Diagnostic Accuracy with Total Adenosine Deaminase as a Biomarker for Discriminating Pleural Transudates and Exudates in a Population-Based Cohort Study
Table 3
Measures of diagnostic accuracy of adenosine deaminase assay following selection of the best cut-off point for pleural exudates and transudates according to the maximum positive likelihood ratio in the ROC curve space.
Diagnostic parameter, % (95% CI)
Exudate ()
Transudate ()
Best cut-off (U/L)
≥22.0
<22.0
Sensitivity
44.35 (35.4-53.5)
96.88 (83.78-99.92)
Specificity
96.98 (83.0-99.0)
42.92 (34.01-52.29)
Positive predictive value
98.11 (88.0-99.0)
29.83 (26.48-33.42)
Negative predictive value
31.0 (27.0-34.0)
98.27 (89.06-99.75)
Positive likelihood ratio
14.64 (2.11-101.9)
1.7 (1.44-2.01)
Negative likelihood ratio
0.57 (0.50-0.70)
0.07 (0.01-0.49)
Diagnostic odds radio
23.26 (3.08-176.75)
24.28 (144.0-4.10)
Predictive accuracy
54.25 (46.0-62.0)
53.77 (45.57-61.83)
Disease prevalence
79.0
21.0
None of the 95% confidence intervals (CIs) overlap 1 (percentages) or 0 (absolute values), indicating the differences are significant at .