EUS-FNA versus Biliary Brushings and Assessment

Kofi Oppong
1
, Dan Raine
1
, Manu Nayar
1
, Viney Wadehra
2
,
Subramaniam Ramakrishnan
3
, Richard M Charnley
4
Departments of
1
Gastroenterology and
4
HPB Surgery, Freeman Hospital.
Newcastle upon Tyne, United Kingdom.
2
Department of Cytopathology, Royal Victoria Infirmary.
Newcastle upon Tyne, United Kingdom.
3
Department of Gastroenterology, Warrington Hospital.
Warrington, United Kingdom
ABSTRACT
Context Individuals with suspected malignant biliary obstruction commonly undergo ERCP for drainage and tissue sampling via
biliary brushings. EUS with EUS-FNA facilitates staging and potentially more accurate tissue sampling. Objective The aim is to
compare the diagnostic performance of EUS-FNA and ERCP with biliary brushings (ERCP-BB) in the diagnosis of pancreatobiliary
carcinoma and the utility of combining the two procedures under conscious sedation. Design Retrospective analysis of a
prospectively maintained database. Patients Thirty-seven patients with suspected malignant obstructive jaundice underwent 39
paired procedures, either combined (n=22) or within a few days (n=17). Results Using strict cytological criteria the sensitivity of
EUS-FNA in the diagnosis of malignancy was 52.9% (95% CI: 35.1-70.2%) versus 29.4% (95% CI: 15.1-47.5%) for ERCP-BB.
Combining the two tests improved sensitivity to 64.7% (95% CI: 46.5-80.3%) which was significantly better than ERCP-BB alone
(P=0.001) but not EUS-FNA alone (P=0.125). When both procedures were performed under the same conscious sedation, there was
a significant difference (P=0.031) between the sensitivity of EUS-FNA (52.6%; 95% CI: 28.9-75.6%) and that of ERCP-BB (21.1%;
95% CI: 6.1-45.6%). When both procedures were performed together the mean±SD in-room time was 79±14 min (range: 45-105
min). Two of the patients (9.1%) had a complication. Conclusions In patients undergoing EUS-FNA and ERCP-BB under the same
sedation, EUS-FNA was significantly more sensitive in diagnosing malignancy. Combining the results of both tests improved
diagnostic accuracy. Combining therapeutic ERCP and EUS-FNA under the same conscious sedation is feasible, with a complication
rate similar to that of ERCP alone.
INTRODUCTION
The majority of patients presenting with signs and
symptoms of malignant biliary obstruction are
ultimately not suitable for surgery, either due to locally
invasive or metastatic disease, or because of co-
morbidities. Only 10-15% of pancreatic cancers [1] and
20-49% of cholangiocarcinomas are resectable [2, 3].
For inoperable patients therapy with either
chemotherapy or radiotherapy requires a definitive
pathological diagnosis. In addition, the quality of
information provided to the patient regarding prognosis
is severely compromised if the exact diagnosis is
unknown.
Prior to the development of endoscopic ultra-
sonography (EUS), ERCP with biliary brush cytology
(ERCP-BB) was the initial investigation of choice for
cytopathological diagnosis in jaundiced patients with
suspected pancreaticobiliary malignancy. Although this
technique has a specificity approaching 100%,
sensitivity for malignancy has been reported as 15-65%
in strictures secondary to pancreatic cancer and 44-
80% in strictures due to cholangiocarcinomas [4, 5].
Overall sensitivity is the order of 42% [5].
EUS allows excellent visualisation of the pancreas and
adjacent organs and has evolved as a sensitive staging
modality for pancreatobiliary malignancy [6, 7, 8]. The
addition of EUS-guided fine needle aspiration (EUS-
FNA) allows cytological diagnosis of pancreatic
masses. This has been shown, in many published
series, to be highly accurate in diagnosing pancreatic
masses [9, 10, 11]. In a retrospective multicenter study
Received June 15th, 2010 - Accepted August 11th, 2010
Key words
Carcinoma, Pancreatic Ductal; Cholangio-
pancreatography, Endoscopic Retrograde; Endosonography
Abbreviations ERCP-BB: ERCP with biliary brush cytology
Correspondence Kofi Oppong
Level 6 Medical Unit 1, Freeman Hospital, Newcastle upon Tyne,
NE7 7DN, United Kingdom
 

Page 2
JOP. J Pancreas (Online) 2010 Nov 9; 11(6):560-567.
JOP. Journal of the Pancreas - http://www.joplink.net - Vol. 11, No. 6 - November 2010. [ISSN 1590-8577]
561
EUS-FNA was diagnostic of malignancy in 71% of
solid pancreatic masses [12]. This conveys a significant
advantage over traditional ERCP-based cytology.
The ERCP-BB and EUS-FNA data however may not
be directly comparable.
In contrast to studies relating to biliary brush cytology,
the majority of published series of EUS-FNA with high
sensitivities have had a cytopathologist or cyto-
pathology technician present in the room [9], allowing
multiple sampling to occur until adequate tissue has
been aspirated. Increasingly, non-operable patients will
have both an ERCP for palliation and biliary brushings,
and where available an EUS for staging and FNA often
on separate occasions.
In this study, we compare these two diagnostic
modalities in a population of jaundiced patients
referred to a tertiary referral centre in which the clinical
suspicion of malignant obstruction was high. In
addition the utility of combining the two procedures
under the same (conscious) sedation was assessed.
