HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Figures Only Full Text (PDF) Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Nael, K. Articles by Finn, J. P. Search for Related Content PubMed PubMed Citation Articles by Nael, K. Articles by Finn, J. P.

Supraaortic Arteries: Contrast-enhanced MR Angiography at 3.0 T—Highly Accelerated Parallel Acquisition for Improved Spatial Resolution over an Extended Field of View¹

¹ From the Department of Radiological Sciences, David Geffen School of Medicine, University of California Los Angeles, 10945 Le Conte Ave, Suite 3371, Los Angeles, CA 90095-7206 (K.N., J.P.V., W.B.P., T.O.M., J.P.F.); and Siemens Medical Solutions, Malvern, Pa (G.L.). Received November 2, 2005; revision requested December 21; revision received February 3, 2006; accepted March 7; final version accepted May 5. Address correspondence to K.N. (e-mail: nkambiz{at}mednet.ucla.edu).

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION ADVANCES IN KNOWLEDGE References

 
Purpose: To prospectively use 3.0-T breath-hold high-spatial-resolution contrast material–enhanced magnetic resonance (MR) angiography with highly accelerated parallel acquisition to image the supraaortic arteries of patients suspected of having arterial occlusive disease.

Materials and Methods: Institutional review board approval and written informed consent were obtained for this HIPAA-compliant study. Eighty patients (44 men, 36 women; age range, 44–90 years) underwent contrast-enhanced MR angiography of the head and neck at 3.0 T with an eight-channel neurovascular array coil. By applying a generalized autocalibrating partially parallel acquisition algorithm with an acceleration factor of four, high-spatial-resolution (0.7 x 0.7 x 0.9 mm = 0.44-mm³ voxels) three-dimensional contrast-enhanced MR angiography was performed during a 20-second breath hold. Two neuroradiologists evaluated vascular image quality and arterial stenoses. Interobserver variability was tested with the coefficient. Quantitation of stenosis at MR angiography was compared with that at digital subtraction angiography (DSA) (n = 13) and computed tomographic (CT) angiography (n = 12) with Spearman rank correlation coefficient (R_s).

Results: Arterial stenoses were detected with contrast-enhanced MR angiography in 208 (reader 1) and 218 (reader 2) segments, with excellent interobserver agreement ( = 0.80). There was a significant correlation between contrast-enhanced MR angiography and CT angiography (R_s = 0.95, reader 1; R_s = 0.87, reader 2) and between contrast-enhanced MR angiography and DSA (R_s = 0.94, reader 1; R_s = 0.92, reader 2) for the degree of stenosis. Sensitivity and specificity of contrast-enhanced MR angiography for detection of arterial stenoses greater than 50% were 94% and 98% for reader 1 and 100% and 98% for reader 2, with DSA as the standard of reference. Vascular image quality was sufficient for diagnosis or excellent for 97% of arterial segments evaluated.

Conclusion: By using highly accelerated parallel acquisition, the described 3.0-T contrast-enhanced MR angiographic protocol enabled visualization and characterization of the majority of supraaortic arteries, with diagnostic or excellent image quality (97% of arterial segments) and diagnostic values comparable with those obtained by using CT angiography and DSA for detection of arterial stenoses.

© RSNA, 2007

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION ADVANCES IN KNOWLEDGE References

 
Arterial occlusive disease commonly affects the supraaortic arteries and results in cerebral infarction and stroke. Currently, digital subtraction angiography (DSA) is the reference standard for evaluation of the craniocervical vasculature. DSA, however, has a 2.5% risk of transient ischemic attack and a small (<1%) risk of permanent neurologic deficit (1–3). Recent advances in three-dimensional contrast material–enhanced magnetic resonance (MR) angiography and x-ray computed tomographic (CT) angiography are changing the diagnostic algorithm in patients with arterial occlusive disease, such that DSA is increasingly being reserved for intervention or for cases in which noninvasive test results are equivocal.

The success of CT angiography in the evaluation of the carotid arteries (4–7), in large part, results from its simplicity and high spatial resolution. Recent advances in multisection technology have led to a great increase in the performance of CT angiography and to the setting of a higher standard for alternative noninvasive imaging techniques. Three-dimensional contrast-enhanced MR angiography has been successfully applied to the evaluation of the carotid and vertebrobasilar circulations (8–11), but the competing requirements for coverage and acquisition speed force a compromise in spatial resolution relative to multisection CT angiography.

Developments in parallel imaging techniques (12–14) create the potential to significantly improve the performance of contrast-enhanced MR angiographic applications in terms of spatial resolution, speed, or coverage, or all three. The introduction of 3.0-T whole-body imaging systems into clinical practice, with the promise of improved signal-to-noise ratios (SNRs) compared with those with 1.5 T, is advantageous when highly accelerated parallel imaging is considered (15). If its potential is realized, 3.0-T imaging might enhance the performance of contrast-enhanced MR angiography to the point where its spatial resolution can rival that of CT angiography, without the known drawbacks of CT angiography: radiation exposure, nephrotoxicity, and sensitivity to artifacts from dental amalgam and vascular calcifications. Thus, the purpose of our study was to prospectively use 3.0-T breath-hold high-spatial-resolution contrast-enhanced MR angiography with highly accelerated parallel acquisition to image the supraaortic arteries of patients who were suspected of having arterial occlusive disease.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION ADVANCES IN KNOWLEDGE References

 
An employee (G.L.) of Siemens Medical Solutions, Malvern, Pa, assisted in MR sequence development and imaging protocol design. The other authors, however, had full control over the data and information submitted for publication.

