The Effect of Transpupillary Thermotherapy on the Human Macula

Transpupillary thermotherapy (TTT) was introduced by investigators from the Netherlands in 1995 as an alternative treatment for choroidal melanoma.^1,2 Since that time, TTT has been used to treat small choroidal melanomas, and preliminary results indicating that TTT can control small melanomas with follow-up of 5 or more years have been published.^3,4 However, localized retinal destruction, vascular occlusions, and nerve fiber bundle defects are commonly associated with effective treatment of small melanomas with TTT. Despite these observed retinal complications, some investigators have recently reported that TTT, using the same laser intensity to treat choroidal melanoma, may successfully treat occult subfoveal choroidal neovascularizations in patients with age-related macular degeneration without observing deleterious retinal complications.⁵ The encouraging results in pilot studies with TTT in the management of occult choroidal neovascular membranes has led to the development of a multicenter prospective randomized clinical trial (Transpupillary Thermotherapy [TTT] of Occult Subfoveal Choroidal Neovascularization in Patients With Age-Related Macular Degeneration Trial) in which patients with subfoveal choroidal neovascular membranes are randomized to a sham treatment or a treatment with a single 60-second exposure of infrared light from the diode laser (810 nm) using a beam diameter of 3 mm and 800 mW of power (Elias Reichel, MD, oral and written communication, January 12, 2000). The unique opportunity afforded by a patient scheduled for enucleation for a malignant melanoma located in the nasal choroid led to this experiment in which infrared light from a diode laser was directed to the macula through a contact lens using the variables identical to those recommended in the TTT of Occult Subfoveal Choroidal Neovascularization in Patients With Age-Related Macular Degeneration Trial.

A 65-year-old woman was referred to the Department of Ophthalmology of the Mayo Clinic, Rochester, Minn, on March 8, 2000, because of a growing pigmented choroidal lesion in the left eye that had been observed to increase in thickness from 2.3 to greater than 4 mm during an interval of 9 years.

The visual acuity was 20/20 OD and 20/25 + 3 OS. The right eye was normal. Results of examination of the left eye showed a normal anterior segment and a pigmented, elevated choroidal lesion, measuring 9 × 9 mm in base dimension, located approximately 4 mm superonasal to the disc (Figure 1). The ultrasonographic studies demonstrated a solid, dome-shaped tumor (B-scan) with low internal reflectivity (A-scan) consistent with the diagnosis of melanoma. The thickness of the lesion was 4.4 mm. A subretinal fibrotic plaque overlying the central portion of the tumor and a secondary retinal detachment overlying the nasal portion of the tumor were seen. Subretinal fluid extended approximately 1 disc diameter beyond the nasal periphery of the mass. Small accumulations of exudates at the nasal boundary of the tumor were also seen. The retina in the central macular region appeared subtly thickened on biomicroscopy findings.

The diagnosis of actively growing malignant melanoma was made, and definitive therapy was recommended. Therapeutic options were discussed. Brachytherapy was encouraged, but the patient was concerned about the potential for continuing problems with the eye and strongly desired an enucleation.

We obtained institutional review board approval for the experiment. The patient was fully informed regarding the use of infrared laser light in the management of macular disease and the possibility that the treatment could cause an alteration of the pigment epithelium and retina overlying the choroidal target tissue. She agreed to have her retina exposed to light from the infrared laser for 60 seconds using a power of 800 mW. She also agreed to undergo color fundus photography and fluorescein angiography of the retina before and after light exposure.

After documenting the appearance of the tumor (Figure 1) and the macula by means of slitlamp biomicroscopy color photographs and fluorescein angiography before TTT, the macula of the left eye was exposed to a 3-mm beam of infrared laser light (810 nm) using 800 mW of power. The laser beam was delivered to the posterior pole through a standard fundus contact lens (Carl Zeiss, Inc, Thornwood, NY), exposing the macula to light from the infrared laser for 60 seconds. Visual acuity was measured with a standard Snellen chart, and the central field was evaluated with an Amsler grid 3 minutes after light exposure. Five days after laser exposure, we reexamined the eye. Best corrected visual acuity was determined, and the central visual field was evaluated with an Amsler grid. Biomicroscopy and ophthalmoscopy were performed, and the appearance of the posterior pole was documented with color fundus photography and fluorescein angiography. The eye was enucleated approximately 3 hours later and fixed in a 0.1M phosphate-buffered solution containing 4% paraformaldehyde and 1% glutaraldehyde. The retina and choroid in the macula were dissected en bloc, fixed in solution, and embedded for transmission electron microscopy. Light microscopy of this region was not performed. The remaining tissue was examined by means of light microscopy.

