Gadolinium-enhanced 3D MRA prior to isolated hepatic perfusion for metastases

Citation
H. Zeh et al., Gadolinium-enhanced 3D MRA prior to isolated hepatic perfusion for metastases, J COMPUT AS, 23(5), 1999, pp. 664-669
Citations number
22
Categorie Soggetti
Radiology ,Nuclear Medicine & Imaging","Medical Research Diagnosis & Treatment
Journal title
JOURNAL OF COMPUTER ASSISTED TOMOGRAPHY
ISSN journal
03638715 → ACNP
Volume
23
Issue
5
Year of publication
1999
Pages
664 - 669
Database
ISI
SICI code
0363-8715(199909/10)23:5<664:G3MPTI>2.0.ZU;2-R
Abstract
Purpose: Isolated hepatic perfusion (IHP) is a new treatment for patients w ith isolated unresectable liver metastases, which can result in a partial o r complete response in similar to 75% of patients. Preoperative knowledge o f hepatic arterial anatomy is important to adequately perfuse the liver. Di gital subtraction angiography (DSA) is currently used to identify the hepat ic arterial anatomy. The purpose of this study was to determine if MR angio graphy (MRA) could replace DSA prior to IHP. Method: Twenty-seven patients scheduled to undergo MP underwent MRA with a contrast-enhanced 3D time-of-flight gradient echo sequence. Both maximal in tensity projections (MIPs) and source coronal images were used to evaluate thr images. The results of the MRA were interpreted by two readers who were blinded to the surgical results. The first 17 patients also underwent DSA, and a separate comparison was made with those results. Anatomy was charact erized as either normal hepatic arteries (NHAs), normal vasculature with an accessory left hepatic artery (aLHA), or a replaced right hepatic artery ( rRHA). Results: MRA correctly detected all 22 patients with NHAs but also identifi ed aLHAs, of which only 2 were confirmed surgically. MRA correctly detected all five rRHAs. MIP images alone accurately depicted the hepatic arterial anatomy in only 9 of 27 (33%), usually because significant vessels were not visualized or their origin could not be determined. Source coronal images were required to accurately determine the anatomy in all patients. Among th e 17 patients who underwent DSA, MRA detected 14 of 14 with NHA and 3 of 3 with rRHA. Six aLHAs were identified by MRA and five were confirmed by DSA. Conclusion: Enhanced 3D MRA is an accurate method of depicting the hepatic arterial supply. In comparison to surgery, MRA overestimates thr number of aLHAs, but this may be because these small vessels are not detected at surg ery. Based on the results of this study, DSA has been replaced by MRA in th e planning of IHP at our institution. A better display of MRA images is nee ded as MIP images were usually insensitive for the small caliber arteries s upplying the liver.