ABSTRACT
BACKGROUND: Latent tuberculosis infection (LTBI) co-infected with human immunodeficiency virus (HIV) is more likely to develop into active tuberculosis (ATB), recombinant Mycobacterium tuberculosis fusion protein ESAT6/CFP10 (EC-Test) is a latest developed method for LTBI. Compared with the interferon γ release test assays (IGRAs), the diagnostic performance of EC-Test to LTBI screening in HIV needs to be evaluated. METHODS: A population-based multicenter prospective study was conducted in Guangxi Province, China. The baseline data was collected and LTBI were measured by QuantiFERON-TB Gold In-Tube (QFT-GIT), EC-Test and T-cell spot of the TB assay (T-SPOT.TB). RESULTS: A total of 1478 patients were enrolled. when taking T-SPOT.TB as reference, the value of sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV) and consistency that EC-Test to diagnosis LTBI in HIV was 40.42, 97.98, 85.26, 85.04 and 85.06% respectively; when taking QFT-GIT as reference, the value was 36.00, 92.57, 55.10, 85.09 and 81.13%, respectively. When the CD4 + cell count was <200âcells/µl, the accuracies of EC-Test to T-SPOT.TB and QFT-GIT were 87.12 and 88.89%, respectively; when it was 200 ≤ CD4 + ≤ 500âcells/µl, the accuracies of EC-Test was 86.20 and 83.18%, respectively; when the CD4 + cell count >500âcells/µl, the accuracies of EC-Test were 84.29 and 77.94%, respectively. The incidence of adverse reactions in EC-Test was 34.23% and the serious adverse reactions were 1.15%. CONCLUSION: EC-Test has good consistency compared with IGRAs in detecting LTBI in HIV no matter in different immunosuppression status or different regions, and the safety of EC-Test is also well, suitable for LTBI screening in HIV in high prevalence settings.