Performance of adenosine deaminase in detecting paediatric pleural tuberculosis: a systematic review and meta-analysis

Background Paediatric pleural tuberculosis (TB) is a paucibacillary disease, which increases the difficulty of examination. We aimed to assess the performance of pleural fluid adenosine deaminase (ADA) in the detection of paediatric pleural TB. Methods PubMed, Web of Science Core Collection, Embase and Cochrane Library databases were searched up to 20 December 2021. We used the bivariate and hierarchical summary receiver operating characteristic models to compute pooled estimates for the overall diagnostic accuracy parameters of ADA for diagnosing paediatric pleural TB. Results Eight studies, including 290 pleural fluid samples, met the inclusion criteria. The pooled sensitivity of ADA was 0.85 (95% CI: 0.78-0.90, I-2: 55.63% < 75%) for detecting patients with paediatric pleural TB. A total of 262 pleural fluid samples from four studies were included to differentiate patients with paediatric pleural TB from controls. At a unified cut-off value of 40 U/L, the pooled sensitivity, specificity, positive likelihood ratio, negative likelihood ratio, diagnostic odds ratio and area under the summary receiver operating characteristic curve of ADA were 0.89, 0.58, 2.09, 0.20, 10.48 and 0.89, respectively. Conclusions At a cut-off value of 40 U/L, the overall performance of ADA was good for detecting paediatric pleural TB, with relatively high sensitivity and low specificity. Key message Accurate identification of paediatric pleural TB will help eliminate TB in children. At a cut-off value of 40 U/L, the overall performance of ADA was good for detecting paediatric pleural TB, with relatively high sensitivity and low specificity.