Study objectives: The aims of this study were to describe the different appearances of pleural fluid during thoracentesis and their frequency in relation to diagnosis, and to evaluate the causes and clinical implications of bloody pleural effusions.
Setting: Tertiary care, university-affiliated hospital.
Subjects and methods: Seven hundred fifteen patients with pleural effusion were prospectively assessed from December 1991 to December 1997.
Interventions: The appearance of the fluid was assessed in a glass assay tube containing 10 mL of pleural fluid.
Results: The most common presentations were serous and blood tinged, with 80% of the fluids fitting into one of these categories. The most frequent cause of watery fluid was transudate, although most transudates were classified as serous effusions. There were 59 bloody and 656 nonbloody pleural fluids. The most common cause of bloody pleural effusion (BPE) was malignancy (47%). Fluid with a bloody appearance slightly increased the probability of malignancy in our series (odds ratio [OR], 1.73; 95% confidence interval [CI], 1.01 to 2.94; p = 0.04). Nevertheless, only 11% of the neoplastic effusions were BPE. Other common causes of BPE were posttraumatic (12%) or parapneumonic (10%) pleural effusions. Tuberculosis and transudates were uncommon causes of BPE. Fluid that was bloody in appearance decreased the probability for both diseases (OR, 0.15; 95% CI, 0.04 to 0.57; p = 0.003 and OR, 0.25; 95% CI, 0.06 to 0.95; p = 0.04, respectively).
Conclusions: Serous and blood tinged were the most common presentations of pleural fluid at thoracentesis. Almost half of BPEs were secondary to neoplasms, but only 11% of the neoplastic effusions were BPEs. Other common causes of BPE were parapneumonic and posttraumatic.