Ocular Toxicity of Ethambutol
Ethambutol hydrochloride is one of the first line agents
employed in the treatment of tuberculosis, which remains
an important infectious disease in Hong Kong, classified
by the World Health Organization as "place with
intermediate burden and a good health infrastructure".
Being a commonly used drug, ocular toxicity of
ethambutol has been described since its first use in
treatment of tuberculosis in 1960s1,2, manifesting as optic
neuritis in affected individuals. Data regarding this
potential side effect have been published, yet much
controversy remains, especially on the issue of prevention
of ocular toxicity of ethambutol.
The article aims to give a summary of literature published
on ethambutol ocular toxicity - background, clinical
presentation, toxicity characteristics, management,
monitoring and preventive measures.
We have performed a search in MEDLINE from 1962
through 2005, using the search formula: (ethambutol OR
myambutol) AND ( eye* OR ophthal* OR ocular ) AND (
adverse OR toxic ). All relevant publications in English
A lthough uncommonly seen in patients on standard
dosage, optic neuritis is the most important potential side
effect of ethambutol hydrochloride. Retrobulbar neuritis is
most common, with involvement of either axial fibres or
less commonly, periaxial fibres. Mixed pattern is possible.
3,4 Other rarer side effects of ethambutol include peripheral
neuropathy, cutaneous reactions (rash, pruritis, urticaria
etc.), thrombocytopenia and hepatitis. 5, 6
Exact mechanism of ocular neurotoxic effect of ethambutol
is yet to be identified. Animal studies have demonstrated
ethambutol toxicity on retinal ganglion neurons in rodents.
Hypotheses for its toxicity have been made, including zinc-
chelating effect of ethambutol and its metabolite 7, 8, 9, and
excitotoxic pathway 10.
The onset of ocular symptoms is usually delayed,
occurring months after therapy initiation, though rare
cases of toxicity occurring few days after initiation have
been reported, one on standard dosage of 15mg/kg/day,
another on 25mg/kg/day.11,12 No study has reported onset
after stopping ethambutol. Clinical course can be acute or
chronic and typically, progressive.
A ffected individuals may complain of bilateral
progressive painless visual blurring. Decreased colour
perception may also be experienced. Central vision is
most commonly affected, though other visual field loss
has also been described. Some individuals may be
asymptomatic with abnormalities detected only by vision
On physical examination, both eyes are usually affected
symmetrically. Physical findings can be very variable.
Pupils may be bilaterally sluggish to light with no relative
afferent pupillary defect. V isual acuity drop varies
greatly from nil or minimal reduction to no light
perception. Central scotoma is the most common visual
field defect, but bitemporal defects13 or peripheral field
constriction have been reported. Dyschromatopsia
(abnormal colour perception) may be the earliest sign of
toxicity14, classically documented to be red-green colour
changes. In contrary, the report by Polak et al stated that
blue-yellow defects were the most common and early
defect in patients without any visual symptoms15.
However, the subtle blue-yellow defects could only be
detected using the generally unavailable desaturated
panel of Lanthony and not Ishihara charts nor
Farnsworth-Munsell D-15 test. Fundoscopic examination
is usually normal.
Characteristics of ocular toxicity of
Classically, the ocular toxicity is described as dose and
duration related, and is largely reversible on drug
discontinuation. However, the issue of reversibility is
challenged by many recent studies.
Studies gave a reported incidence of ethambutol related
retrobulbar neuritis of 18% in patients receiving
>35mg/kg/day, 5-6% with 25mg/kg/day and <1% with
15mg/kg/day of ethambutol HCL for more than two
months.3,16No "safe dosage" for ethambutol was reported17,
with toxicity observed at a dose as low as 12.3 mg/kg18. In
Hong Kong, the usual daily dosage is 15mg/kg for adults
and children, therefore, ethambutol ocular toxicity is not
commonly seen in Hong Kong. As the major excretion
pathway of ethambutol is by the kidneys, patients with
poor renal function are at higher risk of ocular toxicities.
