Rethinking the cut-off value for pleural effusion adenosine deaminase and neutrophil-lymphocyte ratio in a setting of low TB incidence

<b>Introduction:</b> Combining adenosine deaminase (ADA) and lymphocyte-neutrophil ratio (LNR) in pleural fluid (PF) has been shown to help differentiate exudates of tuberculous origin from others, in countries with a high prevelance of tuberculosis (TB). Since 2016 the incidence of TB in Portugal has decreased to less than 20/100.000 inhabitants, which may alter the cut-off value of these 2 markers and the usefulness of both. <b>Aims:</b> To ascertain the best possible cut-off value for PF ADA and LNR in pleural TB and the value of combining both markers. <b>Methods:</b> We conducted a cross sectional retrospective study based on the consultation of the clinical records of all patients who performed PF ADA (November 2016 to December 2018). Transudative and repeated samples were excluded. ROC curve was calculated considering a significance of 5%, and test characteristics: sensitivity (S), specificity (E), positive predictive value (PPV) and negative predictive value (NPV). <b>Results:</b> From a total of 200 patients, 136 were included. Patients were mainly male (60.3%); mean age of 70.4 (±15.65) years. Of total, 10 had a final diagnosis of pleural TB; 54 infectious; 45 malignant; 27 other diagnosis. Area under ROC curve was 0.808 for a cut-off value of ADA of ≥ 24.9 U/L [S 70%; E 87% (p&lt;0.001)]; and 0.771 for LNR &gt; 2.85 [S 100%; E 43% (p&lt;0.001)] for pleural TB. When combining both tests (both positive), S 70%; E 30.2%, PPV 7.4%, NPV 92.7% <b>Conclusions:</b> In our sample a newer and lower cut-off value of PF ADA of 24.9U/L would have a stronger power for diagnosing pleural TB. Combining both markers was only useful to exclude the disease.