72 hours and retrobulbar anesthesia. Krypton-red laser was used to obliterate the area of hyperfluorescence and obtain uniform whitening of the overlying retina. Unlike the extrafoveal trials, treatment of blood or blocked fluorescence associated with neovascularization was not required. Treatment extended 100 μ beyond the edge of neovascularization except when closer than 100 μ to the fovea.

Treated patients were seen at 2, 4, and 6 weeks following treatment. Fluorescein angiography was obtained to assess the adequacy of treatment. Treated patients were encouraged to monitor their vision with an Amsler grid and report any new distortion or enlargement of scotoma. If persistence or recurrence of a treated membrane was found and met the initial treatment criteria, additional treatment was promptly applied.

Treated patients were seen at 3 and 6 weeks following treatment and fluorescein angiography was obtained. Additional treatment was promptly applied when persistence or recurrence of a treated membrane was found and met the initial treatment criteria.

This trial investigated the role of laser treatment for new neovascularization directly under the fovea. Patients with well-demarcated subfoveal CNV smaller than 3.5 standard MPS disc areas were randomized to laser photocoagulation or no treatment. Eyes in the treatment group were randomized to receive kryptonred or argon-green laser. Laser treatment covered the entire neovascular lesion and extended 100 μ beyond the edge of the lesion.

Treated patients were seen at 3 and 6 weeks following treatment and fluorescein angiography was obtained. If persistence or recurrence of a treated membrane was found and met the initial treatment criteria, additional treatment was promptly applied.

Group A eyes were characterized by less loss of visual acuity in treated eyes than in untreated eyes at every follow-up examination. These eyes were excellent candidates for treatment.

Group B was characterized by more loss of visual acuity in treated eyes than untreated eyes at 3 months but less loss of visual acuity in treated eyes than untreated eyes at 12 months and thereafter. It was found that eyes treated 1 year after the initial treatment maintain a visual acuity benefit. Alternatively, these eyes may be followed up and treated when the vision starts to decrease or the lesion size starts to increase.

Group C was characterized by relatively low proportions of large decreases in visual acuity in both treated and untreated eyes and by slightly less loss of visual acuity in treated eyes during follow-up. These lesions are good candidates for treatment because the risk of further loss with treatment is low and there is a small treatment benefit.

Group D was characterized by substantially more visual acuity loss in treated eyes than in untreated eyes at 3 months through 18 months and little difference between the two groups thereafter. These lesions are poor candidates for treatment.

The Treatment of AMD with Photodynamic Therapy (TAP) Study Group reported the 1- and 2-year results of two multicenter randomized clinical trials investigating the role of photodynamic therapy (PDT) in treatment of subfoveal CNV secondary to AMD. Eligible patients had subfoveal CNV caused by AMD measuring 5400 μ or less in greatest linear diameter with evidence of some classic CNV and best-corrected visual acuity of about 20/40 to 20/200. Six hundred nine patients were randomly assigned to PDT with Verteporfin or placebo.

Eyes with extrafoveal and juxtafoveal CNV secondary to OHS or idiopathic causes are treated with laser photocoagulation according to the MPS guidelines. Subfoveal CNV secondary to OHS or idiopathic causes is not treated with laser photocoagulation. The Submacular Surgery Trial is investigating the role of surgical removal of CNV for these lesions. Some physicians may consider treating these lesions with PDT.

The Branch Retinal Vein Occlusion Study (BRVOS) was a multicenter, randomized clinical trial designed to investigate efficacy of (1) scatter-laser photocoagulation in the management of neovascularization and vitreous hemorrhage and (2) focal laser for the treatment of macular edema in patients with branch retinal vein occlusions (BRVO).

Treatment consisted of argon-laser photocoagulation applied in a grid pattern over the area of capillary leakage in the macular region identified by fluorescein angiography. Photocoagulation was not extended closer to the fovea than the edge of the foveal avascular zone (FAZ) and not peripheral to the major vascular arcades.

Patients were evaluated at the time of entry into the study and every 4 months thereafter. Additional photocoagulation was applied at each 4-month visit if untreated leaking areas and foveal edema persisted with continued loss of visual acuity. The primary outcome measure for this part of the study was a gain of two or more lines of visual acuity at two or more consecutive 4-month follow-up visits.

