Diagnostic utility of pleural fluid adenosine deaminase in tuberculous pleural effusion in patients with exudative effusion: a retrospective multicenter study (ADATPE)

Background Tuberculous pleural effusion (TPE) remains a diagnostic challenge, particularly in tuberculosis-endemic regions. Pleural fluid adenosine deaminase (pfADA) is widely used as a diagnostic biomarker; however, its optimal cutoff value may vary depending on the epidemiological context. This study aimed to optimise the diagnostic performance of pfADA by determining a new local cutoff, evaluating the conventional threshold (≥40 U/L), and assessing the diagnostic value of the pleural fluid lactate dehydrogenase (LDH)/ADA ratio. Methods The ADATPE study was a retrospective, observational, multicenter analysis involving 276 patients with exudative pleural effusion from three tertiary Malaysian hospitals. Patients were categorized into definite, clinical, and non-TPE groups based on microbiological, histopathological, and clinical criteria. Receiver operating characteristic curve analysis and the Youden Index were used to identify an optimal local pfADA threshold. Results PfADA demonstrated excellent diagnostic accuracy with an area under the receiver operating characteristic curve (AUC) of 0.852. A new local cutoff of ≥36 U/L improved sensitivity to 82.9% while maintaining specificity at 80.6%, outperforming the conventional threshold of ≥40 U/L. The combination of ADA ≥36 U/L and LDH/ADA ratio ≤16 yielded the highest diagnostic specificity (90.78%), positive predictive value (69.84%), and overall accuracy (83.70%). Conclusions A revised pfADA cutoff of ≥36 U/L enhances diagnostic accuracy for TPE in Malaysian patients. PfADA remains a valuable diagnostic tool, but it should be interpreted in conjunction with other clinical and laboratory parameters. The LDH/ADA ratio is a valuable adjunct, particularly in cases that are borderline or inconclusive.