| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Guest Access | Sign In via User Name/Password
|
|
|
|||||||
Guest Access | Sign In via User Name/Password |
|||||||||
Dr. Baumann is Professor of Medicine, Division of Pulmonary and Critical Care Medicine, University of Mississippi Medical Center.
Correspondence to: Michael H. Baumann, MD, FCCP, Professor of Medicine, Division of Pulmonary and Critical Care Medicine, University of Mississippi Medical Center, 2500 North State St, Jackson, MS 39216-4505; e-mail: mbaumann{at}medicine.umsmed.edu
Increasingly, evidence-based publications have put the practice of pulmonary medicine on firmer scientific footing. Physicians should, however, remain vigilant for the "pulmonary myth." Daily practice of medicine generally continues to embrace practices often established by anecdotal information and common sense which are often found, after further investigation, to be flawed. Many of these practices have reached mythic proportions due to continued daily assertion on teaching rounds and their ongoing use in various practice environments. One such example is that the submission of a large volume of pleural fluid for cytologic evaluation, from patients with suspected pleural malignancy, has an improved yield over smaller volumes. This pulmonary practice appears to have been successfully unmasked as a myth in this edition of CHEST (see page 1913) by Sallach and colleagues.
The suggested volume of pleural fluid to be submitted for cytologic evaluation varies greatly in the literature. No accompanying data support the various volumes suggested. The minimum volume noted to be successful, 2 to 3 mL, is accompanied by the "suggestion" that the optimal volume for analysis is 100 mL.1 The largest volume suggested is several hundred milliliters. However, accompanying this suggestion is the comment that this volume may not improve the initial diagnostic yield.2 Older3 and more current4 literature in cytopathology journals compound the issue by providing no direction regarding the appropriate amount of pleural fluid volume needed for cytologic evaluation. Finally, neither the recent American Thoracic Society (ATS) official statement on the management of malignant effusions5 nor the most recent edition of a textbook of pleural diseases6 provide definitive guidance regarding appropriate submission volumes. Into this factual void comes clinical common sense filling the gap with, "if some is good, more must be better." Understandably, many clinicians have assumed that the submission of a large volume of pleural fluid should improve the diagnostic yield in the setting of suspected pleural malignancy.
Sallach and colleagues begin to nicely fill this factual void with a retrospective review of 374 thoracentesis samples from 282 patients. Samples were classified into quartiles based on the fluid volumes submitted, with each quartile having approximately equal numbers of samples. The volumes of fluid assessed were as follows: 
10 mL, 15 to 80 mL, 100 to 775 mL, and 800 to 2,800 mL. All patients who ultimately received a diagnosis of pleural malignancy had proof of diagnosis by cytology or pleural biopsy. Communication with the authors notes confirmation of pleural malignancy in some cases by autopsy. Importantly, neither the diagnostic sensitivity nor the negative predictive value for pleural malignancy were found to depend on the volume of pleural fluid submitted. Hence, submission of as little as 10 mL of pleural fluid appears adequate for diagnosing a pleural malignancy. As might be expected, samples from patients with no history of malignancy had higher negative predictive values than those from patients with a history of malignancy. Inexplicably, samples collected from women had a higher sensitivity for predicting pleural malignancy. No other factors examined had an effect on sensitivity or negative predictive value of each quartile of pleural fluid volume. The study also confirmed earlier findings that a combination of analyses, both cell blocks and smears, increases the diagnostic yield.3
The linear relationship between volume and diagnostic sensitivity for pleural carcinoma was in a positive direction, implying increasing sensitivity with increased pleural volume. This positive relationship was not statistically significant but points to the limits of this retrospective study. As the authors note, a prospective evaluation may find a cut point at a higher pleural fluid volume whereby diagnostic sensitivity could be increased. Meantime, the study, despite being retrospective, has few other limits.
Until a prospective study is performed, the data presented reasonably support that smaller pleural fluid volumes are diagnostically as valuable as larger volumes. This approach has several advantages that are immediately obvious. In the cytopathology laboratory of my institution, the work, time, and expense incurred by large pleural volume assessment is substantial. The largest sample of fluid that can be centrifuged in our laboratory is 50 mL, with eight total slots available in the centrifuge. Hence, at most, 400 mL of fluid can be spun down at one time. Each of the eight samples then has to be further processed by making a cell block out of the resulting cell pellets. Currently, our cytopathology laboratory only supports assessing a single 50-mL pleural fluid sample due to the time and expense involved.
