OBJECTIVE: Controlled trials have failed to show an effect of Mycobacterium vaccae immunotherapy on treatment outcome and mortality in patients with tuberculosis (TB); however, several studies have suggested improvement in radiographic clearing and resolution of cavitary disease. METHODS: To assess the effect of M. vaccae immunotherapy on radiographic healing in pulmonary TB, chest X-rays from three randomized placebo-controlled trials of M. vaccae given as a single injection during the first 2 weeks of treatment were interpreted by a single, masked assessor using a standard scheme. Endpoints were the overall degree of radiographic improvement or deterioration and changes in cavitary disease at the end of antituberculosis treatment and follow-up. RESULTS: Of 1018 patients (478 HIV-infected; 540 HIV-uninfected) with an end of treatment or end of follow-up X-ray analyzed, 496 received M. vaccae and 522 received placebo. There was no difference in radiographic improvement or deterioration or cavitary disease at the end of treatment or follow-up comparing the M. vaccae and placebo groups. Results were similar comparing HIV-infected and HIV-uninfected patients. CONCLUSION: Adjunctive immunotherapy of drug-susceptible pulmonary TB with M. vaccae during the first 2 weeks of treatment did not improve radiographic responses to treatment or resolution of cavitary disease.