C-reactive protein and serum amyloid A levels in discriminating malignant from non-malignant pleural effusion

Introduction: Distinction between malignant and non-malignant pleural effusion is of great importance in the patient management. The aim: We examined the diagnostic value of C-reactive protein (CRP) and serum amyloid A (SAA) in distinguishing different etiologies of pleural effusion and if they could discriminate between malignant and non-malignant pleural effusions. Subject and methods: CRP and SAA levels in both serum and pleural fluid were measured in 92 patients with pleural effusion. Of the 92 patients included in our study; 44 were diagnosed with malignant pleural effusions (group I) [with male to female ratio (M/F) 23/21 and mean age 57.7 ± 11.5 years in the form of mean ± 2SD] and 48 were diagnosed with non-malignant pleural effusion (group II) [with M/F ratio 33/15 and mean age 54.7 ± 10.4 years in the form of mean ± 2SD]. Results: CRP and SAA values were significantly higher in both serum and pleural effusion of malignant vs. non-malignant group (P < 0.003), but there was no statistical significant difference as regards pleural/serum CRP and pleural/serum SAA ratios between the two groups (P = 0.148 and P = 0.453 respectively). A statistically significant positive correlation between pleural fluid CRP and pleural fluid SAA in malignant and non-malignant effusions was detected (r = 0.315 and P = 0.002 respectively). Diagnostic performance of pleural fluid CRP and pleural fluid SAA in both infectious and malignant pleural effusions showed that at a cutoff value of 96.15 μg/ml for CRP; diagnostic sensitivity was 61% and specificity was 45%, while for pleural fluid SAA, a cutoff value of 137.5 μg/ml was associated with 41% sensitivity and 93% specificity. Conclusion: Measurement of SAA and CRP levels in pleural fluid has good diagnostic utility in differentiation between malignant and non-malignant pleural effusion and pleural SAA has a better diagnostic performance than CRP.

Abstract licence: CC BY-NC-SA

A person counts once a day per paper. We keep no cookie and no address: a reader is a hash that changes every day. Any month under five reads shows no number.