14
Alimentary Tract (Esophagus, Stomach, Small Intestine, colon, Rectum, Anus, Biliary Tract)
Fig. 14.13
Helicobacter pylori.
The curved spiral-shaped organisms lie
unattached to epithelium (Papanicolaou x HP).
epithelial cells. By altering the microscope's focus, the small
benign nuclei are recognized within these cells. The absorp-
tive
elements
have
a tall
columnar configuration,
dense
(non-mucus-containing) cytoplasm, and a distinct brush bor-
der, which appears as an accentuation of the thickness of the
apical cell membrane.
Helicobacter
organisms measure 1-3 pm
in length and possess a curved or spiral configuration; typically,
this manifests as an S or C shape. Often, they are located within
clumps of mucus, but they are also found very close to the apical
surface of the glandular cells. According to Schnadig and associ-
ates, the long axis of the organisms typically parallels the long
axis of the mucus strands.116 Davenport warns against examining
excessively thick plugs of mucus as it may lead to a false-positive
identification of
Helicobacter.117
Overall, although cytology can-
not specifically diagnose the diffuse antral gastritis directly
associated with
Helicobacter
infection, the ready recognition of
the organism in cytologic preparations has diagnostic utility.
A number of authors have compared the ability of endo-
scopic brushings with that of biopsies for the identification of
this organism.116-120 Others have compared imprint smears of
biopsies with the histologic preparation.121-123 Various stains
have been used during these studies, including Papanicolaou,
Romanowsky, Gram, and Steiner. Almost unanimously, cyto-
logic preparations have been found to be more sensitive than
histology. Furthermore, some have shown that the organism
can be identified in smears in the endoscopic suite, permitting
early initiation of appropriate therapy. Yet, biopsy should not be
excluded in that the two techniques are complementary and his-
tologic exam provides additional data that cannot be obtained
from the cytologic preparations.
The major reason for the greater sensitivity of brushings over
biopsies for the detection of
H. pylori
probably relates to the
much greater surface area of mucosa sampled by the brush. This
concept has been supported by the results of an investigation of
Ching et al., who used the urea broth test, which is a biochemical
colormetric assay.124 These authors compared the ability of
antral biopsies and brushings to produce a positive broth
test in patients with endoscopic gastritis. Of the patients with
histologic evidence of
Helicobacter
-associated gastritis, positive
test results occurred in 97% with brushings and 76% with biop-
sies. Furthermore, the average time to develop a positive result
was 5 and 60 minutes for brushings and biopsies, respectively.
Another advantage of cytology examination is the assessment
for other organisms that can be identified.
Candida-type
organ-
isms can be seen overlying some gastric ulcers and show char-
acteristic yeast or pseudohyphae.
Candida
is not a true pathogen
but generally represents secondary colonization of benign and
malignant ulcers. Other organisms rarely noted include rod-
and cocci-shaped bacteria which are usually considered con-
taminants on gastric brushings. Clusters of filamentous-type
organisms have been described in both benign and malignant
esophageal and gastric brushings and are often associated with
malignant tumors.125 A rare case of microfilaria diagnosed on
gastric brushing cytology has been reported.126 Cytomegalovirus
and various parasites have also been reported.53,60,127 CMV viro-
cytes are usually associated with systemic disease and are most
commonly seen in patients with AIDS. In gastric brushing, the
number of CMV-infected cells are generally greater than that in
esophageal specimens. Additionally a rare case of gastric zygo-
mycosis has been reported on brushing cytology.128
Peptic Ulcer Disease
Peptic ulcers may be defined as chronic excavating lesions of the
GI mucosa secondary to the destructive actions of hydrochloric
acid and pepsin. At least 98% occur in the stomach and duode-
num, with approximately 80% occurring in the latter site. The
overwhelming majority are etiologically linked to
H. pylori
infec-
tion;115 the urease activity of this organism may initiate a cascade
of biochemical steps culminating in increased acid secretion.
They are probably the most common gastric lesions sampled
for cytologic exam.
Although peptic ulcers may occur in individuals of any age,
they are most frequent in young and middle-aged adults. The
most common clinical presentation includes a gnawing or
burning-like pain in the epigastrium occurring within an hour
or so of consuming a meal. Benign peptic ulcers produce a char-
acteristic endoscopic or gross appearance: a small (generally
less than 3 cm in diameter), round or ovoid, sharply deline-
ated defect with the adjacent mucosa. Classically, the underlying
scar tissue results in a puckering of the neighboring mucosa. It
is widely recognized that neither radiographic procedures nor
endoscopic examination can consistently and accurately distin-
guish between benign and malignant ulcers. Although many
gastric adenocarcinomas present as an ulcerating defect, this
would be unusual for a malignant lesion in the duodenum.
Consequently, brushings of duodenal ulcers are seen much less
often than those from the stomach.
Histologically, the most superficial portion of a peptic ulcer
consists of necrotic debris and neutrophils. Deeper in the wall
are layers of inflamed granulation tissue and scar. The mucosa
surrounding the ulcer actively regenerates in an attempt to heal
the defect. Both biopsies and brushings are obtained from this
interface, which yields a well-developed reparative process. As
described earlier, these regenerating cells are characterized by
enlarged nuclei and prominent nucleoli, but there is relative
preservation of cohesion and polarity (Fig. 14.14). In smears,
these cells are intimately admixed with neutrophilic leukocytes
and granular necrotic debris.
Helicobacter pylori
may also be rec-
ognized in the smears (Fig. 14.13). In general, the benign atypia
can be readily distinguished from adenocarcinoma.45,50,52 How-
ever, in a small proportion of specimens, this distinction can-
not be reliably made. As stated by Lin and Harmata, when the
387
previous page 383 ComprehensiveCytopathology 1104p 2008 read online next page 385 ComprehensiveCytopathology 1104p 2008 read online Home Toggle text on/off