Year : 2008 | Volume
: 25 | Issue : 4 | Page : 138--143
Cells in pleural fluid and their value in differential diagnosis
Rashmi Kushwaha1, P Shashikala2, S Hiremath3, HG Basavaraj3,
1 Safdarjung Hospital, New Delhi, India
2 Department of Pathology, S.S.I.M.R.S. Davangere, India
3 J.J.M. Medical College, Davangere, India
H. No. F 130/4, Second Floor, Gautam Nagar, New Delhi 110 049
Background : Both non-malignant and malignant causes of effusion can be identified by the relatively non-invasive technique of pleural fluid cytology. With this basis the present study on cytology of pleural fluids was taken up. The diagnostic significance of the cytologic study of the fluid may be attributable to the fact that the cell population present in the sediment is representative of a much larger surface area than that obtained by needle biopsy.
Materials and Methods : One hundred samples of pleural fluid were examined for total cell count, cell type and cellular features. They were also subjected to biochemical study to find out the level of protein, glucose and chloride.
Results : A total of 82% samples were exudative and 18% were transudative. Total leukocyte count (TLC) was less than 1000 cells/cu.mm in most (88.89%) of transudative effusions. Overall 52.44% of exudative effusions had TLC greater than 1000 cells/cu.mm. It was noted that 96.88% of tuberculous effusions had more than 50% lymphocytes, 81.25% had protein greater than 5 gm/dl and 90.63% had glucose greater than 60 mg/dl. Approximately 28% of pleural effusions were positive for malignant cells. Most (82%) of malignant effusions were exudative. The primary site could be assessed by cytological examination in 57.14% of malignant effusions.
Conclusions: The most useful test in establishing the diagnosis of pleural effusion is pleural fluid cytology and pleural fluid cell count. Cytologic study of pleural fluid is a complete diagnostic modality which aims at pointing out the etiology of effusion as well as, in certain cases, a means of prognostication of disease process.
|How to cite this article:|
Kushwaha R, Shashikala P, Hiremath S, Basavaraj H G. Cells in pleural fluid and their value in differential diagnosis.J Cytol 2008;25:138-143
|How to cite this URL:|
Kushwaha R, Shashikala P, Hiremath S, Basavaraj H G. Cells in pleural fluid and their value in differential diagnosis. J Cytol [serial online] 2008 [cited 2022 Oct 5 ];25:138-143
Available from: https://www.jcytol.org/text.asp?2008/25/4/138/50799
Patients with an undiagnosed pleural effusion should be evaluated in an individualised stepwise manner. The initial two steps should be relatively non-invasive and include clinical evaluation and cytologic study.
Aspiration of serous cavities is a simple and relatively non-invasive technique to achieve a diagnosis. The information provided by body fluid analysis serves several functions, first, it assists the clinician in formulating, in order of priority, a list of differential diagnoses, second it allows one to follow the result of therapy. 
Thus cytological study of body effusions is a complete diagnostic modality which aims at pointing out the etiology of effusion as well as in certain cases a means of prognostication of the disease process. The diagnostic performance of the cytologic study of the fluid may be attributable to the fact that the cell population present in sediment is representative of a much larger surface area than that obtained by needle biopsy. ,
Materials and Methods
This study on pleural fluid cytology was undertaken in the Department of Pathology, J.J.M. Medical College, Davangere, Karnataka over a period of one year study. Relevant and available clinical information regarding age, sex, symptoms and accompanying signs was obtained from the patients.
An effort was made in this study to process the pleural fluid specimens as expeditiously as possible, the majority were processed immediately. But in a small number, when there was a delay, these specimens were stored in the refrigerator at 4°C. The fluid was divided into two parts, one part was used for cell count and the other part was poured into the centrifuge tubes and centrifuged for 10 minutes at 2000 rpm. The supernatant was poured off. Part of the sediment was transferred to a clean glass slide and mixed with a part of 1% toluidine blue. After placing the cover slip, the slide was observed under the microscope for immediate identification of cell morphology. Then the remaining sediment was transferred with the help of a Pasteur pipette to three slides coated with albumin. One was air dried and stained with Giemsa, the other two were fixed in 95% alcohol for a minimum period of 15 minutes and stained with Hematoxylin and Eosin, Papanicolaou stains.
