To evaluate the effectiveness and safety of Mycobacterium vaccae (MV) in prevention of tuberculosis (TB) among high risk people.

Database of MEDLINE, EMBASE, BIOSIS, SCI, Cochrane Central Register of Controlled Trials, CBM, CNKI and VIP were searched till July 2009. Randomized controlled trials (RCTs) or non-randomized controlled clinical trials (CCTs) investigating MV as interventions in people at high risk of TB were identified for critical appraisal. Two reviewers independently performed data extraction and quality assessment. Effectiveness of MV was summarized in different group of risk people through RevMan 5.0 by The Cochrane Collaboration.

Thirteen studies were included. Risk difference (RD) of protection index (PI), its 95% confidence interval (95%CI) and the P value were as following: MV vs. Isoniazid (INH): 0.02 (−0.01, 0.05) (P=0.12); MV vs. (INH plus RFT): 0.00 (−0.00, 0.00) (P=1.00); MV vs. Blank: 0.04 (0.00, 0.08) (P=0.03) for soldiers with PPD strong positive; 0.00 (−0.00, 0.00) (P=0.05) for students with PPD strong positive; 0.20 (0.05, 0.36) (P=0.01) for aged people of clinical cured pulmonary TB, and 0.08 (0.01, 0.14) (P=0.03) for type 2 diabetes mellitus. In HIV-infected people, The Risk Ratio (RR) of MV vs. CV (control vaccine) of positive stimulation index (SI) (≥3) in lymphocyte proliferation assays (LPA) to Mycobacterium vaccae sonicate (MVS) was 2.39 with 95% CI (1.56, 3.66), P<0.0001. Immunization had no adverse effects on CD4 cell count or HIV viral load. The most frequent adverse effects of MV were induration and sore arm.

Available evidence shows that MV is effective in preventing TB in PPD strong positive/type 2 diabetes mellitus/aged people of clinical cured pulmonary TB, and is safe, well-tolerated and effective in inducing biologically relevant immune response against TB in HIV-infected patients. High-quality trials aimed at different groups of high risk people are encouraged.

Journal of Infection, volume 60, issue 5, pp 320-330