Diagnostic Cytology
These granules may be so numerous that they crowd the cyto-
plasm. We have observed lipid-containing neutrophils in serous
effusions from patients with a wide variety of neoplastic and
non-neoplastic conditions, an observation confirmed by reports
describing them in pleural effusions caused by tuberculosis,
rheumatoid disease, and various neoplasms.28-30 Whenever neu-
trophils contain lipid droplets, they are also likely to be found
in the cytoplasm of histiocytes and mesothelial cells in the effu-
sion. These lipidic granules are not visible in Papanicolaou-
stained smears or cell-block preparations.
Key features of neutrophilic leukocytes
• Readily identifiable in all types of preparation;
• Cytoplasm granules not prominent;
• Nuclei usually trilobed; and
• Possibly necrotic in inflammatory lesions.
Eosinophilic Leukocytes
Eosinophilic Pleural Effusion
Several comprehensive reviews during the early days of cyto-
diagnosis attest to the fascination that eosinophilic effusions,
especially pleural effusions, have held for clinicians.31 How-
ever, the arbitrary criterion for designating a pleural effusion as
eosinophilic has varied from an effusion with as few as 5% to
one with at least 50% of the cells being eosinophils; we have
adopted the criterion of a concentration of 10% or more as used
by Koss.32
Eosinophilic pleural effusion has been reported in association
with a wide variety of conditions: allergy, autoimmune disorders,
pneumonia (including viral pneumonia), pulmonary infarct,
fungal infection, parasitic infection, malignant neoplasms, pul-
monary tuberculosis, artificial and spontaneous pneumothorax,
and hemothorax, including that induced by thoracic trauma.14,31,33
Eighty-one of the 127 (64%) cases of eosinophilic pleural effu-
sion analyzed by Spriggs and Boddington and 8 of 30 (27%) ana-
lyzed by Veress and co-workers were associated with some form
of thoracic trauma.14,33 Apart from accidental trauma, these series
include the trauma associated with therapeutic pneumothorax,
thoracotomy, and repeated aspiration of pleural effusion.
Most of these examples of trauma would have been associ-
ated with some degree of hemothorax, suggesting that red blood
cells exert a chemotactic influence on eosinophil migration into
the pleural cavity. However, Spriggs and Boddington, noting
that hemothorax, frequently mentioned as a cause of pleural
eosinophilia, is often associated with pneumothorax, raised
the question of whether the eosinophilia is induced by air in
the pleural cavity.14 Even if blood does stimulate pleural eosi-
nophilia, it is important to remember that a high proportion of
eosinophilic pleural effusions are not bloodstained.34
In keeping with air as the stimulator of pleural eosinophilia,
the literature from the era of therapeutic pneumothorax well
documents the appearance of eosinophils in the resulting effu-
sions.35-37 This observation prompted Spriggs to reexamine pleu-
ral fluids from seven cases of spontaneous pneumothorax, of
which five were eosinophilic.38 He concluded that pleural eosi-
nophilia is the normal reaction of the pleura to the introduction
of air into the pleural cavity and suggested that the reaction is
more likely a result of suspended particles of animal or vegeta-
ble origin than of the air itself.
When pneumothorax can be excluded, the commonest
causes of eosinophilic pleural effusion seem to be pulmonary
Fig. 19.38 Smear of empyema fluid containing numerous neutrophilic
leukocytes. Many have become necrotic, manifested as light gray-blue,
fluffy background material (Papanicolaou x MP).
infarct, pneumonia, and neoplasm.9 Hodgkin's disease is rarely
associated with eosinophilic pleural effusion. Only a minority
of cases can be attributed to recognizable hypersensitivity states,
including parasitic infestation. Another cause of pleural eosi-
nophilic effusion is the benign type of effusion associated with
the inhalation of asbestos;39 in a series of 60 cases, 26% showed
eosinophilia of various degrees.40
When all of the possible causes of eosinophilic pleural effu-
sions have been eliminated, there is a substantial residue of cases
in which the cause cannot be identified. For example, in 11 of 30
cases of Veress and co-workers and 11 of 23 consecutive cases of
eosinophilic pleural effusion in our laboratory (an aggregate of
42%), no data adequately accounted for the effusions.33
Despite the uncertainty surrounding the pathogenesis of
idiopathic eosinophilic pleural effusions, they seem to have a
favorable prognosis even in patients with a previous history of
cancer. This was well illustrated by Veress and co-workers, who
found that in 22 of their 30 patients, including 6 with a past
history of cancer, the effusions eventually disappeared.33 Six of
the 30 patients died of myocardial infarction, and two were una-
vailable for follow-up. Their series also showed a predominance
of males with eosinophilic pleural effusions, in keeping with
previous observations.14,34,41-43
Because it has not been possible to identify a single linking
pathologic mechanism behind the formation of eosinophilic
pleural effusion, it has been suggested that these effusions exist
in two forms: an effusion of relatively acute onset related to an
allergic reaction or to thoracic trauma and the more chronic
form with a longer clinical course.14,44,45
Eosinophilic Peritoneal Effusion
Eosinophilic peritoneal effusions are rare; we have no record
of such an effusion in our laboratory. Examples of eosinophilic
peritoneal effusion associated with malignant neoplasm, vari-
ous allergic states, parasitic infection, eosinophilic gastroenteri-
tis, and chronic peritoneal dialysis have been reported.46-59 It is
possible that one or more of the agents used in peritoneal dia-
lysate, such as antiseptic, talc, particles of tubing, and peritoneal
catheters, provoke a hypersensitivity reaction characterized by
an efflux of eosinophils. In addition to eosinophilia, chronic
peritoneal dialysis may stimulate mesothelial hypertrophy and
previous page 524 ComprehensiveCytopathology 1104p 2008 read online next page 526 ComprehensiveCytopathology 1104p 2008 read online Home Toggle text on/off