Poor concordance between interferon-γ release assays and tuberculin skin tests in diagnosis of latent tuberculosis infection among HIV-infected individuals

BACKGROUND A new generation of diagnostic tests, the interferon-gamma release assays (IGRAs), have been developed for the detection of latent tuberculosis infection (LTBI). Limited data are available on their use in HIV-infected persons. METHODS A cross-sectional study was carried out at 2 HIV clinics in Atlanta to assess the utility of two IGRA tests (T-SPOT.TB [TSPOT] and QuantiFERON-TB Gold in Tube [QFT-3G]) compared to the tuberculin skin test (TST). RESULTS 336 HIV-infected persons were enrolled. Median CD4 count was 335 cells/microl and median HIV viral load was 400 copies/ml. Overall, 27 patients (8.0%) had at least 1 positive diagnostic test for LTBI: 7 (2.1%) had a positive TST; 9 (2.7%) a positive QFT-3G; and 14 (4.2%) a positive TSPOT. Agreement between the 3 diagnostic tests was poor: TST and TSPOT, [kappa = 0.16, 95% CI (-0.06, 0.39)], TST and QFT-3G [kappa = 0.23, 95% CI (-0.05, 0.51)], QFT-3G and TSPOT [kappa = 0.06, 95% CI (-0.1, 0.2)]. An indeterminate test result occurred among 6 (1.8%) of QFT-3G and 47 (14%) of TSPOT tests. In multivariate analysis, patients with a CD4 < or = 200 cells/microl were significantly more likely to have an indeterminate result [OR = 3.6, 95% CI (1.9, 6.8)]. CONCLUSION We found a low prevalence of LTBI and poor concordance between all 3 diagnostic tests. Indeterminate test results were more likely at CD4 counts < or = 200 cells/microl. Additional studies among HIV-infected populations with a high prevalence of TB are needed to further assess the utility of IGRAs in this patient population.