|| Checking for direct PDF access through Ovid
In this study, we conducted a meta-analysis to systematically compare the diagnostic accuracy of IGRAs performed for extrasanguinous body fluids with that performed for blood in the diagnosis of TB.Multiple English and Chinese databases were searched up to November 2017. Studies that complied with the guidelines for the Quality Assessment of Diagnostic Accuracy Studies and used QuantiFERON-TB Gold In-Tube and/or T-SPOT.TB (ELISPOT) assays on both blood and extrasanguinous body fluids were included. Statistical analysis was performed using Stata 12.0 software. Since publication bias is a concern in the meta-analysis of diagnostic studies, we tested for this using Begg's funnel plots.Among the 1332 articles searched from the databases, 24 articles met the inclusion criteria, which included 1040 samples in the patient group and 1044 samples in the control group. For extrasanguinous body fluids, the pooled sensitivity, specificity, positive likelihood ratio (PLR), negative likelihood ratio (NLR), diagnostic odds ratio (DOR), and summary receiver operating characteristic (SROC) area under the curve (AUC) were 87% (95% CI: 0.81–0.91), 89% (95% CI: 0.84–0.93), 8.22 (95% CI 5.38–12.56), 0.15 (95% CI: 0.10–0.21), 44.92 (95% CI: 25.61–78.81), and 0.94 (95% CI: 0.92–0.96), respectively. For peripheral blood, these values were 83% (95% CI: 0.79–0.87), 74% (95% CI: 0.68–0.79), 3.17 (95% CI 2.63–3.84), 0.23 (95% CI: 0.19–0.29), 12.99 (95% CI: 10.19–16.57), and 0.86 (95% CI: 0.82–0.89), respectively.IGRAs performed on extrasanguinous body fluids exhibited a better diagnostic accuracy compared with IGRAs performed on peripheral blood for diagnosing TB.