The impact of changing environmental factors--disease, nutrition and a high-dose multi-drug treatment regimen--on isoniazid (INH) elimination kinetics in children of both sexes and various ages was investigated. Thirteen children (mean age 2.3 years), hospitalized for the treatment of tuberculous meningitis, participated in the trial. Although all the children had protein-energy malnutrition, none had marasmus or kwashiorkor. After an oral dose of 20 mg/kg of INH, the concentrations in plasma were determined by the liquid chromatographic method of Lacroix et al. The 2-hour post-dose isoniazid concentration, the apparent first-order elimination rate constant and the corresponding INH half-life were determined in each child on two occasions 6 months apart. All comparisons were tested for significance using the Wilcoxon matched-pair signed-ranks test. There was no significant difference in any of the pharmacokinetic parameters of INH in our patients evaluated at the extremes of the 6-month term of treatment. It was apparent that changing conditions of disease and nutrition and a high-dosage, multi-component antimicrobial agent regimen over a 6-month period of treatment did not significantly influence INH elimination parameters. The trend evident in the pharmacokinetic profile of isoniazid in our children supports a trimodal distribution of acetylator phenotypes.