Jean-Michel Serfaty, MD Ergin Atalar, PhD Jerome Declerck, PhD Perry Karmakar, MS Harald H. Quick, MS Kendrick A. Shunk, MD Alan W. Heldman, MD Xiaoming Yang, MD, PhD Index terms: Angiography, contrast media, 50.121412, 50.12142, 54.121412, 54.12142 Gadolinium, 50.121412, 50.12142, 54.121412, 54.12142 Magnetic resonance (MR), contrast enhancement, 50.121412, 50.12142, 54.121412, 54.12142 Magnetic resonance (MR), guidance Magnetic resonance (MR), vascular studies, 50.121412, 50.12142, 54.121412, 54.12142 Radiology 2000; 217:290 –295 1 From the Departments of Radiology and Biomedical Engineering, (J.M.S., E.A., P.K., H.H.Q., X.Y.), Biomedical En- gineering (J.D.), and Division of Cardiol- ogy (K.A.S., A.W.H.), Johns Hopkins Uni- versity School of Medicine, Outpatient Center, Room 4243, 601 North Caroline St, Baltimore, MD 21287-0845. Re- ceived May 26, 1999; revision requested July 19; final revision received December 15; accepted January 17, 2000. Sup- ported in part by a Lavoisier Program grant (J.M.S.), a Socie ´te ´ Francaise de Ra- diologie grant (J.M.S.), National Insti- tutes of Health RO1HL61672 grant (E.A.), and National Institutes of Health R29HL57483 grant (E.A.). Address cor- respondence to X.Y. (e-mail: xyang @mri.jhu.edu). © RSNA, 2000 Author contributions: Guarantors of integrity of entire study, J.M.S., E.A., X.Y.; study concepts, J.M.S., E.A., X.Y.; study design, J.M.S., E.A., X.Y., J.D.; definition of intellectual con- tent, J.M.S.; literature research, J.M.S.; ex- perimental studies, J.M.S., X.Y., A.W.H., P.K., E.A., K.A.S; data acquisition, J.M.S., E.A., H.H.Q.; data analysis, J.M.S.; manu- script preparation and editing, J.M.S.; manuscript review, J.M.S., E.A., J.D., H.H.Q., K.A.S., X.Y. Real-time Projection MR Angiography: Feasibility Study 1 Intraarterial injections of small doses of gadopentetate dimeglumine were combined with a fast spoiled-gradi- ent-echo magnetic resonance (MR) sequence to obtain real-time projec- tion angiographic images of the rab- bit aorta and canine coronary arter- ies. Arterial filling and washout, as well as venous and perfusion phases, were clearly displayed, demonstrat- ing that arterial fluoroscopy in which an MR technique is used is feasible. Central to the success of vascular inter- ventional procedures, such as translumi- nal angioplasty or embolization, is the accurate depiction of the vascular tree and interventional instruments. Such de- piction has been accomplished by means of x-ray fluoroscopy. However, the disad- vantages of x-ray fluoroscopy, including the toxicity of ionizing radiation, the risk of adverse reaction due to the use of io- dinated agents (1,2), the inability to im- age cross-sectionally, and the difficulty of characterizing surrounding soft tissue (3), have led to the exploration of alter- native strategies. The potential to perform accurate two- dimensional and three-dimensional an- giography (4), obtain high-spatial-resolu- tion MR images with excellent soft-tissue contrast (5,6), and track interventional devices (7–9) make magnetic resonance (MR) imaging a potential alternative. By allowing the imaging of plaque morphol- ogy and composition, as well as the as- sessment of related organ function dur- ing a single vascular intervention, MR imaging may modify the management of vascular interventions and improve the clinical outcome for patients with ath- erosclerosis. However, conventional MR angiogra- phy sequences (time-of-flight angiogra- phy, phase-contrast angiography, and con- trast medium– enhanced angiography), when used for the guidance of vascular intervention devices through the arterial tree (10 –12), have general limitations. First, acquisition times are longer than 1 second, which precludes obtaining infor- mation about the local hemodynamic condition. Second, long postprocessing analysis is required to generate road maps; and third, contrast-enhanced an- giography necessitates the use of large amounts of contrast medium, which pre- cludes the possibility of multiple injec- tions because of increased background signal intensity. Recently, some authors have reported that two-dimensional pro- jection techniques coupled with intrave- nous injections of gadopentetate dime- glumine can depict arterial flow in large arteries (13,14). The goal of this study was to evaluate whether the combination of selective in- traarterial injections of small doses of ga- dopentetate dimeglumine with a fast two- dimensional gradient-echo sequence could generate sufficient information to guide vascular interventions and allow analysis of local hemodynamic conditions. A two- dimensional projection technique with no section selection was used to avoid prob- lems relative to arterial localization and to enable imaging of a tortuous vessel in a single image. MATERIALS AND METHODS Sequence Optimization: Preliminary Study In a preliminary study, the pulse se- quence and the gadopentetate dimeglu- mine concentration were optimized by using phantoms to maximize gadopente- tate dimeglumine signal intensity and background signal suppression. The con- trast medium used was gadopentetate dimeglumine (Magnevist; Berlex, Wayne, NJ). In vitro phantom experiments, which were performed with a 1.5-T imager (Signa; GE Medical Systems, Milwaukee, Wis), showed that the highest image contrast-to- 290