For hemorrhagic fluids, glacial acetic acid or 0.1 N HCl was used as hemolysing agent and then they were processed. For cell count one drop of fluid was mixed with a drop of toluidine blue and the cells were counted in an improved Neubauer counting chamber. For all cases biochemical analysis of protein, sugar and chloride was done. Wherever possible bacteriological culture of pleural fluid was done.
The medical records of all 100 patients with pleural effusion were available for review. The ages of the patients ranged from four years to 75 years, with maximum cases (29.36%) in the sixth decade. Male preponderance of effusion was noted; the ratio of male to female being 1.2:1. Of all the effusions, 31.19% were hemorrhagic; 26.61% turbid, 23.85% clear, 11.93% purulent, 3.67% viscous and 2.75% opalescent. Transudates comprised 18% of cases. Most of them were secondary to cirrhosis and congestive cardiac failure [Table 1].
Overall 82% of pleural effusions were exudative in nature. The most frequent cause of exudative effusion was tuberculosis (39.02%) followed by malignancy (30.49%), paramalignancy (8.54%), pneumonia (7.32%), and empyema (4.89%). In 53.57% of malignant effusions there were predominantly lymphocytes. All samples of parapneumonic effusions had polymorphonuclear cells as the predominant cell type. Approximately 82% of effusions had protein level greater than 3 gm%, hence were exudative [Table 2].
Most of the transudative effusions were clear in appearance and most of the exudative effusions were hemorrhagic.
Clinically 39 cases were diagnosed to be tuberculous but with cytological and biochemical study only 32 cases were confirmed to be of tuberculosis. Approximately 96.88% of samples of tuberculous effusion had lymphocyte count greater than 50% [Figure 1a], 81.25% had protein level greater than 5 gm% and 90.63% had pleural fluid glucose greater than 60 mg/dl [Table 3]. In the present study we found one case of eosinophilic pleural effusion. It had 15% eosinophils.
Malignant cells were present in pleural fluid of 28 patients. Out of them 89.29% samples were exudative and 10.71% were transudative. 71.43% of malignant effusions were hemorrhagic.
Diagnosis of metastatic adenocarcinoma was made in pleural fluid cytological examination of nine cases. All the fluids showed sheets, clusters and acini of large tumor cells having large dense nuclei and pale cytoplasm. Around 55% of patients were in the sixth decade. It was noted that 88.89% were exudative effusions and all samples were hemorrhagic. Diagnosis of metastatic papillary carcinoma was made in four samples. Smears showed clusters and papillary groups of large abnormal epithelial cells having large dense nuclei and conspicuous nucleoli. Nuclear chromatin was coarse. A few psammoma bodies were seen. 50% of patients were in the sixth decade. All samples were exudative and hemorrhagic [Figure 1b].
Mesothelioma was diagnosed in six samples. Mesothelial cells were arranged singly or in three dimensional groups forming spherical, morule-like configuration with knobby borders. Single atypical cells were also present. Atypical nuclear features like variation in nuclear size, shape, multi nucleation, hyperchromasia and enlarged nucleoli were seen. All samples were exudative and 66.66% samples were viscous in appearance [Figure 2a].
