Background. Detection of metastatic liver disease and malignant involvement
of major peripancreatic vessels is important to determine resectability of
pancreatic malignancy. Computed tomography with arterial portography (CTAP
) is the most sensitive method for detection of colorectal liver metastases
; it can also detect malignant vascular involvement. We have assessed CTAP
in patients with pancreatic cancer considered suitable for resection after
standard ultrasonography (US) and computed tomography (CT) examination.
Method. CTAP was performed in 18 patients (8 with a biliary stent). All pat
ients had previous US and CT with no dear evidence of irresectability. Find
ings of CTAP were compared with the prior CT and with findings at operation
or clinical progress.
Results, CTAP suggested liver metastases in 7 patients. Three were confirme
d at operation or at follow-up (sensitivity for detection of metastases in
CT negative patients of 75%). There were 4 false-positive assessments (spec
ificity, 71%). One further patient developed liver metastases within 6 mo a
fter resection (1 false-negative). Nine patients had vascular involvement a
t operation. There was 1 false-positive and one false-negative assessment (
sensitivity, 89% and specificity, 89%). CTAP detected vascular involvement
in 4 patients in whom it was not detected by CT.
Conclusion, This preliminary study suggests that CTAP is a sensitive test f
or detection of liver metastases and vascular involvement in patients with
pancreatic malignancy, This invasive test should be reserved for patients w
ho are considered operable on the basis of other preoperative tests.