METHODS
As in-room cytology was not available in our unit
during the study period. All patients having either
ERCP or EUS within a few days of each other
requiring a tissue diagnosis underwent both biliary
brushings and FNA. Also, in order to expedite the
investigation of patients requiring both ERCP (for
biliary drainage and brush cytology) and EUS for
staging and FNA, we sought to combine the procedures
under the same sedation whenever possible. All
procedures were performed for clinical indication and
not to research protocol. Clinical management
decisions were made through our multi-disciplinary
team process. All patients had a thin-slice multidetector
pancreatic protocol CT study prior to endoscopic
intervention.
ERCP and EUS data was recorded prospectively as
part of our ongoing quality monitoring. Retrospective
review was also performed of clinical case notes and
electronic patient record. All patients in whom ERCP-
BB and EUS-FNA were performed sequentially under
the same sedation and those in whom the second
procedure was performed prior to availability of the
results of the first sampling procedure were identified.
A total of 38 patients (21 male, 17 female; mean age
62.4 years, range 26-87 years) between February 2004
to May 2007 met the above criteria. All the patients
presented with obstructive jaundice and had either an
indeterminate biliary stricture or a mass in the head of
pancreas, with a requirement for biliary drainage and a
formal tissue diagnosis for the purpose of planning
treatment (surgical or conservative). One patient with a
final diagnosis of lymphoma was excluded from further
analysis. Therefore, 37 individuals underwent 39
paired procedures (two individuals had two paired
procedures); 22 paired procedures were performed
sequentially under the same sedation and 17 paired
procedures were performed within a few days (17
cases; mean interval: 2 days; range: 1-4 days).
All procedures were performed under conscious
sedation (administered by the endoscopist) with
midazolam and pethidine.
ERCP and biliary brushings were performed or
supervised by either K.O. or R.M.C., both of whom are
highly experienced biliary endoscopists (more than
2,000 ERCP) using TJF200, TJF240 and TJF260
duodenoscopes (Olympus UK, Southend on Sea,
United Kingdom). All EUS-FNA procedures were
performed or supervised by K.O. (more than 500
pancreatic
EUS-FNA).
When
performed
simultaneously, ERCP was generally performed first as
the priority was to ensure biliary drainage. Brushings
were taken using a standard cytology brush
(M0054500; Boston Scientific, St Albans, United
Kingdom) and standard technique. The brush catheter
was advanced over a wire and under fluoroscopic
control to the lower margin of the stricture. The brush
was then advanced and retracted a minimum of three
times, the catheter removed, the brush wiped on a glass
slide and the slide and brush tip sent for cytological
assessment. EUS and EUS-FNA was performed using
an echoendoscope (EG383OUT; Pentax, Slough,
United Kingdomand ultrasound workstation (EUB
6500; Hitachi Medical Systems, Wellingborough,
United Kingdom). Wilson Cook Predominantly 22G
needles were used, while 25G needles were used on
occasion (Cook Ireland Ltd., Limerick, Ireland). A
standard technique was used. The mass was identified
and after staging assessment and the use of Doppler to
assess for vessels, the FNA needle was passed into the
lesion under EUS control. Suction was used and the
needle moved within the tumour for 6-10 throws. The
needle was removed and the stylet replaced to express
tissue onto cytology slides, these were air dried and
stained when dry with Grunwald Giemsa stain. Needle
rinsings were processed as a cytospin using
Papanicolaou stain. During the course of the study
liquid-based cytology (SurepathTM; Bioscience
Healthcare, Nottingham, United Kingdom) was
introduced for both FNA and BB cytology and was
performed as well as the preparation of conventional
slides.
A cytopathologist was not present in the endoscopy
suite for any of the procedures.
The endoscopy room time (duration from patient entry
into the endoscopy room to their exit) was recorded.
This time comprises brief preparation and positioning
prior to the procedure, the procedure time and then
brief repositioning time prior to taking the patient to
the recovery area.
The results of the ERCP and EUS procedures were
compared with the following reference methods of
tissue diagnosis:
1) surgical histology or other biopsy methods (e.g.
percutaneous sampling of the primary tumour);
2) any positive cytology result combined with clinical
follow-up that provided further evidence of
malignancy;

Page 3
JOP. J Pancreas (Online) 2010 Nov 9; 11(6):560-567.
JOP. Journal of the Pancreas - http://www.joplink.net - Vol. 11, No. 6 - November 2010. [ISSN 1590-8577]
562
3) clinical, biochemical and radiological follow-up
until death or for at least two years if there was no
pathological or radiological evidence of malignancy.
ETHICS
Written informed consent was obtained from all
patients prior to the procedures. All procedures were
done as a part of standard patient care and not to a
research protocol and data collection was performed as
part of our ongoing clinical audit (quality monitoring).
Therefore institutional review body approval was not
required. Normal NHS Clinical Audit Practice was
observed. All aspects of the study were conducted in
accordance with the Declaration of Helsinki 1964, as
revised in Tokyo 2004.
STATISTICS
Sensitivity, specificity, positive and negative predictive
values, and accuracy were determined for ERCP biliary
brush cytology, EUS-FNA and a combination of the
two diagnostic methods. Diagnostic accuracy was
defined as the frequency of cases correctly classified.
The exact 95% confidence intervals (95% CI) of
frequencies were calculated by means of the binomial
distribution [13]. Differences in diagnostic sensitivity
and accuracy between pairs of the three methods were
identified by using the McNemar test and the statistical
significance was assessed at the 0.05 level (two-tailed).
Statistical analysis was performed using SISA [14] and
Medcalc (version 11.3.1.0; http://www.medcalc.be/).
RESULTS
The mean bilirubin was 250 μmol/L (reference range:
0-19 μmol/L). All patients had a final definitive
diagnosis achieved (Table 1). In 13 patients this was by
surgery, 17 patients had diagnostic cytology and
underwent no further intervention, 3 patients had
subsequent diagnostic histology or cytology and 4
patients had clinical follow-up until death (n=2) or for
greater than two years establishing benign disease
(n=2).
