Commercial nucleic acid amplification tests in tuberculous meningitis--a meta-analysis.

Although nucleic acid amplification tests (NAATs) promise a rapid, definitive diagnosis of tuberculous meningitis, the performance of first-generation NAATs was suboptimal and variable. We conducted a meta-analysis of studies published between 2003 and 2013, using the Quality Assessment of Diagnostic Accuracy Studies-2 (QUADAS-2) tool to evaluate methodological quality. The diagnostic accuracy of newer commercial NAATs was assessed. Pooled estimates of diagnostic accuracy for commercial NAATs measured against a cerebrospinal fluid Mycobacterium tuberculosis culture-positive gold standard were sensitivity 0.64, specificity 0.98, and diagnostic odds ratio 64.0. Heterogeneity was limited; P value = 0.147 and I(2) = 33.85%. The Xpert MTB/RIF® test was evaluated in 1 retrospective study and 4 prospective studies, with pooled sensitivity 0.70 and specificity 0.97. The QUADAS-2 tool revealed low risk of bias, as well as low concerns regarding applicability. Heterogeneity was pronounced among studies of in-house tests. Commercial NAATs proved to be highly specific with greatly reduced heterogeneity compared to in-house tests. Sub-optimal sensitivity remains a limitation.