Background and purpose: Malignant and infectious diseases are the most frequent causes of pleural effusion. Malignant diseases produce pleural effusion through different mechanisms: lymphatic and capillary obstruction, pneumonia or atelectasia. Therefore absence of tumor cells in aspirated effusion, highly declines the sensivity of the cytologic study. For this reason, the use of tumor markers to improve the capability of differentiating malignant effusion from benign types is widespread.Materials and methods: 100 patients with pleural effusion were allocated in four groups: I: malignant (positive pleural biology or catology. N=21). II: paramalignant (lnown case of a cancer, but negative cytology or biopsy. N=9). III: Empyema/paraneumonics (clinicalparacillinically proved.) IV: Benign process (CHF, CRF, Liver disease ... n=58). CA 15-3, CA 19-9 and CA 125 quantified using an Ag and ELISA technology.Results: Calculated cut off values were: 500 U/ml for CA 125,35 U/ml for CA 19-9 and 35 U/ml for CA 15-3. Sensivity, specifity, efficiency, PPV and NPV were as follows: CA 15-3 (80%, 90%, 89%, 82%, 91%)CA 19-9 (67%, 89%, 83%, 74%, 86%), CA 125 (60%, 83%, 78%, 60%, 82%)Conclusion: We suggest that using tumor markers in pleural fluid, especially CA 15-3 would greatly increase the diagnostic effectiveness of malignant effusions.