Ks. Naik et al., COMPARISON OF DYNAMIC CONTRAST-ENHANCED MRI AND DOPPLER ULTRASOUND INTHE PREOPERATIVE ASSESSMENT OF THE PORTAL VENOUS SYSTEM, British journal of radiology, 70, 1997, pp. 43-49
The purpose of this study was to compare dynamic contrast enhanced MRI
(DCEMR) with Doppler ultrasound (US) in the assessment of portal veno
us anatomy and to analyse the causes of discrepancy. Over a 1 year per
iod, 97 patients undergoing assessment prior to hepatic surgery underw
ent imaging of the liver and portal venous system using US with colour
and spectral Doppler and MRI with axial T-2 weighted spin echo (SE) a
nd coronal oblique T, weighted rapid gradient echo (GRE) imaging befor
e and immediately after bolus injection of Gd-DTPA (0.1 mmol kg(-1)).
When the US and MRI findings were discrepant, the images were reviewed
by two observers and compared with surgical findings. US and DCEMR we
re concordant in 90 patients (portal vein patent in 80, occluded in 10
). In three patients with cirrhosis and gross ascites the portal vein
was reported as occluded on US and patent on MRI; surgery confirmed th
e MRI findings. In one patient the portal vein was patent on US but no
t on MRI, but there was a 3 week interval between the examinations. In
three patients the portal vein was patent on US, but MRI detected occ
lusion of intrahepatic portal vein branches in two, and encasement of
an intrahepatic branch in the third case. Spontaneous splenorenal shun
ts were seen in 15 patients only on MRI; varices were seen in 39 patie
nts on MRI and in 22 patients on US. Both US and DCEMR contribute to t
he pre-operative assessment of the portal venous system. MRI provides
additional information over US in assessing intrahepatic portal branch
es and detecting varices and splenorenal shunts, and is recommended fo
r all surgical candidates and in patients with abnormal portal venous
anatomy and equivocal US findings.