Transcript investigation and diagnosis of periampullary tumours

INVESTIGATION AND
DIAGNOSIS OF
PERIAMPULLARY TUMOURS
Andrew Barclay – Austin Health
PREAMBLE
• Pancreatic cancer is the fifth leading cause of cancer
•
•
•
•
•
deaths. It has a poor prognosis with a 5-year survival rate
of 3%.
95% of pancreatic carcinomas derive from the exocrine
portion and therefore considered as adenocarcinomas
Other tumours include islet cell tumours or lymphomas
75% of pancreatic cancers are considered to be
periampullary.(1)
Periampullary tumours may originate from the head of
pancreas, ampulla of Vater (4%), or mucosa of duodenum
or bile duct itself (cholangiocarcinoma).
Tend to present earlier (mass effect) and therefore are
usually small (<3cm) making imaging difficult.
DIAGNOSIS
• Clinical and investigative
• Symptoms of pancreatic cancer
• Pain (over 70%): tends to be a dull epigastric pain (+/- back pain)
which is worse in a supine position. Initially episodic and related
with meals but later becomes severe and persistent. However more
than 50% of patients with early pancreatic cancer have little or no
pain.(2)
• Weight loss: loss of 7 to 10kg within a few weeks. Often preceeds
all other symptoms
• Jaundice (75%): deep and progressive
DIAGNOSIS
• More infrequently patients will present with pancreatitis, portal
hypertension, GI bleeding, new-onset diabetes, migratory
thrombophlebitis, polyarthralgia, or subcutaneous fat necrosis.
• Rarely: hypercalcemia, hypoglycaemia, carcinoid syndrome, ACTH
and inappropriate ADH production
• Risk factors: 5% chronic pancreatitis (26 fold - 4% risk over 20yrs(6) ),
30% smoking (2 fold),
diabetes (2 fold)(5),
5-10% hereditary: hereditary pancreatitis (50 fold), MEN,
HNPCC, VPL, FAP, Gardner syndrome
DIAGNOSIS
• Periampullary tumours however tend to present at an
earlier stage with obstructive jaundice (70-80%) (11) and
are therefore more likely to be resectable.
• Physical examination: excoriations, jaundice, palpable
gallbladder (50%).
• Incidental finding on imaging modalities or upper
endoscopy
INVESTIGATIVE
•
•
•
•
•
•
•
•
•
•
LFT’s, CA 19-9
transabdominal ultrasound
triple-phase constrast enhanced helical CT
MRI, MRCP, MRA
ERCP
selective angiography
PET scanning
endoscopic ultrasound
percutaneous fine-needle aspiration
staging laparoscopy and laparascopic ultrasound
ISSUES
• Is this cancer?
• If the presumptive diagnosis is periampullary cancer, is it
resectable or does it only require a palliative procedure?
• Non-resectablity is determined by metastasis, vascular or
extra-pancreatic spread.
- more specifically no direct extension to the coeliac or
superior mesenteric arteries and a patent SMV-portal
vein confluence.
Blood Tests
• LFT’s/component of conjugated bilirubin – helpful in
determining obstruction in jaundiced patients
• CA 19-9
- most extensively studied
- nonspecific (elevated in other gastrointestinal
carcinomas)
- elevated in benign conditions(3)
- only useful as follow-up surveillance
JAUNDICE
• First line imaging includes transabdominal ultrasound or high
resolution CT scanning
• Both detect dilated intrahepatic and extrahepatic ducts
confirming obstruction
Ultrasound is an inexpensive and noninvasive tool. In regard to pancreatic
cancer its strength is in diagnosing liver
metastasis (1cm) and lymph node
involvement. With the addition of
colour doppler it can be 80% accurate
in diagnosing vascular invasion.(7)
However it remains highly operator
dependant.
ERCP
• Endoscopic retrograde cholangio-pancreatography is usually
•
•
•
•
the next step in evaluating jaundiced patients.
Both investigative and therapeutic
Considered sensitive with 90-97% of people with pancreatic
cancer having a positive ERCP
Note the complications: technical failure (10-15%),
pancreatitis, sepsis, perforation and bleeding
Ampullary and duodenal cancers can be visualised and
samples taken
- cytological brushings, FNA, biopsy
- not conclusive in detecting malignancy with
only 35-65% sensitivity reported in the
literature.(7)(11)
ERCP
• Mucosal changes of cancer can be difficult to distinguish
from the erythema and oedema secondary to a passed
gallstone both on visualisation and on microscopy.
