OCR Text |
Show Bitemporal Visual Field Defects in Ethambutol-Induced Optic Neuropathy Richard C. Kho, MD, Majed Al-Obailan, MD, Anthony C. Arnold, MD Background: Ethambutol-induced optic neuropathy is well documented and most frequently associated with central or cecocentral scotomas. We designed a study to char-acterize the subset of patients who exhibit bitemporal visual field defects. Methods: A computer search was performed for patients evaluated in a university academic neuro-ophthalmology consultative practice to identify those with the diagnosis of ethambutol-induced optic neuropathy. Clinical features were tabulated, including dose and duration of etham-butol use, time to onset of visual loss, initial and follow-up visual acuities, automated perimetry, optic disc appear-ance, and MRI features. Assessments for bitemporal visual field defect with alignment on vertical midline and for visual improvement after discontinuing ethambutol were performed. Results: Nineteen cases of ethambutol-induced optic neuropathy were identified; All but 2 eyes demonstrated visual field defects worse in the temporal fields, most with margination along the vertical midline with superimposed central or cecocentral scotomas. Six cases (12 eyes) showed bitemporal defects with such margination without superimposed scotomas. Median time to onset of visual loss was 6.0 months. Visual improvement occurred (of 17 cases with data available) by at least 3 Snellen lines in 17 of 34 eyes (50%); mean visual acuity improvement was 3.74 lines (median, 3.0). Visual improvement by at least 3.0 decibels (dB) mean deviation (MD) on automated perimetry occurred in 27 of 34 eyes (79%); mean im-provement in MD was 7.82 dB (median, 7.86). Median follow-up was 8.0 months. None had MRI abnormality in the chiasmal region. Conclusion: Bitemporal visual field defects are common in ethambutol-induced optic neuropathy. The pattern may mimic chiasmal compression, and neuroimaging is required. It may reflect susceptibility to toxicity of chiasmal crossing fibers. Journal of Neuro-Ophthalmology 2011;31:121-126 doi: 10.1097/WNO.0b013e318205a148 2011 by North American Neuro-Ophthalmology Society Ethambutol is a well-documented cause of optic neu-ropathy, with dose-related severity (1). Previous reports have most frequently described a pattern of visual field loss common in toxic optic neuropathy, with symmetric bilateral central or cecocentral scotomas. Isolated cases of bitemporal visual field loss have been described but have been considered unusual. The degree to which the field loss pattern mimicked chiasmal compression rather than bilateral temporal field loss due to cecocentral depression has been unclear. We reviewed cases of ethambutol-induced optic neuropathy to assess the frequency of bitemporal vi-sual field loss and to establish whether the defects present showed features of chiasmal injury or were more consistent with individual bilateral optic nerve damage. METHODS A computer search was performed for patients evaluated in the consultative practice of 1 investigator (A.C.A.) in the years from 1986 to 2010, coded as ‘‘optic neuropathy, toxic.'' Inclusion criteria included evidence of progressive, bilateral, simultaneous visual field loss within the central 30 , dyschromatopsia, diminished pupillary light responses, normal or pale optic discs, and the use of ethambutol at the time of visual loss. Exclusion criteria included history of vasculitis, demyelinating disease, or other cause of optic neuritis; optic disc edema; and optic nerve or chiasmal abnormality visible on MRI. Clinical features were tabu-lated, including dose and duration of ethambutol preceding visual loss, initial and follow-up visual acuities, automated perimetry (Humphrey Field Analyzer, Carl