Systematic review: Comparison of Xpert MTB/RIF, LAMP and SAT methods for the diagnosis of pulmonary tuberculosis.

Technological advances in nucleic acid amplification have led to breakthroughs in the early detection of PTB compared to traditional sputum smear tests. The sensitivity and specificity of loop-mediated isothermal amplification (LAMP), simultaneous amplification testing (SAT), and Xpert MTB/RIF for the diagnosis of pulmonary tuberculosis were evaluated. A critical review of previous studies of LAMP, SAT, and Xpert MTB/RIF for the diagnosis of pulmonary tuberculosis that used laboratory culturing as the reference method was carried out together with a meta-analysis. In 25 previous studies, the pooled sensitivity and specificity of the diagnosis of tuberculosis were 93% and 94% for LAMP, 96% and 88% for SAT, and 89% and 98% for Xpert MTB/RIF. The I(2) values for the pooled data were >80%, indicating significant heterogeneity. In the smear-positive subgroup analysis of LAMP, the sensitivity increased from 93% to 98% (I(2) = 2.6%), and specificity was 68% (I(2) = 38.4%). In the HIV-infected subgroup analysis of Xpert MTB/RIF, the pooled sensitivity and specificity were 79% (I(2) = 72.9%) and 99% (I(2) = 64.4%). In the HIV-negative subgroup analysis for Xpert MTB/RIF, the pooled sensitivity and specificity were 72% (I(2) = 49.6%) and 99% (I(2) = 64.5%). LAMP, SAT and Xpert MTB/RIF had comparably high levels of sensitivity and specificity for the diagnosis of tuberculosis. The diagnostic sensitivity and specificity of three methods were similar, with LAMP being highly sensitive for the diagnosis of smear-positive PTB. The cost effectiveness of LAMP and SAT make them particularly suitable tests for diagnosing PTB in developing countries.