Of the 37 patients, 32 (86.5%) had malignancy;
pancreatic adenocarcinoma (n=29), neuroendocrine
carcinoma (n=2) and cholangiocarcinoma (n=1). Five
patients (13.5%) had a final diagnosis of benign
disease; chronic pancreatitis (n=2), primary sclerosing
cholangitis (n=1), serous cystadenoma (n=1) and GIST
(n=1). The two individuals who had two paired
procedures had malignancies.
EUS-FNA
The mean number of passes was 2.7 (range: 1-6).
Of the 39 EUS-FNA performed, 18 (46.2%) were
diagnostic for malignancy, 7 (17.9%) were suspicious
and 14 (35.9%) negative. All 7 suspicious aspirations
were subsequently proven to have malignancy, by
surgery (n=5), positive biliary brushings and follow-up
(n=1) and clinical follow-up alone (n=1). Of the 14
negative specimens, 5 (35.7%) were proven to be
benign by surgery (n=3) and clinical follow-up (n=2), 9
(64.3%) were proven to have malignancy by surgery
(n=4), positive repeat FNA (n=3), positive biliary
brushings and follow-up (n=1) and clinical follow-up
alone (n=1). Accepting diagnostic cytology only EUS-
FNA had an accuracy of 59.0%, sensitivity of 52.9%,
specificity of 100%, a positive predictive value (PPV)
of 100% and a negative predictive value (NPV) of
23.8% (Table 2A). Including the 7 suspicious cases
gave an accuracy of 76.9%, sensitivity of 73.5%,
specificity of 100%, PPV of 100%, and NPV of 35.7%
(Table 2B).
Table 1. Final diagnosis.
No. of cases
Malignant:
- Pancreatic adenocarcinoma
- Neuroendocrine tumour
- Cholangiocarcinoma
32 (86.5%)
29 (78.4%)
2 (5.4%)
1 (2.7%)
Benign:
- Chronic pancreatitis
- Primary sclerosing cholangitis
- Serous cyst adenoma
- GIST
5 (13.5%)
2 (5.4%)
1 (2.7%)
1 (2.7%)
1 (2.7%)
Table 2. Comparison of ERCP with biliary brushings (ERCP-BB) and EUS-FNA (all procedures; n=39).
Accuracy
Sensitivity
Specificity
Positive
predictive value
Negative
predictive value
A. Only positive cytology classified as diagnostic of malignancy
EUS-FNA
ERCP-BB
Combined
23/39; 59.0% (42.1-74.4)
15/39; 38.5% (23.4-55.4)
27/39; 69.2% (52.4-83.0)
18/34; 52.9% (35.1-70.2)
10/34; 29.4% (15.1-47.5)
22/34; 64.7% (46.5-80.3)
5/5; 100% (47.8-100)
5/5; 100% (47.8-100)
5/5; 100% (47.8-100)
18/18; 100% (81.5-100)
10/10; 100% (69.2-100)
22/22; 100% (84.6-100)
5/21; 23.8% (8.2-47.2)
5/29; 17.2% (5.9-35.8)
5/17; 29.4% (10.3-56.0)
B. Suspicious cytology considered as a positive result
EUS-FNA
ERCP-BB
Combined
30/39; 76.9% (60.7-88.9)
27/39; 69.2% (52.4-83.0)
35/39; 89.7% (75.8-97.1)
25/34; 73.5% (55.6-87.1)
22/34; 64.7% (46.5-80.3)
30/34; 88.2% (72.6-96.7)
5/5; 100% (47.8-100)
5/5; 100% (47.8-100)
5/5; 100% (47.8-100)
25/25; 100% (86.3-100)
22/22; 100% (84.6-100)
30/30; 100% (88.4-100)
5/14; 35.7% (12.8-64.9)
5/17; 29.4% (10.3-56.0)
5/9; 55.6% (21.2-86.3)
Comparison of diagnostic accuracy among different methods a:
A: EUS-FNA vs. ERCP-BB, P=0.077; EUS-FNA vs. Combined, P=0.125; ERCP-BB vs. Combined, P<0.001.
B: EUS-FNA vs. ERCP-BB, P=0.581; EUS-FNA vs. Combined, P=0.063; ERCP-BB vs. Combined, P=0.008.
Comparison of diagnostic sensitivity among different methods a:
A: EUS-FNA vs. ERCP-BB, P=0.119; EUS-FNA vs. Combined, P=0.125; ERCP-BB vs. Combined, P=0.001.
B: EUS-FNA vs. ERCP-BB, P=0.581; EUS-FNA vs. Combined, P=0.063; ERCP-BB vs. Combined, P=0.008.
Numbers in parentheses represent the exact 95% confidence intervals of the frequencies expressed as percentages
McNemar test

Page 4
JOP. J Pancreas (Online) 2010 Nov 9; 11(6):560-567.
JOP. Journal of the Pancreas - http://www.joplink.net - Vol. 11, No. 6 - November 2010. [ISSN 1590-8577]
563
Biliary Brushing
Biliary brushings were obtained in 38 of the 39 cases
(in one case it proved impossible to pass a wire across
the subsequently proven malignant stricture). On an
intention to treat basis this case was treated as a false
negative. In 10 cases (25.6%) cytology was positive, in
12 suspicious (30.8%) and in 17 negative (43.6%). All
12 suspicious cases were subsequently proven to have
malignancy by surgery (n=4), positive FNA (n=5),
positive repeat FNA (n=2) or by clinical follow-up
alone (n=1). Of the 17 negative cases, 5 (29.4%) were
true negatives by surgery (n=3) and clinical follow-up
(n=2) and 12 (70.6%) were false negatives. Accepting
only positive cytology as diagnostic of malignancy, the
accuracy of biliary brushings was 38.5%, sensitivity
29.4%, specificity 100%, PPV 100% and NPV 17.2%
(Table 2A). Including the 12 suspicious cases accuracy
was 69.2%, sensitivity 64.7%, specificity 100%, PPV
100%, and NPV 29.4% (Table 2B).