Zeiss Meditec, Inc, Dublin, CA and Octopus Perimeter, Haag-Streit USA, Mason, OH), and optic disc appearance. UCLA Department of Ophthalmology, Jules Stein Eye Institute, Los Angeles, California; American Eye Center, Mandaluyong City, Phil-ippines; and Department of Ophthalmology, College of Medicine, King Saud University, Riyadh, Saudi Arabia. Supported in part by an unrestricted grant from the Research to Prevent Blindness, Inc. The authors have no conflicts of interest to disclose. Address correspondence to Anthony C. Arnold, MD, UCLA Department of Ophthalmology, Jules Stein Eye Institute, 100 Stein Plaza, Los Angeles, CA 90095; E-mail: arnolda@ucla.edu Kho et al: J Neuro-Ophthalmol 2011; 31: 121-126 121 Original Contribution Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. RESULTS Computer record search revealed 56 cases of toxic optic neuropathy. Twenty-two cases of ethambutol-induced optic neuropathy were detected. Nineteen cases (38 eyes) with sufficient documentation of clinical and neuroradiologic data were included for this study (Table 1; Figs. 1-3). Age ranged from 23 to 84 years; 15 patients were women and 4 were men. Ethambutol dosage ranged from 500 to 1200 mg/day, with duration 3-24 months (mean, 9.1; median, 6.0) prior to documented visual loss; additional antitu-berculous medications were used in all cases, including clarithromycin in 11, rifampin in 11, isoniazid in 7, and azithromycin in 2. Thirty-six eyes (95%) demonstrated visual field loss worse in the temporal hemifields. Thirty-one of 38 eyes (82%) showed visual field loss with some degree of margination along the vertical midline, of which 26 had superimposed central or cecocentral scotomas. Six cases (12 eyes) (32%) revealed bitemporal visual field loss mar-ginating along the vertical midline without superimposed central or cecocentral scotomas. Visual improvement occurred (of 17 cases with data available) by at least 3 Snellen lines in 17 of 34 eyes (50%); mean visual acuity improvement was 3.74 lines (median, 3.0 lines). Visual improvement by at least 3.0 decibels (dB) mean deviation (MD) on automated perimetry occurred TABLE 1. Clinical features of 19 cases with ethambutol-induced optic neuropathy Patient no. Age/Sex VA1 VA2 TX-VA1, m D/C-VA2 Fundus VF 1 Pattern MD1 (dB) MD2 (dB) 1 69/M 20/30 20/20 10 13 NL Temp -8.78 -1.99 20/80 20/20 1-2+ OA Temp -15.71 -5.22 2 76/F 20/100 20/200 6 3 Equiv OA Temp -15.59 - 20/80 20/200 Equiv OA Temp -15.77 3 84/F 20/400 20/40 5 14 NL Temp -17.03 -6.04 CF 20/60 Tr OA Temp + cent -18.83 -9.99 4 63/F 20/200 20/80 12 8 2+ OA Temp -10.13 -5.47 20/200 20/100 2+ OA Gen contr -12.53 -5.54 5 78/F 20/400 20/60 3 6 NL Temp -17.28 -6.88 CF 20/400 NL Temp + cent -16.44 -7.04 6 77/F 20/300 20/100 9 3 NL Temp -8.21 -4.71 20/400 20/80 NL Temp -8.28 -5.62 7 71/F CF 20/200 24 12 Equiv OA Temp + cent -29.54 -8.63 CF 20/200 Equiv OA Temp + cent -26.04 -10.37 8 84/F 20/60 20/25 15 7 NL Sup temp -6.89 -3.24 20/80 20/30 NL Sup temp -5.78 -2.22 9 69/M 20/30 20/25 9 2 NL Sup temp -5.36 -3.73 20/80 20/60 NL Temp -9.92 -6.99 10 23/F - 20/100 4 25 1-2 + OA Temp + mild nasal -25.33 -10.91 20/100 2 + OA Temp + mild nasal -26.41 -10.82 11 66/F 20/200 20/30 4 4 NL Temp scot -1.21 +0.44 20/200 20/40 NL Temp scot -1.66 -0.01 12 72/F 20/80 20/30 4 3 NL Temp + cent -7.72 -2.35 CF 20/40 Tr OA Temp + cent -9.70 -2.84 13 76/F 20/300 20/40 9 26 Equiv OA Temp -14.49 -3.97 20/400 20/40 Equiv OA Temp + cent -12.53 -4.69 14 64/M 20/200 20/50 5 11 NL Temp + cent -10.53 -2.15 CF 20/60 Equiv OA Cent + gen -12.44 -1.16 15 69/F 20/200 20/200 16 5 1-2 + OA Temp + cent -10.53 -3.87 20/400 20/100 1 + OA Temp + cent -13.14 -5.27 16 75/F 20/400 - 24 3 1-2 + OA Temp -17.52 - 20/400 1 + OA Temp + mild nasal -22.54 17 56/F 20/200 20/80 4 120 1-2 + OA Temp -12.89 -13.43 CF 20/400 2-3 + OA Temp + cent -17.67 -16.65 18 59/F CF 20/400 6 18 1 + OA Sup temp -11.78 -3.21 CF CF 1 + OA Temp -15.48 -4.05 19 76/M 20/400 20/50 3 16 NL Temp + cent -13.64 -0.45 CF 20/50 NL Temp + cent -11.43 -0.22 VA, visual acuity; TX-VA1, time from treatment to initial VA; D/C-VA2, time from discontinuing medication to final VA; VF, visual field; CF, count fingers; OA, optic atrophy; temp, temporal defect; cent, central defect; contr, contraction; sup, superior defect; scot, scotoma; gen, generalized; NL, normal; Equiv, equivocal; Tr, trace; MD, mean deviation. 122 Kho et al: J Neuro-Ophthalmol 2011; 31: 121-126 Original Contribution Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. in 27 of 34 eyes (79%); mean improvement in MD was 7.82 dB (median, 7.86 dB). Time from discontinuing ethambutol to final visual measurement ranged from 2 to 120 months (mean, 15.7 months; median, 8.0 months) after discontinuing ethambutol. DISCUSSION Ethambutol-induced optic neuropathy has been reported to most commonly result in bilateral and relatively symmetric central or cecocentral visual field defects. While typical FIG. 1. Case 1. Top: Initial automated perimetry demonstrates bitemporal visual field defects with alignment on vertical midline in each eye. Bottom: Follow-up automated perimetry 13 months later shows near complete resolution of defect in the right eye and partial resolution in the left eye. Kho et al: J Neuro-Ophthalmol 2011; 31: 121-126 123 Original Contribution Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. FIG. 2. Case 19. Top: Initial automated perimetry demonstrates bitemporal visual field defects with alignment on vertical midline in both eyes. There are superimposed central scotomas in the left eye more severe than in the right eye. Bottom: Follow-up automated perimetry 12 months later shows near complete resolution of field defects in both eyes. 124 Kho et al: J Neuro-Ophthalmol 2011; 31: 121-126 Original Contribution Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. cecocentral scotomas may mimic field loss from compressive chiasmal damage, they are usually readily distinguishable from the latter in their lack ofmargination along the vertical midline and their restriction to the central portion of the temporal visual fields. Rarely, focal posterior chiasmal compression may produce bitemporal hemianopic central scotomas. Bitemporal visual field defects that respect the vertical midline are con-sidered to result from preferential damage of crossing chiasmal fibers, most commonly from compressive lesions. Twenty-two anecdotal cases of bitemporal visual field defects secondary to ethambutol toxicity have been pre-viously reported (1-15). These present a mix of perimetric techniques and results including both large cecocentral scotomas and clear bitemporal defects aligning on the vertical midline. To our knowledge, no other medication currently in use has been associated with this pattern. We add 19 cases in which high-resolution neuroimaging ruled out idiopathic or infectious chiasmal inflammation, tuberculoma, and unrelated compressive lesion and in which quantitative static perimetry allowed precise analysis of defects. In our cases, the most common pattern of field loss was bitemporal depression with some degree of margination on the vertical midline and superimposed cecocentral defect (Figs. 2, 3). Six cases showed no super-imposed scotomas and were highly suggestive of chiasmal injury (Fig. 1). Although sampling bias may have played a part, in this series, the predominant pattern of loss was not pure cecocentral or central scotoma but varying patterns of bitemporal loss. Artifactual margination of visual field defects along both vertical and horizontal meridians has