| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Vascular and Interventional Radiology |
1 From the Department of Radiology (L.V.H., T.D.J., H.B., L.S., D.V., R.O., G.M.) and the Division of Hypertension and Cardiovascular Rehabilitation (R.F.), University Hospitals, Katholieke Universiteit Leuven, Herestraat 49, B-3000 Leuven, Belgium. Received April 15, 1998; revision requested July 1; final revision received April 15, 1999; accepted July 12. Address reprint requests to L.V.H. (e-mail: lieven.vanhoe@hnbe.com).
 |
Abstract |
|---|
 TOP Abstract IntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
MATERIALS AND METHODS: In a prospective study, 43 patients underwent time-resolved MR angiography (acquisition time per data set, 7 seconds). The patients also underwent single-phase high-spatial-resolution MR angiography (acquisition time, 27 seconds) (n = 6), conventional angiography (n = 7), or both (n = 30). No bolus timing study was performed for time-resolved MR angiography. Image quality (presence of artifacts, ability to prevent venous overlap on arterial phase images, contrast enhancement) and demonstration of anatomic variants (renal arterial and venous variants, vena caval anomaly, visceral arterial variants) and vascular diseases were assessed.
RESULTS: Time-resolved MR angiographic images were characterized by fewer and less severe artifacts, less overlap of enhancing veins, and better contrast enhancement than were single-phase MR angiographic images (P < .05). The mean sensitivity and specificity were 90% (nine of 10) and 100% (173 of 173), respectively, for detection of arterial anatomic variants and 93% (28 of 30) and 100% (324 of 325), respectively, for detection of disease. The technique also proved to be reliable for demonstration of venous disease.
CONCLUSION: In comparison with current non–time-resolved MR angiographic techniques, time-resolved MR angiography is more robust and easier to perform and allows simultaneous evaluation of arterial and venous disease.
Index terms: Arteries, abnormalities, 961.1312 • Arteries, MR, 95.129411, 95.12942, 9*.129411, 9*.12942, 9*.129432 • Magnetic resonance (MR), vascular studies, 9*.12942 • Renal arteries, stenosis or obstruction, 961.721 • Veins, MR, 95.129411, 95.12942, 95.12943, 9*.129411, 9*.12942, 9*.12943
|
Introduction |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
The technical factors that determine the quality of 3D contrast-enhanced MR angiographic images are well understood (5). To optimize image contrast, the maximum concentration of gadolinium should be present during acquisition of central k-space lines. In practice, optimal timing requires estimation of the contrast material travel time by considering clinical parameters (age, patient history) or, preferably, by using the test bolus method (6,7). Another approach, called automatic triggering, provides an even more elegant solution to the timing problem but has been implemented for a limited number of magnets (8). In theory, a final approach could be to collect 3D image data sets so rapidly that at least one data acquisition is aligned with the arterial phase of the contrast material bolus (5). Besides eliminating the timing problem, such a "time-resolved" approach could provide temporal information about relative rates of contrast enhancement of arteries, parenchymal tissue, and veins. Furthermore, time-resolved MR imaging during a single breath hold helps eliminate spatial misregistration among data sets and makes subtraction technically feasible.
As early as 1996, Korosec et al (9) proposed a technique for time-resolved 3D contrast-enhanced MR angiography; their technique was based on elements such as temporal interpolation of k-space views and an increased sampling rate with low spatial frequencies. Korosec et al used a technique (not commercially available) that permits reconstruction of a series of 3D image sets with an effective temporal frame rate of one volume every 2–6 seconds. The only disadvantages of this method are the requirement of an off-line workstation and the long processing times.
Ongoing improvements in gradient strength now allow further reduction in repetition and echo times. By using state-of-the-art equipment, independent 3D MR data sets can now be obtained in less than 10 seconds. Unfortunately, given a specific gradient strength, each further improvement in temporal resolution is accompanied by a penalty in spatial resolution and vice versa. The optimal compromise between temporal and spatial resolution in the different applications of 3D contrast-enhanced MR angiography remains to be determined with comparative clinical studies.