Metastatic mucin secreting carcinoma was seen in a pleural fluid sample from a 60-year-old female patient who was clinically diagnosed to have malignant ovarian tumor. The fluid was exudative in nature. Smears showed clusters of abnormal epithelial cells having mucin filled cytoplasm and eccentric large nuclei with conspicuous nucleoli [Figure 2b]. Metastatic small cell carcinoma was seen in the pleural fluid sample of a 75-year-old male patient who was clinically diagnosed to have tuberculosis. The fluid was exudative in nature. Smears showed scattered and loose cohesive clusters of small abnormal epithelial cells two to two-and-a-half times the size of lymphocytes, having large dense nuclei with coarse chromatin and scant cytoplasm. Nuclear and cytoplasmic molding was seen [Figure 3a]. Anaplastic large cell lymphoma was seen in the pleural fluid sample of a 12-year-old female patient. The fluid was transudative in nature and hemorrhagic. Smears showed lymphocytes, neutrophils and macrophages. There were scattered and loose aggregates of large pleomorphic densely staining cells with cleaved nuclei and pale cytoplasm. Non-Hodgkins lymphoma was diagnosed in the pleural fluid sample of a 50-year-old male patient with clinical diagnosis of tuberculosis. The fluid was exudative in nature. Smears showed predominantly large lymphoid cells with cleaved nuclei, speckled chromatin and scanty pale vacuolated cytoplasm [Figure 3b].
Out of 28 samples of malignant pleural effusions, six (21.43 %) had the primary site in the pleura, two (7.14%) each in breast and ovary, two (7.14%) had effusion positive for lymphoma/ leukaemia, one (3.57%) had effusion due to small cell lung carcinoma and the same number of patients had the primary site in the cervix.
Clinically 39 samples were diagnosed to be tuberculous in origin but tuberculosis was cytologically confirmed in 32 cases. Among the remaining seven cases of clinically suspected tuberculous effusion, two were of metastatic adenocarcinoma, one each of mesothelioma, metastatic papillary carcinoma, metastatic small cell carcinoma, non-Hodgkin's lymphoma and metastatic carcinoma undifferentiated. Out of 25 clinically suspected malignant effusions 18 were diagnosed to be malignant after cytological examination. The remaining seven samples turned out to be paramalignant. 100% samples of clinically diagnosed pneumonia were confirmed to be of pneumonia on cytological examination [Table 4].
The relative ease of pleural fluid aspiration, analysis and cytological examination has kept alive the search for a test to unequivocally differentiate the various causes of effusion. The cytological examination of body effusion is a complete diagnostic modality which aims at pointing out the etiology of effusions. The diagnostic performance of the cytologic study of the fluid may be attributable to the fact that the cell population present is representative of a much larger surface area than that obtained by needle biopsy. , The present study deals with the accuracy of diagnosis on the basis of contemporary cytological features and cell count. A total of 100 samples of pleural fluid were studied.
The clinical presentation of malignant tumor is an indispensable, effective aid to a cytopathologist and may increase the yield by several times.  So an attempt was made to study the relationship between clinical and cytological diagnosis.
One study showed that the incidence of dull chest pain is higher in malignant disease, while pleuritic chest pain is higher in patients with benign disease. Symptoms of less than seven days occur mostly in benign diseases. Marel et al. ,  also reported similar findings.
The present study showed slight male preponderance with female to male ratio of 1:1.22. Our study is in concordance with the study done by Romero et al .  In our study most of the effusions (82%) were exudative in nature. Transudates comprised 18% of cases. Most of malignant effusions were exudative except three cases which were transudates which could be explained on the basis of concomitant anemia and hypoproteinemia. In a study done by Sherwani et al. ,  two cases of malignancy had transudates due to similar reason.
The most frequent cause of exudative effusion was tuberculosis (39.02%) followed by malignancy (30.49%), paramalignant effusion (8.54%), pneumonia (7.49%) and empyema (4.89%). The remaining cases were due to trauma, liver disease, hepatic abscess, pelvic abscess, myocardial infarction, and pancreatitis. Hence our study is in accordance with the study done by Alusi. 