Between July 1977 and February 1984, 139 eligible eyes from 139 patients were recruited: 71 randomized to laser and 68 to observation. The average duration of follow-up was 3.1 years.

Eyes at risk for development of neovascularization were enrolled in group I and those at risk for vitreous hemorrhage were enrolled in group II. To be eligible, patients must have had a BRVO 3 to 18 months earlier with at least five disc areas of retinal involvement and sufficient clearing of intraretinal hemorrhage to permit safe laser photocoagulation. In addition, group I eyes were required to have no neovascularization, and group II eyes were required to have disc or peripheral neovascularization. Eligible eyes in each group were randomized to scatter laser photocoagulation or to observation.

Treatment consisted of argon laser to the involved segment to achieve “medium” white burns (200 to 500 μ in diameter) spaced one burn width apart, extending no closer than two disc diameters from the center of the fovea.

Patients were evaluated at the time of entry into the study and every 4 months thereafter. The outcome measure for group I was neovascularization, defined as any amount of surface or preretinal new vessel formation. The outcome measure for group II was vitreous hemorrhage, defined as any amount of preretinal or vitreous hemorrhage.

The Central Retinal Vein Occlusion Study (CRVOS) was a multicenter, randomized clinical trial designed to evaluate (1) the timing of panretinal photocoagulation (PRP) for ischemic central vein occlusion; (2) the role of macular grid-pattern photocoagulation for macular edema with reduced visual acuity resulting from CRVO; (3) the natural history of eyes with CRVO. This study was not designed to evaluate the efficacy of PRP in the treatment of anterior segment neovascularization, that is, neovascularization of the iris or angle (NVI/NVA), in patients with CRVO.

To be eligible, patients must have had CRVO of less than 1 year in duration with at least 10 disc areas of retinal nonperfusion. Visual acuity of light perception or better and intraocular pressure lower than 30 mm Hg were required. Eligible eyes were randomized to either immediate PRP or laser only if NVI/NVA developed. Prophylactically treated eyes received additional PRP if NVI/NVA developed.

Treatment consisted of 1000 to 2000 evenly spaced laser spots applied to achieve moderately intense white burns (500 to 1000 μ in diameter) spaced 0.5 to 1 burn apart, avoiding retinal hemorrhage or retinal vessels. Treatment was applied no closer than two disc diameters from the center of the fovea and no closer than 500 μ nasal to the disc.

Patients were evaluated at the time of entry in the study and monthly for the first 6 months after entry. After the first 6 months, patients were seen at 2 months and every 4 months thereafter. At each visit, undilated slit-lamp examination and gonioscopy were performed to examine for anterior segment neovascularization. The primary outcome of the study was anterior segment neovascularization defined as two clock hours of NVI or any NVA.

From August 1988 to August 1992, 181 eligible eyes from 180 patients were recruited. Ninety eyes were randomized to prophylactic PRP and 91 eyes to PRP when anterior neovascularization developed. The follow-up period was 3 years, except for patients recruited after February 28, 1991, who were followed up until February 28, 1994.

To be eligible, patients must have had CRVO of at least 3 months in duration with visual acuity between 20/50 and 5/200 as a result of macular edema without capillary nonperfusion involving the fovea and documented by fluorescein angiography. Eligible eyes were randomized to either grid-pattern laser treatment or to observation.

From August 1988 to August 1992, 155 eyes of 155 eligible patients were recruited; 75 were randomized to grid-pattern laser treatment and 78 to observation. The follow-up period was 3 years, except for patients recruited after February 28, 1991, who were followed up until February 28, 1994.

At the initial examination, it is important to perform undilated slit-lamp examination and gonioscopy to evaluate for NVI/NVA. If NVI/NVA is present at the initial examination, PRP should be considered. Intraocular pressure should be measured because of the reported association of CRVO and glaucoma. Patients are recommended to be under general medical care, with yearly measurement of blood pressure. Color photography and fluorescein angiography are not essential but might be useful in providing a good baseline and identifying the perfusion status. In patients with bilateral simultaneous CRVOs, the possibility of hyperviscosity should be considered. Patients with unilateral CRVO are notified that the risk of a vascular occlusion in the fellow eye is about 1% per year.

Treatment with PRP is instituted only at the first sign of neovascularization. After PRP, the patient is monitored closely every 2 to 4 weeks to ensure that the NVI is not advancing. If NVI recurs, photocoagulation should be added in untreated areas of the retina.