A false-positive tuberculin skin test (TST) may be a result of T-cell sensitivity due to Bacille Calmette-Guerin (BCG) vaccination or exposure to non-tuberculous mycobacteria, thus leading to unnecessary isoniazid preventive therapy, especially in low-risk populations. Unlike TST, T-SPOT.TB is not confounded by BCG vaccination or exposure to most of the other non-tuberculous mycobacteria, because this assay is based on enumeration of interferon-gamma-secreting T cells in response to Mycobacterium tuberculosis-specific antigens. We compared the TST with T-SPOT.TB with respect to different TST cut-off points in healthy unexposed BCG-vaccinated schoolchildren. A total of 209 children between 6 and 10 years of age with a TST induration of 0 (n = 50), 10 - 14 (n = 45), 15 - 19 (n = 95) and > or =20 mm (n = 19) were enrolled. Among TST-positive subjects, only 26 (23%) were positive with T-SPOT.TB, and T-SPOT.TB was positive in 4, 7, 20 and 42% of children with TST indurations of 0, 10 - 14, 15 - 19 and > or =20 mm, respectively. We suggest that confirmation of a positive TST by the interferon-gamma-based test would reduce unnecessary preventive therapy significantly in healthy unexposed BCG-vaccinated children.