PART TWO
Diagnostic Cytology
to explain the patient's clinical picture or endoscopic features
cannot be rendered, then endoscopy may have to be repeated.
This greatly increases the expense. Finally, in some laboratories
cytology has a more rapid turn-around time than do the concur-
rently obtained biopsy specimens. This is clearly the case with
on-site evaluation of submucosal aspirates.
Despite these advantages, cytology is currently underutilized
in patients with GI or hepatobiliary disease. Overall GI cytology
is still extremely useful in the detection of infections and pre-
malignant and malignant lesions. A common theme presented
throughout this chapter is the complimentary value of cytology
and histology in providing an increased diagnostic yield.
Specimen Collection and Preparation
Relative to many other body sites, the examination of exfolia-
tive cytologic specimens from the alimentary tract has lagged
behind for a number of reasons. In part, this has been due to a
lack of satisfactory methods for collecting large numbers of well-
preserved cells, both normal and neoplastic, from these hollow
organs.1
For several decades, cellular samples were procured
from both the upper and lower GI tracts by lavage methods.
These procedures entailed instilling large volumes of fluid (usu-
ally saline) orally or rectally, and then maneuvering the patient
into various positions so that the liquid would make contact
with the entire mucosal surface of the organ being sampled.
Largely, these methods were not highly popular with medical
personnel and/or the patients. Generally, these washing proce-
dures are no longer used for the detection of neoplasms of the
upper GI tract. Renewed interest in the use of washing cytology
to detect adenocarcinomas of the large intestine has received
limited attention.8-10
As mentioned earlier, the current diagnostic era for GI tract
cytology was ushered in by the flexible fiberoptic endoscope,
which permits visually directed brushings of any grossly evident
mucosal abnormality. Briefly, the cardinal component of this
endoscope is the fiber bundle that transmits the images. This bun-
dle is composed of hundreds of delicate glass fibers, each com-
posed of two types of glass, the outer one having a lower refractive
index. Light travels down the fiber bundle to create the image that
is magnified in the ocular portion of the endoscope. The flexible
nature of this endoscope allows it to traverse through the tortu-
ous lumen of the GI tract. This also allows the scope to undergo
retroflexion so that the viewer may look "backward." Endoscopes
possess a specific channel through which biopsy forceps and cyto-
logic brushes may be passed. Brushes are most often composed of
nylon and may be either disposable or reusable following sterili-
zation. Most brushes are protected by a sheath that prevents loss
of cellular material as it is removed from the endoscope. Several
studies have demonstrated that the diagnostic yield by cytologic
methods is enhanced when the brushings are procured prior to
endoscopic biopsy.11,12 The yield for histologic diagnosis is not
affected by the order in which the specimens are obtained.
Although very rarely utilized, salvage cytology has been advo-
cated by some as a quicker and easier cytologic method than
brushings.13-15 Usually obtaining the salvage specimen occurs
after endoscopic biopsies of lesions suspected of being malig-
nant. The goal of salvage cytology is to retrieve and process
material present on the external surface of the biopsy forceps
dislodged during withdrawal of the forceps through the biopsy
channel. The channel is flushed with saline (or occasionally
fixative), and cellular material is collected in a mucus trap. The
cellular material in the trap is concentrated and processed either
as smears or cell blocks.
The most recent addition to the diagnostic armamentarium
has been the use of transmucosal fine-needle aspiration (FNA)
biopsies. Transmucosal FNA of intramucosal GI masses or tumors
in adjacent organs has enjoyed increased use due to the develop-
ment of endoscopic ultrasound (EUS). This technique also allows
real-time visualization of the mass and surrounding structures for
more accurate placement of the aspirating needle.16-19 This is gen-
erally used for detectable submucosal masses and/or infiltrative
lesions.16,20-26 This allows sampling of cellular material deeper in
the wall of the GI tract that cannot be obtained by either lavage
or brushings. The aspirated material may be processed for direct
smears and/or cell blocks or liquid-based preparations (LBP).
Several authors have advocated combining endoscopic ultra-
sound with the aspiration procedure.16 This technique allows
aspiration of deeper submucosal or extensive lesions that cannot
be reached by brushing or lavage cytology due to intact overlying
mucosa. Transmural FNA cytology has been especially useful in
diagnosing submucosal neoplasms such as lymphomas, carci-
noid tumors, stromal tumors, and stenotic tumors or lesions that
are diffusely infiltrative or necrotic masses.
Although it is not widely practiced in the United States, some
authors strongly support use of the tissue biopsy specimen to
produce cytologic preparations.27-30 Imprint cytology examines
touch preparations of biopsy specimens prior to tissue fixa-
tion; the biopsy is then processed routinely for histology. These
imprints do not hinder the subsequent histologic evaluation of
the biopsies. Crush preparations, on the other hand, consume
the tissue as it is smeared between slides.
Material obtained by cytologic brushing is most often pre-
pared as direct smears on slides. As with brushings of other body
sites, very rapid fixation is imperative. Some authors prefer to
rinse material from the bristles of the brush; the sediment so
collected is then concentrated in smears.31,32 Similarly, cellular
material obtained by lavage is concentrated and smeared or col-
lected on filters. Material obtained by salvage cytology, suction,
and transmucosal aspiration may be processed as direct smears
and/or cell block preparations.33
Most workers use the Papanicolaou stain on alcohol fixed
material. Some authors prefer other stains such as a Romanowsky
modification or hematoxylin and eosin.34 The application of
cytochemical, immunocytochemical, and image morphometry
has been limited in alimentary tract materials.
A major disadvantage to cytology specimens obtained by
brushings on direct smears is air-drying artifact and smear thick-
ness with cellular overlap. Recently the use of automated LBPs
including ThinPrep or SurePath preparation systems have been
introduced, reducing both the drying and overlap problems. The
advantages of semi-automated LBPs include better cell prepara-
tion, easier to interpret smears due to less overlap, thickness,
and a clearer background, resulting in a shorter screening time.
Another advantage is the availability of material for ancillary
studies. Some disadvantages include higher cost, expense of
equipment and reagents, more labor-intensive preparation time,
and increased time for the learning for LBPs. Wang et al. found
ThinPrep slides at least comparable to direct smears in cytologic
examination of brushings.35
In addition to visually directed brushings, nonendoscopic or
"blindly obtained" material may be retrieved by other abrasive
techniques. The use of brushes that have been passed through
374
previous page 370 ComprehensiveCytopathology 1104p 2008 read online next page 372 ComprehensiveCytopathology 1104p 2008 read online Home Toggle text on/off