PART TWO
Diagnostic Cytology
one that contains only a few. In the former situation, neoplasm
presumably erupted through the mesothelium, perhaps in many
areas, whereas when neoplastic cells are few, the neoplasm may
be confined mainly to subserosal connective tissue, penetrating
the overlying mesothelium in only small areas.
Effusions containing neoplastic cells usually contain benign
cells, which frequently outnumber the neoplastic cells. Meso-
thelial cells may be stimulated to undergo hypertrophy and
hyperplasia, and their number may greatly exceed that of the
neoplastic cells. In addition, such an effusion may contain mac-
rophages and lymphocytes. Such an effusion, apart from the
presence of neoplastic cells, occasionally is entirely lymphocytic.
Purulent effusions containing neoplastic cells are not common
and denote coexisting inflammation of either the serous mem-
brane or the underlying parenchyma or both. Hemorrhagic
effusions are a result of vascular congestion of the subserosal
connective tissue caused by the presence of neoplasm and are
approximately twice as common in neoplastic effusions as in
non-neoplastic effusions.2
identification of Neoplastic cells
In most situations, such as in cervicovaginal cytology, the cyto-
logic diagnosis of cancer is carried out against a background of
benign cells of the type that gave rise to the cancer cells. Further-
more, often found is an accompanying range of cells interme-
diate between the benign and malignant. In such a situation,
identification of cells that are cancerous depends on knowing
the range of deviation of cells normal to the specimen. In serous
effusions, the situation is different in that with the exception of
mesotheliomas and a few ovarian carcinomas, neoplastic cells
have a totally different appearance from any benign cells that
accompany them.
It is well recognized that the diagnosis of neoplastic cells in
an effusion does not depend on any single morphologic crite-
rion or constellation of criteria. One may gain the impression
from textbooks that the diagnosis of neoplasm in an effusion
depends on finding abnormal mitotic figures as well as cells that
are large with a high nucleocytoplasmic ratio and that possess
large, hyperchromatic, irregularly shaped nuclei with prominent
nucleoli. However, these criteria cannot always be applied to the
diagnosis of neoplastic cells in effusions. For example, the cells
of small-cell (oat cell) carcinoma of the lung and some adeno-
carcinomas are small, the cells of malignant mesotheliomas and
some adenocarcinomas may possess profuse cytoplasm with a
resulting low nucleocytoplasmic ratio, and the nuclei of mes-
otheliomas and some adenocarcinomas may not be irregular in
shape or hyperchromatic.
Obviously, in making a diagnosis of neoplastic cells in a
serous effusion, the observer subconsciously uses a large number
of criteria, some subtle, in arriving at a correct diagnosis, and the
application of these criteria depends on familiarity with the type
of specimen, the type of preparation, and the context in which
the diagnosis is being made. Essentially, one is attempting to
recognize cells that are alien to the type of specimen, a find-
ing that with few exceptions allows them to be recognized as
cancerous.
Almost all examples of neoplastic cells in serous effusions are
recognizable on routine preparations. Supplementary methods
of identifying cancer cells are occasionally used. However, they
are virtually never used to make a decision about whether the
cells are neoplastic or nonneoplastic; instead, they are applied
to demonstrate certain characteristics of particular types of
neoplasm. Immunocytochemistry has its most fruitful appli-
cation in distinguishing between adenocarcinoma and malig-
nant mesothelioma and in confirming a suspected diagnosis of
melanoma in cells that are amelanotic. The use of histochem-
istry is more or less limited to the demonstration in certain
cells of mucin, glycogen, melanin, or hemosiderin; however,
it rarely has any real practical value. Electron microscopy does
not enable one to discriminate between neoplastic and non-
neoplastic cells, although it may have some use in this respect
in selected cases. Its use is to demonstrate distinctive features
of certain types of neoplastic cells, such as osmiophilic bodies
in metastatic bronchioloalveolar cell carcinoma, melanosomes
in the cells of amelanotic melanoma, dense core granules in
neuroendocrine neoplasms, intracytoplasmic lumina in cer-
tain metastatic adenocarcinomas, and short microvilli with
central filaments in gastric and colonic adenocarcinomas. For
an account of the application of electron microscopy to cells
in serous effusions, readers should consult the monograph by
Bedrossian.6 Cytogenetic analysis can demonstrate abnormal
karyotypes in neoplastic cells, but it is too laborious, expensive,
and esoteric for routine application.
Key features of identification of neoplastic cells
• Identification does not depend on any single
morphologic criterion or constellation of criteria;
• Neoplastic cells almost always recognizable in routine
preparations; and
• Electron microscopy cannot distinguish between
neoplastic and non-neoplastic cells.
Differential Diagnosis of Types of neoplasms and
Determination of Primary Sites of neoplasms
Most neoplasms in serous effusions are readily classifiable in
terms of the type of neoplasm that gave rise to them. This is
especially true of adenocarcinomas, small-cell carcinomas,
keratinizing squamous cell carcinomas, melanomas, and lym-
phomas. Adenocarcinomas are by far the commonest type of
neoplastic cells to be found in serous effusions, and with experi-
ence, it may be possible to suggest with accuracy the primary site
of the neoplasm, especially if given information about the sex
of the patient and the site from which the fluid was obtained.
Several studies attested to this, but it is not clear how much suc-
cess in this type of divination rests solely on morphologic fea-
tures of the adenocarcinoma cells and how much depends on
a knowledge of probabilities when informed of the sex of the
patient and the site of origin of the fluid.170-172 Despite some
success in deducing the origin of adenocarcinomas from their
cells exfoliated into effusions, there are always cases in which
the deduction turns out to be wrong. From a practical point of
view, determining the site of origin of adenocarcinoma cells is
an interesting but largely redundant exercise because the clinical
background of a patient usually provides the best clues about
the origin of the cells.
Key features related to types of neoplasm
• Most neoplasms are readily classifiable as to their type;
• It is sometimes possible to suggest accurately the
primary site of a neoplasm; and
• Clinical background usually provides the best clues as
to origin of neoplastic cells.
550
previous page 544 ComprehensiveCytopathology 1104p 2008 read online next page 546 ComprehensiveCytopathology 1104p 2008 read online Home Toggle text on/off