Patients
Eighty consecutive patients who were suspected of having cerebrovascular atherosclerotic disease and who met our study criteria were prospectively enrolled in this study between May 2004 and February 2005. The mean age was 66.0 years ± 12.7 (standard deviation), and the range was 44–90 years; there were 36 women (mean age, 64.6 years ± 13.9; range, 44–86 years) and 44 men (mean age, 67.3 years ± 11.6; range, 49–90 years). All patients had symptoms of cerebrovascular atherosclerotic disease in the prior 6 months, and these symptoms included transient ischemic attack, minor disabling ischemic stoke, amaurosis fugax, or progressive dizziness. Exclusion criteria included all standard contraindications to MR imaging (pacemaker, claustrophobia, contrast agent reaction, implanted metallic devices). Twenty-five patients underwent subsequent imaging; 12 patients underwent follow-up CT angiography within 5–20 days after contrast-enhanced MR angiography, and 13 patients underwent DSA within 1–15 days after contrast-enhanced MR angiography. Our Health Insurance Portability and Accountability–compliant study received institutional review board approval; written informed consent was obtained from all patients.

MR Angiography
All MR angiographic studies were performed with a 3.0-T whole-body MR system (Magnetom Trio; Siemens Medical Solutions, Erlangen, Germany) equipped with eight receiver channels and a fast gradient system (peak gradient amplitude, 40 mT/m; slew rate, 200 [mT · m^–1]/msec). Patients were placed supine on the imaging table and were advanced head first into the magnet bore. An electronic power injector (Spectris; Medrad, Pittsburgh, Pa) was used for contrast material injection. For signal reception, a commercially available eight-channel neurovascular array coil (Invivo, Orlando, Fla) with 13 coil elements distributed over the head and neck was used. A 2-mL timing bolus of gadodiamide (Omniscan; Amersham-GE Healthcare, Princeton, NJ) was injected at a rate of 1.2 mL/sec and was used to measure transit time from the arm vein to the carotid arteries. The mean delay time was 16 seconds (range, 12–23 seconds).

Subsequently, high-spatial-resolution contrast-enhanced MR angiography was performed in the coronal plane by using a fast spoiled gradient-echo sequence (Table 1). An asymmetric k-space sampling scheme was applied in all three planes to minimize the echo and acquisition times, and zero interpolation was performed to facilitate partial Fourier transform (75% along all three planes). Parallel imaging was performed with a generalized autocalibrating partially parallel acquisition algorithm that was based on autocalibrating simultaneous acquisition of spatial harmonics and parallel acquisition (14). An acceleration factor of four was used, with 24 reference k-space lines for calibration in the left-to-right phase-encoding direction. One hundred twelve partitions with a section thickness of 0.9 mm (interpolated to 0.7 mm) were selected for complete coverage of the carotid and vertebrobasilar circulations. The k-space matrix was 576 x 404 over a 390 x 282-mm FOV and resulted in voxel dimensions of 0.7 x 0.7 (in plane) x 0.9 mm.

All contrast-enhanced MR angiographic studies were performed successfully. No study had to be repeated because of technical problems. The mean examination time, defined as the time from patient entry into the MR imaging suite until departure, was 34 minutes (range, 30–48 minutes). Once in the magnet bore, average table time per patient was 14 minutes.

CT Angiography
CT angiographic images were obtained in 12 patients with a 16-section CT scanner (Somatom Sensation 16; Siemens Medical Solutions, Forchheim, Germany). Helical data were acquired with a 0.7-mm section collimation and a table speed of 8.6 mm/sec for 17 seconds starting at the third thoracic vertebra and proceeding to the cranial vertex. An FOV of 180 mm and a section thickness of 0.5 mm allowed a spatial resolution of 0.35 x 0.35 x 0.7 mm. Total coverage was 240 mm for a total of 480 reconstructed transverse sections. The helical acquisition was initiated after the start of the administration of the bolus of contrast medium, which was determined by using automated bolus detection software (CARE Bolus; Siemens Medical Solutions, Erlangen, Germany). The contrast agent injection protocol included intravenous injection of 90 mL of nonionic contrast medium (Omnipaque 350; Amersham-GE Healthcare) at a rate of 3 mL/sec.

DSA Imaging
In 13 patients, DSA was performed with insertion of the catheter via the femoral artery by an individual (T.O.M.) with 25 years of experience. The Seldinger technique was employed and involved aortography of the aortic arch, followed by selective catheterization of the common carotid artery and/or the vertebral artery. Images were obtained in the anteroposterior and lateral projections and in two oblique projections (–45° and +45°) for each catheterization, with spatial resolution of 0.15 x 0.15 mm. The injected volume of nonionic contrast medium was 7 mL for each injection (Omnipaque 300; Amersham-GE Healthcare). DSA images for a total number of 214 arterial segments were available on a computer workstation.

Image Analysis
After data acquisition, image processing for both contrast-enhanced MR angiography and CT angiography was performed with a three-dimensional workstation (Leonardo; Siemens Medical Solutions, Malvern, Pa) and standard commercial software by using a maximum intensity projection (MIP) and multiplanar volume reconstruction algorithm. One reader (K.N.), who had 4 years of experience and was not involved in the subsequent image analysis, performed all reconstructions. All of the reformatted data, as well as the source images, were available at the workstation for image analysis.

