Abstract | BACKGROUND: METHODS: We searched PubMed, Embase and Cochrane Library database from the earliest available date of indexing through April 30, 2019. The study quality was evaluated using the Quality Assessment of Diagnostic Accuracy Studies (QUADAS2) checklist. We determined the sensitivities and specificities across studies, calculated positive and negative likelihood ratios (LR+ and LR-) and constructed summary receiver operating characteristic curves parameters. RESULTS: Across six results from five studies (415 patients), the pooled sensitivity for IGRA methods was 0.94 (95% confidence interval [CI]; 0.87-0.98) with heterogeneity (χ2 = 69.9, P = .01) and a pooled specificity of 0.94 (95% CI; 0.75-0.94) without heterogeneity (χ2 = 41.1, P = .13). Likelihood ratio (LR) syntheses gave an overall positive likelihood ratio (LR+) of 16.8 (95% CI; 8.0-35.4) and negative likelihood ratio (LR-) of 0.06 (95% CI; 0.03-0.13). The pooled diagnostic odds ratio was 278 (95% CI; 114-6806). CONCLUSIONS:
|
Authors | Hyun T Seo, Yun S Kim, Hye S Ock, Lae H Kang, Ki S Byun, Doo S Jeon, Seong J Kim |
Journal | International journal of clinical practice
(Int J Clin Pract)
Vol. 74
Issue 5
Pg. e13479
(May 2020)
ISSN: 1742-1241 [Electronic] India |
PMID | 31927772
(Publication Type: Journal Article, Meta-Analysis)
|
Copyright | © 2020 John Wiley & Sons Ltd. |
Topics |
- Humans
- Interferon-gamma Release Tests
(methods)
- Odds Ratio
- Pericarditis, Tuberculous
(diagnosis)
- ROC Curve
- Sensitivity and Specificity
|