In this study, we focused on 3D contrast-enhanced MR angiography of the abdominal vessels. A time-resolved MR technique that allowed acquisition of successive independent data sets with a 7-second frame rate was compared with a single-phase high-spatial-resolution technique with an imaging time of 27 seconds (Fig 1). Whereas the test bolus technique was used for individualized bolus timing at single-phase MR angiography, no timing sequence was used at time-resolved MR angiography. Image quality and detectability of anatomic variants and arterial and venous disease were compared.
|
|
MATERIALS AND METHODS |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
A first group of 26 patients was included in a prospective study that compared time-resolved 3D contrast-enhanced MR angiography, single-phase 3D contrast-enhanced MR angiography, and conventional angiography. Twenty-four of these patients were referred to the angiography department because of a clinical suspicion of renovascular hypertension; two patients were referred for preoperative evaluation of possible pancreatic carcinoma. Inclusion of patients in this comparative study was based on fulfillment of the following criteria: (a) ability and willingness to give written informed consent to participate in the study, (b) availability for evaluation with the different techniques, and (c) absence of contraindications for MR imaging. For practical reasons, MR angiography and conventional angiography were always performed on the same day, and conventional angiography was always performed first. The bed rest period after angiography was 3–6 hours. Time-resolved and single-phase MR angiography were performed in random order and were separated by at least 1 hour. Institutional review board approval and informed consent were obtained.
A second group of 10 patients was referred for MR angiography for a variety of reasons and was included in a more limited comparative study of time-resolved MR angiography and single-phase MR angiography. Again, inclusion criteria were ability and willingness to give written informed consent to participate in the study, availability to be examined with the two techniques, and absence of contraindications for MR imaging. In four of these patients, conventional angiography was considered to be clinically indicated within a few days or weeks after MR angiography; in these patients, MR angiography and conventional angiography were performed on different days, with 1–12 days (mean, 4 days) between examinations. Institutional review board approval and informed consent were also obtained for this study.
Finally, 37 patients referred for abdominal MR angiography did not fulfill the criteria for inclusion in one of the two aforementioned comparative studies. These patients were examined with time-resolved MR angiography only. Seven of these patients also underwent conventional angiography because of clinical indications. In this subgroup of patients, MR angiography and conventional angiography were always performed on different days, with 3–15 days (mean, 6 days) between examinations.
In summary, of the 43 patients included in this study, 30 were examined with all three techniques (time-resolved MR angiography, single-phase MR angiography, and conventional angiography), and 13 were examined with two techniques (time-resolved MR angiography and conventional angiography [n = 7] or time-resolved and single-phase MR angiography [n = 6]).
MR Imaging
All time-resolved and single-phase MR angiographic images were obtained with a 1.5-T MR system (Vision; Siemens Medical Systems, Erlangen, Germany) with a gradient-switching capability of 25 mT/m and a rise time of 300 µsec. A body phased-array coil was used. The coil was placed to cover the volume of interest.
First, gradient-echo scout MR images (15/6 [repetition time msec/echo time msec]; flip angle, 30°; matrix, 128 x 256; field of view, 500 mm; section thickness, 10 mm; five sections) were obtained in three planes. Second, T2-weighted half-Fourier acquisition single-shot turbo spin-echo MR images (
/60, 378 [repetition time msec/effective echo times msec]; echo spacing, 4 msec; matrix, 160 x 256; bandwidth, 650 Hz/pixel) were obtained during quiet breathing. The purpose of obtaining these images was to use them as "localizer" images to permit accurate positioning for acquisition of the MR angiographic slab. Two interleaved series of 16 sections were acquired. The acquisition time was 700 msec per section, and the section thickness was 5 mm. Fat suppression was not applied.
MR angiographic images were always obtained in the coronal plane. Contrast material (0.1 mmol/kg gadopentetate dimeglumine, Magnevist; Schering, Berlin, Germany) was injected by means of an MR power injector (Spectris, model SBT 200; Medrad, Pittsburgh, Pa) through a 20-gauge angiographic catheter placed in an antecubital vein. Before the start of the study, patients were informed about the breath-hold requirements. They were instructed to hyperventilate prior to data acquisition and to hold their breath as long as possible during data acquisition.
Time-resolved MR imaging.—The time-resolved MR angiography sequence was a radio-frequency spoiled gradient-echo acquisition (3.2/1.1; flip angle, 30°; bandwidth, 650 Hz/pixel; matrix, 140 x 256; average field of view, 300 x 350 mm; average pixel size, 2.19 x 1.37 mm; slab thickness, 80 mm; number of partitions, 16; effective section thickness after interpolation, 2.5 mm; ratio of rectangular field of view, 7:8; acquisition time, 7 seconds). Gadopentetate dimeglumine was injected at a rate of 2.5 mL/sec. The imaging delay time was 10 seconds. Six successive independent data sets were obtained during a total imaging time of 42 seconds.
