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 Kramer, H. Articles by Reiser, M. F. Search for Related Content PubMed PubMed Citation Articles by Kramer, H. Articles by Reiser, M. F.

Cardiovascular Screening with Parallel Imaging Techniques and a Whole-Body MR Imager¹

¹ From the Institute of Clinical Radiology, Ludwig-Maximilians-University, Klinikum Grosshadern, Marchioninistrasse 15, D-81377 Munich, Germany (H.K., S.O.S., K.N., A.H., A.S., B.J.W., O.D., M.F.R.); and Siemens Medical Solutions, Erlangen, Germany (E.W., B.K.). Received April 3, 2004; revision requested June 15; revision received August 27; accepted October 4. Address correspondence to H.K. (e-mail: harald.kramer{at}med.uni-muenchen.de).

	ABSTRACT

TOP ABSTRACT INTRODUCTION Materials and Methods Results Discussion References

 
The purpose of this study was to integrate parallel acquisition techniques into a comprehensive whole-body cardiovascular screening protocol to image all relevant organ systems without compromising spatial or temporal resolution. The study was approved by the institutional review board, and oral and written informed consent was obtained from each subject. Fifty subjects underwent whole-body magnetic resonance imaging that included imaging of heart, blood vessels, brain, lungs, and abdominal organs with a standard eight-channel imager. Image quality and pathologic findings were evaluated by two readers. The same protocol was then implemented with a new 32-channel whole-body imager. Depiction of 1476 (73.2%) of 2016 vessel segments was rated as good to excellent, and that of 1744 (86.5%), as without venous overlay. Interobserver agreement was good in evaluation of image quality and excellent in evaluation of pathologic findings. Acquisition time was reduced significantly (P < .05) with use of the whole-body imager and parallel acquisition techniques, which provided high-quality fast cardiovascular imaging.

© RSNA, 2005

	INTRODUCTION

TOP ABSTRACT INTRODUCTION Materials and Methods Results Discussion References

 
In most industrialized countries, cardiovascular diseases are still the most common cause of morbidity and mortality (1). The most threatening manifestations are internal carotid artery stenosis, coronary artery disease, and renal artery stenosis, which may have the potentially fatal consequences of stroke, myocardial infarction, and renovascular hypertension, respectively. Mortality statistics show that malignant diseases, predominantly lung cancer and colon cancer, are the next most serious category. Magnetic resonance (MR) imaging holds promise for use in screening for these diseases because of its lack of ionizing radiation, excellent soft-tissue contrast, and ability to depict both morphologic structure and function (2,3).

Currently, a leading role has been established for MR imaging in the diagnostic work-up of various organ systems with various techniques, including MR angiography of the carotid, renal, and peripheral arteries (4,5). In addition, MR imaging is considered the reference standard for assessment of cardiac function. Moreover, with the use of perfusion imaging techniques, MR imaging is very useful for the functional assessment of significant coronary artery disease. The definitive assessment of these organ systems with MR imaging, however, is often complex and time consuming. Because of different requirements in regard to coil configuration, section positioning, and contrast medium application, it is often impossible to integrate state-of-the-art evaluations of different organ systems into a single comprehensive examination with an acceptable acquisition time.

Previously, investigators in several studies (2,3,6) demonstrated the feasibility of a whole-body imaging examination that includes cardiovascular screening as well as screening for malignant disease. In all these studies, dedicated rolling platforms were used in combination with fast imaging techniques to cover the entire body within a short acquisition time. This approach, however, requires substantial compromises with regard to spatial and temporal resolution of the different organ systems. Recent multicenter trials have shown that, even with state-of-the-art high-spatial-resolution imaging, MR angiography of the renal arteries has limited accuracy (7–10). In the Renal Artery Diagnostic Imaging Study in Hypertension, sensitivity and specificity values of less than 80% were found for MR angiography. Similarly sobering findings were recently published for MR angiography of the carotid arteries in comparison with new high-resolution Doppler ultrasonography (US) with amplitude-weighted power imaging (11). In addition, some areas of the body, such as the lungs, continue to offer a challenge for MR imaging because of fast signal decay and pronounced respiratory and cardiac motion effects.

Parallel imaging is a recently introduced technique that uses the spatial distribution of the MR signal received by different receiver coils with various anatomic orientations. Mainly, this technique has the potential to increase spatial or temporal resolution without greatly increasing acquisition time. While this method has already gained wide clinical use for the dedicated evaluation of certain organ systems, its use for whole-body imaging has been limited (12), predominantly because it has been difficult to maintain the well-defined positioning of the various receiver coils necessary for parallel imaging of the entire body during table movement.