The difference of and between biliary brushings and
EUS-FNA was not statistically significant both when
only positive cytology as diagnostic of malignancy
(accuracy: P=0.077; sensitivity: P=0.119; Table 2A) or
including the suspicious cases (accuracy and
sensitivity: P=0.581; Table 2B) were considered.
Combination
An analysis was performed to investigate the utility of
combining the results of brushings and EUS-FNA.
Accepting only positive cytology as diagnostic of
malignancy, gave an accuracy of 69.2%, sensitivity of
64.7%, specificity of 100%, PPV of 100%, and NPV of
29.4% (Table 2A). This was significantly more
accurate and sensitive (P<0.001 and P=0.001,
respectively) than brushings alone but not EUS-FNA
alone (P=0.125). Accepting positive and suspicious
cytology as diagnostic gave an accuracy of 89.7%,
sensitivity of 88.2%, specificity of 100%, PPV 100%,
and NPV 55.6% (Table 2B). This was again
significantly more accurate and sensitive than biliary
brushings alone (P=0.008) but not EUS-FNA alone
(P=0.063).
Simultaneous Procedures
Of the 39 procedures, 22 (56.4%) were performed
simultaneously. There were 12 female and 8 male
patients (1 male and 1 female underwent a dual
procedure twice). Mean age was 60.5 years (range: 26-
77 years). One procedure was under general
anaesthetic. The other 21 procedures were performed
under conscious sedation with pethidine and
midazolam administered by the endoscopist.
Tissue Sampling
EUS-FNA was performed in all 22 combined
procedures, the mean number of passes being 2.9 with
a range of 1-6. Biliary brushings were obtained in 21
procedures. In one case as stated previously it proved
impossible to pass a wire across the subsequently
proven malignant stricture. On an intention to treat
basis this case was treated as a false negative. For
EUS-FNA, 10 positive specimens (45.5%) were
obtained and 3 suspicious (13.6%). There were 9
negative aspirates (40.9%), 6 of which were false
negatives (66.7%). For biliary brushings there were 4
positive samples (18.2%), 6 suspicious (27.3%) and 12
negative samples (54.5%), 9 of which were false
negatives (75.0%). Test performance either accepting
positive cytology only as diagnostic, or including
suspicious samples is shown in Table 3. When only
positive cytology was accepted as diagnostic there was
a significant difference between ERCP-BB and EUS-
FNA (accuracy and sensitivity: P=0.031; Table 3A)
and the combination of EUS-FNA and ERCP-BB was
significantly more accurate and sensitive than ERCP-
BB alone (P=0.031; Table 3A). No significant
differences were observed among the three methods
when suspicious cytology was considered as a positive
result (Table 3B).
Procedure Duration and Sedation
The mean±SD in-room time for 19 of the 21 (data
unavailable for 2 patients) simultaneous procedures
under conscious sedation was 79±14 min (range: 45-
105 min). For comparison the mean duration of 44
Table 3. Comparison of ERCP with biliary brushings (ERCP-BB) and EUS-FNA (simultaneous procedures only; n=22).
Accuracy
Sensitivity
Specificity
Positive
predictive value
Negative
predictive value
A. Only positive cytology classified as diagnostic of malignancy
EUS-FNA
ERCP-BB
Combined
13/22; 59.1% (36.4-79.3)
7/22; 31.8% (13.9-54.9)
13/22; 59.1% (36.4-79.3)
10/19; 52.6% (28.9-75.6)
4/19; 21.1% (6.1-45.6)
10/19; 52.6% (28.9-75.6)
3/3; 100% (29.2-100)
3/3; 100% (29.2-100)
3/3; 100% (29.2-100)
10/10; 100% (69.2-100)
4/4; 100% (39.8-100)
10/10; 100% (69.2-100)
3/12; 25.0% (5.5-57.2)
3/18; 16.7% (3.6-41.4)
3/12; 25.0% (5.5-57.2)
B. Suspicious cytology considered as a positive result
EUS-FNA
ERCP-BB
Combined
16/22; 72.7% (49.8-89.3)
13/22; 59.1% (36.4-79.3)
18/22; 81.8% (59.7-94.8)
13/19; 68.4% (43.5-87.4)
10/19; 52.6% (28.9-75.6)
15/19; 79.0% (54.4-94.0)
3/3; 100% (29.2-100)
3/3; 100% (29.2-100)
3/3; 100% (29.2-100)
13/13; 100% (75.3-100)
10/10; 100% (69.2-100)
15/15; 100% (78.2-100)
3/9; 33.3% (7.5-70.1)
3/12; 25.0% (5.5-57.2)
3/7; 42.9% (9.9-81.6)
Comparison of diagnostic accuracy among different methods a:
A: EUS-FNA vs. ERCP-BB, P=0.031; EUS-FNA vs. Combined, P=1.000; ERCP-BB vs. Combined, P=0.031.
B: EUS-FNA vs. ERCP-BB, P=0.453; EUS-FNA vs. Combined, P=0.500; ERCP-BB vs. Combined, P=0.063.
Comparison of diagnostic sensitivity among different methods a:
A: EUS-FNA vs. ERCP-BB, P=0.031; EUS-FNA vs. Combined, P=1.000; ERCP-BB vs. Combined, P=0.031.