Single-phase MR imaging.—First, the travel time for contrast material from the injection site to the region of interest was determined with the test bolus injection technique, as previously described (6). A 2-mL dose was used for this purpose. The contrast material injection rate was 2.5 mL/sec. The shortest possible echo time was used for the diagnostic study. However, in comparison with the time-resolved technique, we had to use a decreased bandwidth and somewhat longer repetition and echo times (4.4/1.4; flip angle, 30°; bandwidth, 390 Hz/pixel; matrix, 192 x 512; average field of view, 300 x 400 mm; average pixel size, 1.56 x 0.78 mm; slab thickness, 80 mm; number of partitions, 32; section thickness after interpolation along the z axis, 1.25 mm; ratio of rectangular field of view, 6:8; acquisition time, 27 seconds). The contrast material injection rate was 2.5 mL/sec.
Postprocessing.—All MR angiograms were postprocessed at the MR imager console. Maximum intensity projection (MIP) data were calculated for each image set. Multiplanar reformatting was applied only if native images and/or MIP images were suggestive of the presence of arterial disease. In time-resolved MR angiography, an additional set of images was calculated by subtracting arterial phase images from images that showed adequate venous enhancement. This additional set of images was used to evaluate for the presence of venous abnormalities.
Conventional Angiography
Intraarterial digital subtraction angiography was performed by using a transfemoral 5-F pigtail catheter and the Seldinger technique. Injection of 120–160 mL of 300 mg/mL ioxaglate (Hexabrix; Guerbet, Roissy, France) at a flow rate of 15 mL/sec was performed with the catheter tip placed in the abdominal aorta. Anteroposterior and lateral projections were obtained in all cases. If necessary, left and right oblique projections were obtained and/or selective renal arterial catheterization was performed by using a catheter with a curved tip. In seven patients, selective catheterization of the celiac trunk and/or superior mesenteric artery was performed. In five of these patients, more distal branches (gastroduodenal artery, dorsal pancreatic artery, or great pancreatic artery) also were selectively catheterized. In five patients referred for evaluation of peripancreatic vessels, images of the splenic vein and portosplenic confluence also were obtained during the portal venous phase.
Image Evaluation and Data Analysis
All images were independently evaluated by three radiologists (L.V.H., T.D.J., L.S.). All MR images were printed on film and stored on optical disc. MR angiographic images were evaluated without knowledge of the results obtained at conventional angiography.
In the first image evaluation session, time-resolved MR angiographic images in 43 patients were assessed. As a preliminary step, those images that best showed the arterial anatomy were selected (ie, one of the six available sets of images was selected). These images were used for further evaluation. Images obtained with more delayed acquisitions were reviewed to help detect venous abnormalities. Single-phase MR images obtained in 36 patients were evaluated at a second session, which occurred 3 weeks after the first session. To prevent bias, images in patients were presented in random order. Again, although attention was mainly focused on the arterial system, incidentally discovered venous abnormalities also were recorded. Finally, conventionalangiograms in 37 patients were evaluated at a third session, which occurred 2 weeks after the second session.
Image quality.—Qualitative evaluation of the MR angiographic images was based on the following criteria: (a) presence of artifacts, (b) ability to selectively display arterial anatomy (without venous superimposition), and (c) arterial contrast enhancement. For artifacts, images were assigned a score of 0 if artifacts were absent; 1, if artifacts were mild; 2, if artifacts were moderate; and 3, if artifacts were severe. For superimposition of enhancing veins, images were assigned a score of 0 if superimposition was absent; 1, if present but without negative effects on the display of arterial anatomy; and 2, if present with negative effects on the display of arterial anatomy. For blood vessel contrast enhancement, images were assigned a score of 0 if contrast enhancement was poor; 1, if moderate; 2, if good; and 3, if excellent.
Mean scores were used for further analysis. The statistical significance of differences between time-resolved MR angiography and single-phase MR angiography was determined by using the Wilcoxon signed rank test.
Detection of arterial variants and disease.—All anatomic variants and abnormalities (stenosis, occlusion, aneurysmal dilatation, etc) were recorded. Grading of vascular stenoses was based on interactive evaluation of cine-loop display of native images, MIP images, and transverse reformatted images at the computer console. Measurements were not obtained.