Further, the data supporting small-volume diagnostic thoracentesis help preclude issues related to large volume thoracentesis, such as reexpansion pulmonary edema and the development of significant negative pleural pressures. The ATS management statement on malignant pleural effusions notes that the volume of fluid that can be removed at thoracentesis is unknown. It is suggested that pleural fluid pressure monitoring be done during large volume removal,5
not a common practice nor one easily pursued. The development of excess negative pleural pressures (< - 20 cm of water pressure7
or - 19 cm of water pressure8
during pleural fluid drainage of 1 L or 0.5 L, respectively) should be avoided and may herald the presence of a trapped lung.7
8
Safe pleural fluid removal may continue as long as pleural pressures do not fall below - 20 cm H2O of pressure.7
The ATS statement, however, notes that 1 to 1.5 L of fluid may be removed at one sitting without pressure monitoring, as long as dyspnea, chest pain, or severe cough do not develop. However, if the patient presents without contralateral or with ipsilateral mediastinal shift, precipitous falls of pleural pressure may occur during thoracentesis and removal of < 300 mL of fluid is suggested.5
All of these concerns are obviated with a small volume diagnostic thoracentesis.
Why does common sense fail regarding the value of a large volume of pleural fluid? The limited yield of closed pleural biopsy in the setting of pleural malignancy provides a clue. Pleural fluid cytology yields a positive result in up to 87% of cases, while pleural biopsy is generally < 60%.9 The comparatively limited yield of pleural biopsy in the setting of pleural malignancy appears due to several issues: the focal nature of pleural involvement, areas inaccessible by pleural biopsy,10 and less frequent involvement of the parietal pleura than the visceral pleura.11 By contrast, the exfoliated malignant cells from these nonhomogeneously distributed parietal and visceral pleural sites are distributed within the malignant pleural effusion and should be found with equal likelihood within 10 mL or 1,000 mL of pleural fluid. However, these malignant cells may degrade and repeat thoracentesis may increase the diagnostic yield due to the exfoliation of fresher, more readily recognized malignant cells.6 Therefore, repeat thoracentesis in the setting of a suspected malignant pleural effusion may increase the cytologic yield.12 13
The trend, albeit not statistically significant, toward increased yield with increasing volumes of pleural fluid if proven in a prospective study might be explained as follows. Pleural fluid cells may not be homogenously distributed within the pleural fluid. Cells may settle in a gravity-dependent gradient that results in a nonhomogenous distribution that creates an environment whereby a larger volume will more likely recover malignant cells. This seems feasible in patients who often have limited activity and remain recumbent for long periods due to their underlying malignancy.
What should the diagnostic approach to pleural malignancy be today in light of currently available information? Diagnostic thoracentesis should remain the mainstay, but use of smaller fluid volumes is substantially supported by the study by Sallach and colleagues. Samples should be assessed by both cell blocks and smears.3 A negative initial thoracentesis finding should lead to a repeat thoracentesis and even perhaps a third thoracentesis.13 One of these repeat procedures might reasonably include a closed pleural biopsy, particularly if thoracoscopy is not readily available. Combined pleural fluid cytology and closed pleural biopsy can have a yield of up to 90% for pleural malignancy.13 If suspicion remains high and the patient is informed of the available options, including expectant monitoring, a diagnostic and potentially therapeutic thoracoscopy is a reasonable next step given a yield of > 90% for malignancy in this setting.9 Tempering the entire process should be the knowledge, appropriately shared with the patient, that mortality in diagnosed pleural malignancy may be as high as 48% in 6 months.14
To conclude, physicians should draw on the current study and other similar studies as an impetus to reevaluate their daily practice for their use of unsupported myths vs research-supported fact in the care of their patients. Such a reassessment will lead to further valuable studies such as that by Sallach and colleagues, and ultimately better care for our patients.
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