The arterial system from the aortic arch to the proximal intracranial circulation was divided into 22 segments. These segments included the following: segment 1, aortic arch; segment 2, brachiocephalic trunk; segments 3 and 4, bilateral subclavian arteries; segments 5 and 6, bilateral common carotid arteries; segments 7 and 8, bilateral internal carotid arteries; segments 9 and 10, bilateral external carotid arteries; segments 11 and 12, bilateral superficial temporal arteries; segments 13 and 14, bilateral anterior cerebral arteries (A1); segments 15 and 16, bilateral middle cerebral arteries (M1); segment 17, anterior communicating artery; segments 18 and 19, bilateral vertebral arteries; segment 20, basilar artery; and segments 21 and 22, bilateral posterior cerebral arteries (P1).

The image quality of each visualized arterial segment on contrast-enhanced MR angiograms was evaluated independently by two experienced neuroradiologists (J.P.V. and W.B.P., with 10 and 5 years of experience, respectively) by using a scale of 1–4 (grade 1, arterial segment was poorly visible with substantial blurring and/or artifacts; grade 2, arterial segment was visible with moderate blurring and/or artifacts [not sufficient for confident diagnosis]; grade 3, good arterial enhancement and minimal blurring and/or artifacts [sufficient image quality for confident diagnosis]; and grade 4, excellent image quality and sharply defined borders [information was of diagnostic quality]).

The image quality of an arterial segment was rated to be diagnostic (grade of 3) if there was clear discrimination between the vessel lumen and background, there were sharp vessel boundaries, there was uniform intraluminal signal intensity, and there were sharply defined lesion boundaries. Image quality was considered nondiagnostic if there was inadequate vessel enhancement or definition.

Contaminating venous signal was assigned a score on a scale of 0–2 (score 0, none or minimal; score 1, mild to moderate and did not interfere with diagnosis; and score 2, severe and did interfere with diagnosis).

For evaluation of arterial disease, both contrast-enhanced MR angiograms and CT angiograms were evaluated by each reader (the same neuroradiologists who evaluated image quality) independently. Evaluation sessions for CT angiograms and MR angiograms were performed at least 4 weeks apart to avoid possible recall bias. The order of presentation of the images from the examinations was randomized, and each reader was blinded to the patient's medical history, clinical data, and results of other imaging studies. Separate image reading sessions were organized for both readers by the study coordinator (K.N.), who attended all reading sessions. The readers were instructed to use the postprocessed data initially and, if required, to use the source data for interactive reformatting.

DSA images were reviewed at a computer workstation by the same two readers, who reached agreement with consensus. Both readers were blinded to the patient's name and clinical information, as well as to the findings of the MR angiographic examination. The severity of arterial disease was evaluated by using a scale of 1–4 (score 1, vessel irregularity with <10% luminal narrowing; score 2, mild stenosis with 10%–50% luminal narrowing; score 3, significant stenosis with 51%–99% luminal narrowing; and score 4, occlusion). When two or more stenoses were detected in the same vessel segment, the most severe stenosis was used for assignment of a score and analysis. The presence of any additional vascular abnormality, artifact, or incidental finding was recorded for each individual.

Statistical Analysis
A Wilcoxon rank sum test was used to evaluate the significance of the differences in image quality scores assigned by the readers. P < .05 was used as the criterion to indicate a statistically significant difference. Generalized estimating equation analysis was used to assess that all measurements in the same person were correlated exchangeably. The exchangeable working correlation structure was used with the logit link and binomial distribution to identify the agreement for detection of arterial stenoses by using contrast-enhanced MR angiography versus DSA and by using contrast-enhanced MR angiography versus CT angiography.

The degree of interobserver agreement for the image quality score assigned and for the detection of arterial stenoses within and between each imaging modality (contrast-enhanced MR angiography and CT angiography) were determined by calculating the coefficient (poor agreement, = 0; slight agreement, = 0.01–0.20; fair agreement, = 0.21–0.40; moderate agreement, = 0.41–0.60; good agreement, = 0.61–0.80; and excellent agreement, = 0.81–1.00) (16). The relationship between contrast-enhanced MR angiography and CT angiography and that between contrast-enhanced MR angiography and DSA in terms of categories of stenosis were analyzed by using the Spearman rank correlation coefficient (R_s). The sensitivity and specificity of MR angiography for detection of a stenosis of more than 50% were calculated for each reader, with DSA as the standard of reference.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION ADVANCES IN KNOWLEDGE References

 
Evaluation of Vascular Image Quality on Contrast-enhanced MR Angiograms
A majority (97%) of supraaortic arterial segments were identified by both readers with grades for definition in the diagnostic range (Fig 1). Reader 1 evaluated 56 arterial segments as having insufficient definition for diagnosis (grades 1 and 2) and 1704 segments as having diagnostic definition (grades 3 and 4). Reader 2 evaluated 52 arterial segments as having insufficient definition for diagnosis (grades 1 and 2) and 1708 segments as having diagnostic definition (grades 3 and 4). There was no significant difference in the image quality grades assigned by the two readers (P = .58). Analysis with the coefficient revealed good interobserver agreement ( = 0.74) (Table 2).

View larger version (112K): [in this window] [in a new window] [Download PPT slide]   Figure 1a: (a) Coronal and (b) sagittal thick (40-mm) MIP reformations of the entire supraaortic arteries and oblique sagittal (c) left and (d) right thin (10-mm) MIP reformations of the carotid bifurcations from contrast-enhanced MR angiography (3/1.1; flip angle, 20°; bandwidth, 720 Hz/pixel) in a 68-year-old man show atherosclerotic irregularity in the aortic arch and ectatic distal common carotid arteries (arrows). Three-dimensional data with voxel dimensions of 0.7 x 0.7 x 0.9 mm were acquired in 20 seconds during breath hold.

