Effect of treatment on contagiousness of patients with active pulmonary tuberculosis.

In view of the important consequences with regard to policies for respiratory isolation of hospitalized patients with active tuberculosis, the duration of contagiousness after initiation of effective therapy was reviewed. All relevant English-language literature was reviewed to identify in vitro, animal experimental, and epidemiological evidence regarding contagiousness of patients with active tuberculosis after initiation of therapy. Based on in vitro evidence of numbers of tubercle bacilli initially present and rapidity of reduction-with therapy, it can be predicted that patients whose respiratory secretions are initially smear-negative but culture-positive should no longer have viable bacilli detectable by culture within 2 weeks. Based on the same evidence, it can be predicted that, after 2 weeks of therapy, almost all smear-positive patients will remain culture-positive, and more than one half will remain smear-positive. There are few epidemiological studies of this issue, most of which have had major methodological weaknesses. None of the results from these studies can be considered relevant to the hospital environment, where the majority of workers are uninfected and patients are potentially immunocompromised. Animal and in vitro evidence suggest that patients with active tuberculosis remain contagious at least 2 weeks after the initiation of therapy. Patients with smear-positive disease are likely to be contagious much longer. There is no relevant and valid epidemiological evidence regarding this issue.