Integrated parallel acquisition techniques (PAT) with autocalibration of coils within each acquisition allow the use of almost every combination of different receiver coils and permit repositioning of the table during the various stages of image acquisition (13). Thus, the aim of this study was to incorporate integrated PAT into a comprehensive whole-body cardiovascular screening protocol to image all relevant organ systems without compromising spatial or temporal resolution.

	Materials and Methods

TOP ABSTRACT INTRODUCTION Materials and Methods Results Discussion References

 
Study Subjects
The study was approved by the institutional review board, and oral and written informed consent was obtained from each of the study subjects. The study was not industry-supported in any way, and none of the authors who had control over the inclusion of data were employees of industry at the time of the study.

Fifty participants in a screening program between April and December 2003, including 48 men and two women with a mean age of 54 years (age range, 41–63 years), were included in the study. All individuals underwent routine yearly screening for cardiovascular and malignant disease with a standardized set of conventional techniques, including Doppler US of the carotid arteries, US of the abdominal organs and vessels, electrocardiography while at rest and during stress testing with an ergometer, echocardiography, chest radiography, and laboratory tests that included a complete blood cell count, differential blood cell count, and determination of prostate-specific antigen level. All examinations were performed by the company medical officer or by staff at a medical center that specializes in health screening examinations. When possible, the results of these examinations were compared with findings at MR imaging. All individuals included in the study were considered healthy on the basis of their self-reported impression and the objective evaluation by the referring physician. All individuals referred were from two large international companies. After the examination, all of them answered a questionnaire about their subjective opinion of the examination and its duration.

Screening was performed with two different MR imaging systems: Initially, in 42 subjects (group 1), a standard clinical imager equipped with eight receiver channels (Magnetom Sonata Maestro Class; Siemens Medical Solutions, Erlangen, Germany) was used. Later, eight subjects (group 2) were evaluated with a new whole-body MR imager equipped with 32 receiver channels (Magnetom Avanto; Siemens Medical Solutions). With both systems, it was possible to use integrated PAT (iPAT; Siemens Medical Solutions).

Application of Integrated PAT
Parallel imaging exploits differences in the MR signal intensity received simultaneously by multiple surface coils with different spatial sensitivity profiles. With parallel imaging techniques, the number of phase-encoding steps can be reduced without decreasing the spatial resolution or matrix size of the acquired image. Aliasing artifacts, which would occur with conventional image reconstruction, are corrected by using the parallel acquisition data and the coil profiles of the different receiver coils. This type of image reconstruction can be performed in the image domain by using data acquired with sensitivity encoding or in the k-space domain by using data obtained with simultaneous acquisition of spatial harmonics. The signal for the measurement of the coil profiles is obtained in a reference acquisition, which usually is performed at the beginning of the examination. A reference acquisition, however, requires that the position of the coils be fixed, which is not ordinarily possible with table movement. With integrated PAT, the signal for the coil profiles is measured during each individual image acquisition by acquiring a few additional lines of k-space, which are termed reference lines. The calibration of the receiver coils is therefore intrinsic to integrated PAT, and this calibration process, which is automatically performed for each acquisition, accommodates table movement and allows repositioning of the subject between acquisitions. This procedure also offers the possibility of flexible use of many receiver coils and, therefore, extended anatomic coverage. In addition, the main directions for parallel imaging can be chosen with virtually any orientation of the acquisition plane.

The general advantages of parallel imaging are increased temporal and spatial resolution without an increase in acquisition time, or shorter acquisition times and therefore shorter breath-hold phases for the subject. Another important advantage is the reduction of blurring and geometric distortions with single-shot imaging sequences (eg, half-Fourier single-shot turbo spin echo), an improvement that results from substantial reductions in echo train length and echo time. A second effect of the echo time reduction is the decrease in signal decay in organs such as the lungs, which have a very short T2*. As reported previously, the generalized autocalibrating partially parallel acquisition algorithm based on automatic simultaneous acquisition of spatial harmonics is also less susceptible to aliasing artifacts from the margins of the field of view propagated in the center of the image than are techniques based on sensitivity encoding (13,14).

Whole-Body MR Imaging Protocol
MR imaging was performed with two different systems: initially (group 1, 42 subjects), with the system equipped with eight receiver channels; later (group 2, eight subjects), with the whole-body MR system equipped with 32 receiver channels. It was possible to use integrated PAT with both systems. All 50 individuals successfully completed the study. No adverse events occurred.

