Setting: Ioannina University Hospital, Ioannina, Greece.
Objective: To evaluate the value of adding an interferon-gamma release assay (IGRA) to the tuberculin skin test (TST) for detecting latent tuberculous infection (LTBI) in a Greek university hospital among health care workers (HCWs) predominantly vaccinated with bacille Calmette-Guérin (BCG).
Design: Of 788 HCWs enrolled, 68.1% were BCG-vaccinated. A TST ⩾ 10 mm was considered positive and was followed by the QuantiFERON-TB(®) Gold In-Tube assay (QFT-GIT) in a two-step strategy.
Results: Of the enrolled HCWs, 36.4% were TST-positive, of whom only 14.4% were IGRA-positive. Agreement between the tests was poor (κ = 0.019; 95%CI -0.014-0.05, P = 0.355). Both TST and IGRA positivity increased with TST diameter, from 5.7% in TST 10-14 mm to 48.8% in TST ⩾20 mm. TST-positive, IGRA-negative results were most likely in younger, recently BCG-vaccinated HCWs (84.6% in those aged 20-29 years) and less likely in older HCWs (45% in those aged 50-59 years). The two-step strategy would have been more cost saving compared to the TST-only approach if adherence to LTBI treatment in our cohort had been ⩾24%.
Conclusions: Poor overall agreement between TST and QFT-GIT was found. Use of IGRA as a second step in TST-positive cases offers an appropriate tool for LTBI detection among BCG-vaccinated HCWs in low-TB-incidence settings.