View larger version (48K): [in this window] [in a new window] [Download PPT slide]   Figure 1b: (a) Coronal and (b) sagittal thick (40-mm) MIP reformations of the entire supraaortic arteries and oblique sagittal (c) left and (d) right thin (10-mm) MIP reformations of the carotid bifurcations from contrast-enhanced MR angiography (3/1.1; flip angle, 20°; bandwidth, 720 Hz/pixel) in a 68-year-old man show atherosclerotic irregularity in the aortic arch and ectatic distal common carotid arteries (arrows). Three-dimensional data with voxel dimensions of 0.7 x 0.7 x 0.9 mm were acquired in 20 seconds during breath hold.

View larger version (51K): [in this window] [in a new window] [Download PPT slide]   Figure 1c: (a) Coronal and (b) sagittal thick (40-mm) MIP reformations of the entire supraaortic arteries and oblique sagittal (c) left and (d) right thin (10-mm) MIP reformations of the carotid bifurcations from contrast-enhanced MR angiography (3/1.1; flip angle, 20°; bandwidth, 720 Hz/pixel) in a 68-year-old man show atherosclerotic irregularity in the aortic arch and ectatic distal common carotid arteries (arrows). Three-dimensional data with voxel dimensions of 0.7 x 0.7 x 0.9 mm were acquired in 20 seconds during breath hold.

View larger version (44K): [in this window] [in a new window] [Download PPT slide]   Figure 1d: (a) Coronal and (b) sagittal thick (40-mm) MIP reformations of the entire supraaortic arteries and oblique sagittal (c) left and (d) right thin (10-mm) MIP reformations of the carotid bifurcations from contrast-enhanced MR angiography (3/1.1; flip angle, 20°; bandwidth, 720 Hz/pixel) in a 68-year-old man show atherosclerotic irregularity in the aortic arch and ectatic distal common carotid arteries (arrows). Three-dimensional data with voxel dimensions of 0.7 x 0.7 x 0.9 mm were acquired in 20 seconds during breath hold.

Additional anatomic anomalies detected at contrast-enhanced MR angiography were as follows: bovine arch (n = 5), direct origin of the vertebral artery from the aortic arch (n = 4) (Fig 2), three arterial fenestrations (two vertebral and one basilar) (Fig 3), and fetal origins of the posterior cerebral arteries (n = 8).

View larger version (59K): [in this window] [in a new window] [Download PPT slide]   Figure 2: Sagittal oblique thin (20-mm) MIP from contrast-enhanced MR angiography (2.9/1.1; flip angle, 21°; bandwidth, 720 Hz/pixel) in 55-year-old man with headache and dizziness shows direct origin of the vertebral artery, proximally duplicated (arrows), from the aortic arch.

View larger version (164K): [in this window] [in a new window] [Download PPT slide]   Figure 3: Coronal MIP from contrast-enhanced MR angiography (3/1.2; flip angle, 20°; bandwidth, 720 Hz/pixel; number of partitions, 112; voxel size, 0.44 mm3; imaging time, 20 seconds) in 56-year-old man evaluated for atherosclerosis shows proximal basilar arterial fenestration (arrow).

Reader 2 identified disease in 218 arterial segments, which included 77 segments with vessel irregularities (luminal narrowing of <10%), 80 segments with 10%–50% stenosis, 53 segments with 51%–99% stenosis, and eight segmental occlusions. Image analysis was based on the default postprocessed images, except for analysis of 22 arterial segments in which interactive reformatting was performed by reader 2.

The overall interobserver agreement for detection of arterial stenoses was good ( = 0.80; 95% confidence interval: 0.76, 0.84) (Table 3). The agreement between the two readers for determination of nonocclusion versus occlusion was excellent ( = 1.00).

With CT angiography, reader 1 identified disease in 62 arterial segments, which included 30 segments with arterial irregularities (luminal narrowing of <10%), 23 segments with 10%–50% stenosis, eight segments with 51%–99% stenosis, and one segmental occlusion. Reader 2 identified disease in 64 arterial segments, which included 31 arterial irregularities (luminal narrowing of <10%), 23 segments with 10%–50% stenosis, nine segments with 51%–99% stenosis, and one segmental occlusion. The interobserver agreement for classification of degree of stenosis by using CT angiography was excellent ( = 0.90; 95% confidence interval: 0.84, 0.96).

The intraobserver agreement between CT angiography and contrast-enhanced MR angiography for the detection of arterial occlusive disease was excellent (Figs 4, 5) for both reader 1 ( = 0.89; 95% confidence interval: 0.83, 0.95) and reader 2 ( = 0.86; 95% confidence interval: 0.79, 0.93).

View larger version (134K): [in this window] [in a new window] [Download PPT slide]   Figure 4: Images in 78-year-old man with history of stroke show significant eccentric segmental stenosis of the left internal carotid artery (arrow). Left: Sagittal oblique thin (10-mm) MIP from contrast-enhanced MR angiography (3/1.2; flip angle, 20°; bandwidth, 720 Hz/pixel; FOV, 390 mm; number of partitions, 112; reconstructed section thickness, 0.7 mm; imaging time, 20 seconds). Right: MIP in same view from CT angiography (FOV, 180 mm; reconstructed section thickness, 0.5 mm; imaging time, 17 seconds).

View larger version (108K): [in this window] [in a new window] [Download PPT slide]   Figure 5: Images in 64-year-old woman with history of transient ischemic attack show severe stenosis (arrow) of the origin of the left external carotid artery. Left: Sagittal oblique thin (10-mm) MIP from contrast-enhanced MR angiography (3/1.2; flip angle, 20°; bandwidth, 720 Hz/pixel; FOV, 390 mm; number of partitions, 112; interpolated section thickness, 0.7 mm; imaging time, 20 seconds). Right: MIP in same view from CT angiography (FOV, 180 mm; reconstructed section thickness, 0.5 mm; imaging time, 17 seconds).