Other factors predisposing subjects to ethambutol toxicity
include diabetes and optic neuritis related to tobacco and
alcohol. 19, 20
Manifestation of ocular toxicity is usually delayed, which
Dr. Alvin Kwok MD(HKU), MD(CUHK), FRCS(Edin), FCSHK, FCOphthHK,
?FHKAM(Ophthalmology), PDip Biostat & Epidem(CUHK), MBBS(HKU)
President, Hong Kong Ophthalmological Society
Dr. Alvin Kwok
VOL.1 1 NO.2 FEBRUARY 2006
generally does not develop until after treatment for at
least 1.5 months.13 Variable mean interval between onset
of therapy and toxic effects were reported, from three to
five months.3, 17, 21 Manifestations of toxicity as late as 12
months after therapy initiation were reported.22,23
However, these report series were of small scales with
Views on the issue of reversibility of ethambutol toxicity
are divided. Although classically described as reversible
on discontinuation of ethambutol (with visual acuity
recovery over period of weeks to months), studies have
reported permanent visual impairment without recovery
in some patients w ith prompt ethambutol
discontinuation, within a follow up period ranging from
six months to three years. 13, 17, 20, 21, 23 No risk factor was
identified for the poor visual recovery, although a study
showed statistically significant difference in visual
recovery between groups of over 60 years old and below
60 years old.17 Even in patients with visual improvement
after therapy discontinuation, complete recovery was not
always achieved.17,23 Progressive worsening of vision after
ethambutol discontinuation was also documented.23
However, these report series are again of small scale with
unknown generalisability, and only patients with severe
visual deficits were recruited. Furthermore, queries were
made on the possible contribution of isoniazid induced
ocular toxicity to the observed irreversibility, as isoniazid
is not withdrawn together with ethambutol in the affected
Upon recognition of ethambutol induced ocular toxicity,
the drug must be immediately discontinued and patient
referred to ophthalmologists for further evaluation.
Therapy discontinuation is the only effective
management currently, which can stop the progression of
vision loss and allow recovery of vision. Some authors
recommend that when severe ocular toxicity occurs, both
isoniazid and ethambutol are to be stopped immediately
and consider adding other anti-tuberculous agents to
control tuberculosis. Isoniazid should also be stopped 6
weeks after stopping ethambutol if ocular toxicity is not
severe but with no vision improvement. 23
Recommendations on monitoring
and preventive measures
Several international guidelines have been published to
suggest measures for prevention and early detection of
ethambutol induced ocular toxicity. In August 2002, a
Tuberculosis and Chest Service of the Department of
Health of Hong Kong Special A dministrative Region
(Table 1), based upon available clinical information,
international guidelines and local experts' experiences25, 26, 27.
However, there is no well-established agreement on the
24 on this issue was published by the
The Joint Tuberculosis Committee of the British Thoracic
Society (Table 2)25 and the American Thoracic Society26
recommend routine visual acuity assessment prior to
starting ethambutol, but no longer recommend visual
acuity assessment during follow up. The A merican
Thoracic Society also recommends performing assessment
of red-green colour perception test prior to treatment.
Hong Kong's guideline recommends baseline vision test
for both visual acuity and red-green colour perception by
the use of Snellen chart and Ishihara chart, respectively24
which do not require ophthalmologist consultation.
Another unsettled issue is the clinical effectiveness of
regular visual tests to achieve early ocular toxicity
detection. Most guidelines do not recommend regular
visual acuity assessment25, 26, where visual acuity may be
normal in the early stages. Hong K ong's guideline
suggests that regular visual acuity test may be
considered in patients with risk factors, especially with
high dose ( 25mg/kg/day ) or prolonged treatment with
ethambutol. However, with increasing evidence of
comparatively better sensitivity of colour vision defect in
early toxicity detection3,14,15, regular colour vision test
with Ishihara chart may be considered instead for high-
risk patients, without the need of ophthalmologist
referral. A descriptive study of colour vision test in 42
patients with systemic tuberculosis receiving
ethambutol28, however, showed that 15 of 42 patients with
high total error scores at the Farnsworth-Munsell 100 test
had normal colour vision measured by Ishihara
pseudoisochromatic plates. The sensitivities of different
colour vision tests are yet to be determined.
With the usual daily dosage of 15mg/kg/day for patients
under the care of Tuberculosis and Chest Service in Hong
Kong5, incidence of ethambutol induced ocular toxicity is
below 1%3,16. Given the low incidence of toxicity and the
yet unknown sensitivity of regular visual tests in early
toxicity detection, the cost effectiveness of regular vision
tests (even colour vision test) on patients on dosage of
15mg/kg/day is questionable.
It is unlikely that any guidelines will completely remove
the risk of optic neuritis with ethambutol. Many a time,
the course of ethambutol induced ocular toxicity is
unpredictable. Measures to ensure a high level of
awareness of this potential adverse effect in both medical
staff and patients seems to be the best current preventive
method available. Medical staff should regularly ask
about vision change in patients taking ethambutol and
provide education to ensure that all patients know to stop
ethambutol immediately and seek prompt medical advice
if visual symptoms occur.