The details of the initial whole-body MR imaging protocol are listed in Table 1. At the beginning of each part of the examination (in the upper and the lower parts of the body), localizer images were acquired as benchmarks for the different examinations. The subject first underwent half-Fourier RARE imaging of the lungs with high spatial resolution, followed by imaging of cardiac function and perfusion. Next, high-spatial-resolution three-dimensional (3D) gadolinium-enhanced MR angiography of the carotid arteries and morphologic imaging of the brain were performed, followed by the application of a 3D gradient-echo sequence in the thorax to image the lungs after contrast agent injection. To enable detection of myocardial infarction, delayed contrast-enhanced images of the heart were acquired 13–17 minutes after administration of the contrast agent. For the first part of the examination, a special coil configuration was used that consisted of a head coil, several spinal coil elements, and four anteriorly and laterally positioned elements from two dual-element surface coils (Fig 1). This configuration allowed parallel imaging throughout the upper part of the body with a minimum of four receiver coils. During MR imaging of the thorax, carotid arteries, and brain, the subject was repositioned three times; repositioning was possible because of the built-in autocalibration of the coils before each acquisition. In the second part of the examination, the subject was repositioned feet first in the magnet, and a combination of 16 anterior receiver coils and eight posterior spinal receiver coils were used to cover the entire body from the chest to the feet. Again, all MR angiographic examinations were performed with integrated PAT, followed by T1-weighted FLASH and T2-weighted half-Fourier RARE imaging of the liver and kidneys with navigator-corrected parallel acquisitions. The respiratory-gating navigator was placed in the middle of the right diaphragm. This examination included only one postcontrast phase because the dynamic phase of contrast enhancement was primarily used for MR angiography.

View larger version (104K): [in this window] [in a new window] [Download PPT slide]   Figure 1a. (a) Diagrams show coil configuration for multistation parallel imaging with eight-channel MR imager, with upper (left: spinal array and large-field-of-view adapter, two body coils, and head coil) and lower (right: spinal array and large-field-of-view adapter, two body coils, and peripheral angiographic array) stations. (b) Diagram shows optimal coil configuration for whole-body imaging with MR system equipped with 32 independent receiver channels and with simultaneous connection of 76 coil array elements (matrix coils) for complete head-to-toe coverage. This coil configuration (total imaging matrix) allows full flexibility in whole-body parallel imaging. All matrix coils are designed for parallel imaging in three directions.

View larger version (59K): [in this window] [in a new window] [Download PPT slide]   Figure 1b. (a) Diagrams show coil configuration for multistation parallel imaging with eight-channel MR imager, with upper (left: spinal array and large-field-of-view adapter, two body coils, and head coil) and lower (right: spinal array and large-field-of-view adapter, two body coils, and peripheral angiographic array) stations. (b) Diagram shows optimal coil configuration for whole-body imaging with MR system equipped with 32 independent receiver channels and with simultaneous connection of 76 coil array elements (matrix coils) for complete head-to-toe coverage. This coil configuration (total imaging matrix) allows full flexibility in whole-body parallel imaging. All matrix coils are designed for parallel imaging in three directions.

View larger version (22K): [in this window] [in a new window] [Download PPT slide]   Figure 2. Diagram of echo train for each of three different applications of the half-Fourier RARE (HASTE) MR imaging sequence. Top: Echo time without integrated PAT or reference acquisition. Middle: With integrated PAT and integrated reference acquisition, echo time is substantially shorter. Bottom: With integrated PAT and external reference acquisition, echo time is shorter than that without integrated PAT and reference acquisition but slightly longer than that with integrated reference acquisition, and lung signal intensity is increased. rs = Reference acquisition.

View larger version (87K): [in this window] [in a new window] [Download PPT slide]   Figure 3. Real-time single-breath-hold short-axis cardiac MR images acquired with the generalized autocalibrating partially parallel acquisition algorithm (51.8/0.92 [repetition time msec/echo time msec]; section thickness, 8 mm; matrix, 128 x 128). With this technique, as many as 11 sections were acquired with temporal resolution of 48 msec and mean breath-hold time of 18 seconds. Images were obtained in left ventricular myocardium from base to apex for detection of wall motion abnormality and calculation of end-systolic and end-diastolic volume.

Contrast-enhanced MR Angiography with Parallel Imaging
All stations were imaged with a parallel imaging factor of two, which permitted an isotropic spatial resolution of less than 1.5 mm for images acquired at each station. For visualization of the carotid arteries during imaging of the upper body (as described previously), a separate test-bolus acquisition was performed after injection of gadopentetate dimeglumine at a dose of 0.1 mmol/kg and at a rate of 1.5 mL/sec. For MR angiography of the thorax, abdomen, and peripheral arteries, a biphasic injection protocol was used that began with the injection of 10 mL of contrast material at a rate of 1 mL/sec, followed by another 15 mL of contrast material injected at a rate of 0.5 mL/sec. The use of the whole-body MR imaging system enabled a reduction in the volume of contrast material because a monophasic injection could be used for imaging of both the abdominal aorta and blood vessels in the thigh.