For evaluation purposes, the arterial tree was divided into the following segments: aorta, superior mesenteric artery and its branches, celiac trunk and its branches, renal arteries, and iliac arteries and grafts. If one or more vessel segments were not visible on images obtained with either MR angiography or conventional angiography, these segments were not used for further data analysis. Left and right renal arteries and iliac arteries were considered separately. On time-resolved MR angiograms, 183 vessel segments were available for further analysis, including 37 aortas, 16 superior mesenteric arteries, 14 celiac trunks, 62 renal arteries, and 54 iliac arteries and grafts. On single-phase MR angiograms, 149 segments were available for analysis, including 30 aortas, 13 superior mesenteric arteries, 11 celiac trunks, 53 renal arteries, and 42 iliac arteries and grafts.
Arterial stenoses were classified in four groups: 0%–49% stenosis, 50%–69% stenosis, 70%–99% stenosis, and occlusion. For the purpose of reporting, this classification was used only for renal arterial stenoses because of the small number of stenoses detected in other vessel segments. In cases of disagreement among observers, a final diagnosis was determined by means of consensus. The sensitivity and specificity of time-resolved MR angiography and single-phase MR angiography were calculated, with the results of conventional angiography as the reference standard. The statistical significance of differences between the two MR angiographic techniques was determined by using the McNemar test.
Detection of venous disease.—Venous abnormalities visible on images obtained with time-resolved MR angiography, single-phase MR angiography, or conventional angiography were recorded. In addition, clinical files, surgical reports, and results obtained with other imaging studies (computed tomography [CT], endoscopic or transabdominal ultrasonography [US]) were systematically reviewed to evaluate whether venous anatomic variants or disease had been detected. Venous anatomic variants and abnormalities could thus be confirmed with findings from CT (n = 4), conventional angiography (n = 1), endoscopic US (n = 2), transabdominal US (n = 1), and surgery (n = 1).
We also assessed whether subtraction MR images allowed better visualization of venous abnormalities. This assessment was performed by means of side-by-side comparison of images obtained without and those obtained with subtraction. The value of subtraction images was assigned a score of 0 if there was no added value; 1, if demonstration of anatomic variants or disease was improved; and 2, if subtraction images were necessary for detection.
|
RESULTS |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
Image Quality
The comparative results of time-resolved and single-phase MR angiography are given in Table 1. Time-resolved MR angiography yielded significantly better scores for artifact (P < .001), venous overlap (P < .05), and contrast enhancement (P < .05). In the evaluation of time-resolved images, concordant results (identical score assigned by all three observers) for artifact, venous overlap, and arterial contrast enhancement were obtained in 28 (78%), 31 (86%), and 30 (83%) of 36 patients, respectively. In the evaluation of single-phase MR images, concordant results for artifact, venous overlap, and arterial contrast enhancement were obtained in 32 (89%), 30 (83%), and 28 (78%) patients, respectively.
|
|
|
|
|
|
|
|
The results obtained with time-resolved MR angiography and single-phase MR angiography are given in Table 2 and can be summarized as follows: One of 10 anatomic variants (a small accessory renal artery) was overlooked on images obtained with time-resolved and single-phase MR angiography. All other accessory renal arteries were well demonstrated (Fig 5).
|
|
|
|
Detection of Venous Anatomic Variants and Disease
The following venous anatomic variants were found: circumaortic left renal vein (n = 3), polar right renal vein crossing a ureteropelvic junction stenosis (n = 1), and atresia or chronic obstruction of the infrarenal segment of the inferior vena cava with retroperitoneal collateral vessels (n = 1). Venous abnormalities included occlusion of the superior mesenteric and/or portal veins (n = 2), displacement of a renal vein and occlusion of the inferior vena cava (n = 1), and acute thrombosis of the left iliac vein (n = 1). All lesions were well depicted on images obtained with time-resolved MR angiography (Figs 6, 7). The distribution of scores for the added value of subtracted images was as follows: A score of 0 was assigned for images in three patients, and a score of 1 was assigned for images in seven patients. Thus, in all but three patients, venous abnormalities were more clearly depicted on subtraction images. The abnormalities could, however, also be detected on nonsubtracted (source) images.
|
|
|
|
|
|
|
DISCUSSION |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
These results are remarkable in view of the relatively low spatial resolution of the images obtained with time-resolved MR angiography. Indeed, voxels in time-resolved MR angiographic images were nearly five times larger than those in single-phase images. Our results suggest that these limitations in spatial resolution may be outweighed by specific advantages offered by the high temporal resolution of time-resolved imaging. This may be related to several factors.
