Utility of tumour markers in the diagnosis of neoplastic pleural effusion.

Approximately 20% of pleural effusions are caused by neoplastic processes. Although cytology is the most specific routine diagnostic procedure, its sensitivity of 50-60% is insufficient, and thus diagnosis is usually carried out by more invasive techniques such as pleural biopsy, thoracoscopy or thoracotomy. The object of this study is to evaluate the use of determining some tumour markers in pleural fluid obtained by thoracocentesis for diagnosis of neoplastic pleural effusion. Patients (271) with pleural effusions were classified in five groups: I: neoplasms n = 88; II: tuberculosis n = 63; III: parapneumonics n = 53; IV: miscellaneous exudates n = 39 and V: transudates n = 28. The tumour markers studied were: carcinoembryonic antigen (CEA), CA 125, squamous cell carcinoma antigen (SCC), and neuron specific enolase (NSE). The tumour makers had the following diagnostic efficiencies for neoplastic origin of the pleural effusion: CEA 76% (sensitivity 31%, specificity 93%); CA 125 66% (70% and 61%); SCC 65% (48% and 80%) and NSE 53% (30% and 89%). The diagnostic efficiencies for pulmonary neoplastic origins were 68% for NSE (sensitivity 83%, specificity 53%); 65% for SCC (54% and 75%); 63% for CEA (80% and 48%) and 61% for CA 125 (79% and 42%). We believe that the routine testing of tumour markers in pleural fluid obtained by thoracocentesis would greatly increase diagnostic effectiveness and could avoid the practice of more aggressive diagnostic techniques on the patient.