tend to persist for several months in amiodarone-induced optic neuropathy, whereas in nonarteritic anterior ischemic optic neuropathy, these usually resolve more quickly. After discontinuation of amiodarone, visual acuity and visual field deficits tend to stabilize. Although the mechanism of amiodarone optic neuropathy is unknown, ultrastructural changes in the human optic nerve illustrate a primary lipidosis.2 Selective accumulation of intracytoplasmic lamellar inclusions in large optic nerve axons may mechanically or biochemically decrease axoplasmic flow. Resultant optic nerve head edema may persist if transport is inhibited (perhaps as long as several months after discontinuation of amiodarone, which has up to a 100-day half-life3). Although we believe the above features are more typical of amiodarone-induced optic neuropathy, any form of optic neuropathy in a patient receiving amiodarone should cause concern. Current data are insufficient to recommend specific guidelines; however, we perform an ophthalmic examination before the patient begins taking amiodarone and every 6 months thereafter. If the diagnosis of amiodarone-induced optic neuropathy is made and therapeutic alternatives exist, then amiodarone should be discontinued. Any patient reporting visual disturbance while using amiodarone should be promptly examined by an ophthalmologist.
FIGURE 1. Fundus photograph of the right eye showing blonde fundus, fine macular drusen, and disk pallor.
METHODS:
A 59-year-old man had visual loss attributable to paraneoplastic retinopathy and died of metastatic cutaneous melanoma. His eyes were studied by conventional histopathologic techniques. RESULTS: Histopathologic examination of both eyes disclosed a marked reduction in the density of bipolar neurons in the inner nuclear layer; photoreceptor cell neurons in the outer nuclear layer were normal. Ganglion cells were present, although many showed evidence of transsynaptic atrophy. CONCLUSION: The histopathologic changes observed are consistent with clinical, immunologic, and electrophysiologic data that implicate the bipolar cell as the major site of the paraneoplastic process in cutaneous melanomaassociated retinopathy. (Am J Ophthalmol 1999;127: 612– 614. © 1999 by Elsevier Science Inc. All rights reserved.)
REFERENCES
1. Feiner LA, Younge BR, Kazmier FJ, Stricker BH, Fraunfelder FT. Optic neuropathy and amiodarone therapy. Mayo Clin Proc 1987;62:702–717. 2. Mansour AM, Puklin JE, O’Grady R. Optic nerve ultrastructure following amiodarone therapy. J Clin Neuroophthalmol 1988;8:231–237. 3. Latini R, Tognoni G, Kates RE. Clinical pharmacokinetics of amiodarone. Clin Pharmacokinet 1984;9:136 –156.
Cutaneous Melanoma-Associated Paraneoplastic Retinopathy: Histopathologic Observations John W. Gittinger, Jr, MD, and Thomas W. Smith, MD
C
PURPOSE:
To describe the retinal histopathology of paraneoplastic retinopathy associated with cutaneous melanoma.
Accepted for publication Nov 9, 1998. From the Departments of Ophthalmology (J.W.G.), Neurology (J.W.G., T.W.S.), and Pathology (T.W.S.), University of Massachusetts Medical School, Worcester, Massachusetts. Inquiries to John W. Gittinger, Jr, MD, Department of Ophthalmology, Room S7-834, University of Massachusetts Medical Center, Worcester, MA 01655-0322; fax: (508) 856-5084; e-mail: John.Gittinger @Banyan.UMMED.EDU
612
AMERICAN JOURNAL
UTANEOUS MELANOMA-ASSOCIATED PARANEOPLAS-
tic retinopathy was first recognized by Berson and Lessell1 in a man with a history of malignant melanoma of the skin presenting with night blindness and photopsias. The reduced b-wave on his electroretinogram was similar to the pattern observed with congenital stationary night blindness with myopia and contrasted with the total extinction that correlates with loss of the photoreceptors in the previously described paraneoplastic retinopathy associated with carcinoma.2 Subsequent cases of paraneoplastic retinopathy associated with melanoma have disclosed variable clinical features.3 Preservation and loss of central vision have been reported, but in general, unlike paraneoplastic retinopathy associated with carcinoma, the visual loss in paraneoplastic
OF
OPHTHALMOLOGY
MAY 1999
FIGURE 2. Photomicrographs of normal retina from near the fovea (left) and of a corresponding area of the retina with cutaneous melanoma-associated paraneoplastic retinopathy (right). In the eye with melanoma-associated paraneoplastic retinopathy, the photoreceptors are preserved, but the bipolar and ganglion cell layers are reduced (hematoxylin and eosin, 340).