ERCP
• Kimchi NA et al. The contribution of endoscopy and
•
•
•
•
biopsy to the diagnosis of periampullary tumours. Tel
Aviv University
928 patients referred to institute for ERCP
26 patients ended up having a malignancy
Sensitivity
- on appearance (90%), on biopsy (81%)
Specificity
- on appearance (33%), on biopsy (50%)
ERCP
• Cholangiogram: in regard to cancer the cholangiogram can demonstrate mass
•
•
lesions distal to endoscopic visualisation and may also demonstrate duct dilatation
post obstruction
Pancreatogram
- single, irregular abrupt stricture of the pancreatic duct
- gradual occlusion of the main duct
- alteration of side branches near the tumour such as fragmentation
- displacement of the main pancreatic duct
- irregularly pooled contrast (within necrotic tumour)
- ‘double duct sign’ strictured CBD and pancreatic duct
Pancreatitis: irregular and tortuous with multiple stenosis
Shemesh et al. The role of endoscopic retrograde cholangiopancreatography in
differentiating pancreatic cancer with chronic pancreatitis.
Retrospective study on 10 patients with cancer and pancreatitis and 45 patients
with only pancreatitis – all ten easily distinguished on pancreatogram
CT
• CT (as with MRI) can be used for diagnosis and to assess
tumour resectability
• The addition of triple-phase contrast enhanced spiral CT
has improved this imaging modality. Multislice CT also
considerably improves the images obtained. It also allows
more accurate examination of the SMA which has an
oblique course.
• It is used to assess the presence of a mass, extrapancreatic
involvement, vascular involvement, lymph node
involvement and liver metastasis. It detects resectability in
70-80% and its accuracy in determining vascular
involvement has largely replaced visceral angiography.(12)
• Lentschig MG et al. The value of 3-phase spiral CT and
magnetic resonance tomography in preoperative diagnosis
of pancreatic carcinoma. Radiology 1996; 36: 406-412
• Prospective study on 28 patients (18 unresectable)
• Sensitivity: unresectability 94% for both CT and MRI
resectability 80% CT and 70% MRI
vascular involvement 82% CT and 62% MRI
Conclusion of the study was that both techniques had
similar clinical value.
• Phoa SS et al. Spiral computed tomography for preoperative staging
of potentially resectable carcinoma of the pancreatic head. Brit J
Surgery 1999; 86:789-794
• Prospective study on 56 patients using spiral CT (5mm)
• Cantaalano C et al. Pancreatic Carcinoma: the role of
high-resolution multislice spiral CT in the diagnosis and
assessment of resectability. Eur Radiology 2003; 13:149156
• Prospective study on 44 patients using 1mm intervals
• Diagnosis (sens/spec/accuracy) 97% 80% 96%
Unresectability (sens/spec/accuracy) 96% 86% 93%
MRI
• MRI is generally considered as equivalent to CT
in assessing pancreatic cancer.
• Drawbacks: expensive/metal implants
• Sterner E et al. Imaging of pancreatic neoplams:
comparison of MRI and CT. AJR Am J
Roentgenol 1989; 152:487-491
- assessed 32 patients with pancreatic cancer
using MRI and CT and found no clinical benefit
(subjectively noted that MRI was slightly clearer)
MRI
• However MRI technology continues to improve and the
addition of MRCP and MRA bringing a new dimension to
this modality.
• Trede M et al. Ultrafast magnetic resonance imaging
improves the staging of pancreatic tumors. Annals of
Surgery 1997; 226(4): 393-407
• 58 patients compared using an ‘all in one’
MRCP/MRI/MRA combination with CT, angiography, and
transabdominal ultrasound.
• Included both pancreatic (35 within the pancreatic head),
distal CBD (2) and periampullary (9) cancers: note 8
patients were mananged non-operatively
UMRI
+
Extrapancreatic spread 93% 97%
Liver metastasis
Lymph node involv.
US
+
-
CT
+
-
68% 94%
83% 70%
100% 92%
80% 89%
71% 90%
75% 86%
100% 50%
57% 83%
This particular study also showed
that UMRI had superior accuracy in
detecting vascular involvement
(89%) than CT (79%) or
angiography (68%).
Angiography
• Visceral angiography with selective arterial injection of the coeliac
and superior mesenteric arteries with venous phase studies had been
considered as the best in demonstrating vascular anatomy and tumour
encasement or occlusion.
• Dooley WC et al. Is preoperative angiography useful in patients with
periampullary tumors? Ann Surgery 1990; 211:649-655
• Found that 77% of patients with normal vessels on angiography were
resectable while those with major vessel occlusion were unresectable.