B: EUS-FNA vs. ERCP-BB, P=0.688; EUS-FNA vs. Combined, P=0.500; ERCP-BB vs. Combined, P=0.125.
Numbers in parentheses represent the exact 95% confidence intervals of the frequencies expressed as percentages
McNemar test

Page 5
JOP. J Pancreas (Online) 2010 Nov 9; 11(6):560-567.
JOP. Journal of the Pancreas - http://www.joplink.net - Vol. 11, No. 6 - November 2010. [ISSN 1590-8577]
564
consecutive EUS-FNA for pancreatic masses during a
3-month period of this study was 45±14 min (range:
15-65 min) and for ERCP (51 consecutive cases)
56±25 min (range: 20-125 min). The average dose of
midazolam administered was 8.5 mg (range; 2.0-12.5
mg) and of pethidine 47 mg (range 25-50). No
flumazenil was given.
Complications
There were two complications from the simultaneous
procedures (9.1%). One patient had a mild attack of
pancreatitis (4.5%) and one patient (4.5%) failed to
drain and required a stent change a week later.
DISCUSSION
The pre-test probability of malignancy is high and has
been reported as greater than 90% in a patient
presenting with a pancreatic mass and jaundice [15].
The differential diagnosis of pancreatic adeno-
carcinoma includes focal chronic pancreatitis, auto-
immune pancreatitis and pancreatic neuroendocrine
tumour [16, 17]. An accurate preoperative diagnosis is
desirable but is not absolutely necessary because most
patients will undergo attempted resection based on the
presence of a mass and obstructive jaundice. Insisting
on a confirmed diagnosis of malignancy in fit patients
with an operable suspicious lesion in the head of the
pancreas can delay surgery and allow the tumour to
progress. In elderly and comorbid patients, however,
confirmation of the diagnosis preoperatively should be
sought more keenly since surgery carries a greater risk
and is best avoided in those without malignancy. CT is
the first investigation of choice in patients with
suspected pancreatic cancer. The absence of a mass on
CT, however, is an indication for EUS which has an
important role in clarifying the diagnosis prior to
surgery [6]. In those patients with unresectable disease,
chemotherapy is the treatment of choice. In patients
with benign disease (other than autoimmune
pancreatitis) surgery is often the most appropriate
treatment, since symptoms caused by biliary
obstruction and duodenal stenosis can be effectively
treated. Chemotherapy is contra-indicated in those who
may have benign disease and therefore positive
histology or cytology is essential in those patients who
are to be treated by non-surgical treatments.
Establishing the diagnosis in pancreatic cancer is not
easy. The tumour is commonly confined to the
pancreas and it is unusual for the tumour to erode an
epithelial surface such as stomach or duodenum. A
lesion in the head of the pancreas will usually cause
obstructive jaundice due to its proximity to the bile
duct. Extrinsic compression of the bile duct, however,
will cause obstruction even before the tumour erodes
onto the biliary epithelium. In this case malignant cells
may not be present within the lumen of the bile duct.
Prior to the advent of EUS-FNA, biliary brush cytology
was the foremost method of obtaining a cytological
diagnosis in patients presenting with jaundice and a
biliary stricture. EUS-FNA is well documented to
achieve high accuracy and sensitivity in the diagnosis
of pancreatic masses. EUS-FNA has shown high
sensitivity where prior biliary brushings or
percutaneous biopsies have been negative or non
diagnostic [2, 11]. The prior sampling procedure may
have been performed in the referring hospital however,
and read by a cytopathologist without particular
expertise in pancreatic and biliary cytology. The
experience of the cytopathologist has previously been
shown to have significant impact on the accuracy of the
report [18]. In addition, the majority of EUS-FNA
studies reporting high accuracy have utilised an in-
room cytopathologist, which has been shown to
improve diagnostic yield [19, 20]. This is not available
in many units. In a recent large retrospective series [21]
EUS-FNA (utilising strict cytological criteria) had an
overall sensitivity of 77% in the diagnosis of pancreatic
neoplasia. In this study immediate cytological
examination was available for 43.8% of procedures and
significantly improved test performance. Our study
reports the comparison between EUS-FNA and biliary
brush cytology (39 procedures) and the utility of
sequential procedures under the same sedation (22
procedures). In the subset having simultaneous
procedures, ERCP, biliary brushings and stent insertion
as well as EUS with EUS-FNA were performed
utilising endoscopist-administered conscious sedation.
All EUS-FNA was performed by one operator and
ERCP by two. Both brush and aspiration cytology were
reported by the same experienced pancreatobiliary
cytopathologists. A previous study by Rosch et al. [22]
has directly compared ERCP and EUS for the tissue
diagnosis of biliary strictures. In this study there were
50 paired procedures of which 12 were performed
during the same session. No difference was found
between ERCP-BB and EUS-FNA in the detection of
malignancy although there was a numerical advantage
to ERCP-BB in the detection of biliary malignancy and
EUS-FNA for pancreatic masses. There was no in-
room cytopathologist and it is not stated in the paper as
to what sedation was used for the 12 procedures done
in one session.
Two other studies have reported on the utility of
combining ERCP and EUS-FNA in the same procedure
[23, 24]. Both of these studies utilised propofol
sedation administered by an anaesthetist. In the study
by Ross et al. [23] 87 of the 112 patients had a FNA
whilst biliary brushings were obtained in 54. EUS-
FNA had an overall accuracy of 86.7% and a
sensitivity of 83.3%, while brush cytology had a
sensitivity of 13% and an accuracy of 48%. In the Ross
et al. study [23] EUS-FNA was performed first and
brushings only taken if there was a delay in reporting
the EUS-FNA cytology or if FNA cytology was not
positive and is therefore not a true comparison of the
two tests. In a study by Tarantino et al. [24], 72
patients underwent combined ERCP and EUS, EUS-
FNA being performed on 25 of the patients. It is not
reported whether biliary brushings were obtained. An
in-room cytopathologist was available for analysis of

Page 6
JOP. J Pancreas (Online) 2010 Nov 9; 11(6):560-567.
