| OCR Text |
Show ORIGINAL CONTRIBUTION Multifocal Electroretinographic Abnormalities in Ethambutol-Induced Visual Loss Yao Liu, MD, Marc J. Dinkin, MD, John I. Loewenstein, MD, Joseph F. Rizzo, III, MD, and Dean M. Cestari, MD Abstract: Two patients who developed decreased visual acuity after several months of ethambutol treatment for Mycobacterium avium-intracellulare infection had bitemporal visual field defects that suggested optic chiasm damage. Multifocal electro-retinography (mfERG) disclosed markedly low-amplitude responses at fixation and in the regions corresponding to the visual field defects. These results suggested that the visual dysfunction might be entirely attributable to retinal rather than optic nerve toxicity. These are the first reported patients to show mfERG abnormalities that correspond to bitemporal visual field defects and add to the growing evidence that ethambutol damages the retina. (J Neuro-Ophthalmol 2008;28:278-282) Ethambutol is a first-line drug used to treat mycobacterial infection. Bilateral and symmetrical loss of vision due to optic neuropathy is a well-known complication of ethambutol therapy (1). Animal studies have confirmed that ethambutol causes demyelination and axonal swelling of the optic nerves. Use of the recommended therapeutic dose of 15-25 mg/kg/day causes visual loss in 1% of patients treated with ethambutol (1). The majority of those who develop ethambutol toxicity have permanent visual loss that can be severe. In one consecutive case series of 7 patients with ocular ethambutol toxicity (2), less than 50% of patients experienced recovery of their visual acuity to better than 20/200 after an average of 8 months after discontin-uation of the drug. No prognostic tests are available to assess which patients have a higher risk of permanent visual loss. Neuro-Ophthalmology Service (YL, JFR, DMC) and Retina Service (DMP), Massachusetts Eye and Ear Infirmary, Harvard Medical School, Boston, Massachusetts; and Department of Neurology (MJD), Brigham and Women's Hospital, Harvard Medical School, Boston, Massachusetts. Address correspondence to Dean M. Cestari, MD, Department of Neuro-Ophthalmology, 9th Floor Massachusetts Eye and Ear Infirmary, 243 Charles Street, Boston, MA 02114; E-mail: dean_cestari@meei. harvard.edu In addition to causing an optic neuropathy, etham-butol probably also causes visual loss due to retinal dysfunction (3-6). Retinal dysfunction has been identified through abnormal electro-oculogram (EOG) light/dark ratios and first-order kernel amplitudes with the multifocal electroretinography (mfERG) technique. Yen et al (3) found abnormal EOG ratios in patients presenting with etham-butol toxicity. Lai et al (4) later reported a patient with ethambutol-induced bilateral central scotomas and demon-strated dramatic bilateral global depression of first-order mfERG waveforms that improved 3 months after the drug was discontinued. Kardon et al (5) described two patients treated with ethambutol who developed cecocentral and central scotomas, respectively. Both patients demonstrated bilateral diffuse depression of first-order mfERG wave-forms that were most prominent centrally. The degree of mfERG waveform depression in both patients was signifi-cant compared with a database of normal subjects tested at another center. Both patients had significant recovery of visual loss after discontinuation of ethambutol. Behbehani et al (6) described four patients who developed central or cecocentral scotomas while being treated with ethambutol. By aggregating their patients' mfERG response amplitudes and averaging the responses by centrally distributed rings of hexagonal testing areas, the authors found significant first-order mfERG depressions in the most central testing zone compared with a database of 10 normal subjects tested in their center. In the majority of these case reports, results of full-field ERG was normal, suggesting that only the central retina is damaged by ethambutol. We report two patients who developed bitemporal visual field defects while taking ethambutol. In both patients, mfERG testing demonstrated bilateral decreased amplitudes in the nasal macular region, which presumably accounted for the bitemporal visual field loss. CASE REPORTS Case 1 A 48-year-old woman presented to the Massachusetts Eye and Ear Infirmary (MEEI) with a 2-month history of 278 J Neuro-Ophthalmol, Vol. 28, No. 4, 2008 Bitemporal Visual Field Deficits J Neuro-Ophthalmol, Vol. 28, No. 4, 2008 progressively blurred vision, difficulty identifying colors, and peripheral visual field loss. Her medical history was significant for pulmonary Mycobacterium avium-intracellulare infection for which she was being treated with clarithromycin and 16 mg/kg/day ethambutol. Upon her report of visual loss, ethambutol was promptly