Background Tuberculosis (TB) is still a serious global public health concern, being essential a rapid and accurate diagnosis of infected individuals. The aim of this study was to estimate the sensitivity of the interferon-γ release assays (IGRA) and tuberculin skin test (TST) in patients diagnosed with active TB and the agreement between the tests.

Methods Retrospective cohort study carried out using data from the Portuguese National Tuberculosis Surveillance system, from 2008 to 2015. The study included all TB cases with an IGRA and TST result (n=727). The IGRA test used in the patients enrolled in the study was the QuantiFERON-TB Gold In-Tube (Qiagen). Sensitivity was calculated with 95% confidence interval (95%CI) for each test separately and in combination (IGRA and TST-5 mm or IGRA and TST-10 mm) and outcomes were compared using McNemar’s test. Kappa coefficient (k) was used to evaluate the agreement between IGRA and TST test results.

Results The mean age of the patients was 47.9 years (± standard deviation 20.0 years), ranging from less than a 1 year to 91 years, with the age group 16–64 years representing the majority of cases. IGRA, TST-5 mm and TST-10 mm were positive in 82.4%, 84.5% and 78.4% of the TB patients. These results imply that 128 (17.6%), 112 (15.4%) and 157 (21.6%) patients with an active TB diagnosis were not identified by IGRA, TST-5 mm and TST-10 mm respectively. The difference between IGRA and TST was only was only statistically significant between IGRA and TST-10 mm (p=0.021). Agreement between IGRA and TST-5 mm was k=0.402 (p<0.001) with a concordance rate of 83.5% and between IGRA and TST-10 mm was k=0.351 (p<0.001) with a concordance rate of 79.5%. Combine sensitivity of IGRA plus TST-5 mm and IGRA plus TST-10 mm was 91.7% and 90.6%, respectively.

Conclusion IGRA tests showed a high sensitivity, however lower than the TST with a 5 mm cutoff. The level of agreement between IGRA and TST with either cut-offs was poor, with 16.5% of the patients showing different outcomes between IGRA and TST-5 mm and 20.5% between IGRA and TST-10 mm. This significant increase in sensitivity when results from both tests were combined suggests that the use of the two tests together could promote the identification of more cases of infection than if used separately and in substitution of one another. This could be especially important in countries were latent infection is the primary source of TB cases.