DSA showed disease in 53 of 214 arterial segments, which included 18 segments with mild irregularities (<10% stenosis), 17 segments with 10%–50% stenosis, 16 segments with 51%–99% stenosis, and two segmental occlusions. The distribution of arterial segments with significant disease (>50% luminal narrowing) was as follow: two segmental occlusions (one in the vertebral artery and one in the internal carotid artery) and 16 significant stenoses of 51%–99% (one in the brachiocephalic trunk, four in the common carotid artery, four in the internal carotid artery, three in the external carotid artery, two in the vertebral artery, and two in the middle cerebral artery).

When we compared contrast-enhanced MR angiography with DSA, reader 1 underestimated significant stenosis in one segment and overestimated mild stenosis in two segments, whereas reader 2 overestimated mild stenosis in three arterial segments. More luminal irregularities (<10% stenosis) were noted with contrast-enhanced MR angiography than were noted with DSA (seven and eight additional segments for readers 1 and 2, respectively). The sensitivity and specificity of contrast-enhanced MR angiography for the diagnosis of a stenosis greater than 50% were, respectively, 94% and 98% for reader 1 and 100% and 98% for reader 2.

There was a significant correlation between DSA and contrast-enhanced MR angiography (Figs 6–8) for the degree of stenosis (reader 1: R_s = 0.94, P < .001; reader 2: R_s = 0.92, P < .001) (Table 5).

View larger version (168K): [in this window] [in a new window] [Download PPT slide]   Figure 6: Images in 83-year-old woman with history of transient ischemic attack show mild stenosis (grade 2, 10%–50%) at the origins of the left internal carotid artery (black arrow) and significant stenosis (grade 3, 51%–99%) at the origin of the left external carotid artery (white arrow). Left: Coronal MIP from contrast-enhanced MR angiography (2.9/1.1; flip angle, 20°; bandwidth, 720 Hz/pixel; FOV, 390 mm; number of partitions, 112; reconstructed section thickness, 0.7 mm; imaging time, 20 seconds). Right: Sagittal oblique view from DSA shows image subtraction artifact (arrowhead).

View larger version (48K): [in this window] [in a new window] [Download PPT slide]   Figure 7a: Images in 65-year-old woman with history of stroke show severe stenosis of the proximal left M1 segment (arrows). (a, b) Coronal thin (20-mm) MIPs and (c) coronal multiplanar reconstruction from contrast-enhanced MR angiography (3/1.2; flip angle, 20°; bandwidth, 720 Hz/pixel; FOV, 390 mm; number of partitions, 112; reconstructed section thickness, 0.7 mm; imaging time, 20 seconds) and (d) sagittal oblique view from DSA.

View larger version (71K): [in this window] [in a new window] [Download PPT slide]   Figure 7b: Images in 65-year-old woman with history of stroke show severe stenosis of the proximal left M1 segment (arrows). (a, b) Coronal thin (20-mm) MIPs and (c) coronal multiplanar reconstruction from contrast-enhanced MR angiography (3/1.2; flip angle, 20°; bandwidth, 720 Hz/pixel; FOV, 390 mm; number of partitions, 112; reconstructed section thickness, 0.7 mm; imaging time, 20 seconds) and (d) sagittal oblique view from DSA.

View larger version (35K): [in this window] [in a new window] [Download PPT slide]   Figure 7c: Images in 65-year-old woman with history of stroke show severe stenosis of the proximal left M1 segment (arrows). (a, b) Coronal thin (20-mm) MIPs and (c) coronal multiplanar reconstruction from contrast-enhanced MR angiography (3/1.2; flip angle, 20°; bandwidth, 720 Hz/pixel; FOV, 390 mm; number of partitions, 112; reconstructed section thickness, 0.7 mm; imaging time, 20 seconds) and (d) sagittal oblique view from DSA.

View larger version (98K): [in this window] [in a new window] [Download PPT slide]   Figure 7d: Images in 65-year-old woman with history of stroke show severe stenosis of the proximal left M1 segment (arrows). (a, b) Coronal thin (20-mm) MIPs and (c) coronal multiplanar reconstruction from contrast-enhanced MR angiography (3/1.2; flip angle, 20°; bandwidth, 720 Hz/pixel; FOV, 390 mm; number of partitions, 112; reconstructed section thickness, 0.7 mm; imaging time, 20 seconds) and (d) sagittal oblique view from DSA.

View larger version (178K): [in this window] [in a new window] [Download PPT slide]   Figure 8: Images in 72-year-old woman with history of stroke show multiple segmental significant stenosis (grade 3, 51%–99%) at the distal cervical portion of the right vertebral artery and occlusion of the distal cervical segment of the left vertebral artery (arrow). Left: Coronal oblique thin MIP from contrast-enhanced MR angiography (3/1.2; flip angle, 19°; bandwidth, 720 Hz/pixel; FOV, 390 mm; number of partitions, 112; reconstructed section thickness, 0.7 mm; imaging time, 20 seconds). Right: Projection in same view from DSA.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION ADVANCES IN KNOWLEDGE References

By using a multichannel neurovascular array coil and highly accelerated parallel acquisition (generalized autocalibrating partially parallel acquisition algorithm with an acceleration factor of four), images with 0.44-mm³ voxels were generated in a routine breath hold and yielded high definition of arterial anatomy from the aortic arch to the circle of Willis. To our knowledge, this is the first report of contrast-enhanced MR angiography applied to the supraaortic circulation with the specified spatial resolution, speed, and coverage. The reliability of the contrast-enhanced MR angiographic protocol was validated in 80 patients, and its diagnostic performance was validated in a subpopulation of patients with reference to DSA and CT angiography. The positive effect of decreased voxel size on image quality and delineation of stenosis is well known (17), and findings in our study were indicative of this effect. The obtained overall sensitivity and specificity for evaluation of significant arterial disease exceeded 90%, and these values were in accordance with the results of previously published studies with imaging at 1.5 T (11,18,19).