Transudative effusions are usually characterised by a majority of lymphocytes or other mononuclear cells. In our study also 83.33% samples of transudative effusion had more than 50% lymphocytes. When exudative effusions are considered, all but one (96.86%) of tuberculous effusions had more than 50% small lymphocytes. This pattern of lymphocyte predominance was observed in malignant (44%) and paramalignant (71.93%) effusions. Thus our study correlates with the study done by Light and Erozan.  The pattern of predominantly polymorphonuclear cells was seen in most effusions secondary to pneumonia, post myocardial infarction, empyema, hepatic abscess, and pelvic abscess.
Our study is in accordance with the study done by Epstein et al.  and showed that the majority of tuberculous effusions had more than 50% lymphocytes, 81.25% had greater than 5 gm/dl of protein and 90.63% had glucose greater than
50 gm/dl. Aggarwal et al. ,  showed that tuberculous effusions rarely contain more than 5% mesothelial cells which is in agreement with our study.
The presence of predominantly polymorphonuclear cells in pleural fluid indicates that the fluid is the result of acute pleural inflammation, hence raising the probability of pneumonia with effusion. In the present study we got a case of a 55-year-old male patient with a history of trauma, where 15% eosinophils were found in the pleural fluid. So a diagnosis of eosinophilic pleural effusion was made. Fluid was hemorrhagic in nature. This can be attributed to the fact that the blood in the pleural space acts as an eosinotactic substance. Undoubtedly, there are other factors involved since a significant number of eosinophilic effusions are non-hemorrhagic and not all hemorrhagic effusions are eosinophilic.  Air in pleural space has been shown to be another cause of eosinophilic inflammatory response.
It is possible to diagnose the type and source of malignant tumor cells in serous effusions with an overall accuracy of 50%. In the present study, out of 28 samples with malignancy, the primary site could be confirmed on cytology in 16 (57.14%) of cases. In our study the most common primary site was the pleura (21.43%) followed by the breast, ovary and lymphoma / leukaemia, each comprising 7.14%. Prostate, gall bladder, female genital tract and lung made up 3.57%.
In the present study pleural fluid cytology was not helpful in ascertaining the cause of pleural effusion in 12 (42.86%) of patients, thereby indicating its limitations.
Malignant pleural effusion was found to affect females more than males (2.1:1), which is in agreement with the study done by Sears et al . 
A single variable which strongly predicts malignancy is "bloody fluid".  In our study 68.97% of hemorrhagic effusions were positive for malignancy. But all hemorrhagic fluids need not be due to malignancy and non-hemorrhagic fluids can have malignant cells.
In the present study mesothelioma was the second most common malignancy accounting for six samples (21.43%). Mesothelial cells were arranged singly or in three dimensional groups. These cell clusters had spherical morule like configuration. Single atypical mesothelial cells were also present. Atypical nuclear features, such as variation in nuclear size, shape, multinucleation, hyperchromasia and enlarged nucleoli were seen. The cells had dense perinuclear cytoplasm, centrally placed nuclei and slit like spaces between some cells. There was also focal vacuolisation of the cytoplasm and a prominent lymphocytic infiltrate. Grandos et al. ,  showed similar findings in cytologic analysis of fluid from malignant mesothelioma.
In the present study there was only one case of metastatic small cell carcinoma of the lung in a 75-year-old male patient. Grossly, the fluid was hemorrhagic. Smears showed neoplastic cells arranged singly or in small clusters. Individual cells were about two to two-and-half times the size of small lymphocytes, with scant cytoplasm and hyperchromatic nuclei. Nucleoli were inconspicuous. Nuclear molding was noted and best appreciated within cohesive groups. Chhieng et al.  and Khan et al. ,  observed similar features in their study.
We had two cases of pleural effusion due to metastasis from breast carcinoma. Pleural fluid was hemorrhagic in both cases. Smears showed cells arranged in cohesive, tight balls with smooth borders ("proliferation spheres" or "cannon balls") [Figure 3c]. The cells were cytologically innocuous. The cytoplasm was distinctly outlined. The nucleus was moderately hyperchromatic and uniform in size. Spieler et al.  confirmed similar findings in his study.