been documented with theHumphrey perimeter, presumably due to the algorithm for initialization by quadrant (16). It is possible that this factor contributed, in some cases, to the prominent margination seen in our cases. However, the correlation of these cases with similar features previously reported with various perimeters and with the experimental data described belowsuggests that it represents more than artifact.Moreover, in 1 case (Fig. 3; Case 17), Octopus perimetry, which does not rely on the same initializing algorithm and is not susceptible to this artifact, revealed similar margination. Experimental animal studies by several investigators have confirmed histopathologic evidence of central chiasmal damage as an early sign of ethambutol toxicity (16-20). A single human study of brain and eye histopathology also indicated focal chiasmal demyelination (16). No study to FIG. 3. Case 17. Initial Octopus automated perimetry illustrates bitemporal visual field defects with alignment along vertical midline and superimposed central scotoma in the left eye. Kho et al: J Neuro-Ophthalmol 2011; 31: 121-126 125 Original Contribution Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. date, however, has confirmed the etiology for specific in-volvement of crossing fibers. Lessell (19) postulated factors such as 1) selective deposition in chiasm and adjacent optic nerves and 2) local variations in microvascular supply, contact with cerebrospinal fluid, enzymatic activity, and glial cell populations, any of which might predispose to chiasmal damage. His subsequent study suggested that chiasmal glial subpopulations did indeed differ from those in more distal optic nerves (21). Our findings corroborate these studies. Several authors have addressed evidence that etham-butol may be directly toxic to the retina. Heng et al (22) studied ethambutol toxicity in rats and concluded that ganglion cells were specifically susceptible via an ex-citotoxic (glutamate-mediated) pathway. Zoumalan et al (23) also concluded that ganglion cell injury was a sig-nificant mechanism for visual loss from ethambutol, citing patterns of retinal nerve fiber loss as measured by optical coherence tomography. Lai et al (24), Kardon et al (25), and Liu et al (15) have reported abnormalities of the multifocal electroretinogram in cases of ethambutol-as-sociated visual loss, suggesting a primary injury to deeper layers of retina. Three of the reported cases demonstrated bitemporal visual field loss but were more in keeping with a cecocentral pattern. Cases in our series predominantly revealed a pattern more suggestive of preferential in-volvement of chiasmal crossing fibers, with more wide-spread visual field loss and with margination on the vertical midline. These findings in aggregate suggest that damage from ethambutol may involve several regions: deeper retinal layers (given reports of multifocal elec-troretinographic findings) papillomacular retinal nerve fibers (documented by optical coherence tomography) and crossing fibers within the chiasm (supported by bi-temporal visual field loss demonstrated in our cases). Previous reports vary regarding prognosis for visual recovery after discontinuation of the drug. In general, prognosis has been said to depend on the degree of optic nerve injury at discovery. In the only other large series of cases with follow-up data (26), 6 of 10 eyes (60%) showed visual acuity im-provement by at least 2 Snellen lines, with 7 of 9 (78%) showing visual field improvement.Our data in a larger number of cases indicate a similar favorable prognosis with relatively early detection. Some 50% showed visual acuity, and 79% showed visual field improvement. Available data on dose of ethambutol were insufficient to correlate with visual loss and recovery. REFERENCES 1. Carr RE, Henkind P. Ocular manifestations of ethambutol. Arch Ophthalmol. 