SESSION TITLE: Pleural Disease Posters
SESSION TYPE: Original Investigation Poster
PRESENTED ON: Wednesday, October 29, 2014 at 01:30 PM - 02:30 PM
PURPOSE: Our objective was to determine the usefulness of pleural effusion C reactive protein levels in the diagnosis of pleural effusions.
METHODS: A comparison of serum and pleural fluid C-reactive protein (CRP) levels in different subgroups of 286 patients with pleural effusion was made. We assessed prospectively the sensitivity, specificity, positive and negative predictive values, accuracy, Youden index, likelihood ratio and ROC curve of the test, for a period from February 2008 to November 2011.
RESULTS: Among 286 patients with pleural effusion, 67 patients were included in the transudate group, 219 patients were included in the exudate group. In transudates the cut-off value of pleural fluid CRP ≤ 15 mg/L had a Youden index of 0.678 and the area under curve = 0.86 comparing with exudative pleural effusions. In malignant pleural effusions, the cut-off value of pleural fluid CRP ≤ 20mg/L had a Youden index of 0.728 and the area under curve = 0.89 comparing with tuberculous effusions. In tuberculous effusions, the cut-off value of pleural fluid CRP >20mg/dl had a Youden index of 0.45 and the area under curve=0.96 comparing with malignant effusions. The values of pleural fluid/blood CRP ratios had a very small Youden index and the area under curve in all subgroups of patients with pleural effusion.
CONCLUSIONS: Levels of CRP in exudative pleural effusions less than 20 mg/L are strongly suggestive of malignant effusion and chronic tuberculous effusion.A CRP pleural fluid level > 20mg/L almost excludes transudative pleural effusion while the levels of CRP above 30mg/L are suggestive of an inflammatory etiology and almost exclude malignant pleural effusion.
CLINICAL IMPLICATIONS: Low values of CRP in exudative pleural effusions suggest the differential diagnosis between tuberculous chronic pleural effusion and malignant effusion.CRP values higher than 40 almost exclude the malignant etiology of pleural effusion.
DISCLOSURE: The following authors have nothing to disclose: Perlat Kapisyzi, Dhimitraq Argjiri, Anila Mitre, Genc Burazeri, Anila Aliko, Jeta Beli, Ylli Vakeflliu, Roland Kore, Elenka Shehu, Holta Tafa, Regina Hasa, Ornela Nuredini, Mirela Tabaku, Eritjan Tashi, Richard Light
No Product/Research Disclosure Information