Accuracy of enzyme-linked immunospot assay for diagnosis of pleural tuberculosis: a meta-analysis.

Current methods for diagnosing tuberculous pleurisy are poor. Some studies have explored the diagnostic value of a pleural effusion enzyme-linked immunospot (ELISPOT) assay, but its accuracy remains controversial. Therefore, we performed a meta-analysis of the existing evidence on the ability of the ELISPOT assay to diagnose tuberculous pleurisy. We systematically searched PubMed, Google Scholar, and EMBASE databases for studies measuring the sensitivity, specificity, and other measures of accuracy of the pleural effusion ELISPOT assay for diagnosis of tuberculous pleurisy. A total of nine studies were identified and subjected to meta-analysis, giving the following pooled values for diagnostic accuracy: sensitivity, 0.93 [95% confidence intervals (95%CI) = 0.90 to 0.95]; specificity, 0.90 (95%CI = 0.86 to 0.93); positive likelihood ratio, 8.21 (95%CI = 4.00 to 16.84); negative likelihood ratio, 0.11 (95%CI = 0.06 to 0.19); diagnostic odds ratio, 88.26 (95%CI = 33.81 to 230.43); and the area under the curve, 0.9599 (standard error of the mean, 0.0134). The available evidence suggests that pleural effusion ELISPOT assay is sufficiently accurate to diagnose tuberculous pleurisy as a stand-alone technique. In fact, it appears to be superior to assays based on adenosine deaminase and gamma interferon for screening patients and confirming the diagnosis of tuberculous pleurisy.