retinopathy associated with melanoma is less often progressive and does not usually respond to corticosteroids. Electrophysiologic and immunologic studies point to the bipolar cell as a site of the paraneoplastic damage.4 A 59-year-old man had a malignant melanoma removed from his left forearm in 1990. A local recurrence was excised in 1992, and in 1994, he developed a left retroperitoneal mass. A needle-biopsy specimen of the mass yielded tissue consistent with malignant melanoma. He was treated with interleukin-2 and interferon. In October, 1995, he had a seizure. Computed tomography of the head showed intracranial masses. About the same time, he noted difficulty with his vision, especially at night. He received whole brain irradiation. Two months later, he was referred for evaluation of progressive visual loss. His visual acuity was RE: 20/30 and LE: 6/200. He missed five-and-one-half American Optical pseudoisochromatic plates with his right eye and could not see the control with his left eye. Goldmann visual fields demonstrated constriction in the right eye and constriction with a central scotoma in the left eye. The left eye had a relative afferent pupillary defect. Examination of the right eye with VOL. 127, NO. 5
a dilated pupil disclosed a blonde fundus with fine macular drusen and disk pallor (Figure 1). A full field flash electroretinogram demonstrated scotopic wave forms at high intensity that consisted of a mildly reduced a-wave and markedly reduced b-wave. B-wave amplitudes were 25% to 30% of normal in the right eye and 10% to 15% of normal in the left eye. Melanoma-associated paraneoplastic retinopathy was diagnosed, and the patient was followed up without treatment or change in his vision. He died in June 1995, and both eyes were fixed in 10% buffered formalin and processed for routine paraffin section. The most striking histopathologic observation was marked reduction in the thickness of the inner nuclear layer and the density of the bipolar neuron nuclei (Figure 2). By contrast, the photoreceptor cell nuclei in the outer nuclear layer were normal. No inflammatory cells were present. Most of the ganglion cells had reduced perikaryal cytoplasm and some nuclear shrinkage and condensation of the chromatin, suggesting transsynaptic degeneration. Glial fibrillary acidic protein immunohistochemistry disclosed gliosis in the ganglion and bipolar cell layers. The
BRIEF REPORTS
613
optic nerves had preserved axonal structure with prominent gliosis. The few histopathologic studies of congenital stationary night blindness have confirmed the presence of rods, but bipolar cell abnormalities have not been recognized.5 If an eye with uncomplicated congenital stationary night blindness becomes available, it will be important to look at the bipolar cells to see if there is a reduction in number similar to that described here in paraneoplastic retinopathy associated with melanoma. Histopathologic observations of eyes with paraneoplastic retinopathy associated with carcinoma have already confirmed the clinical, electrophysiologic, and immunologic evidence that pointed to the photoreceptors as the primary site of paraneoplastic damage. This case supplies similar evidence that the bipolar cells are the primary site in paraneoplastic retinopathy associated with melanoma.
laser followed by one pass of the CO2 laser. With the Student t test, we compared skin re-epithelialization time and duration of erythema with those of a previous group of 25 patients who had undergone eyelid resurfacing with only the CO2 laser (two passes). A pathologist evaluated all skin biopsy specimens. RESULTS: Combining both lasers shortened re-epithelialization time (7 vs 12 days, P 5 .04) and the duration of erythema (2.5 vs 7.0 weeks, P 5 .02). Histopathologic examination disclosed less coagulative dermal damage with the combined laser protocol. CONCLUSION: The different biophysical properties of these two lasers can be combined in a periorbital resurfacing protocol to minimize both clinical and histopathologic morbidity. (Am J Ophthalmol 1999;127:614–616. © 1999 by Elsevier Science Inc. All rights reserved.)
