Customize your JAMA Network experience by selecting one or more topics from the list below.
Copyright 2003 American Medical Association. All Rights Reserved. Applicable FARS/DFARS Restrictions Apply to Government Use.2003
Subfoveal choroidal neovascularization (CNV) caused by age-related macular degeneration is the leading cause of irreversible vision loss in Americans 65 years or older. Photodynamic therapy (PDT) with verteporfin (Visudyne; CIBA Vision Corp, Duluth, Ga) has been shown to retard vision loss compared with placebo in eyes with predominantly classic and purely occult CNV lesions.1,2 Verteporfin generates reactive oxygen species when illuminated with light at a wavelength of 689 nm, which results in occlusion of choroidal new vessels.3,4 Visual disturbances, including decreased visual acuity, visual field defect, or otherwise abnormal vision, are reported by 22% to 42% of patients after PDT.2,3 Severe vision loss, defined as the loss of at least 20 letters of visual acuity, has been reported in 4.4%(10/225) of patients with purely occult CNV.2 Eighty percent (8/10) of patients who suffered severe vision loss had occult lesions.2 Visual acuity improved to varying degrees in these patients.2 We report 2 cases of a marked exudation associated with vision loss within days after PDT.
An 83-year-old patient, diagnosed as having age-related macular degeneration associated with occult subfoveal CNV in the right eye (Figure 1), underwent PDT according to a standard protocol with the use of a laser (Coherent Opal Photoactivator; Coherent, Inc, Santa Clara, Calif). Pretreatment best-corrected visual acuity was 20/60 in the right eye. The patient reported decreased visual acuity immediately after PDT. When the patient was examined 5 days after treatment, the best-corrected visual acuity was 20/400. Fundus examination demonstrated a marked increase in submacular fluid. Fluorescein angiography demonstrated early hypofluorescence of the treated area with late patchy hyperfluorescence corresponding to the entire area of treatment (Figure 2). The retinal circulation appeared normal throughout the study. Indocyanine green angiography demonstrated diffuse hyperfluorescence in the late phase throughout the treated area (Figure 3). A course of observation was undertaken. Six weeks after treatment, the best-corrected visual acuity was 20/50 with almost complete resorption of the subretinal fluid. Three months after treatment, the best-corrected visual acuity was 20/50 and fundus examination showed a dry subfoveal retinal pigment epithelial(RPE) scar. Fluorescein angiography showed recovery of normal fluorescence and slightly less leakage from the neovascularization (Figure 4).
Case 1. Pretreatment angiogram after fluorescein injection, demonstrating occult subfoveal choroidal neovascularization.
Case 1. A, Fluorescein angiogram 5 days after photodynamic therapy demonstrating intact retinal circulation and choroidal hypofluorescence in the area of laser exposure. B, Late frame showing patchy leakage in the same area.
Case 1. Indocyanine green angiogram 5 days after photodynamic therapy, 20 minutes after injection, demonstrating irregular subretinal leakage corresponding to the area of laser exposure.
Case 1. Fluorescein angiogram 3 months after photodynamic therapy showing recovery of normal choroidal fluorescence and slightly less leakage from the neovascularization than at baseline.
A 68-year-old patient with age-related macular degeneration developed metamorphopsia and vision loss in the right eye. Pretreatment best-corrected visual acuity was 20/60 in the right eye. Examination showed central subretinal fluid, exudate, and hemorrhage. Fluorescein angiography demonstrated an occult subfoveal CNV in the right eye. The patient underwent PDT with verteporfin according to the standard protocol with the use of a laser (Coherent Opal Photoactivator). He noted severely decreased vision 2 days after PDT. On examination 3 days after treatment, the best-corrected visual acuity was 4/200. There was a marked increase in subretinal fluid with no change in subretinal hemorrhage. Angiography showed partial hypofluorescence of the CNV lesion with late patchy hyperfluorescence of the entire treated area. The retinal circulation was intact, with a notable absence of retinal vascular staining. One week after treatment, there was an improvement in the subretinal fluid and hemorrhage. Improvement continued for 2 months after treatment, with only scant subretinal hemorrhage and fluid remaining and a best-corrected visual acuity of 20/70.
The sudden accumulation of fluid in the subneurosensory retinal space after PDT may arise from several sources. Retinal vascular injury or subsequent inflammation could cause dysfunction of retinal vascular endothelial tight junctions, resulting in retinal edema and accumulation of subretinal fluid. In our cases, however, angiography demonstrated no occlusion or staining of retinal vessels. A second possible cause is increased leakage from the CNV. Angiography in this series demonstrated less hyperfluorescence from the CNV and a diffuse area of hyperfluorescence corresponding to the entire treated area.
Animal models have demonstrated damage to the RPE and choriocapillaris after PDT with verteporfin.4 We postulate that sudden fluid accumulation after PDT occurs as a result of dysfunction of the RPE pump and/or increased permeability of the choroid. Studies have demonstrated that RPE cells possess low-density lipoprotein receptors that are targeted by the liposomal preparations of verteporfin.4,5 Reactive oxygen species can be generated by verteporfin within RPE cells, resulting in cell damage and impaired fluid transport. Damage of the RPE in primates was described after PDT with verteporfin.4 It is interesting to note that both of these cases were occult lesions. Neovascularization under the RPE could predispose the RPE cells to unusual damage by causing up-regulation in low-density lipoprotein receptors or simply by making the cells less resistant to oxidative stress. Another possible mechanism may be that with predominantly classic CNV, the lesion itself may cast a shadow on the RPE and thereby lower the irradiance of the RPE compared with occult lesions.
Both cases in our series were observed. The subretinal fluid resolved slowly during 2 to 3 months, with return of vision to near pretreatment levels. Because neither eye developed a geographic area of atrophy, a temporary RPE dysfunction that resolved over time without significant RPE cell death may have been involved.
Photodynamic therapy has become a mainstay in treatment of the subfoveal form of age-related macular degeneration. Although PDT has very few vision-threatening complications, the clinician should be aware of the possibility of sudden vision loss caused by a marked exudative response, particularly when dealing with occult lesions.
Corresponding author: Eric R. Holz, MD, Cullen Eye Institute, Baylor College of Medicine, 6565 Fannin, Mail Code NC-205, Houston, TX 77030 (e-mail: email@example.com).
Holz ER, Linares L, Mieler WF, Weinberg DV. Exudative Complications After Photodynamic Therapy. Arch Ophthalmol. 2003;121(11):1649–1652. doi:10.1001/archopht.121.11.1649