1. A multicentre study of the early bactericidal activity of anti-tuberculosis drugs.
- Author
-
Sirgel FA, Donald PR, Odhiambo J, Githui W, Umapathy KC, Paramasivan CN, Tam CM, Kam KM, Lam CW, Sole KM, and Mitchison DA
- Subjects
- Adult, Anti-Infective Agents pharmacology, Antibiotics, Antitubercular pharmacology, Colony Count, Microbial, Hong Kong epidemiology, Humans, India epidemiology, Isoniazid pharmacology, Microbial Sensitivity Tests, Ofloxacin pharmacology, Rifampin pharmacology, South Africa epidemiology, Sputum microbiology, Tuberculosis, Pulmonary epidemiology, Antitubercular Agents pharmacology, Mycobacterium tuberculosis drug effects, Tuberculosis, Pulmonary microbiology
- Abstract
The early bactericidal activities (EBAs) of 300 mg isoniazid, 18.5 mg isoniazid, 600 mg rifampicin and 800 mg ofloxacin given daily to 262 patients with newly diagnosed pulmonary tuberculosis in Cape Town, Nairobi, Madras and Hong Kong were measured by counting cfu and total acid-fast bacilli in sputum collections taken pre-treatment (S1), at 2 days (S3) and at 5 days (S6). In Cape Town, Nairobi and Madras, the cfu findings suggested that isoniazid produced a massive kill, perhaps of actively growing organisms, during the first 2 days (mean S1-S3 EBAs of 0.636-1.006) but was almost inactive thereafter (mean S3-S6 EBAs of 0.000-0.081), whereas rifampicin maintained moderate activity against slowly growing organisms throughout the 5 days (mean S3-S6 EBAs of 0.242-0.305). This finding suggests that EBAs measured during the 2-5 day interval might be able to assess the sterilizing activity of drugs. Ofloxacin had moderately high mean S1-S3 EBAs of 0.130-0.391. However, in Hong Kong rifampicin appeared to be the most bactericidal drug from the start, possibly because patients had more chronic disease. A method of adjusting the cfu EBAs using total counts was devised which decreased the variability between patients within a treatment group without altering the mean cfu EBA. This resulted in a large gain in precision in Hong Kong, suggesting that their estimates were greatly affected by type II variation, due to dilution of pus by saliva and bronchial secretions, whereas small or no gains were obtained in the other three centres, suggesting that the main cause of variability was type I, due to other factors.
- Published
- 2000
- Full Text
- View/download PDF