Being one of the safest first-line anti-tuberculous agents,
ethambutol HCl is commonly prescribed for patients with
tuberculosis. Optic neuritis is a rare, yet, most important
side effect of ethambutol, which the mechanism of
toxicity is still under investigation. This ocular toxicity is
dose and duration related3. Though classically described
as reversible, irreversibility of vision change was also
reported in several case series13,17,21,23. A lthough
international guidelines on prevention and early
detection of ethambutol induced ocular toxicity have
been published, views on use of regular vision tests for
early toxicity detection are still divided.
Classified by World Health Organization as a place with
intermediate tuberculosis burden and good health
infrastructure, Hong Kong is in a good position to look
into the unanswered questions of ethambutol induced
ocular toxicity, where studies on reversibility of ocular
toxicity, sensitivity of various types of vision tests in early
ocular toxicity detection, cost effectiveness of vision
monitoring as compared with patient education alone can
VOL.1 1 NO.2 FEBRUARY 2006
Upon commencement of anti-TB treatment, patients should be
feasibility and contraindications of using EMB. In situations where
there is an increased risk of ocular toxicity, the benefit of using EMB
should be carefully balanced against its risk. The availability, efficacy
and toxic profile of alternate drugs should be taken into account in
the choice of an effective treatment regimen. EMB may be
contraindicated or dosage reduction may be indicated in the some
Impaired baseline vision may make visual monitoring difficult.
However, in conditions like refractive error, which is correctable
with the use of spectacles, and mild cataract that is unlikely to
affect visual changes rapidly, continuous monitoring of vision can
be conducted during treatment with EMB. EMB should be avoided
in patients with significantly reduced vision.
Patients with difficulty in appreciating and reporting visual
symptoms or changes in vision, like young children, persons with
language difficulties, may also make visual monitoring difficult.
Impaired renal function can predispose to the development of
EMB-related ocular toxicity. Hence, renal function should be
checked upon commencement of anti-TB treatment.
Recommendations on dosage adjustment of EMB in the case of
renal impairment have been described in the recent local TB
For all patients undergoing treatment with anti-TB drugs that
includes EMB, health education should be provided to them on the
visual side effects of the drug and a high level of awareness of this
potential side effect should be emphasised during treatment. The
patients should be advised that, in case visual symptoms arise, the
drug should be stopped immediately and they should report
promptly to the health care staff. The offering of such advice to the
patients should be recorded in the medical notes. In case it is
necessary to prescribe EMB to young children or patients with
language difficulties, appropriate advice should similarly be given to
parents or other family members. The use of written instructions or
education pamphlets would be beneficial.
Baseline vision tests for visual acuity and red-green colour perception
(e.g., using Snellen chart and Ishihara chart) should be conducted
before starting treatment. There is controversy about the use of
regular visual test although this may be considered in certain patients
with risk factors, especially when a high dose (25 mg/kg/day, see
below) of EMB is used or the treatment is prolonged.
With normal renal function, the recommended daily dose for EMB is
15 mg/kg/day throughout the course of anti-TB treatment.5 However,
the use of a higher dose of 25 mg/kg/day may be considered in certain
conditions like severe cavitatory TB, drug-resistant TB, or retreatment
cases. This higher dose should not be given for more than two
months. Ideal body weight should be used in calculations for obese
During medical consultations in the course of anti-TB treatment
including EMB, all patients should be assessed clinically for
symptoms of visual disturbance. Enquiring monthly about visual
symptoms is advisable.
Directly observed treatment (DOT), apart from ensuring treatment
adherence, also allows health care workers to monitor the patients
closely for such symptoms.
Table 1. Recommendations in 'Preventive Measures Against
Drug-Induced Ocular Toxicity During A nti-tuberculous
Treatment (General Recommendations) [Extracted from
Annual Report (Suppl) 2002, Tuberculosis and Chest Service,
Department of Health, Hong Kong]
Special precautions and pretreatment screening point (1)
Because of the possible (but rare) toxic effects of ethambutol on the eye,
it is recommended that visual acuity should be tested by Snellen chart
before it is first prescribed. The drug should only be used in patients
who have reasonable visual acuity and who are able to appreciate and
report visual symptoms or changes in vision. The notes should record
that the patient has been told to stop the drug immediately if such
symptoms occur, and to report to the physician. The general
practitioner should also be informed of this. In small children and in
those with language difficulties ethambutol should be used where
appropriate, with the above advice given to parents or other family
Table 2. British Thoracic Society Guidelines - Chemotherapy
and management of tuberculosis in the United K ingdom:
*Joint Tuberculosis Committee of the British Thoracic Society.
