Abstract
BACKGROUND: There is little information concerning community-based prevalence of latent tuberculosis infection (LTBI) using T-cell based interferon-γ (IFN-γ) release assays (IGRAs), particularly in TB endemic settings. In this study, the prevalence of LTBI in the Afar pastoral community was assessed using QuantiFERON-TB Gold In-Tube (QFTGIT) and tuberculin skin tests (TST). METHODS: A community-based cross-sectional survey of LTBI involving 652 apparently healthy adult pastoralists was undertaken in the pastoral community of Amibara District of the Afar Region between April and June 2010. RESULTS: The prevalence of LTBI was estimated as 63.7% (363/570) using QFTGIT at the cut-off point recommended by the manufacturer (≥0.35 IU/ml IFN-γ), while it was 74.9% (427/570) using a cut-off point≥0.1 IU/ml IFN-γ. The QFTGIT-based prevalence of LTBI was not significantly associated with the gender or age of the study participants. However, the prevalence of LTBI was 31.2% (183/587) using TST at a cut-off point≥10 mm of skin indurations, and it was higher in males than females (36.8% vs. 23.5%, X2=11.76; p<0.001). There was poor agreement between the results of the tests (k=0.098, 95% CI, 0.08-0.13). However, there was a positive trend between QFTGIT and TST positivity (X2=96.76, P<0.001). Furthermore, individuals with skin indurations≥10 mm were 13.6 times more likely to have positive results using QFTGIT than individuals with skin indurations of 0 mm (adjusted OR=13.6; 95%CI, 7.5 to 24.7, p<0.001). CONCLUSIONS: There is currently no agreed gold standard for diagnosis of LTBI. However, the higher prevalence of LTBI detected using QFTGIT rather than TST suggests that QFTGIT could be used for epidemiological studies concerning LTBI at the community level, even in a population unreactive to TST. Further studies of adults and children will be required to assess the effects of factors such as malnutrition, non-tuberculosis mycobacterial infections, HIV and parasitic infections on the performance of QFTGIT.