Color photographs of the macular region showed no obvious abnormality (Figure 2A). The pretreatment fluorescein angiogram showed an early pattern of complex vascular loops within the tumor and patchy leakage of dye from these sites leading to early patchy tissue staining of the tumor. Later frames showed diffuse staining of the retina overlying the tumor. Dye leakage from fluorescein-incompetent capillaries in the perifoveal region was seen, which produced an incomplete pattern of cystoid edema visible in the late frames of the study and arclike areas of diffuse intraretinal staining superior to the fovea, temporal to the fovea, and to a lesser extent inferotemporal to the fovea (Figure 2 B and Figure 2 D). The largest dimension of the staining site had a diameter of approximately 3 mm (before and after laser exposure).

During and immediately after transpupillary exposure of the macula to light from the infrared laser, no discernible alteration in the ophthalmoscopic appearance of the fundus was seen. Three minutes after light exposure, the patient noticed a bluish discoloration in the central field of vision that she outlined on the Amsler grid. The best-corrected distance visual acuity 3 minutes after light exposure was 20/100. When the patient returned 5 days later, the central visual acuity had recovered to the pretreatment level (20/25), but she still recognized a faint bluish discoloration in the central part of her vision on Amsler grid testing. She reported that this dyschromatopsia had been decreasing in intensity each day since the exposure. Examination of the fundus 5 days after laser exposure showed no discernible change in the appearance of the fundus. The fluorescein angiogram at the 5-day follow-up visit demonstrated findings identical to those seen in the pretreatment angiogram (Figure 3).

Light microscopy of the enucleated eye showed a malignant melanoma located in the posterior choroid, nasal to the optic disc, that formed a mass measuring 10 × 10 × 3 mm and consisted predominantly of epithelioid cells. A serous detachment of the sensory retina was seen overlying the nasal portion of the tumor. A plaque of fibrous tissue was visible over the central portion of the tumor, and cystoid degeneration was present in the overlying retina (Figure 4A and B).

Results of the ultrastructural examination of the macular and paramacular region, which had been dissected and embedded for transmission electron microscopy, showed retinal pigment epithelial cells with numerous cytoplasmic granules of lipofuscin and melanofuscin with round and irregular shapes rather than the usual oval-shaped melanosome granules (Figure 4 C). Focal disruption of cellular membranes and dispersion of pigment granules among outer segments of photoreceptor cells were seen. Vacuolation and distension of the outer segments of the photoreceptors were also observed, with partial disintegration of the lamellar structure with a rare thumbprintlike configuration (Figure 4 D). The underlying choriocapillaris showed congestion, but the vessels were normal, with intact walls and normal endothelial cells. The larger choroidal vessels also appeared normal (Figure 4 E).

Transpupillary thermotherapy has been used in recent years as a therapeutic alternative in the management of some choroidal melanomas.^1-4 We have shown that effective treatment of selected small choroidal melanomas almost always leads to a profound field defect because of destruction of the photoreceptors and nerve fibers in the retina overlying the treated tumor.⁴ When used for treatment of a choroidal melanoma, the same variables recommended by the ongoing multicenter randomized study of occult choroidal neovascular membranes in age-related macular degeneration (800 mW, 60 seconds of exposure, and 3-mm beam diameter) are usually associated with extensive retinal damage, causing a localized scotoma and, frequently, a wedge-shaped field defect as a result of nerve fiber bundle destruction. Therefore, some concern exists that these variables have been chosen for treatment of choroidal neovascular membranes. Our patient with a malignant melanoma located in the nasal choroid, scheduled for enucleation, consented to have her macula exposed to light from the diode laser using the variables cited above. In this patient's affected eye, subtle intraretinal edema involved the posterior pole, presumably related to the actively growing melanoma located in the superior nasal fundus. This intraretinal edema was recognized in the macula before exposure to the laser light. The melanoma itself showed complex vascular patterns on fluorescein angiography with extensive leakage of dye that diffused into the overlying retina, where heavy fluorescein staining was seen in the late part of the fluorescein study.

