Background and objective Pentraxin-3 (PTX-3) is a relatively new marker of inflammation that has not been previously tested in pleural effusions. We aimed to assess whether PTX-3 is an accurate biomarker of parapneumonic effusions (PPE) and whether it discriminates complicated (CPPE)from non-complicated PPE. Methods The concentrations of pleural fluid PTX-3 were measured by a commercial enzyme-linked immunosorbent assay in a prospective cohort of 84 patients with pleural effusions, including 24 PPE, 40 malignant, and 20 miscellaneous exudative effusions. The area under the curve quantified the overall diagnostic accuracy of the test. A multivariate logistic regression analysis selected pleural fluid biochemistries predictive of PPE. Results Median pleural fluid PTX-3 levels were higher in PPE than in both malignant effusions and other exudates (32.4ng/mL vs 6.7ng/mL, and 8.5ng/mL, respectively, P<0.001). PTX-3>12ng/mL yielded 88% sensitivity, 73% specificity, likelihood ratio positive 3.3 and likelihood ratio negative 0.17 for diagnosing PPE, with an area under the curve of 0.855 (95% CI: 0.7690.941). In the multivariate analysis, pleural PTX-3 levels remained associated with increased diagnostic odds for PPE (odds ratio 17.7, 95% confidence interval: 3.785.1, P<0.001). There was a non-significant trend towards higher pleural PTX-3 levels in CPPE as compared with non-complicated. Conclusions High concentrations of PTX-3 in pleural effusions are very sensitive to differentiate PPE from non-PPE. However, they do not seem to differentiate uncomplicated-complicated from CPPE differentiation.