DETROIT—Historically, large amounts of pleural fluid were thought to be necessary for accurate diagnosis of malignancy. A recent retrospective chart analysis refutes this long-held belief, however—finding that small sample sizes have just as much predictive value as large ones.[1]

Investigators from the Henry Ford Hospital Heart and Vascular Institute in Detroit examined 374 pleural fluid samples taken from 282 patients within a nine-month period. Demographic information, volume of pleural fluid sampled, and pathologic and clinical diagnoses were obtained from the patients’ medical records.

Pleural fluid samples were grouped into quartiles based on volume of fluid collected. Sensitivity and negative predictive value were calculated and then compared among quartiles. In addition, the effect of demographic factors and medical history on sensitivity and negative predictive value were considered for each quartile.

METHOD OF PROCESSING IMPORTANT

Pleural malignancies were diagnosed in 99 patients, from whom a total of 132 samples had been collected. The volume of pleural fluid taken ranged from a mean of 6.2 mL in quartile 1 to 1,129.7 mL in quartile 4. No differences in sensitivity or negative predictive value were found between any two quartiles.

Processing pleural fluid using both cell blocks and smears was associated with more frequent identification of malignant cells than was either method alone. For example, of the 57 specimens that were positive for malignancy, six would have been missed if smears analysis alone had been performed, and seven would have been falsely labeled negative if cell blocks alone were used.

RECOMMENDATIONS CONFLICT

The literature regarding pleural fluid sampling is conflicting and lacks supporting data, the investigators noted. Published papers recommend taking from as little as 2 to 3 mL of fluid to as much as 250 mL or more. However, the Henry Ford findings suggest that when malignant cells are present, they can be identified in small volumes of fluid as often as in large volumes. In short, sensitivity is not dependent on the volume of pleural fluid extracted.

Taking a large amount of pleural fluid carries with it the risks of pneumothorax and re-expansion pulmonary edema, and it may be significantly more painful for the patient. Taking small samples eliminates these risks.

In an editorial, Michael H. Baumann, MD, Professor of Pulmonary and Critical Care Medicine at the University of Mississippi Medical Center in Jackson, hailed the study as having “successfully unmasked a myth.”[2] He observed that the popular view of pleural sampling has been “if some is good, more must be better.”

Aside from the risks involved in taking large samples, cytopathology laboratories often have limits on how much pleural fluid can be processed. Using small samples would cut down on costs and processing time. Dr. Baumann suggested a reassessment of clinical practices that are not supported by research findings.

—Gale Jurasek

References
1. Sallach SM, Sallach JA, Vasquez E, et al. Volume of pleural fluid required for diagnosis of pleural malignancy. Chest. 2002;122:1913-1917.
2. Baumann MH. A pulmonary myth unmasked? Chest. 2002;122:1875-1877.

Return to table of contents