JOP. Journal of the Pancreas - http://www.joplink.net - Vol. 11, No. 6 - November 2010. [ISSN 1590-8577]
565
the FNA in both of these studies. In the study by Ross
et al. [23] suspicious aspirates were excluded from
analysis and not considered as either diagnostic or false
negative. In the Tarantino et al. study [24] the
cytological criteria for a diagnostic aspirate is not
stated. In our study (using strict cytological
classification) the performance of biliary brushings
(accuracy 38%) although in line with the reported
literature is at the lower end of the range. The
performance of EUS-FNA appears at the lower end of
the range (accuracy 59%) possibly because of
adherence to strict diagnostic cytology criteria and the
absence of in-room cytopathology as compared to other
studies [9, 23, 24]. If suspicious cytology is considered
diagnostic as has been reported in a number of studies
[21, 25] the performance characteristics are improved
(sensitivity 73%, accuracy 77%) and in line with such
studies. Another approach in early studies was to
exclude nondiagnostic specimens in addition to
classifying suspicious cytology as positive [26]. In our
study, if nondiagnostic aspirates are excluded and
suspicious cytology considered diagnostic, the
sensitivity, negative predictive value, and accuracy of
EUS-FNA are 80%, 40% and 83%, respectively.
In the present study we have documented that when
performed under the same sedation, by the same
cytopathologist, using strict cytological diagnostic
criteria and in the absence of in-room cytopathology
EUS-FNA is more sensitive than ERCP with biliary
brushings in the diagnosis of pancreatobiliary
carcinoma. We have also demonstrated that combining
the results of the FNA and brushings significantly
improves the result compared to ERCP-BB alone and
that if suspicious cytology is classified as a positive
diagnosis the combined test performance is comparable
(accuracy 90%, sensitivity 88%, negative predictive
value 56%) to that reported for EUS-FNA in units with
in-room cytopathology.
A concern with performing EUS and ERCP during the
same session has been the risk of complication [27,
28]. ERCP is associated with a complication rate of 4-
13% [29, 30, 31, 32, 33] with a reported pancreatitis
rate that varies from 1% to 13%. [29, 32, 33, 34]. EUS-
FNA is associated with a lower risk of post-procedure
pancreatitis less than 1% [35, 36] and has an overall
lower reported complication rate of 2.5-5% [36, 37]. In
our series the overall complication rate was 9.1%, all of
which were mild, which is comparable to the
complication rate reported for ERCP only and the
10.5% complication rate reported by Ross et al. [23] in
their study of combined EUS and ERCP. We had one
case of pancreatitis (4.5%). We have previously
reported a 3.0% post-ERCP pancreatitis rate in our unit
[38].
A potential concern when combining two potentially
lengthy procedures under the same sedation
particularly using endoscopist administered conscious
sedation is of an overly prolonged procedure and
excess sedation. The endoscopy room time in this study
(79±14 min; range: 45-105 min; 19 procedures) was
very similar to the procedure time reported by Ross et
al. [23] of 74.6±30 min (range: 25-148 min) and the
time of 58.6±16.4 min (range: 30-91 min) in the study
by Tarantino et al. [24]. In contrast to these previous
studies all but one of the 22 simultaneous procedures in
our series were performed under conscious sedation.
The mean midazolam dose was 8.5±3.7 mg, this is
significantly more (P=0.00017) than the mean dose of
6 mg that we documented in an audit of 482 ERCP
procedures from our institution [38]. Concerns have
been raised about prolonged sedation during ERCP in
the elderly [39]. In this small series there was no
complication attributable to sedation. We however feel
that the future of such complex endoscopy lies in
deeper sedative techniques allied to more intensive
monitoring [40].
As a retrospective review from one unit there are
limitations to this study; it did not include consecutive
patients requiring ERCP and EUS, the numbers are
relatively small and no formal assessment of patient
comfort during the procedures was made.
Notwithstanding this, we have documented in the same
cohort a significant statistical difference between EUS-
FNA and ERCP-BB in the diagnosis of
adenocarcinoma. The combined results were
comparable to the best results of EUS-FNA with in-
room cytology .We suggest that in the era of EUS-
FNA, biliary brushings at the time of ERCP should not
be abandoned even when the procedure is performed
simultaneously with EUS-FNA (particularly if in-room
cytology is not available).
We also believe this study provides further evidence of
the feasibility of combining EUS-FNA with ERCP and
documents that this is possible under conscious
sedation. Combining the procedures has potential
benefits in shortening investigation time and lag to
therapy [41].
Conflict of interest The authors have no potential
conflict of interest
References
1. Butturini G, Stocken DD, Wente MN, Jeekel H, Klinkenbijl JH,
Bakkevold KE, et al. Influence of resection margins and treatment on
survival in patients with pancreatic cancer: meta-analysis of
randomized controlled trials. Arch Surg 2008; 143:75-83. [PMID:
18209156]
2. Chamberlain RS, Blumgart LH. Hilar cholangiocarcinoma: a
review and commentary. Ann Surg Oncol 2000; 7:55-66. [PMID:
10674450]
3. Mansfield SD, Barakat O, Charnley RM, Jaques BC,
O'Suilleabhain CB, Atherton PJ, Manas D. Management of hilar
cholangiocarcinoma in the North of England: pathology, treatment,
and outcome. World J Gastroenterol 2005; 11:7625-30. [PMID:
16437689]
4. Fogel EL, deBellis M, McHenry L, Watkins JL, Chappo J,
Cramer H, et al. Effectiveness of a new long cytology brush in the
evaluation of malignant biliary obstruction: a prospective study.
