How long are TB patients infectious?

CMAJ 2000;163(2):157

Tuberculosis (TB) is spread by the coughing up of minute droplets smaller than 2 µm. Suspension of these droplets as droplet nuclei necessitates the evaporation of any moisture in less than a fraction of a second. This causes the droplet nucleus to shrink to less than a thousandth of its original size. The concentration of anti-TB drugs in the saliva and bronchial secretions is the same as it is in the blood. With the evaporation of the moisture the dried-out tubercle bacillus in the droplet nucleus is exposed to a thousand-fold increase in the concentration of the drugs.

Schwartzman and Menzies quoted several papers by Richard Riley and his colleagues, dealing mainly with the infectiousness of untreated TB and the use of ultraviolet light in the control of infection. They failed to quote other papers by Riley and colleagues relating to the infectiousness of patients with TB once effective treatment is started.^2,3 Riley and colleagues found that the infectiousness of untreated patients with drug-susceptible organisms was much greater than that of patients on chemotherapy.

About the same time, Wallace Fox and coworkers showed that the tuberculin conversion rates of the close contacts of patients with open cavitary TB being treated with standard chemotherapy were the same regardless of whether the patients were treated in hospital or at home.^4–6 The only contacts who developed a positive tuberculin test or TB per se demonstrated a positive test either at the time of, or within 1 month of, diagnosis of the case. This implies they had inhaled tubercle bacilli before starting treatment and before the tuberculin test had time to convert. These observations made it clear that anti-TB therapy rendered patients virtually noninfectious within 2 weeks or so; it also persuaded most jurisdictions to eliminate compulsory segregation of subjects being treated for TB and removed the need for sanatoria.

Perhaps it will come as a shock to Schwartzman and Menzies to note the following statements in a highly regarded recent textbook: "for practical purposes patients can be regarded as being noninfectious two weeks after the start of treatment"⁷ and "only untreated patients with sputum positive pulmonary TB are likely to be infectious."⁸

D. Ahmad
Internist
London Health Sciences Centre
London, Ont.
W.K.C. Morgan
Respirologist (ret'd)
London, Ont.

Send a letter to the editor Envoyez une lettre à la rédaction

1. Schwartzman K, Menzies D. Tuberculosis: 11. Nosocomial disease. CMAJ 1999;161(10):1271-7. [MEDLINE]
2. Sultan LU, Nyka W, Mills CC, O'Grady F, Wells W, Riley RL. Tuberculosis disseminatory: a study of the variability of aerial infectivity of tuberculus patients. Am Rev Respir Dis 1960;82:358-69.
3. Riley RL, Mills CC, O'Grady F, Sultan Lu, Wittstadt S, Shuvpuri DN. Infectiousness of air from a tuberculosis ward. Am Rev Respir Dis 1961;83:511-25.
4. Fox W. The chemotherapy and epidemiology of tuberculosis. I. Lancet 1962;2:415-72.
5. Fox W. The chemotherapy and epidemiology of tuberculosis. II. Lancet 1962;2:473-7.
6. Fox W. The scope of the controlled clinical trial. Illustrated by studies of pulmonary tuberculosis. Bull World Health Organ 1971;45:559-72. [MEDLINE]
7. McNicol MW, Campbell IA, Jenkins PA. Clinical features and management of tuberculosis. In: Brewis A, Corrin B, Geddes DM, Gibson JG, editors. Respiratory medicine. Philadelphia: WB Saunders; 1995. p. 823.
8. Davies PDO. The control of tuberculosis. In: Brewis A, Corrin B, Geddes DM, Gibson JG, editors. Respiratory medicine. Philadelphia: WB Saunders; 1995. p. 833.

© 2000 Canadian Medical Association or its licensors