State-of-the-art ultrasonography is as accurate as helical computed tomography and computed tomographic angiography for detecting unresectable periampullary cancer
Mm. Morrin et al., State-of-the-art ultrasonography is as accurate as helical computed tomography and computed tomographic angiography for detecting unresectable periampullary cancer, J ULTR MED, 20(5), 2001, pp. 481-490
Objective. To compare the ability of state-of-the-art ultrasonography with
that of helical computed tomography and computed tomographic angiography in
detecting unresectable periampullary cancer. In most patients periampullar
y cancer is unresectable because of either distant metastasis or local vasc
ular involvement. The advent of gray scale and color Doppler ultrasonograph
y has improved the ability of ultrasonography to detect vascular involvemen
t. Methods. Twenty-three consecutive patients with periampullary cancer wer
e enrolled for prospective staging of their disease by comparing helical co
mputed tomography and computed tomographic angiography with gray scale and
color Doppler ultrasonography of the abdomen. Portal vein, superior mesente
ric vein, splenic vein, and superior mesenteric artery involvement was grad
ed 0 to 4, grade 0 being no vascular involvement and grade 4 being total oc
clusion of the vessel. Agreement between ultrasonography and computed tomog
raphic angiography for determining vascular involvement was measured by X-2
analysis. Results. Two patients (9%) were excluded because excessive overl
ying bower gas hampered the ability of ultrasonography to visualize the pan
creas. For the remaining 21 patients, there was significant agreement betwe
en ultrasonography and computed tomographic angiography for detecting vascu
lar involvement in ail vessels (P < .001; portal vein, kappa = 0.67; superi
or mesenteric vein, kappa = 0.67; splenic vein, kappa = 0.85; and superior
mesenteric artery, kappa = 0.59). Ultrasonography was in agreement with com
puted tomographic angiography in all cases of unresectability. Both modalit
ies were equally poor in preoperatively showing lymphadenopathy and metasta
ses. Conclusions. Provided that there is adequate visualization on ultrason
ography of the head of the pancreas in the periampullary region, then state
-of-the-art gray scale and color Doppler ultrasonography are as accurate as
helical computed tomography and computed tomographic angiography for detec
ting the unresectability of periampullary cancer. If performed as the initi
al investigation and the region of the pancreatic head is clearly shown, an
d if vascular encasement or occlusion or distant metastasis is identified,
further investigations are unnecessary.