Objective: To determine whether measurement of the complement activation products SC5b-9 and C3a-desArg in pleural fluid can reliably differentiate tuberculous from malignant pleural effusions.
Design: Twenty-four patients with tuberculous pleuritis, 29 with malignant pleural effusion, and 30 control subjects with transudates were enrolled in the study. SCSb-9 and C3a-desArg were measured in pleural fluid using commercial ELISA tests, and their performances were evaluated using receiver operating characteristic (ROC) analysis.
Results: Patients with tuberculous pleuritis had higher mean levels of pleural SC5b-9 (5052 microg/L) and C3a-desArg (7436 microg/L) than those with malignant effusions (1048 and 2835 microg/L, respectively), whereas only SC5b-9 concentrations in the latter were comparable with controls. The area under the ROC curve (AUC) was 0.84 for SC5b-9 and 0.81 for C3a-desArg. Pleural SC5b-9 showed an accuracy of 80.8%, compared with 78.8% for C3a-desArg, when cut-off points of 1500 and 4500 microg/L, respectively, were used. Using a stepwise logistic regression model, the combination of pleural SCSb-9 > or =1500 microg/L, age < or =35 years, and pleural monocyte percentage > or =90% provided the highest accuracy for tuberculous pleurisy (88.5%, AUC 0.95).
Conclusion: This pilot study suggests that pleural SC5b-9 is clinically useful for differentiating tuberculous and malignant pleural effusions.