This natural history study seeks to determine some of the mycobacterial and host factors involved in the failure of antituberculous chemotherapy, disease recurrence, and the development of multidrug resistance by M. tuberculosis. Despite optimal treatment with directly-observed short-course therapy (DOTS), about 5-10% of compliant patients with (cured) tuberculosis relapse, usually within a year after completion of therapy. In Korea, where DOTS is not practiced, the relapse rate has been reported to be 15 to 20 %. In individual patients, failure to eradicate disease contributes directly to the development of drug-resistance and to low overall cure rates. While factors such as patient drug compliance and HIV status have been extensively studied in relation to rates of relapse; host genetic factors and the specific relevance of the infecting mycobacterial strain have not yet been investigated in detail. Identification of patient characteristics and specific strains of M. tuberculosis that are associated with relapse and the evolution of drug resistance would greatly facilitate the development of treatment protocols that might avert these complications.

Our study population will consist of patients with pulmonary tuberculosis receiving treatment at the National Masan Tuberculosis Hospital (NMTH), the largest tertiary referral center for patients with previously treated and drug-resistant tuberculosis in the Republic of Korea. The hospital also serves as a site where the local population (approximately 1,000,000) is treated as inpatients for new cases of tuberculosis. Healthy volunteers will be selected as the genotypic control population. All patients withnew cases of tuberculosis enrolled in the study will be followed by periodic chart review and data extraction during their treatment by NMTH. For 2.5 years after completing drug treatment, patient recurrence will be queried using the recently implemented tuberculosis registration system to determine the 2 year recurrence rate for TB, and the patient's M. tuberculosis isolates will be cultured upon readmission and analyzed to distinguish between relapse and re-infection.

Study patients will be asked to volunteer to have 20 mL of blood collected by hospital staff during normal admission testing, as well as a 5 to 30 mL sample of sputum for acid-fast bacilli (AFB) smear and culture. Patients will have their sputum cultured for M. tuberculosis by the hospital diagnostic laboratory and isolates will be tested for drug resistance (DR). The research staff will use molecular DR tests to confirm the hospital's agar-growth DR results. In addition, subjects will be asked a series of medical history questions including history of prior tuberculosis, antituberculous treatment, disease contacts, and risk factors associated with tuberculosis; and will be asked to give consent to allow clinical research staff to abstract treatment regimens and results from their inpatient and outpatient medical charts for the duration of treatment.