Conventional methods for diagnosing tuberculous pleurisy (TB pleurisy) are either invasive or have a long turn-around-time. Performances of pleural adenosine deaminase (ADA), ADA2, interferon-gamma (IFN-γ), and interferon-gamma release assays (IGRA) as diagnostic tools for TB pleurisy were evaluated.

Eighty-eight patients with lymphocyte-predominant pleural exudates between June 2010 and March 2011, including 31 with clinically diagnosed TB pleurisy, were prospectively studied. Pleural ADA and ADA2 activity were measured by colorimetric method, IFN-γ levels by enzyme-linked immuno-sorbent assay, and IGRA by enzyme-linked immuno-spot (T-SPOT.TB) assay.

Pleural ADA, ADA2, and IFN-γ levels, but not the proportion of positive T-SPOT.TB assay, were significantly higher in patients with TB pleurisy than in those without TB pleurisy. The area under the receiver-operating-characteristic (ROC) curve was 0.920, 0.893, 0.875, and 0.544 for IFN-γ, ADA2, ADA, and T-SPOT.TB assay, respectively. The combination of ADA ≥40 IU/L and IFN-γ ≥75 pg/mL yielded a specificity of 100%.

Pleural ADA, ADA2 and IFN-γ, but not T-SPOT.TB assay, are all sensitive and specific for TB pleurisy. In patients with lymphocyte-predominant pleural exudates, ADA ≥40 IU/L and IFN-γ ≥75 pg/mL in pleural effusion imply a very high probability of TB pleurisy.