BACKGROUND: Mortality rates of HIV-infected patients treated for tuberculosis remain high. This study aimed to assess the effect on mortality of immunotherapy with single-dose SRL172 added to standard antituberculosis chemotherapy in such patients. METHODS: The double-blind trial enrolled 1229 patients aged 18-60 years, who had never received antiretroviral treatment and who presented with newly diagnosed, sputum-smear-positive pulmonary tuberculosis to referral centres in Lusaka, Zambia, and Karonga, Malawi. Both HIV-positive and HIV-negative patients were enrolled, to avoid stigmatisation. Participants were randomly assigned a single injection of SRL172 or matching placebo within 2 weeks of starting 8 months of antituberculosis chemotherapy and followed up for at least 12 months. The primary endpoint was time to death in the HIV-infected population. Analyses were based on 760 HIV-positive patients after exclusion of 84 patients with errors in storage of the injection, no bacteriological confirmation, or no HIV result. FINDINGS: Of 760 HIV-infected patients, 374 received SRL172 and 386 received placebo. SRL172 did not cause any serious adverse events. The follow-up rate was 88% at 12 months in both groups. Of the HIV-positive patients, 109 (19.5 per 100 person-years) of 372 assigned SRL172 and 107 (19.3 per 100 person-years) of 386 assigned placebo died. In the Cox's regression analysis, stratified by centre, the hazard ratio of deaths (SRL172/placebo) was 1.03 (95% CI 0.79-1.35). There was no evidence of benefit to the group assigned SRL172. INTERPRETATION: Immunotherapy with single-dose SRL172 as an adjunct to standard antituberculosis treatment in HIV-positive adults with pulmonary tuberculosis had no significant effect on survival or bacteriological outcome, though the treatment was safe and well tolerated.