Lavelle M T, Lee V S, Rofsky N M, Krinsky G A, Weinreb J C
Department of Radiology, Division of Body MRI, New York University Medical Center, 530 First Ave, New York, NY 10016, USA.
Radiology. 2001 Feb;218(2):389-94. doi: 10.1148/radiology.218.2.r01fe31389.
To assess the accuracy of an interpolated breath-hold T1-weighted three-dimensional (3D) gradient-echo (GRE) magnetic resonance (MR) imaging sequence with near-isotropic pixel size (</=2.3 mm) for evaluation of hepatic arterial anatomy variants during dynamic liver parenchymal imaging and to report patterns of hepatic arterial anatomy.
Liver MR imaging, including an interpolated breath-hold 3D GRE sequence with fat suppression (4.2/1.8 [repetition time msec/echo time msec], 12 degrees flip angle), was performed in 207 consecutive patients before and after gadopentetate dimeglumine administration. Of the 207 patients, 202 (98%) had technically satisfactory studies clearly defining the hepatic arterial system. The first contrast material-enhanced GRE acquisition was timed for optimal arterial enhancement with a timing examination. In a retrospective review, hepatic arteries were evaluated on the basis of arterial phase images interpreted by two independent readers using transverse source images complemented by multiplanar reconstructions. Twenty-three patients also underwent digital subtraction angiography, which was a reference standard for comparison.
Conventional hepatic arterial anatomy was demonstrated in 135 (67%) of 202 patients. In the 23 patients with angiographic correlation, no discrepancy was noted between MR imaging and digital subtraction angiographic findings.
Hepatic arterial anatomy can be reliably demonstrated during liver parenchymal imaging with an optimally timed contrast-enhanced isotropic 3D GRE sequence.