PURPOSE

Diagnosing extrapulmonary tuberculosis (EPTB) - including pleural, peritoneal, pericardial, meningeal forms - remains challenging due to the insufficient sensitivity of smear microscopy (SM), mycobacteriological culture, and nucleic acid amplification test (NAAT). The Adenosine Deaminase (ADA) assay has potential as a diagnostic tool for EPTB, but its performance in high-income countries is poorly documented. This study aimed to evaluate the diagnostic performance of ADA for microbiologically confirmed EPTB in such a setting.

METHODS

We retrospectively analyzed data from all patients undergoing ADA testing in our hospital network in Paris area between May 2014 and April 2024. Microbiological confirmation (positive SM, culture, or NAAT) from the same sample site served as the reference standard.

RESULTS

Among 363 ADA assays (352 patients), 69% were pleural fluid, 18% peritoneal, < 1% pericardial, 11% CSF. For pleural fluid, ADA at a threshold of 30 U/L demonstrated 92% sensitivity (CI 80-98%), 75% specificity (CI 68-81%), 47% PPV (CI 37-57%), and 97% NPV (CI 94-99%). For peritoneal fluid, sensitivity, specificity, PPV, and NPV were 77% (CI 46-95%), 81% (CI 69-91%), 50% (CI 27-73%), and 94% (CI 82-99%), respectively. Raising the ADA threshold to 60 U/L improved specificity to 92% in pleural fluid (CI 87-95%) and 85% in peritoneal fluid (CI 73-93%). Combining ADA with other biomarkers showed no added diagnostic value.

CONCLUSION

ADA testing is a rapid and practical tool for EPTB diagnosis. In pleural and peritoneal fluids, a threshold < 30 U/L effectively excludes EPTB, while a threshold > 60 U/L supports initiating treatment pending culture results.