Gastrointest Endosc 2006; 63:71-7. [PMID: 16377319]
5. de Bellis M, Sherman S, Fogel EL, Cramer H, Chappo J,
McHenry L Jr, et al. Tissue sampling at ERCP in suspected
malignant biliary strictures (Part 2). Gastrointest Endosc 2002;
56:720-30. [PMID: 12397282]

Page 7
JOP. J Pancreas (Online) 2010 Nov 9; 11(6):560-567.
JOP. Journal of the Pancreas - http://www.joplink.net - Vol. 11, No. 6 - November 2010. [ISSN 1590-8577]
566
6. Mansfield SD, Scott J, Oppong K, Richardson DL, Sen G,
Jaques BC, et al. Comparison of multislice computed tomography
and endoscopic ultrasonography with operative and histological
findings in suspected pancreatic and periampullary malignancy. Br J
Surg 2008; 95:1512-20. [PMID: 18942059]
7. Kala Z, Valek V, Hlavsa J, Hana K, Vanova A. The role of CT
and endoscopic ultrasound in pre-operative staging of pancreatic
cancer. Eur J Radiol 2007; 62:166-9. [PMID: 17344007]
8. Dewitt J, Devereaux BM, Lehman GA, Sherman S, Imperiale
TF. Comparison of endoscopic ultrasound and computed tomography
for the preoperative evaluation of pancreatic cancer: a systematic
review. Clin Gastroenterol Hepatol 2006; 4:717-25. [PMID
16675307]
9. Eloubeidi MA, Chen VK, Eltoum IA, Jhala D, Chhieng DC,
Jhala N, et al. Endoscopic ultrasound-guided fine needle aspiration
biopsy of patients with suspected pancreatic cancer: diagnostic
accuracy and acute and 30-day complications. Am J Gastroenterol
2003; 98:2663-8. [PMID: 14687813]
10. Agarwal B, Abu-Hamda E, Molke KL, Correa AM, Ho L.
Endoscopic ultrasound-guided fine needle aspiration and
multidetector spiral CT in the diagnosis of pancreatic cancer. Am J
Gastroenterol 2004; 99:844-50. [PMID: 15128348]
11. Gress F, Gottlieb K, Sherman S, Lehman G. Endoscopic
ultrasonography-guided fine-needle aspiration biopsy of suspected
pancreatic cancer. Ann Intern Med 2001; 134:459-64. [PMID:
11255521]
12. Savides TJ, Donohue M, Hunt G, Al-Haddad M, Aslanian H,
Ben-Menachem T, et al. EUS-guided FNA diagnostic yield of
malignancy in solid pancreatic masses: a benchmark for quality
performance measurement. Gastrointest Endosc 2007; 66:277-82.
[PMID: 17643700]
13. Exact confidence limits for p. Geigy Scientific Tables, 7th Ed.
Ciba-Geigy Ltd., Basle: Switzerland, p85-98.
14. Quantitative Skills. Simple Interactive Statistical Analysis
(SISA). Quantitative Skills; Consultancy for Research and Statistics.
Hilversum: The Netherlands.
15. Mansfield SD, Lochan R, Jaques K, Scott J, Bennett MK,
O'Sullivan CB, et al. Outcome of patients presenting with suspected
pancreato-biliary malignancy. 38th Europoean Pancreatic Club.
Volume 6. Tampere, Finland. Pancreatology 2006; 6:358-9.
16. To'o KJ, Raman SS, Yu NC, Kim YJ, Crawford T, Kadell BM,
Lu DS. Pancreatic and peripancreatic diseases mimicking primary
pancreatic neoplasia. Radiographics 2005; 25:949-65. [PMID:
16009817]
17. Chari ST, Takahashi N, Levy MJ, Smyrk TC, Clain JE, Pearson
RK, et al. A diagnostic strategy to distinguish autoimmune
pancreatitis from pancreatic cancer. Clin Gastroenterol Hepatol 2009;
7:1097-103. [PMID 19410017]
18. Eltoum IA, Chhieng DC, Jhala D, Jhala NC, Crowe DR,
Varadarajulu S, Eloubeidi MA. Cumulative sum procedure in
evaluation of EUS-guided FNA cytology: the learning curve and
diagnostic performance beyond sensitivity and specificity.
Cytopathology 2007; 18:143-50. [PMID: 17388936]
19. Jhala NC, Jhala DN, Chhieng DC, Eloubeidi MA, Eltoum IA.
Endoscopic ultrasound-guided fine-needle aspiration. A
cytopathologist's perspective. Am J Clin Pathol 2003; 120:351-67.
[PMID 14502798]
20. Klapman JB, Logrono R, Dye CE, Waxman I. Clinical impact of
on-site cytopathology interpretation on endoscopic ultrasound-guided
fine needle aspiration. Am J Gastroenterol 2003; 98:1289-94.
