The influence of intravenous drug use and HIV infection in the transmission of tuberculosis

To investigate whether smear-positive pulmonary tuberculosis (TB) patients present a different risk of TB transmission according to their HIV status, in an area where the majority of HIV-infected patients studied were intravenous drug users (IVDU). A case-control study was performed on smear-positive pulmonary TB patients diagnosed between 1990 and 1993 for whom a contact study had been performed. Patients with and without HIV infection were matched by age (+/- 3 years), sex and hospital of diagnosis. A micro-epidemic was defined if two or more secondary cases were detected from the same index case. Data were analysed comparing the percentage of contacts with TB in both groups. Thirty-six secondary cases were detected in 436 contacts of 124 HIV-infected TB patients, whereas only 24 were identified in 624 contacts of 124 HIV-seronegative TB patients [odds ratio (OR), 2.14; 95% confidence interval (CI), 1.22-3.77; P = 0.004]. Comparing the contacts of HIV-infected patients, 34 secondary cases of TB were detected in 371 contacts of 97 IVDU, whereas only two cases were detected in 85 contacts of 27 non-IVDU (OR, 4.19; 95% CI, 1.09-15.95). HIV-seropositive index cases were observed to cause more micro-epidemics than seronegative cases (eight versus four), indicating that micro-epidemic cases were more frequent in the contacts of HIV-infected subjects (27 out of 36 versus 10 out of 24; OR, 1.41; 95% CI, 1.41-12.49; P = 0.009). All index cases in eight micro-epidemics were HIV-seropositive IVDU, whereas only four micro-epidemics were generated by HIV-seronegative subjects, none of whom were IVDU. Smear-positive pulmonary TB patients were more likely to transmit TB if they were HIV-infected, as evidenced by the role of IVDU in generating micro-epidemics.