1962;67:566-571. 2. Asayama T. Two cases of bitemporal hemianopsia due to ethambutol. Jpn J Ophthalmol. 1969;23:1209-1212. 3. Leibold JE. Drugs having a toxic effect on the optic nerve. Int Ophthalmol Clin. 1971;11:137-157. 4. Lessell S. Toxic and deficiency optic neuropathies. In: Smith JL, Glaser JS, eds. Neuro-Ophthalmology, volume 8. St Louis, MO: CV Mosby; 1973:21-37. 5. Bronte-Stewart J, Pettigrew AR, Foulds WS. Toxic optic neuropathy and its experimental production. Trans Ophthalmol Soc U K. 1976;96:355-358. 6. Arruga J. Le test de la ‘‘desaturation'' subjective de la couleur dans le diagnostic de l'atteinte de la voie optiques anterieure par l'ethambutol. Bull Soc Ophtalmol Fr. 1982; 82:189-192. 7. DeVita EG, Miao M, Sadun AA. Optic neuropathy in ethambutol-treated renal tuberculosis. J Clin Neuroophthalmol. 1987;7:77-83. 8. Fledelius HC, Petrera JE, Skjodt K, Trojaborg W. Ocular ethambutol toxicity. Acta Ophthalmol. 1987;65: 251-255. 9. Woung LC, Jou JR, Liaw SL. Visual function in recovered ethambutol optic neuropathy. J Ocul Pharmacol Ther. 1995; 11:411-419. 10. Tsai RK, Lee YH. Reversibility of ethambutol optic neuropathy. J Ocul Pharmacol Ther. 1997;13:473-477. 11. Sivakumaran P, Harrison AC, Marschner J, Martin P. Ocular toxicity from ethambutol: a review of four cases and recommended precautions. N Z Med J. 1998;111: 428-430. 12. Melamud A, Kosmorsky GS, Lee MS. Ocular ethambutol toxicity. Mayo Clin Proc. 2003;78:1409-1411. 13. Kocabay G, Erelel M, Tutkin IT, Ecder T. Optic neuritis and bitemporal hemianopsia associated with isoniazid treatment in end-stage renal failure. Int J Tuberc Lung Dis. 2006;10:1418-1419. 14. Lim S-A. Ethambutol-associated optic neuropathy. Ann Acad Med Singapore. 2006;35:274-278. 15. Liu Y, Dinkin MJ, Loewenstein JI, Rizzo JF, Cestari DM. Multifocal electroretinographic abnormalities in ethambutol-induced visual loss. J Neuroophthalmol. 2008; 28:278-282. 16. Stewart JF. Automated perimetry and malingerers. Can the Humphrey be outwitted? Ophthalmology. 1995;102: 27-32. 17. Schmidt IG. Central nervous system effects of ethambutol in monkeys. Ann N Y Acad Sci. 1966;135: 759-774. 18. Schmidt IG, Schmidt LH. Studies on the neurotoxicity of ethambutol and its racemate for the rhesus monkey. J Neuropathol Exp Neurol. 1966;25:40-67. 19. Lessell S. Histopathology of experimental ethambutol intoxication. Invest Ophthalmol. 1976;15:765-769. 20. Shiraki H. Neuropathy due to intoxication with anti-tuberculous drugs from neuropathological viewpoint. Adv Neurol Sci. 1973;17:120-125. 21. Lessell S. The histology and histochemistry of the rat's optic nerve and chiasm. Am J Ophthalmol. 1977;84: 681-688. 22. Heng JE, Vorwerk CK, Lessell E, Zurakowski D, Levin LA, Dreyer EB. Ethambutol is toxic to retinal ganglion cells via an excitotoxic pathway. Invest Ophthalmol Vis Sci. 1999; 40:190-196. 23. Zoumalan CI, Sadun AA. Optical coherence tomography can monitor reversible nerve-fibre layer changes in a patient with ethambutol-induced optic neuropathy. Br J Ophthalmol. 2007;91:839-840. 24. Lai TYY, Chan W-M, Lam DSC, Lim E. Multifocal electroretinogram demonstrated macular toxicity associated with ethambutol related optic neuropathy. Br J Ophthalmol. 2004;88:774-775. 25. Kardon RH, Morrisey MC, Lee AG. Abnormal multifocal electroretinogram (mfERG) in ethambutol toxicity. Semin Ophthalmol. 2006;21:215-222. 26. Lee EJ, Kim S-J, Choung HK, Kim JH, Yu YS. Incidence and clinical features of ethambutol-induced optic neuropathy in Korea. J Neuroophthalmol. 2008;28:269-277. 126 Kho et al: J Neuro-Ophthalmol 2011; 31: 121-126 Original Contribution Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. |