• However it is invasive and with recent studies demonstrating the
accuracy of CT and MRI in respect to vascular involvement and
anatomy it is generally not considered essential.
• Murugiah M et al. The role of selective visceral
•
•
•
•
•
angiography in the management of pancreatic and
periampullary cancer. World J Surg 1993; 17:796-800
Prospective study of 46 patients with pancreatic/
periampullary tumours (no evidence of disseminated
disease on Ultrasound/CT)
Wrongly diagnosed hepatic metastasis (7/9)
False positive result of 15% (in detecting irresectability)
False negative result of 48% (in dectecting resectability)
Study concluded that angiography was poor in determining
resectability.
Endoscopic Ultrasound
• High-frequency inducer placed in the gastric and duodenal
lumen to image the pancreas. This method is claimed to be
superior than other methods in determining resectability
and local lymph node involvement. It is able to diagnose
cancers of 5mm in size but has difficulty distinguishing
between pancreatic cancer and pancreatitis.
• While it does offer the possibility of FNA it obviously is
unable to detect distant metastasis and does have difficulty
determining tumour encasement of the SMA.(8)
• Current studies suggest a role in diagnosing small tumours. However
it remains operator dependant.
• Muller MF et al. Pancreatic tumors: evaluation with endoscopic US,
CT, and MR imaging. Radiology 1994; 190:745-751
• 49 patients with suspected pancreatic cancer (16 of which were less
than 3cm). EUS had a superior accuracy of 94% compared to
conventional CT (67%) and MRI (83%) in the diagnosis.
• Midwinter MJ et al. Correlation between spiral computed tomography,
endoscopic ultrasonography and findings at operation in pancreatic
and ampullary tumours. British Journal of surgery 1999; 86:189-193
• 48 patients with 34 primary lesions found operatively. EUS diagnosed
33/34 while CT only found 26/34. In regard to determining vascular
involvement both were equivalent except that EUS was less effective
in diagnosing SMA involvement. Conclusion of the study was that
EUS had a role in diagnosing small tumours.
Percutaneous fine-needle aspiration
•
•
•
•
•
There are two disadvantages to this procedure.
It can produce seeding of the tumour(9) along the needle tract or
intraperitoneally
The second consideration is that if an operation is planned anyway, a
negative FNA will not alter management.
Its main benefit is to diagnose cancer in a patient who is not an
operative candidate (for subsequent palliation) or diagnose a
suspected pancreatic lymphoma.
Enayati PG et al. Traverso LW, Galagan K, et al. The meaning
of equivocal pancreatic cytology in patients thought to have
pancreatic cancer. Am J Surg. 1996; 171:525-528.
- retrospective study on 224 patients
- 50% of those with atypical/non-malignant cytology had
cancer
Laparoscopy
• None of the previous methods are accurately able to assess peritoneal
•
•
•
•
and omental metastases which can be only 1 to 2mm size
30% of liver metastasis are smaller than 2cm and may not be detected
routinely(10).
Thus diagnostic laparoscopy can be performed in attempt to decrease
the number of unnecessary laparotomies. The lesser sac and
mesenteric root can also be explored.
However special expertise is required and there is a related morbidity
rate of 2.5%(11).
Warshaw et al. Laparoscopy in the staging and planning for
pancreatic cancer. Am J Surg 1986; 151:76-80
- 40 patients with known pancreatic cancer who were thought to be
resectable. 14 out of 17 cases with intrabdominal metastasis were
detected at diagnostic laparoscopy.
• Pisters PWT et al. Laparoscopy in the staging of
pancreatic cancer. Brit J Surgery 2001; 88:325-337
• Review article examining the English literature (19962000) on staging laparoscopy commented that conclusions
are difficult due to the inconsistent use of modern spiral
CT. However it does conclude that routine use of
laparoscopy (especially as a separate anaesthetic) is not
appropriate given the available evidence. They suggest
staging laparoscopy in those with larger tumours,
equivocal radiological evidence of abdominal metastasis
(including low-volume ascites), and changes in albumin
and CA19-9.
• Brooks A et al. The Value of Laparoscopy in the
•
•
•
•
•
Management on Non-Pancreatic Periampullary Tumors.
(2001)
139 patients examined with spiral CT
127 patients having an adequate laparoscopy
122 patients had malignancy
Diagnostic laparoscopy decreased the incidence of an
unnecessary laparotomy by 8%
(4 with liver metastasis, 4 with peritoneal seeding, 2 other)
Authors argued for selective use of diagnostic laparoscopy
Laparoscopic Ultrasound
• This technique improves the accuracy of staging
laparoscopy in detecting local extension of tumour and
liver metatasis.
