BACKGROUND AND OBJECTIVE: Although QuantiFERON-TB Gold (QFT-G) has been approved for the diagnosis of latent tuberculosis infection (LTBI), there are limited data regarding its performance in routine clinical practice. The aim of this study was to compare QFT-G 'In Tube' results, based on stimulation with Mycobacterium tuberculosis-specific antigens, with tuberculin skin test (TST) results in an unselected hospital-based Greek population. METHODS: This was a prospective study of inpatients and outpatients in the University Hospital of Larissa, Greece. Demographic, clinical and microbiological data were collected and correlated to the QFT-G 'In Tube' results. Agreement was tested using the Q-test; kappa coefficient and confounding factors were adjusted for by univariate and multivariable logistic regression. RESULTS: Among the 191 patients with valid results for both tests, the concordance between the TST and the QFT-G was 68.5%, with a kappa value of 0.38 (95% CI 0.2-0.5). In 27 (14.1%) patients with active tuberculosis, QFT-G 'In Tube' yielded 85.1% positive results compared with 74% for the TST (P = 0.45). CONCLUSIONS: Overall agreement between the QFT-G 'In Tube' assay and the TST in this unselected hospital population was low, mainly due to discordant TST-positive/QFT-negative results.
BACKGROUND AND OBJECTIVE: Although QuantiFERON-TB Gold (QFT-G) has been approved for the diagnosis of latent tuberculosis infection (LTBI), there are limited data regarding its performance in routine clinical practice. The aim of this study was to compare QFT-G 'In Tube' results, based on stimulation with Mycobacterium tuberculosis-specific antigens, with tuberculin skin test (TST) results in an unselected hospital-based Greek population. METHODS: This was a prospective study of inpatients and outpatients in the University Hospital of Larissa, Greece. Demographic, clinical and microbiological data were collected and correlated to the QFT-G 'In Tube' results. Agreement was tested using the Q-test; kappa coefficient and confounding factors were adjusted for by univariate and multivariable logistic regression. RESULTS: Among the 191 patients with valid results for both tests, the concordance between the TST and the QFT-G was 68.5%, with a kappa value of 0.38 (95% CI 0.2-0.5). In 27 (14.1%) patients with active tuberculosis, QFT-G 'In Tube' yielded 85.1% positive results compared with 74% for the TST (P = 0.45). CONCLUSIONS: Overall agreement between the QFT-G 'In Tube' assay and the TST in this unselected hospital population was low, mainly due to discordant TST-positive/QFT-negative results.