Bottom line: 50-60 mL samples are associated with approximately the same diagnostic accuracy as larger volume samples.
Prospective study to determine the volume of pleural fluid required to diagnose malignancy. Swiderek J, Morcos S, Donthireddy V, Surapaneni R, Jackson-Thompson V, Schultz L, Kini S, Kvale P. Chest. 2009 Apr;135(4):999-1001. doi: 10.1378/chest.08-2002
Thoracentesis was performed on 121 patients with pleural effusion. Fluid from each thoracentesis was didvided into 10 mL, 60 mL, and 150+ mL samples. Cytologist was blinded to sample size being evaluated. Samples were subjected to cytospin/direct smear and to cell block preparation. RESULTS: Table 3 summarizes sensitivities, specifities and predictive values for cytospin/direct smear. For 60 mL v. 150+ mL sizes, sensivity: 62.8% v. 69.2%, specificity: 89.7% v. 86.2%, NPV: 47.3% v. 51.0%. No statistically significant difference between 60 mL and 150+ mL sample sizes. Table 4 summarizes similar findings when both the cytospin/direct smear and cell block preparation were considered. With addition of cell block preparation, Positive predictive value approached statistical significance with 97.7% for the 60 mL sample and 94.0% PPV for the 150+ mL sample.
A prospective study of the volume of pleural fluid required for accurate diagnosis of malignant pleural effusion. Abouzgheib W, Bartter T, Dagher H, Pratter M, Klump W. Chest. 2009 Apr;135(4):999-1001. doi: 10.1378/chest.08-2002. Epub 2008 Nov 18.
Thoracentesis performed on 44 patients with pleural effusion and either suspected malignancy or historyof malignancy. From each thoracentesis, fluid divided into 50 mL sample and other fluid (mean volume 890 mL, 250-1900 mL.) Cytologist blinded to sample tested. RESULTS: Fluid for 23 of the patients was positive for malignant cells in the 250+ mL fluid sample and negative for the fluid from the other 21 patients. In all cases (both positive and negative) the findings for the 60 mL sample had the same result as the 250+ mL sample (i.e., 100% agreement.)