Objectives: C-reactive protein (CRP) which is synthetized by hepatocytes is an acute phase protein and its serum level increases within 6-9 hours after infection or tissue damage. We investigated its usefulness as a marker of bacterial infection in patients with pleural effusion.
Methods: We studied the usefulness of pleural fluid C-reactive protein measurement in a population of 72 patients with pleural effusion, by means of an immunoturbidimetric method (Hitachi 717, Boheringer Mannheim). A comparison of serum and pleural effusion C-reactive protein levels in different subgroups of patients with effusion was made.
Results: According to preset diagnostic criteria, 19 patient effusions were classified as transudates and the mean (+/- 5 D) pleural fluid CRP [5.3 (+/- 7.8) mg per liter, p < 0.001] were significantly lower than those in the exudate effusions group. Among the 53 patients with exudate effusion, eight were caused by neoplastic disease and the pleural fluid CRP mean (29.3 +/- 16.1 mg per liter, p < 0.001) were significantly lower than those in exudates from parapneumonic effusions (122.7 +/- 48.0 mg per liter, p < 0.001) and than those in the exudates from patients with effusion associated with tuberculosis (67.8 +/- 32.1 mg per liter, p < 0.001). Moreover, all but two transudates had a C-reactive protein lower than 10 mg/L, whereas only two exudates with tuberculosis origin had a C-reactive protein value lower than 10 mg/L, instead all pleural-fluid C-reactive protein from exudates with pneumonia were greater than 10 mg/L. We had found a correlation between the pleural and serum C-reactive protein (r = 0.6884, p < 0.0001). And transudates tended to have lower ratios of pleural to serum CRP (0.26) than exudates (0.55), and malignant effusions had lower ratios (0.37) than pneumonic and tuberculous effusions (0.52, 0.58).
Conclusions: Pleural fluid CRP > 10 mg per liter had good sensitivity (82%), specificity (87.5%) and predictive value of positivity (95.5%) in the diagnosis of exudate effusions and higher CRP-levels may prove to be a practical, accurate and rapid method for differentiating pneumonic effusions and effusions associated with tuberculosis from others. It can be considered that quantitative immunoturbidimetric assay of pleural-fluid C-reactive protein will be a useful diagnostic tool to differentiate pleural effusions with bacterial origin from others.