Carr RE, Henkind P. Ocular manifestations of ethambutol. Arch
Ophthalmol 1962; 67:566-571.
2. Barron GJ, Tepper L, Iovine G. Ocular toxicity from ethambutol.
Am J Ophthalmol 1974; 77:256-260.
Citron K M, Thomas GO. Ocular toxicity from ethambutol
(editorial). Thorax 1986; 41:737-739.
Chen L, Liang Y. Optic nerve neuropathy by ethambutol toxicity.
Zhonghua Jie He He Hu Xi Za Zhi 1999; 22:302-304.
Consensus statement of Tuberculosis Control Coordinating
Committee of Hong K ong Department of Health and the
Tuberculosis Subcommittee of the Coordinating Committee in
Internal Medicine of the Hospital Authority of Hong Kong -
Chemotherapy of Tuberculosis in Hong Kong, 2001.
British National Formulary ed. 48, September 2004.
Shindler KS, Zurakowski D, Dreyer EB. Caspase inhibitors block
zinc-chelator induced death of retinal ganglion cells. Neuroreport
2000 14; 11:2299-2302.
Yoon YH, Jung KH, Sadun AA, Shin HC, Koh JY. Ethambutol-
induced vacuolar changes and neuronal loss in rat retinal cell
culture: mediation by endogenous zinc. Toxicol Appl Pharmacol
Kahana LM. Toxic ocular effects of ethambutol. Can Med Assoc J
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.
V.K.K. Chatterjee, D. R. Buchanan, A.I. Friedman, M. Green.
Ocular toxicity following ethambutol in standard dosage. Br J Dis
Chest 1986; 80:288-91
Harris S. Schild, Barry C. Fox. Rapid-onset reversible ocular
toxicity from ethambutol therapy. Am J Med 1991; 90:404-406.
Melamud A, Kosmorsky GS, Lee MS. Ocular Ethambutol Toxicity.
Mayo Clinic Proc 2003; 78:1409-1411.
Trusiewicz D. Farnsworth 100-hue test in diagnosis of ethambutol
- induced damage to optic nerve. Ophthalmologica 1975; 171:425-
Polak BCP, Leys M, van Lith GHM. Blue-yellow color vision
changes as early symptoms of ethambutol oculotoxicity
Ophthalmologica 1985; 191:223-226
Leiboid JE. The ocular toxicity of ethambutol ad its relation to
dose. Ann New York Acad Sci 1966; 135:904-909.
Tsai RK, Lee YH. Reversibility of ethambutol optic neuropathy. J
Ocul Pharmacol Ther 1997;13:473-477.
Choi SY , H wang JM. Optic neuropathy associated with
ethambutol in Koreans. Korean J Ophthalmol 1997; 11:106-110.
Murray, F.J. US Public H ealth Service experience with
ethambutol. International Congress of Chemotherapy 1967,
Vienna 6, 339.
Chuenkongkaew W, Samsen P, Thanasombatsakul N. Ethambutol
and optic neuropathy. J Med Assoc Thai. 2003 Jul;86(7):622-5.
Kumar A, Sandramouli S, Verma L, et al. Ocular ethambutol
toxicity: is it reversible? J Clin Neuroopthalmol 1993; 13:15-17.
Citron LM. Ethambutol: a review with special reference to ocular
toxicity. Tubercle 1969; 50(suppl):32-36.
P. Sivakumaran, Adrian C Harrison, et al. Ocular toxicity from
ethambutol: a review of 4 cases and recommended precautions.
NZ Med J 1998; 111:428-430.
Preventive measures against drug-induced ocular toxicity during
anti-tuberculous treatment (general recommendations) [Extracted
from A nnual Report (Suppl) 2002, TB & Chest Service,
Department of Health, Hong Kong]
Peter Ormerod, Ian Campbell, Vas Novelli, et al. Chemotherapy
and Management of Tuberculosis in the United K ingdom:
Recommendations 1998 of Joint Tuberculosis Committee of the
British Thoracic Society. Thorax 1998; 53: 536-548.
A merican Thoracic Society. Treatment of tuberculosis and
tuberculosis infection in adults and children. Am J Respir Crit Care
Med 1994; 149:1359-1374.
National Health and Medical Research Council. Tuberculosis in
Australia and New Zealand into the 1990s. Canberra: Australian
Government Publishing Service; 1989: p50-51.
Kaimbo WK, Bifuko ZA, Longo MB, Dralands L, Missotten L.
Color vision in 42 Congolese patients with tuberculosis receiving
ethambutol treatment. Bull Soc Belge Ophtalmol. 2002;(284):57-61.
VOL.1 1 NO.2 FEBRUARY 2006