[PMID: 12818271]
21. Turner BG, Cizginer S, Agarwal D, Yang J, Pitman MB, Brugge
WR. Diagnosis of pancreatic neoplasia with EUS and FNA: a report
of accuracy. Gastrointest Endosc 2010; 71:91-8. [PMID: 19846087]
22. Rosch T, Hofrichter K, Frimberger E, Meining A, Born P,
Weigert N, et al. ERCP or EUS for tissue diagnosis of biliary
strictures? A prospective comparative study. Gastrointest Endosc
2004; 60:390-6. [PMID: 15332029]
23. Ross WA, Wasan SM, Evans DB, Wolff RA, Trapani LV,
Staerkel GA, et al. Combined EUS with FNA and ERCP for the
evaluation of patients with obstructive jaundice from presumed
pancreatic malignancy. Gastrointest Endosc 2008; 68:461-6. [PMID:
18384788]
24. Tarantino I, Barresi L, Di Pisa M, Traina M. Simultaneous
endoscopic ultrasound fine needle aspiration and endoscopic
retrograde cholangio-pancreatography: Evaluation of safety. World J
Gastroenterol 2007; 13:3861-3. [PMID: 17657842]
25. Mitsuhashi T, Ghafari S, Chang CY, Gu M. Endoscopic
ultrasound-guided fine needle aspiration of the pancreas:
cytomorphological evaluation with emphasis on adequacy
assessment, diagnostic criteria and contamination from the
gastrointestinal tract. Cytopathology 2006; 17:34-41. [PMID:
16417563]
26. Faigel DO, Ginsberg GG, Bentz JS, Gupta PK, Smith DB,
Kochman ML. Endoscopic ultrasound-guided real-time fine-needle
aspiration biopsy of the pancreas in cancer patients with pancreatic
lesions. J Clin Oncol 1997; 15:1439-43. [PMID: 9193337]
27. Mergener K, Jowell PS, Branch MS, Baillie J.
Pneumoperitoneum complicating ERCP performed immediately after
EUS-guided fine needle aspiration. Gastrointest Endosc 1998;
47:541-2. [PMID: 9647385]
28. Di Matteo F, Shimpi L, Gabbrielli A, Martino M, Caricato M,
Esposito A, et al. Same-day endoscopic retrograde
cholangiopancreatography
after
transduodenal
endoscopic
ultrasound-guided needle aspiration: do we need to be cautious?
Endoscopy 2006; 38:1149-51. [PMID: 17111340]
29. Cheng CL, Sherman S, Watkins JL, Barnett J, Freeman M,
Geenen J, et al. Risk factors for post-ERCP pancreatitis: a
prospective multicenter study. Am J Gastroenterol 2006; 101:139-47.
[PMID: 16405547]
30. Vandervoort J, Soetikno RM, Tham TC, Wong RC, Ferrari AP
Jr, Montes H, et al. Risk factors for complications after performance
of ERCP. Gastrointest Endosc 2002; 56:652-6. [PMID: 12397271]
31. Christensen M, Matzen P, Schulze S, Rosenberg J.
Complications of ERCP: a prospective study. Gastrointest Endosc
2004; 60:721-31. [PMID: 15557948]
32. Masci E, Toti G, Mariani A, Curioni S, Lomazzi A, Dinelli M,
Minoli G, et al. Complications of diagnostic and therapeutic ERCP: a
prospective multicenter study. Am J Gastroenterol 2001; 96:417-23.
[PMID: 11232684]
33. Loperfido S, Angelini G, Benedetti G, Chilovi F, Costan F, De
Berardinis F, et al. Major early complications from diagnostic and
therapeutic ERCP: a prospective multicenter study. Gastrointest
Endosc 1998; 48:1-10. [PMID: 9684657]
34. Freeman ML, DiSario JA, Nelson DB, Fennerty MB, Lee JG,
Bjorkman DJ, et al. Risk factors for post-ERCP pancreatitis: a
prospective, multicenter study. Gastrointest Endosc 2001; 54:425-34.
[PMID: 11577302]
35. Eloubeidi MA, Gress FG, Savides TJ, Wiersema MJ, Kochman
ML, Ahmad NA, et al. Acute pancreatitis after EUS-guided FNA of
solid pancreatic masses: a pooled analysis from EUS centers in the
United States. Gastrointest Endosc 2004; 60:385-9. [PMID:
15332028]
36. Eloubeidi MA, Tamhane A, Varadarajulu S, Wilcox CM.
Frequency of major complications after EUS-guided FNA of solid
pancreatic masses: a prospective evaluation. Gastrointest Endosc
2006; 63:622-9. [PMID: 16564863]
37. Adler DG, Jacobson BC, Davila RE, Hirota WK, Leighton JA,
Qureshi WA, et al. ASGE guideline: complications of EUS.
Gastrointest Endosc 2005; 61:8-12. [PMID: 15672049]
38. Ramakrishnan S, Crosbie J, Gray H, Panter S, Mansfield SD,
Charnley RM, et al. One year prospective audit of ERCP across one
Hospital Trust. Gut 2005; 54(Suppl 2):a1-a117. n n]
39. Fisher L, Fisher A, Thomson A. Cardiopulmonary complications
of ERCP in older patients. Gastrointest Endosc 2006; 63:948-55.
[PMID: 16733108]

Page 8
JOP. J Pancreas (Online) 2010 Nov 9; 11(6):560-567.
JOP. Journal of the Pancreas - http://www.joplink.net - Vol. 11, No. 6 - November 2010. [ISSN 1590-8577]
567
40. Qadeer MA, Vargo JJ, Dumot JA, Lopez R, Trolli PA, Stevens
T, et al. Capnographic monitoring of respiratory activity improves
safety of sedation for endoscopic cholangiopancreatography and
ultrasonography. Gastroenterology 2009; 136:1568-76. [PMID:
19422079]
41. Peter S, Eloubeidi MA. Feasibility of combined EUS-FNA and
ERCP for obstructive jaundice from presumed pancreatic
malignancy. Nat Clin Pract Gastroenterol Hepatol 2009; 6:132-3.
[PMID: 19204740]

There are no products listed under this category.