dis-continued after 7 months of therapy. Our examination revealed best-corrected visual acuities of 20/50 in the right eye and 20/60 in the left eye. She correctly identified 7/8 and 5/8 Ishihara color plates in the right and left eyes, respectively. Automated (Humphrey) visual field testing revealed bilateral supero-temporal defects that respected the vertical meridian and extended to fixation (Fig. 1A). The probability map of the total deviation plot did, however, cross the vertical meridian, which made it unlikely for the pattern of visual loss to be caused solely by a chiasmal lesion. Findings for the remainder of the ophthalmologic examination, including ophthalmoscopy, were normal. Results of brain and orbit MRI were normal. A full-field ERG, including scotopic flash, photopic flash, and 30-Hz flicker, recorded according to the International Society for Clinical Electrophysiology of Vision standards, showed normal results. We performed mfERG on the VERIS system (Electro-Diagnostic Imaging, San Mateo, CA) using 103 hexagons. Pupils were dilated with tropicamide and phenylephrine. Burian-Allen bipolar contact lens electrodes were used. The technician monitored fixation throughout the test. Seven-minute recordings were performed in 30-second segments. Results were compared, hexagon by hexagon, to data averaged from 30 normal subjects who had undergone testing in our laboratory. Multiple hexagons showed responses with significantly reduced amplitudes in the central and nasal macular regions in both eyes that corresponded to the visual field defects (Fig. 1B-D). Seven months after the discontinuation of ethambutol, the patient's visual acuities had recovered to 20/30 in both eyes. Case 2 A 78-year-old woman presented to the MEEI with a 4-week history of blurred vision, decreased color vision, new floaters, and soreness in her left eye. Her past ocular history included cataract surgery in both eyes and a branch retinal vein occlusion in the left eye. She had nonmetastatic breast cancer, hypertension, and pulmonary M. avium-intracellulare infection. Medications included ethambutol, clarithromycin, rifampin, anastrozole, zolpidem, and felodipine. She had begun 19.6 mg/kg/day ethambutol 6 months before the onset of her visual symptoms, at which time ethambutol was discontinued. On our examination, best-corrected visual acuities were 20/80 in the right eye and 20/400 in the left eye. She was able to identify only the test plate on Ishihara color testing. Automated Humphrey visual field testing demon-strated decreased sensitivity in the temporal paracentral aspect of the visual field in both eyes (Fig. 2A). As in Case #1, the threshold abnormalities of the total deviation plot did cross the vertical meridian, which made it unlikely for the pattern of visual loss to be solely attributable to a chiasmal lesion. Dilated stereoscopic ophthalmoscopy revealed temporal pallor of both optic discs. Retinal pig-ment epithelial changes were noted in the nasal midper-ipheral retina in the right eye and in the nasal peripheral area in the left eye. Several flame hemorrhages were noted along the inferior temporal arcade in the left eye. Findings for the remainder of the ophthalmologic examination were normal. Results of brain and orbit MRI were normal. The cancer-associated retinopathy autoantibody was negative. Full-field and mfERG studies were performed as in Case #1. The bright scotopic responses showed reduced B wave amplitudes. The remainder of the full-field responses was normal. On the mfERG, multiple hexagons showed reduced amplitude responses in the central and nasal macular regions in both eyes, corresponding to the visual field defects and in a similar distribution to that of Case #1 (Fig. 2B-D). On follow-up examination 7 months after the discontinuation of ethambutol, the patient's visual acuities had improved to 20/30 in both eyes. DISCUSSION Toxicity to the afferent visual pathway from ethambutol has been attributed to damage to the optic nerve or chiasm (1,2). Our two cases add to the growing evidence that ethambutol can also produce a retinopathy that may partly or entirely account for the decreased vision, dyschromatopsia, and visual field defects. Previous reports (3-6) have described patients with ethambutol retinopathy who presented with central or cecocentral visual field defects and evidence of central retinal dysfunction on mfERG. Our patients had bitemporal visual field defects and mfERG revealed deficits that localized with the pattern of bitemporal visual field loss. Given that mfERG results are highly dependent on the ambient conditions of the testing facility, the inclusion of control values from subjects tested in our own facility (rather than from a database of normal mfERG results tested elsewhere) fortifies our findings. It is unlikely