Researchers in several previous reports addressed contrast-enhanced MR angiography of the carotid circulation at 1.5 T. Much of the early work focused on small FOVs (approximately of 220 mm), in combination with modest image matrices (approximately of 256 x 256), to achieve reasonable spatial resolution over the cervical carotid arteries (9,19,20). To some extent, this focus reflected the limitations of receiver coil arrays and image reconstruction hardware available at the time. Because of the systemic nature of atherosclerosis, the presence of tandem or concurrent arterial disease in supraaortic arteries is common (21,22); therefore, accurate visualization of the entire supraaortic arterial tree is desirable.

There are several reports in which the investigators indicated the successful implementation of breath-hold contrast-enhanced MR angiography for imaging of the supraaortic arteries at 1.5 T (23). Our work describes a similar approach with imaging at 3.0 T but takes advantage of the higher baseline SNR to apply parallel acquisition with fast acceleration factors. The use of parallel acquisition in this context allows increased spatial resolution and greatly increased coverage, which enhances performance by a factor of almost four compared with the performance with a nonparallel acquisition.

Several workers (24–26) advocate specific k-space trajectories, notably elliptic-centric reordering or radial k-space acquisition, for contrast-enhanced MR angiography. These techniques have been shown to facilitate venous suppression in contrast-enhanced MR angiography by ensuring that the central k-space points are acquired when only arteries are enhanced. It is important, however, to ensure that image acquisition does not precede the peak of the contrast agent bolus; otherwise, the study can be degraded. Willinek et al (27) reported an MR angiographic protocol for evaluation of supraaortic arteries at 1.5 T in which the central segment of k-space is encoded in random order. This use of random order is postulated to lessen the likelihood of edge artifact sometimes seen with elliptic centric acquisition. Willinek and colleagues achieved a voxel size of 0.81 x 0.81 x 1 mm = 0.66 mm³ over a 350-mm FOV during a 58-second image acquisition (27). The choice of a specific k-space trajectory or reordering scheme is, to some extent, a matter of user preference and experience and each may have advantages and trade-offs.

In our study of imaging at 3.0 T, we chose a linear asymmetric k-space trajectory, whereby the center of k-space is acquired one-third of the way through the measurement (approximately 7 seconds into the acquisition). This point is calculated to coincide approximately with the peak of arterial contrast enhancement, which is based on the initial timing bolus. We did not aim for full venous suppression at all costs but rather for maximal arterial signal intensity in a short overall acquisition period. In our study, venous enhancement was mild to absent in most cases and rarely interfered with diagnosis (0.6%). This result was achieved with our use of a test bolus to measure the contrast agent timing window and the use of a short image acquisition period.

The drawback of parallel acquisition is a degradation in SNR, mainly based on the degree of k-space undersampling and the coil array geometry known as the g factor (13,28). The SNR penalty is proportional to the square root of the acceleration factor, A, and the g factor, g, of the coil, which may be expressed with the following: SNR 1/g · A (13). Optimized coil designs and higher receiver channels provide better sensitivity profiles, and the better sensitivity profiles result in g factors closer to one. Theoretical work that describes the ultimate intrinsic SNR for parallel imaging demonstrates that the best possible g factor of one can be achieved for acceleration factors of four or less (29,30). Strategies to preserve SNR include raising the baseline signal by increasing field strength (31–33) and minimizing noise amplification by using array coils with more channels and improved geometry and sensitivity profiles (34,35), as in our study.

There are also potential challenges for contrast-enhanced MR angiography at a higher field strength. Magnetic susceptibility effects are known to increase with an increase in the magnetic field strength (36) and can cause image artifacts. In our study, magnetic susceptibility effects were not limiting, probably because of the short echo time (1.2 msec) employed in the contrast-enhanced MR angiographic sequence. Nonetheless, if the local concentration of gadolinium in the ipsilateral subclavian vein remains high during the arterial phase (36), it can degrade the signal in the adjacent subclavian artery, as we observed in our study (10%). It is important to be aware of this phenomenon so as not to overdiagnose subclavian artery disease. Also, preferential injection into a right arm vein protects against the extension of the artifact proximally into the left innominate vein and the subjacent aortic arch.

We acknowledge limitations of our study. Comparison studies (CT angiography or DSA) were not available for all patients, and it is important to interpret the measurements of diagnostic performance of this technique in this context. DSA was performed only in patients in whom therapeutic intervention was planned. This selection bias may be partially responsible for the high values for sensitivity and specificity in our study. At our institution (as at many others), however, this selection bias represents the standard of practice, as DSA is no longer routinely performed in all patients with supraaortic atherosclerosis.

In conclusion, by using highly accelerated parallel acquisition, the described 3.0-T contrast-enhanced MR angiographic protocol enabled visualization and characterization of the majority of for supraaortic arteries, with diagnostic or excellent image quality (97% of arterial segments) and diagnositic values comparable with those obtained by using CT angiography and DSA for detection of arterial stenoses. Further studies are required to confirm the accuracy of the technique in a broader clinical setting.

