Effect of the tuberculostaticum ethambutol and stimulus intensity on chromatic discrimination in man

In goldfish it has been shown that ethambutol shifts the threshold for wavelength discrimination without affecting the absolute sensitivity of the cones. In this study we demonstrate that a similar colour vision disturbance occurs in tuberculosis patients treated with ethambutol. After 2 months of ethambutol treatment, chromatic discrimination was measured with a computerized forced two choice (CD) test with isoluminant coloured stimuli and with three other colour vision tests: the Ishihara, the Oscar and the Lanthony Desaturated 15 Hue tests. The scores of the patient group (n = 19) on these four colour vision tests were compared with the scores of a group of control subjects (n = 33) and a group of congenital red/green colour-blind subjects (n = 5). A reduction of the stimulus intensity of 1 log unit caused a significant reduction in red/green chromatic discrimination, measured with the CD test in both, control subjects and patients. This intensity dependent reduction was significantly greater for patients than for controls. In this respect, man and goldfish behave similarly. Furthermore, the CD test showed the same ethambutol-induced reduction in chromatic discrimination at low intensity for the blue/green part of the spectrum. This has not been measured in goldfish. The origin of this ethambutol-induced colour vision disturbance must be at a post-photoreceptor site, because the Ishihara and Oscar tests, both designed to screen for photoreceptor-based, or primary red/green colour vision disturbances, did not discriminate between patients and control subjects. Thus, as in goldfish, we find that in patients ethambutol shifts the threshold for chromatic discrimination without changing the absolute sensitivity.