We found two cases of metastatic pleural effusion due to primary foci in the ovary. One was diagnosed as metastatic mucin secreting adenocarcinoma and it showed clusters of abnormal epithelial cells with mucin filled cytoplasm and eccentric large nuclei with conspicuous nucleoli. The other case was of metastatic papillary carcinoma. It showed papillary clusters of abnormal epithelial cells with large densely staining nuclei and pale cytoplasm. Psammoma bodies were present.
The present study demonstrates that the most useful test in establishing the diagnosis of pleural effusion is pleural fluid cytology and pleural fluid cell count. Cytologic study of pleural fluid is a complete diagnostic modality which aims at pointing out the etiology of effusion as well as, in certain cases, a means of prognostication of disease process. The diagnostic performance may be attributable to the fact that the cell population present in the sediment is representative of a much larger surface area than that obtained by needle biopsy. Detailed study of cytomorphologic features of various metastatic malignant cells in pleural effusions provide definite clues regarding the primary site.
Thus patients with an undiagnosed pleural effusion should be evaluated in an individualised stepwise manner. The initial two steps should be relatively non-invasive and should include clinical evaluation and cytologic study.
|1||Cheson BD. Clinical utility of body fluid analysis. Clin Lab Med 1985;5:195-208.|
|2||Frist B, Kahan AV, Koss LG. Comparison of the diagnostic values of biopsies of the pleura and cytologic evaluation of pleural fluids. Am J Clin Pathol 1979;72:48-51. |
|3||Sherwani R, Akhtar K, Naqvi AH, Akhtar S, Abrari A, Bhargava R. Diagnostic and prognostic significance of cytology in effusions. J Cytol 2005;22:73-7. |
|4||Marel M, Stastny B, Melinova L, Svandova E, Right RW. Diagnosis of pleural effusions: Experience with clinical studies, 1986-1990. Chest 1995;107:1598-603. |
|5||Romero S, Candela A, Martin C, Hernandez L, Trigo C, Gil J. Evaluation of different criteria for the separation of pleural transudates from exudates. Chest 1993;104:399-404. |
|6||Alusi FA. Pleural effusion in Iraq: A prospective study of 100 cases. Thorax 1986;41:492-3. |
|7||Light RW, Erozan YS, Ball WC. Cells in pleural fluid. Arch Intern Med 1973;132:854-60. |
|8||Epstein DM, Kline LR, Albelda SM, Miller WT. Tuberculous pleural effusions. Chest 1987;91:106-9. |
|9||Aggrawal AN, Gupta D, Jindal SK. Diagnosis of tuberculous pleural effusion. Indian J Chest Dis Allied Sci 1999;41:89-100.|
|10||Kumar UN, Varkey B, Mathai G. Post traumatic pleural fluid and blood eosinophilia. JAMA 1975;234:625-6. |
|11||Sears D, Hajdu SI. The cytologic diagnosis of malignant neoplasms in pleural and peritoneal effusions. Acta Cytol 1987;31:85-97. |
|12||Granados R, Libas ES, Fletcher JA. Cytogenetic analysis of effusions from malignant mesothelioma. Acta Cytol 1994;38:711-7. |
|13||Chhieng DC, Ko EC, Yee HT, Shultz JJ, Dorvault CC, Eltoum IA. Malignant pleural effusion due to small cell lung carcinoma. Diagn Cytopathol 2001;25:356-60. |
|14||Khan N, Sherwani RK, Afroz N, Kapoor. Cytodiagnosis of malignant effusion and determination of primary site. J Cytol 2005;22:107-10. |
|15||Spieler P, Gloor F. Identification of types and primary sites of malignant tumors by examination of exfoliated tumor cells in serous fluids. Acta Cytol 1985;29:753-67.|