Introduction: We investigated whether tumour markers carcinoembryonic antigen (CEA), neuron-specific enolase (NSE), cancer antigen 125 (CA125), and cytokeratin 19 fragment (CYFRA 21-1) in pleural effusions and serum can be used to distinguish pleural effusion aetiology. Materials and methods: During the first thoracentesis, we measured pleural fluid and serum tumour marker concentrations and calculated the pleural fluid/serum ratio for patients diagnosed with pleural effusion, using electrochemiluminescence immunoassays. Receiver operating characteristic (ROC) analysis was carried out and the Hanley and McNeil method was used to test the significance of the difference between the areas under ROC curves (AUCs). In order to detect which tumour marker best discriminates between malignant and non-malignant pleural effusions and to establish the predictive value of those markers, discriminant function analysis (DFA) and logistic regression analysis were utilized. Results: Serum tumour markers CYFRA 21-1 and NSE as well as pleural NSE were good predictors of pleural effusion malignancy and their combined model was found statistically significant (Chi-square = 28.415, P < 0.001). Respective ROC analysis showed significant discrimination value of the combination of these three markers (AUC = 0.79). Conclusions: Serum markers showed superiority to pleural fluid markers in determining pleural fluid aetiology. Serum CYFRA 21-1 and NSE concentrations as well as pleural fluid NSE values had the highest clinical value in differentiating between malignant and non-malignant pleural effusions. The combination of these three markers produced a significant model to resolve pleural effusion aetiology.