A 72-year-old smoker developed cough, hemoptysis, night sweats, weight loss, and fever. Culture of bronchoscopic specimens revealed an atypical mycobacterium. Treatment was initiated with ethambutol, mistakenly prescribed as 1,600 mg for 4 times a day (usual maximum dose for this 70-kg man should have been 1,200 mg daily), along with azithromycin and moxifloxacin. At that time, the patient had no visual symptoms, and baseline ophthalmologic examination, including perimetry, was normal. Renal function also was normal.
Eleven weeks later, the patient developed blurry vision and visual “rippling” in the superotemporal visual field of the left eye. Ethambutol overdose was recognized, and ethambutol, moxifloxacin, and azithromycin were discontinued. The patient developed tingling in his feet, a squeezing sensation in his feet and calves, and gait unsteadiness.
Visual acuity was counting fingers in each eye, without a relative afferent pupillary defect, and he was unable to recognize any of the Ishihara color plates. Automated perimetry showed bitemporal visual field loss with central depression bilaterally (Fig. 1A). The remainder of the examination, including ophthalmoscopy, was normal. Neurologic testing demonstrated diminished vibration sensation in both feet, with intact light touch and proprioception, negative Romberg sign, normal gait and station, and silent plantar responses. Deep tendon reflexes (DTRs) were 1+ in the upper extremities, trace at the knees, and absent at the ankles. Nerve conduction studies (NCS) were normal. Visual evoked potentials showed markedly prolonged latency bilaterally. Magnetic resonance imaging (MRI) showed no abnormalities of the anterior visual pathways.
Eight months after stopping ethambutol, vision was 2/200 in each eye. Visual fields were somewhat improved (Fig. 1B), and MRI showed increased signal in the optic chiasm, extending into the optic tracts (Fig. 1B). The patient had diffuse hyperreflexia, with a bilateral Hoffman sign and nonsustained clonus at the ankles, although plantar responses were flexor. NCS showed predominantly axonal sensory neuropathy. Vibration quantitative sensory testing (VQST) revealed markedly elevated threshold of 7.6 vibration units (normal <3.4 vibration units).
Six months later, visual acuity had improved to 20/60 in the right eye and 20/160 in the left eye. Automated visual field testing showed continued improvement (Fig. 1C), and MRI disclosed near-complete resolution of the hyperintensity in the anterior visual pathways (Fig. 1C). Sensory examination was unchanged, but reflexes were normal. NCS showed slight improvement, and VQST threshold had improved to 3.76 vibration units.
Twenty-six months after discontinuing ethambutol, visual acuity was 20/20 bilaterally, and there was bilateral optic disc pallor. Visual fields were near normal (Fig. 1D), and MRI was unremarkable. Acral numbness and paresthesias persisted, and NCS findings were unchanged.
Our patient received ethambutol at 5 times the usual maximum dose for 11 weeks, an overdose of greater magnitude than has been reported previously, and consequently he was affected earlier and more severely than patients in previous reports. He noted visual loss 11 weeks after beginning treatment, whereas ethambutol in standard doses produces visual loss after a mean duration of 7.31 months of treatment (1).
As is common in neurotoxicology, our patient's findings were symmetric, and symptoms continued to worsen transiently following removal of the offending substance (“coasting”), after which symptoms and neuroimaging abnormalities gradually, but incompletely, improved. The initial examination of our patient did not show hyperreflexia or clonus, whereas subsequent examinations did. Similarly, the initial MRI was unremarkable but ultimately showed hyperintense signal abnormalities in the optic chiasm and proximal optic tracts. MRI findings within the optic chiasm have only recently been reported (2), and our report supports this observation.
In addition to anterior visual pathway involvement, our patient manifested evidence of corticospinal dysfunction including hyperactive DTRs, ankle clonus, and Hoffman sign, which subsequently resolved. Such findings are consistent with pyramidal tract demyelination, which has been observed in monkeys given high doses of ethambutol (3). The predominantly sensory axonopathy seen in our patient is typical of a polyneuropathy induced by ethambutol (4).
1. Lee EJ, Kim SJ, Choung HK, Kim JH, Yu YS. Incidence and clinical features of ethambutol-induced optic neuropathy in Korea. J Neuroophthalmol. 2008;28:267–277.
2. Osaguona VB, Sharpe JA, Awaji SA, Farb RI, Sundaram AN. Optic chiasm involvement on MRI with ethambutol-inducted bitemporal hemianopia. J Neuroopthalmol. 2014;34:155–158.
3. Schmidt IG, Schmidt LH. Studies of the neurotoxicity of ethambutol and its racemate for the rhesus monkey. J Neuropathol Exp Neurol. 1966;25:40–67.
4. Spencer PS. Ethambutol. In: Spencer PS, Schaumburg HH, Ludolph AC, eds. Experimental and Clinical Neurotoxicology. Oxford, United Kingdom: Oxford University Press, 2000:539–541.