Diagnostic accuracy of the Xpert MTB/RIF assay for tuberculous pericarditis: A systematic review and meta-analysis

Objective The purpose of this study was to evaluate the diagnostic efficacy of Xpert MTB/RIF for tuberculous pericarditis (TBP). Methods We searched relevant databases for Xpert MTB/RIF for TBP diagnosis until April 2021 and screened eligible studies for study inclusion. We evaluated the effectiveness of Xpert MTB/RIF when the composite reference standard (CRS) and mycobacterial culture were the gold standards, respectively. We performed meta-analyses using a bivariate random-effects model, and when the heterogeneity was obvious, the source of heterogeneity was further discussed. Results We included seven independent studies comparing Xpert MTB/RIF with the CRS and six studies comparing it with culture. The pooled sensitivity, specificity, and area under the curve of Xpert MTB/RIF were 65% (95% confidence interval, 59–72%), 99% (97–100%), and 0.99 (0.97–0.99) as compared with the CRS, respectively, and 75% (53–88%), 99% (90–100%), and 0.94 (0.92–0.96) as compared with culture, respectively. There was no significant heterogeneity between studies when CRS was the gold standard, whereas heterogeneity was evident when culture was the gold standard. Conclusions The sensitivity of Xpert MTB/RIF for diagnosing TBP was moderate and the specificity was good; thus, Xpert MTB/RIF can be used in the initial diagnosis of TBP.