that the artifacts characteristic of poor fixation on mfERG account for our findings, given that significantly shortened implicit times and loss of the blind spot were not present (7). We cannot exclude a component of optic nerve toxicity in our patients. However, we offer evidence that a retinopathy may be more common than has been 279 J Neuro-Ophthalmol, Vol. 28, No. 4, 2008 Liu et al Left Eye Right Eye FIG. 1. Case 1. A. Humphrey visual fields show bilateral superotemporal visual field deficits. B. Multifocal electroretinographic (mfERG) hexagons are plotted in the same orientation as the visual field. Amplitudes measuring less than 2 SD from the mean of 30 normal subjects are highlighted in black. There is marked blunting of the central responses and those of the nasal maculae, which correspond to the bitemporal visual field defects. C. Three-dimensional topographic maps of the mfERG response amplitudes. D. Trace array of the mfERG first-order kernel responses. 280 q 2008 Lippincott Williams & Wilkins Bitemporal Visual Field Deficits J Neuro-Ophthalmol, Vol. 28, No. 4, 2008 FIG. 2. Case 2. A. Humphrey visual fields show decreased sensitivity in the temporal paracentral aspect of the visual fields. B. Multifocal electroretinographic (mfERG) hexagons are plotted in the same orientation as the visual field. Amplitudes measuring less than 2 SD from the mean of 30 normal subjects are highlighted in black. There is marked blunting of the central responses and those of the nasal maculae, which correspond to the bitemporal visual field defects. C. Three-dimensional topography maps of the mfERG response amplitudes. D. Trace array of the mfERG first-order kernel responses. 281 J Neuro-Ophthalmol, Vol. 28, No. 4, 2008 Liu et al previously recognized in patients who experience toxicity from ethambutol and that this retinal toxicity may mimic a chiasmal lesion. A more accurate approximation of the prevalence of retinopathy will be obtained as mfERGs are performed more often in patients who develop visual loss after ethambutol treatment. The mechanism of ethambutol-induced injury to the retina has yet to be elucidated. Animal studies in the fish retina have shown that ethambutol alters the synaptic connections between horizontal cells and cones (8). Ethambutol can also lead to the degeneration of cone pedicles in the fish retina while sparing the rod pathway. As shown in our patients, the mfERG provides an advantage over full-field ERG testing in that it is possible to assess the pattern of retinal dysfunction and to compare it with the pattern of visual field loss to judge whether the retinopathy accounts for the field loss. The prognosis for recovery of vision with ethambutol toxicity is highly variable. This variability is partly related to whether the primary site of injury is at the level of the retina or the optic nerves. Of the 7 patients with ethambutol retinopathy who were reported previously (3-6), 6 who were seen at follow-up demonstrated either complete or significant recovery of baseline visual acuity (2 Snellen lines or greater). Both of our patients demonstrated substantial recovery of visual acuity despite having experienced ethambutol-induced retinal toxicity with visual acuity as poor as 20/400. A systematic study using mfERGs in patients who experience visual loss with ethambutol therapy would help confirm whether patients with an ethambutol-induced retinopathy have a lower risk of permanent visual loss than those with an optic neuropathy. REFERENCES 1. Lessell S. Toxic and deficiency optic neuropathies. In Smith JL, Glaser JS, eds. Neuro-Ophthalmology: Symposium of the University of Miami and the Bascom Palmer Eye Institute. Vol. 7. St. Louis, CV Mosby; 1973:21-37. 2. Kumar A, Sandramouli S, Verma L, et al. Ocular ethambutol toxicity: is it reversible? J Clin Neuroophthalmol 1993;13:15-7. 3. Yen MY, Wang AG, Chiang SC, et al. Ethambutol retinal toxicity: an electrophysiologic study. J Formosa Med Assoc 2000;99:630-4. 4. Lai TY, Chan WM, Lam DS, et al. Multifocal electroretinogram demonstrated macular toxicity associated with ethambutol related optic neuropathy. Br J Ophthalmol 2005;89:774-5. 5. Kardon RH, Morrisey MC, Lee AG. Abnormal multifocal electro-retinogram (mfERG) in ethambutol toxicity. Semin Ophthalmol 2006; 21:215-22. 6. Behbehani RS, Affel EL, Sergot RC, et al. Multifocal ERG in ethambutol associated visual loss. Br J Ophthalmol 2005;89:976-82. 7. Vrabec TR, Affel EL, Gaughan JP, et al. Voluntary suppression of the multifocal electroretinogram. Ophthalmology 2004;111:169-76. 8. Kohler K, Zrenner E, Weiler R. Ethambutol alters spinule-type synap-tic connections and induces morphologic alterations in the cone pedi-cles of the fish retina. Invest Ophthalmol Vis Sci 1995;36:1046-55. 282 q 2008 Lippincott Williams & Wilkins |