• John TG et al. Carcinoma of the pancreatic head and
periampullary region. Tumor staging with laparoscopy
and laparoscopic ultrasound. Ann Surg 1995; 221:156-164
• Prospective trial of 40 patients. Claims that the ultrasound
component changed the decision regarding resectability in
10 patients (compared to laparoscopy alone). Note that
pre-operative investigations included, ERCP
transabdominal ultrasound and conventional CT.
• Pisters PWT et al. Laparoscopy in the staging of
pancreatic cancer. Brit J Surgery 2001; 88:325337
• This review article also examines the role of
ultrasound at laparoscopy (1995-1998) and
concludes that there is very little evidence
comparing modern spiral CT (with defined limits
of resectability) to the advantages gained by the
addition of ultrasound. Although it concedes that
ultrasound certainly improves the accuracy of
laparoscopy.
Peritoneal lavage and cytology
during laparoscopy
• Nieveen VD, Cytology of Peritoneal lavage performed
during staging laparoscopy for gastrointestinal
malignancies: is it useful? Annals of Surgery 1998;
228:728-733
• 449 patients: of the 28 patients who were cytology positive
- 19 had metastasis at laparoscopy
- 3 had false postive or misleading results
- in 6 cases predicted irresectablitity
(Therefore lacks any practical use)
PET scanning
• A review of the available literature(13)
between 1994 and 1997 suggests PET
scanning is good in diagnosing pancreatic
cancer (accuracy 85-93%).
• However there are few head to head trials
with modern spiral CT, and therefore its role
remains unclear in the diagnosis and staging
of pancreatic carcinoma.
Pancreatic head mass
• Differentiating between chronic pancreatitis
and pancreatic cancer can be difficult
• Even intraoperative biopsy has a reported
10% false negative result
• ERCP and PET scanning may be beneficial
in diagnosis
Bibliography
(1)
(2)
(3)
(4)
(5)
(6)
(7)
Keith D. et l. Current Management of Pancreatic Carcinoma. Annals of
Surgery. 1995; 221:133-148
Hudis C. et al. Pain is not a prominent symptom in most patients with early
pancreas cancer. Proc Am Soc Clin Oncol 1991; 10:1149 note:
retrospecitve study on 72 patients
Freboug W. et al. The evaluation of CA 19-9 antigen level in the early
detection of pancreatic cancer. Cancer 1988; 62:2287-2290 note: 112 out
of 866 patients with benign disease had elevated CA 19-9
Shemesh E et al. The role of endoscopic retrograde
cholangiopancreatography in differentiating pancreatic cancer with chronic
pancreatitis. Cancer 1990; 65:893-896
Everhart J et al. Diabetes mellitus as a risk factor for pancreatic cancer. A
meta- analysis. JAMA 1995; 273(20): 1605-9
Lowenfels AB et al. Pancreatitis and the risk of pancreatic cancer.
International Pancreatitis Study Group. N Engl J 1993; 328(20): 1433-7
note: multicentre trial of 2000 patients
Cipolletta L et al. Pancreatic Head Mass: What can be done? JOP 2000;
1:108-110
(7)
(8)
(9)
(10)
(11)
(12)
(13)
Tomiyama T et al. Assessment of arterial invasion in pancreatic cancer
using color Doppler ultrasonography. Am J Gastroenterol 1996; 91:14101416
Trede M et al. Ultrafast magnetic resonance imaging improves the staging
of pancreatic tumors. Annals of Surgery 1997; 226(4): 393-407
Rashleigh-Bilcher HJC et al. Cutaneous seeding of pancreatic carcinoma
by fine-needle aspiration biopsy. Brit J Radiol 1986; 59:182-183
Ward EM et al. Computed tomographic characteristics of pancreatic
carcinoma: An analysis of 100 cases. Radiographics 1983; 3:547-565
John TG, Greig JD, Carter DC, Garden OJ. Carcinoma of the pancreatic
head and periampullary region. Ann Surg 1995; 221:156-164
note: 40 patients in prospective study
Boris WK et al. Treatment of Resectable and Locally Advanced
Pancreatic Cancer. Cancer Control: Journal of the Moffitt cancer
centre 2000; 7:428-436
Berberat P et al. Diagnosis and Staging of pancreatic Cancer by Positron
Emission Tomography. World J Surgery 1999; 23:882-887