Effect of anti-tuberculosis treatment on the tuberculin interferon-gamma response in tuberculin skin test (TST) positive health care workers and patients with tuberculosis.

SETTING: Public hospital, Victoria, Australia. OBJECTIVE: To evaluate the effect of multidrug treatment and isoniazid (INH) chemoprophylaxis on the tuberculin interferon-y assay (QIFN) in 19 patients with culture-confirmed Mycobacterium tuberculosis and 119 health care workers (HCWs) with tuberculin skin tests (TST) > or =15 mm. DESIGN: Patients with M. tuberculosis were treated with standard medication and tested with QIFN at diagnosis and at regular intervals over a 12-month period. All HCWs, 59 (50%) of whom were prescribed INH chemoprophylaxis, were tested with QIFN at baseline, 2, 4, 6 and 12 months. RESULTS: QIFN results in patients with tuberculosis were consistent and reproducible. At the initial time point QIFN assays were positive for M. tuberculosis in 67%, and once positive, the QIFN assay remained so over the 12-month period. In the HCWS, initial QIFN assays were positive in 73 (61%). During the 12-month study, 91 HCWs had a QIFN assay on at least two occasions. The overall reproducibility between tests was fair (kappa statistic = 0.45), and was little affected by administration of INH. CONCLUSION: These data suggest that although the QIFN assay is generally positive in patients with proven tuberculosis, it does not provide clinically useful information during the first 12 months of treatment with multidrug chemotherapy or INH chemoprophylaxis.