First, a short acquisition time obviously reduces the frequency and severity of respiratory artifact. Second, although the same dose of contrast material and the same injection rate were used in both techniques, time-resolved MR angiography provided higher levels of contrast between the arterial vascular system and background tissue, which makes intuitive sense: Because the contrast material injection time in this study was 5–8 seconds and taking into account a minimum 5–7-second bolus spread (5), imaging with a temporal resolution of 7 seconds implies that the central lines of at least one k space will be filled during peak arterial enhancement.
In single-phase MR angiography, optimal arterial contrast enhancement requires individualized timing. In theory, inaccuracies may be related to technical or interpretative errors associated with the test bolus technique or with intraindividual variability in cardiac output (differences in heart rate at different moments during the study). Other possible explanations for the better scores for vessel contrast enhancement in time-resolved MR angiography are the slightly reduced repetition time (better background suppression) and the decreased image noise (increased contrast-to-noise ratio) resulting from the larger pixel size. The latter effect is counteracted, however, by the use of a higher bandwidth in time-resolved MR angiography.
With respect to the need for breath holding at time-resolved MR angiography, the fact that several independent data sets are successively acquired implies that intermittent breathing does not necessarily degrade the diagnostic quality of the study. With our current setup, two 7-second episodes of breath holding would theoretically allow us to obtain arterial and venous images that are free of artifact. However, we prefer use of a prolonged breath hold because (a) the exact arrival time of the contrast material bolus is not known unless a preliminary timing study is performed, and (b) the quality of subtraction images is dependent on the absence of respiratory misregistration between images obtained at different phases of perfusion.
Besides decreased respiratory artifact and improved arterial contrast, a third effect of the improved temporal resolution in time-resolved MR angiography is that images without marked venous and parenchymal enhancement can be obtained more frequently (in 100% of cases in our study). With a 25–30-second acquisition time, enhancement of renal and peripancreatic veins is more difficult to prevent.
It should be stressed here that the results obtained with time-resolved MR angiography are critically dependent on the size of the vessel studied. In the evaluation of hepatic arterial variations, for instance, it usually is possible to detect a replaced right or main hepatic artery, at least in our experience. On the other hand, smaller arteries such as the left gastric artery and the left hepatic arterial branches usually are too small to be adequately demonstrated.
Besides the advantages directly related to improved quality of images obtained during the first pass of contrast material, time-resolved MR angiography has other important advantages. First, individual bolus timing by means of the test bolus injection technique is no longer necessary. This not only reduces the total examination time but also renders the examination easier to perform. Second, multiphase imaging within a single breath hold offers possibilities for simultaneous detection of arterial and venous disease. Subtraction of arterial phase images from the venous phase images helps highlight the venous anatomy. Important potential applications are noninvasive vascular roadmapping in patients sheduled for partial liver resection or organ transplantation and "all-in-one" evaluation of pancreatic tumors (10). Third, in patients with arterial stenoses, the ability to obtain temporal information about enhancement patterns helps in assessing the hemodynamic significance of a stenosis. In general, time-resolved MR imaging with high temporal resolution theoretically opens possibilities for "functional" perfusion imaging of organs (11). However, this potential application was not addressed in our study.
Ongoing improvements in gradient technology make time-resolved MR imaging easier and less time-consuming. Indeed, as shown in this study, time-resolved MR angiography of the abdominal vasculature can now be successfully performed without the need for an off-line workstation.
This study has some limitations. First, the comparison of single-phase MR angiography and time-resolved MR angiography for detection of venous anatomic variants and abnormalities was biased because single-phase MR angiography in our patients was optimized for detection of arterial, not venous, disease. Furthermore, only one data set was obtained with single-phase MR angiography. In theory, we could have obtained a second data set at a more delayed phase of perfusion to demonstrate venous abnormalities. We did not do this, because we believe that the advantages of breath-hold time-resolved MR imaging for evaluation of the venous system are so obvious that specific comparative studies were not required. Indeed, subtraction images that selectively display venous anatomy can be obtained easily with the time-resolved technique. Selective display of venous anatomy facilitates image interpretation and thus contributes to the acceptance of the technique by nonradiologists.
