Performance of QuantiFERON-TB Gold Plus for detection of latent tuberculosis infection in pregnant women living in a tuberculosis- and HIV-endemic setting.

Abstract

We evaluated the performance of QuantiFERON-TB Gold Plus (QFT-Plus), which includes two Mycobacterium tuberculosis antigen formulations (TB1 and TB2), for detection of latent tuberculosis infection during pregnancy. Eight-hundred-twenty-nine Ethiopian pregnant women (5.9% HIV-positive) were tested with QFT-Plus, with bacteriological sputum analysis performed for women with clinically suspected tuberculosis and HIV-positive women irrespective of clinical presentation. QFT-Plus read-out was categorized according to the conventional cut-off (0.35 IU/ml) for both antigen formulations. In addition, we analysed the distribution of QFT-Plus results within a borderline zone (0.20-0.70 IU/ml), and interferon-γ response in relation to HIV infection and gestational age. Two-hundred-seventy-seven women (33%) were QFT-Plus-positive (HIV-positive 16/49 [33%]; HIV-negative 261/780 [33%]). There was a strong agreement between the two antigen formulations (κ = 0.92), with discordant results in 29 cases (3.5%). Whereas discordant QFT-Plus results were rare in pregnancy, several results with both TB1 and TB2 within the borderline range were observed (11/49 [22%] vs. 43/780 [5.5%] in HIV-positive and HIV-negative women, respectively; p<0.0001). HIV-positive women had lower absolute interferon-γ levels (TB1: 0.47 vs. 2.16 IU/ml; p<0.001, TB2: 0.49 vs. 2.24 IU/ml, p<0.001, considering results ≥0.20 IU/ml) compared to HIV-negative women. QFT-Plus-positive women who submitted samples at later stages of pregnancy had lower mitogen- (p<0.001) but higher TB-antigen-specific (p = 0.031 for TB1, p = 0.061 for TB2) interferon-γ response. Considering their lower capacity to produce TB-specific interferon-γ, a lower cut-off level for defining QFT-Plus-positivity may be considered in HIV-positive pregnant women.