	ADVANCES IN KNOWLEDGE

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION ADVANCES IN KNOWLEDGE References

 

• By using highly accelerated parallel acquisition for supraaortic contrast-enhanced MR angiography at 3.0 T, we successfully acquired images with 0.44-mm³ voxels over an extended field of view in a routine breath-hold period.
  
• Acquisition of a three-dimensional data set with 0.44-mm³ voxels encompassing the entire carotid and vertebrobasilar circulations resulted in diagnostic or excellent image quality for 97% of arterial segments evaluated.
  

	ACKNOWLEDGMENTS

 
We thank James W. Sayre, PhD, for performing the statistical analysis.

	FOOTNOTES

 

Abbreviations: DSA = digital subtraction angiography • FOV = field of view • MIP = maximum intensity projection • SNR = signal-to-noise ratio

See Materials and Methods for pertinent disclosures.

Author contributions: Guarantors of integrity of entire study, K.N., J.P.F.; study concepts/study design or data acquisition or data analysis/interpretation, all authors; manuscript drafting or manuscript revision for important intellectual content, all authors; manuscript final version approval, all authors; literature research, K.N., J.P.V., G.L.; clinical studies, K.N., J.P.V., W.B.P., T.O.M., G.L., J.P.F.; statistical analysis, K.N., W.B.P.; and manuscript editing, K.N., J.P.V., J.P.F.

	References

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION ADVANCES IN KNOWLEDGE References

 