A second limitation involved the method used for injection of contrast material. According to the recommendations of Prince and co-workers (5), the contrast material should be administered with an injection time that is 5–7 seconds shorter than the actual data acquisition time. In our study, application of this principle would involve an injection rate of approximately 1 mL/sec for single-phase MR angiography, whereas time-resolved MR angiography would probably have benefitted from a very high injection rate. For this comparative study, we decided to use identical contrast material injection rates for both acquisitions and to choose an "intermediate" injection rate of 2.5 mL/sec. We made this decision because we believed that the use of different injection rates would have introduced bias in the study because high levels of arterial contrast enhancement would never have been obtained at single-phase MR angiography. We cannot exclude the possibility that our results were influenced by this choice. Further comparative studies are needed to confirm our results and help determine the optimal contrast material injection rate as a function of acquisition time and technique.
We conclude that, despite inherent limitations in spatial resolution, time-resolved 3D contrast-enhanced abdominal MR angiography is a fast, simple, robust, and reliable technique that offers particular advantages over single-phase high-spatial-resolution MR angiography.
|
Footnotes |
|---|
Abbreviations: MIP = maximum intensity projection 3D = three-dimensional
Author contributions: Guarantor of integrity of entire study, L.V.H.; study concepts, L.V.H., T.D.J., H.B., L.S., G.M., R.F.; study design, L.V.H., T.D.J., H.B., L.S., G.M.; definition of intellectual content, L.V.H., T.D.J., H.B., G.M.; literature research, L.V.H., T.D.J.; clinical studies, L.V.H., T.D.J., L.S.; data acquisition, L.V.H., T.D.J., L.S.; data analysis, L.V.H., T.D.J., D.V., L.S.; statistical analysis, L.V.H.; manuscript preparation and editing, L.V.H.; manuscript review, T.D.J., H.B., L.S., D.V., R.O., G.M., R.F.
|
References |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  A. Nchimi, D. Brisbois, R. Materne, T. K. Y. Broussaud, I. Mancini, and P. Magotteaux Free-Breathing Accelerated Gadolinium-Enhanced MR Angiography in the Diagnosis of Renovascular Disease Am. J. Roentgenol., June 1, 2009; 192(6): 1531 - 1537. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. A. Schmidt and R. Morgan Renal Contrast-enhanced MR Angiography: Timing Errors and Accurate Depiction of Renal Artery Origins Radiology, October 1, 2008; 249(1): 178 - 186. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Fenchel, R. Saleh, H. Dinh, M. H. Lee, K. Nael, M. Krishnam, S. G. Ruehm, S. Miller, J. Child, and J. P. Finn Juvenile and Adult Congenital Heart Disease: Time-resolved 3D Contrast-enhanced MR Angiography Radiology, August 1, 2007; 244(2): 399 - 410. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. L. Coenegrachts, R. M. Hoogeveen, J. A. Vaninbroukx, H. T. Bosmans, D. J. Bielen, G. Maleux, F. Maes, P. Hamaekers, R. H. Oyen, and G. J. Marchal High-Spatial-Resolution 3D Balanced Turbo Field-Echo Technique for MR Angiography of the Renal Arteries: Initial Experience Radiology, April 1, 2004; 231(1): 237 - 242. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Masunaga, Y. Takehara, H. Isoda, T. Igarashi, M. Sugiyama, S. Isogai, N. Kodaira, H. Takeda, A. Nozaki, and H. Sakahara Assessment of Gadolinium-Enhanced Time-Resolved Three-Dimensional MR Angiography for Evaluating Renal Artery Stenosis Am. J. Roentgenol., May 1, 2001; 176(5): 1213 - 1219. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. F. Glockner Three-dimensional Gadolinium-enhanced MR Angiography: Applications for Abdominal Imaging RadioGraphics, March 1, 2001; 21(2): 357 - 370. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. N. Shetty, K. G. Bis, A. J. Duerinckx, and V. R. Narra Lower Extremity MR Angiography: Universal Retrofitting of High-Field-Strength Systems with Stepping Kinematic Imaging Platforms—Initial Experience Radiology, January 1, 2002; 222(1): 284 - 291. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
). (b) Coronal MIP image obtained with arterial phase time-resolved MR angiography (3.2/1.1; matrix, 140 x 256) shows narrowing of the gastroduodenal artery (arrow). (c) Coronal MIP image (3.2/1.1; matrix, 140 x 256) calculated after subtraction of venous phase data from arterial phase data shows narrowing (arrow) of the portosplenic confluence due to invasion by the tumor.
