Alimentary Tract (Esophagus, Stomach, Small Intestine, colon, Rectum, Anus, Biliary Tract)
Differential Diagnosis of Biliary Duct Cytology
Large, flat smooth
with smooth edges214
aLoss of polarity208
Nuclear contou rs
Statistically significant features.
fine chromatin and smooth membranes is present. Polarity is,
generally; maintained streaming with generally intact cell bor-
ders present. Nuclei are inconspicuous. Although macronucleoli
can occasionally occur, the smear background may contain bile
pigment, cholesterol, crystals, and varying degrees of inflam-
mation (Table 14.8). Marked reactive changes may be seen
in biliary tract brushings of patients with primary sclerosing
cholangitis (PSC), which can be difficult to distinguish from a
malignant process. This is further complicated by the increased
incidence of dysplasia and cholangiocarcinoma in patients with
PSC. In a study by Ponsioen et al. of cytology brushings from
43 patients with PSC, the sensitivity and specificity for detecting
malignancy were 62 and 89%, respectively.211 Interestingly the
use of P53 and K-ras did not improve accuracy. Layfield reported
false-positive bile brushings in two patients with PSC.212 Lastly,
dysplasia or adenomatous change can be difficult to distinguish
from malignancy.194,195,213 Dysplastic cell groups usually show
more significant cellular crowding and overlapping than reac-
tive reparative cells194 Additionally more significantly nuclear
atypia with increased N/C ratios and abnormal chromatin may
be present in dysplasia.
A recent study analyzed the features of endoscopic biliary
duct brushings in liquid-based preparations (ThinPrep) on 142
patients with bile duct obstructions or pancreatic masses.214
Fifty-six cases (39%) were negative/atypical on cytology but
were suspicious or positive in surgical or biopsy specimen (false
negative). Of the false-negative cases, 9 (16%) were upgraded
to suspicious/positive. Statistically, significant helpful cytomor-
pholic features included three-dimensional micropapillae, ani-
sonucelosis, high N/C ratios, nuclear contour irregularity, and
prominent nucleoli. Interestingly, features not helpful in dis-
tinguishing malignancy included single marked nucleoli, chro-
matin granularity, and necrosis. The overall sensitivity increased
from 31 to 42% based on the statistically significant cytomor-
phologic features reviewed above.
Overall, diagnostic sensitivity of ERCP-obtained cytologic
brushing specimens is generally low, but false-positive diag-
noses are rare. Although it is not a unanimous finding, sensi-
tivity is generally higher in samples from the bile duct than in
those obtained from the pancreatic system.22,192,215 Furthermore,
sensitivity appears to be better for the diagnosis of primary bil-
iary ductal carcinomas (cholangiocarcinomas) than for primary
pancreatic carcinomas or metastases. The latter may obstruct
the duct by impinging on it externally and thus not involve the
Associated with an expanding utilization of multiple endo-
scopic tissue biopsies, the role of brushing cytology in the diag-
nosis of GI tract malignancies in symptomatic patients appears
to be waning in many centers worldwide. However, we expect
to witness continued investigation and clinical utilization of
exfoliative cytology in the screening and surveillance of selected
patients with an elevated risk of developing carcinoma, with the
target being the consistent recognition of epithelial dysplasia.
We also believe that exfoliative cytology will enjoy persistent
use in ERCP-related studies of potential neoplasms of the biliary
Brushing cytology will also continue to be employed to diag-
nose mucosal-based infections, in part due to its inherently
high sensitivity related to sampling of larger surface areas and
its reduced trauma, vis a vis biopsies. This is no small matter in
this era of accentuated immunosuppression.
The arena of greatest growth is with transmural EUS-directed
FNAs of submucosally based and diffusely infiltrative mass
lesions of the GI tract. Use in the diagnosis of lymphomas, GISTs,
and neuroendocrine tumors, currently inaccessible to both the
brush and biopsy forceps in many patients, will only increase.
The addition of ancillary diagnostic procedures on aspirated
cellular material improves accuracy even more.
Geisinger KR, Wang HH, Ducatman BS,
et al. Gastrointestinal cytology.
1991 ; 11 (2):403-441.
Wang HH, Jonasson JG, Ducatman BS.
Brushing cytology of the upper gastro-
intestinal tract. Obsolete or not?
Chambers LA, Clark WE 2nd. The endo-
scopic diagnosis of gastroesophageal
malignancy. A cytologic review.
Geisinger KR. Endoscopic biopsies and
cytologic brushings of the esophagus
are diagnostically complementary.
Kobayashi S, Kasugai T. Brushing cytol-
ogy for the diagnosis of gastric cancer
involving the cardia or the lower esopha-
Achkar E, Carey W. The cost of surveil-
lance for adenocarcinoma complicating
Am J Gastroenterol
The role of endoscopy in the surveil-
lance of premalignant conditions of the
upper gastrointestinal tract. Guidelines
for clinical application.