1. Hankey GJ, Warlow CP, Sellar RJ. Cerebral angiographic risk in mild cerebrovascular disease. Stroke 1990;21:209–222.[Abstract/Free Full Text]
2. Heiserman JE, Dean BL, Hodak JA, et al. Neurologic complications of cerebral angiography. AJNR Am J Neuroradiol 1994;15:1401–1407.[Abstract]
3. American College of Radiology. The ACR practice guideline for the performance of diagnostic cervicocerebral angiography in adults. Reston, Va: American College of Radiology, 2005; 57–70.
4. Marcus CD, Ladam-Marcus VJ, Bigot JL, Clement C, Baehrel B, Menanteau BP. Carotid arterial stenosis: evaluation at CT angiography with the volume-rendering technique. Radiology 1999;211:775–780.[Abstract/Free Full Text]
5. Berg M, Zhang Z, Ikonen A, et al. Multi-detector row CT angiography in the assessment of carotid artery disease in symptomatic patients: comparison with rotational angiography and digital subtraction angiography. AJNR Am J Neuroradiol 2005;26:1022–1034.[Abstract/Free Full Text]
6. Randoux B, Marro B, Koskas F, et al. Carotid artery stenosis: prospective comparison of CT, three-dimensional gadolinium-enhanced MR, and conventional angiography. Radiology 2001;220:179–185.[Abstract/Free Full Text]
7. Josephson SA, Bryant SO, Mak HK, Johnston SC, Dillon WP, Smith WS. Evaluation of carotid stenosis using CT angiography in the initial evaluation of stroke and TIA. Neurology 2004;63:457–460.[Abstract/Free Full Text]
8. Remonda L, Senn P, Barth A, Arnold M, Lovblad KO, Schroth G. Contrast-enhanced 3D MR angiography of the carotid artery: comparison with conventional digital subtraction angiography. AJNR Am J Neuroradiol 2002;23:213–219.[Abstract/Free Full Text]
9. Leclerc X, Gauvrit JY, Nicol L, Pruvo JP. Contrast-enhanced MR angiography of the craniocervical vessels: a review. Neuroradiology 1999;41:867–874.[CrossRef][Medline]
10. Carr JC, Ma J, Desphande V, Pereles S, Laub G, Finn JP. High-resolution breath-hold contrast-enhanced MR angiography of the entire carotid circulation. AJR Am J Roentgenol 2002;178:543–549.[Abstract/Free Full Text]
11. Randoux B, Marro B, Koskas F, Chiras J, Dormont D, Marsault C. Proximal great vessels of aortic arch: comparison of three-dimensional gadolinium-enhanced MR angiography and digital subtraction angiography. Radiology 2003;229:697–702.[Abstract/Free Full Text]
12. Sodickson DK, McKenzie CA, Li W, Wolff S, Manning WJ, Edelman RR. Contrast-enhanced 3D MR angiography with simultaneous acquisition of spatial harmonics: a pilot study. Radiology 2000;217:284–289.[Abstract/Free Full Text]
13. Pruessmann KP, Weiger M, Scheidegger MB, Boesiger P. SENSE: sensitivity encoding for fast MRI. Magn Reson Med 1999;42:952–962.[CrossRef][Medline]
14. Griswold MA, Jakob PM, Heidemann RM, et al. Generalized autocalibrating partially parallel acquisitions (GRAPPA). Magn Reson Med 2002;47:1202–1210.[CrossRef][Medline]
15. Pruessmann KP. Parallel imaging at high field strength: synergies and joint potential. Top Magn Reson Imaging 2004;15:237–244.[CrossRef][Medline]
16. Landis JR, Koch GG. An application of hierarchical kappa-type statistics in the assessment of majority agreement among multiple observers. Biometrics 1977;33:363–374.[CrossRef][Medline]
17. Leclerc X, Nicol L, Gauvrit JY, Le Thuc V, Leys D, Pruvo JP. Contrast-enhanced MR angiography of supraaortic vessels: the effect of voxel size on image quality. AJNR Am J Neuroradiol 2000;21:1021–1027.[Abstract/Free Full Text]
18. Willinek WA, von Falkenhausen M, Born M, et al. Noninvasive detection of steno-occlusive disease of the supra-aortic arteries with three-dimensional contrast-enhanced magnetic resonance angiography: a prospective, intra-individual comparative analysis with digital subtraction angiography. Stroke 2005;36:38–43.[Abstract/Free Full Text]
19. Phan T, Huston J 3rd, Bernstein MA, Riederer SJ, Brown RD Jr. Contrast-enhanced magnetic resonance angiography of the cervical vessels: experience with 422 patients. Stroke 2001;32:2282–2286.[Abstract/Free Full Text]
20. Cloft HJ, Murphy KJ, Prince MR, Brunberg JA. 3D gadolinium-enhanced MR angiography of the carotid arteries. Magn Reson Imaging 1996;14:593–600.[CrossRef][Medline]
21. Rouleau PA, Huston J 3rd, Gilbertson J, Brown RD Jr, Meyer FB, Bower TC. Carotid artery tandem lesions: frequency of angiographic detection and consequences for endarterectomy. AJNR Am J Neuroradiol 1999;20:621–625.[Abstract/Free Full Text]
22. Eliasziw M, Streifler JY, Fox AJ, Hachinski VC, Ferguson GG, Barnett HJ. Significance of plaque ulceration in symptomatic patients with high-grade carotid stenosis. North American Symptomatic Carotid Endarterectomy Trial. Stroke 1994;25:304–308.
23. Yang CW, Carr JC, Futterer SF, et al. Contrast-enhanced MR angiography of the carotid and vertebrobasilar circulations. AJNR Am J Neuroradiol 2005;26:2095–2101.[Abstract/Free Full Text]
24. Fain SB, Riederer SJ, Bernstein MA, Huston J 3rd. Theoretical limits of spatial resolution in elliptical-centric contrast-enhanced 3D-MRA. Magn Reson Med 1999;42:1106–1116.[CrossRef][Medline]
25. Huston J 3rd, Fain SB, Wald JT, et al. Carotid artery: elliptic centric contrast-enhanced MR angiography compared with conventional angiography. Radiology 2001;218:138–143.[Abstract/Free Full Text]
26. Nederkoorn PJ, Elgersma OE, van der Graaf Y, Eikelboom BC, Kappelle LJ, Mali WP. Carotid artery stenosis: accuracy of contrast-enhanced MR angiography for diagnosis. Radiology 2003;228:677–682.[Abstract/Free Full Text]
27. Willinek WA, Gieseke J, Conrad R, et al. Randomly segmented central k-space ordering in high-spatial-resolution contrast-enhanced MR angiography of the supraaortic arteries: initial experience. Radiology 2002;225:583–588.[Abstract/Free Full Text]
28. Sodickson DK, Manning WJ. Simultaneous acquisition of spatial harmonics (SMASH): fast imaging with radiofrequency coil arrays. Magn Reson Med 1997;38:591–603.[Medline]
29. Wiesinger F, Boesiger P, Pruessmann KP. Electrodynamics and ultimate SNR in parallel MR imaging. Magn Reson Med 2004;52:376–390.[CrossRef][Medline]
30. Ohliger MA, Grant AK, Sodickson DK. Ultimate intrinsic signal-to-noise ratio for parallel MRI: electromagnetic field considerations. Magn Reson Med 2003;50:1018–1030.[CrossRef][Medline]
31. Campeau NG, Huston J 3rd, Bernstein MA, Lin C, Gibbs GF. Magnetic resonance angiography at 3.0 Tesla: initial clinical experience. Top Magn Reson Imaging 2001;12:183–204.[CrossRef][Medline]
32. Willinek WA, Born M, Simon B, et al. Time-of-flight MR angiography: comparison of 3.0-T imaging and 1.5-T imaging—initial experience. Radiology 2003;229:913–920.[Abstract/Free Full Text]
33. Nael K, Michaely HJ, Villablanca P, Salamon N, Laub G, Finn JP. Time-resolved contrast enhanced magnetic resonance angiography of the head and neck at 3.0 Tesla: initial results. Invest Radiol 2006;41:116–124.[CrossRef][Medline]
34. Weiger M, Pruessmann KP, Leussler C, Roschmann P, Boesiger P. Specific coil design for SENSE: a six-element cardiac array. Magn Reson Med 2001;45:495–504.[CrossRef][Medline]
35. de Zwart JA, Ledden PJ, van Gelderen P, Bodurka J, Chu R, Duyn JH. Signal-to-noise ratio and parallel imaging performance of a 16-channel receive-only brain coil array at 3.0 Tesla. Magn Reson Med 2004;51:22–26.[CrossRef][Medline]
36. Neimatallah MA, Chenevert TL, Carlos RC, et al. Subclavian MR arteriography: reduction of susceptibility artifact with short echo time and dilute gadopentetate dimeglumine. Radiology 2000;217:581–586.[Abstract/Free Full Text]

This article has been cited by other articles:

				C. K. Kuhl, F. Traber, and H. H. Schild Whole-Body High-Field-Strength (3.0-T) MR Imaging in Clinical Practice * Part I. Technical Considerations and Clinical Applications Radiology, March 1, 2008; 246(3): 675 - 696. [Abstract] [Full Text] [PDF]

				K. Nael, J. P. Villablanca, L. Mossaz, W. Pope, A. Juncosa, G. Laub, and J. P. Finn 3-T Contrast-Enhanced MR Angiography in Evaluation of Suspected Intracranial Aneurysm: Comparison with MDCT Angiography Am. J. Roentgenol., February 1, 2008; 190(2): 389 - 395. [Abstract] [Full Text] [PDF]

This Article Abstract Figures Only Full Text (PDF) Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Nael, K. Articles by Finn, J. P. Search for Related Content PubMed PubMed Citation Articles by Nael, K.