Skip to main content
  • By Michael Vaphiades, DO
    Neuro-Ophthalmology/Orbit

    The authors present the case of a 55-year-old man with pulmonary Mycobacterium avium intracellulare (MAI) infection who developed progressive visual loss due to presumed ethambutol optic neuropathy 11 months after starting on ethambutol, rifampin and isoniazid. After use of the medications was discontinued, the patient's visual acuity and visual field improved over the next 34 months yet he developed progressive bilateral optic disc cupping, disc pallor and diffuse nerve fiber layer loss on optical coherence tomography. The authors conclude that a disparate correlation between structure and function may occur during the recovery phase of ethambutol optic neuropathy, with visual improvement occurring despite progressive structural changes.

    At a baseline visual examination after initiation of ethambutol therapy, the patient had no visual symptoms. He was receiving daily 1800 mg ethambutol (26 mg/kg), 600 mg rifampin and 300 mg isoniazid for MAI infection. A follow-up visit was scheduled for six months later and the patient told to return immediately if he experienced vision loss.

    He did not return until one year later and had developed progressive visual loss over four months. He had continued ethambutol, isoniazid and rifampin, with the daily dosage of ethambutol averaging 24 mg/kg. His visual acuity had decreased to 3/200 in the right eye and 20/200 in the left eye. He could not identify any of the Ishihara color plates with either eye. Kinetic perimetry demonstrated bilateral central field loss. IOP was 18 mm Hg in the right eye and 21 mm Hg in the left eye. The optic discs were normal in appearance. OCT demonstrated normal retinal nerve fiber layer (RNFL) thickness in the right eye and moderate thinning in the left eye, with average thickness of 112.60 and 77.23 µm, respectively.

    He was evaluated for potential causes of bilateral optic neuropathy aside from ethambutol but none were found. The authors say that toxic effects from isoniazid and rifampin cannot be completely excluded. However, they say that ethambutol toxicity is implicated because the average dosage during the period that the patient was lost to follow-up exceeded the current preferred dosage of 15 to17 mg/kg per day.

    Thirty-four months after discontinuing the medications, the patient's vision had improved to 20/ 30 in the right eye and 20/70 in the left eye. IOP was 10 mm Hg in each eye. He was using travoprost once daily bilaterally. The optic discs were diffusely pale and cupped. OCT demonstrated marked diffuse RNFL thinning, with average thickness of 54.65 µm in the right eye and 50.34 µm in the left eye.

    The authors say that the progressive cupping in the patient coupled with decreased RNFL on OCT reflects the cumulative effect of nerve fiber loss in the retinal and prelaminar portion of the optic nerve head. Improvement in visual acuity as the nerve fiber layer progressively thinned suggests that while some axons suffered irreversible damage and underwent apoptosis, the function of the remaining axons improved as the toxic effect of ethambutol waned. Presumably some axons, including those in the papillomacular bundle, did not reach a threshold for apoptosis and were able to survive and recover function. They say that this case highlights the need for caution in predicting visual outcome from optic nerve appearance and OCT structural measures.