T
REFERENCES
1. Berson EL, Lessell S. Paraneoplastic night blindness with malignant melanoma. Am J Ophthalmol 1988;106:307–311. 2. Rizzo JF III, Gittinger JW Jr. Selective immunohistochemical staining in the paraneoplastic retinopathy syndrome. Ophthalmology 1992;99:1286 –1295. 3. Kellner U, Bornfield N, Foerster MH. Severe course of cutaneous melanoma associated paraneoplastic retinopathy. Br J Ophthalmol 1995;79:746 –752. 4. Weinstein JM, Kelman SE, Bresnick GH, Kornguth SE. Paraneoplastic retinopathy associated with antiretinal bipolar cell antibodies in cutaneous malignant melanoma. Ophthalmology 1994;101:1236 –1243. 5. Vaghefi HA, Green WR, Kelley JS, Sloan LL, Hoover RE, Patz A. Correlation of clinicopathologic findings in a patient; congenital night blindness, branch retinal vein occlusion, cilioretinal artery, drusen of the optic nerve head, and intraretinal pigmented lesion. Arch Ophthalmol 1978;96: 2097–2104.
Histologic and Clinical Evaluation of Combined Eyelid Erbium:YAG and CO2 Laser Resurfacing Arthur L. Millman, MD, and Geva E. Mannor, MD, MPH To report the histopathologic and clinical effects of eyelid resurfacing that combines two different lasers. METHODS: A case series of 23 patients who underwent eyelid resurfacing with two passes of the Erbium:YAG PURPOSE:
Accepted for publication Oct 24, 1998. From the Department of Ophthalmology, Mount Sinai Medical Center, New York, New York (A.L.M., G.E.M.), the Department of Ophthalmology, New York Eye and Ear Infirmary, New York, New York. (A.L.M.), and the Ophthalmology Section, Veterans Medical Center, Bronx, New York (G.E.M.). Inquiries to Arthur L. Millman, MD, 345 E 37 St, Ste 212, New York, NY 10016; fax: (212) 697-4902.
614
AMERICAN JOURNAL
HE SUPER-PULSED CO2 AND THE ERBIUM:YAG LASERS
have different biophysical properties. The CO2 laser ablates about 100 mm of skin and leaves an additional 50 mm of thermally damaged tissue.1 The amount of thermally damaged tissue increases up to 300 mm with subsequent passes.1 In contrast, the Erbium:YAG laser is more highly absorbed by water and ablates less skin per pass (20 to 40 mm) with less collateral thermal damage (20 to 30 mm).2 We sought to minimize postoperative morbidity while retaining clinical efficacy by combining these two lasers in a periorbital resurfacing protocol. All patients who had eyelid resurfacing gave informed consent and were selected by clearly established criteria.1,3 In the combined protocol, two passes of an Erbium:YAG laser (DERMA-20; Luxar Corporation, Bothell, Washington) set at 1 J and 10 Hz were delivered through a 3-mm handpiece to ablate the epidermis. Next, a single pass of a super-pulsed CO2 laser (Novapulse ShureScan; Luxar Corporation) set at 300 mJ was delivered to the dermis. Perioperative skin care regimen included systemic antibiotics, systemic antivirals, topical sunblocks, topical bleaching agents, and topical lubrication.1–3 Patients were evaluated daily, for up to 2 weeks, until skin re-epithelialization, and then monthly for 6 months. The combined laser protocol patients were compared by Student t test with a historical control group of 25 previous patients who had eyelid resurfacing with two passes of the same CO2 laser at the same power setting as that used in the combined protocol. The same preoperative and postoperative skin care regimen was used for both patient groups. Both groups of patients were similar in sex and age distribution, and all had similar Fitzpatrick skin types. Eyelid skin biopsy specimens from both patient groups were obtained after re-epithelialization on postoperative days 8 through 12. No patient suffered any permanent complication. Reepithelialization required only 7 days in the combined laser group (range, 4 to 9 days) vs 12 days (range, 9 to 15 days;
OF
OPHTHALMOLOGY
MAY 1999