Laser exposure of the macula failed to produce a clinically recognizable reaction in the retina during the treatment, and no changes were recognized on results of a careful clinical examination 5 days after exposure. A fluorescein angiogram 5 days after the laser exposure showed no difference from the fluorescein angiogram obtained immediately before the laser exposure. Although the central visual acuity was reduced to 20/100 immediately after exposure, 5 days after TTT, the central visual acuity had recovered to the pretreatment visual acuity.

We were unable to see a graying of the retina after exposure to laser energies with the same dosage that ordinarily causes a graying ("take") when directed to a choroidal melanoma. The subtle edema of the macular retina in this case was similar to the mild retinal edema frequently seen with occult choroidal neovascular membranes. However, we do not believe that the presence of retinal edema could explain the observed absence of a take in the retina and/or retinal pigment epithelium during or 5 days after the laser exposure. This absence of a take appears similar to the relative absence of a take that we have observed when attempting to prophylactically treat normal-appearing tissue adjacent to a pigmented choroidal melanoma. Takes are often not well seen in the normal-appearing tissue. For example, a 3-mm beam of laser light that is placed so as to equally straddle the edge of a pigmented choroidal tumor and clinically normal tissue adjacent to it often dramatically outlines the ophthalmoscopically recognizable perimeter of the tumor. The tumor, the overlying retina, and the retinal pigment epithelium become gray white, whereas the adjacent retina overlying normal-appearing choroid frequently remains clinically unchanged or only minimally gray. We believe that the pigment within the tumor is largely responsible for generating a considerable amount of heat, which causes the tumor, the overlying pigment epithelium, and retina to turn gray more readily. In addition, the choriocapillaris and larger vessels in the choroid are altered in the presence of a melanoma, thus decreasing the ability of the choroid to play its role as a heat sink to disperse energy. The absence of a heavy concentration of pigment and the presence of a normal choroid, providing a normal heat sink, can facilitate dispersement of energy, thereby minimizing the potential for thermal damage to the overlying retina.

In this study, we were unable to identify TTT-induced adverse effects in the retina or the pigment epithelium by means of clinical examination or fluorescein angiography. However, we observed histological and ultrastructural abnormalities in the tissue after the eye was enucleated. We believe that these abnormalities can be explained by the presence of the preexisting retinal edema. However, some of the observed abnormalities could have been caused by light alone, as shown by Robertson and Erickson⁶ and Green and Robertson.⁷ Although we looked for evidence of vascular closure or coagulative necrosis in the small capillaries in the choroid, we were unable to demonstrate such changes in the choroid in the foveolar region by means of ultrastructural studies.

The absence of recognizable destruction of the retina and retinal vasculature observed in this single experiment does not ensure that vascular closure and retinal destruction will not occur when TTT is used to treat occult choroidal neovascular membrane. However, in this case, with clinical and angiographic evidence of mild retinal edema and no retinal or subretinal blood, a 60-second exposure of 800 mW using a 3-mm beam diameter did not cause clinically recognizable damage to the macular retina, the retinal vessels, or the underlying choriocapillaris and other choroidal vessels.

This study was supported in part by a grant from Research to Prevent Blindness, Inc, New York, NY, and in part by the Mayo Foundation, Rochester, Minn.

We thank Bonnie Ronken for her secretarial support and Cheryl Hann, MS, for her support with the electron microscopy studies.

Corrresponding author and reprints: Dennis M. Robertson, MD, Department of Ophthalmology, Mayo Clinic, Rochester, MN 55905 (e-mail: robertson.dennis@mayo.edu).