BACKGROUND: Individuals infected with non-tuberculous mycobacteria may elicit false-positive reactions on tuberculin skin testing. The QuantiFERON-TB (QFT) assay utilizes tuberculin and M. avium antigens and, therefore, may be more specific for latent tuberculosis infection. The objective of this study was to investigate the agreement between the QFT and single and dual antigen skin testing for detecting latent M. tuberculosis and assess the impact of cross-reactions from latent infection with other mycobacteria in inner-city injection drug users, a population at high risk for tuberculosis. MATERIAL/METHODS: We studied the agreement of results from skin testing using tuberculin and purified protein derivative-Battey (PPD-B) with the QFT test using tuberculin and Mycobacterium avium sensitin (MAS) in 48 HIV-seronegative injection drug users. RESULTS: The agreement between skin testing and the QFT assay for tuberculin was 73% (kappa = 0.45) and for PPD-B/MAS was 63% (kappa = 0.12). Agreement between skin test tuberculin dominance (tuberculin reaction > or =5 mm greater than PPD-B) and QFT tuberculin dominance (proportional difference between MAS and tuberculin reaction of > or =10%) was 75% (kappa = 0.53). All subjects tuberculin dominant by skin test were also QFT positive for tuberculin. Agreement between skin test Battey dominance and QFT avium dominance was 83% (kappa = 0.12). CONCLUSIONS: Results from the QFT assay and skin testing demonstrated moderate concordance in identifying subjects with latent tuberculous infection, and use of dual antigens did not appreciably improve the agreement between the two methods.

Keywords: Tuberculosis - immunology, Tuberculin Test - statistics & numerical data, Tuberculin - immunology, Substance Abuse, Intravenous, Sensitivity and Specificity, Mycobacterium avium Complex - immunology, Mycobacterium Infections - immunology, Leukocytes, Mononuclear - immunology, Interferon-gamma - blood, Immunologic Tests - statistics & numerical data, Cross Reactions, Antigens, Bacterial