The protocol for imaging of vessels in the foot depended on the height of the subject. In group 1 subjects, the total acquisition range was 140 cm below the diaphragm, and in group 2 subjects, it was 200 cm below the diaphragm. If the height of a subject exceeded this measurement, the vessels in the feet were not imaged.

All MR angiograms were evaluated by two fellows in the department of radiology (H.K., K.N.), who were blinded to each other's evaluation in terms of vessel conspicuity, venous overlay, and artifacts.

Delayed Contrast-enhanced Imaging with Phase-sensitive Inversion-Recovery Technique
A number of techniques other than parallel imaging were implemented to reduce the number of breath holds and the total acquisition time to a minimum. Phase-sensitive inversion recovery is a new technique that allows imaging of late enhancement of myocardial infarction without the time-consuming optimization of inversion time with scout imaging (20). High accuracy has been reported for detection of the extent of an infarcted area with delayed contrast-enhanced imaging, compared with that achieved with state-of-the-art turbo FLASH inversion-recovery sequences. In addition, this technique was found to tolerate suboptimal inversion times and consistently to depict the infarcted area at inversion times ranging from 200 to 800 msec (21). For this study, the technique was further optimized to allow single-shot multisection imaging and the acquisition of nine sections in a single breath hold. Therefore, the process for myocardial late enhancement imaging was reduced to a single breath-hold acquisition with gadopentetate dimeglumine at a dose of 0.15 mmol/kg (Table 2). Consequently, the duration of the entire cardiac examination was shortened to three breath holds.

Image Analysis
The quality of MR angiographic images was independently evaluated by two fellows in cardiovascular imaging (H.K., K.N.). A three-point scale was used to rate image quality in terms of vessel conspicuity (good, moderate, or poor) and in terms of the presence of artifacts and venous overlay (none, moderate, or major). The MR images from all subjects were independently evaluated by two radiologists (S.O.S. and A.H., each with more than 6 years of experience in cardiovascular MR imaging) for the depiction of pathologic entities. The radiologists were blinded to images from the reference examinations. Stenoses on 3D contrast-enhanced MR angiographic images were graded as follows: absent (0%), less than 30%, 30%–50%, greater than 50%, greater than 75%, greater than 90%, or occlusion.

Statistical Analysis
Interobserver variability with regard to image quality and pathologic findings was quantified with the statistic. A value of less than 0.20 denotes consistent disagreement; 0.20–0.40, poor agreement; 0.41–0.60, moderate agreement; 0.61–0.80, good agreement; and more than 0.80, excellent agreement. Intermodality agreement between MR imaging and reference examinations was carried out on a segment-by-segment basis, and, because of the small number of pathologic findings, specificity and sensitivity were not calculated. Differences between the two MR imagers with regard to total acquisition time were assessed by means of an unpaired Mann-Whitney U test. A P value of less than .05 was considered statistically significant. Statistical analysis was performed by using statistical software (SPSS 11.0.1; SPSS, Chicago, Ill).

	Results

TOP ABSTRACT INTRODUCTION Materials and Methods Results Discussion References

 
For subjects in group 1, the mean examination time was 102 minutes ± 23 (standard deviation). In 15 (33%) of the 42 subjects, the implementation of the entire protocol required less than 90 minutes. In only two cases did examination time exceed 135 minutes, because of technical problems with the multiple receiver coils (Fig 4). In group 1, 31 (73%) of 42 subjects reported that the total examination time was very tolerable, and the other 11 (27%) subjects reported that the examination time was too long but still tolerable. After implementation of the protocol with the whole-body imager (group 2), the total examination time was significantly reduced, to a mean of 79 minutes ± 10, compared with 102 minutes ± 23 (P < .05).

View larger version (20K): [in this window] [in a new window] [Download PPT slide]   Figure 4a. Graphs show total examination time with (a) standard clinical MR imaging system (mean, 102 minutes ± 22) and (b) whole-body MR imaging system (mean, 79 minutes ± 9). The whole-body imager enabled a decrease of about 25% in examination time because no subject repositioning was needed and new sequences allowed shorter acquisition times.

View larger version (15K): [in this window] [in a new window] [Download PPT slide]   Figure 4b. Graphs show total examination time with (a) standard clinical MR imaging system (mean, 102 minutes ± 22) and (b) whole-body MR imaging system (mean, 79 minutes ± 9). The whole-body imager enabled a decrease of about 25% in examination time because no subject repositioning was needed and new sequences allowed shorter acquisition times.

View larger version (101K): [in this window] [in a new window] [Download PPT slide]   Figure 5. Transverse and coronal T2-weighted half-Fourier RARE (404/25; section thickness, 6 mm; matrix, 320 x 320) MR images obtained with integrated PAT show lung nodule (arrows) with a diameter of 6 mm, confirmed at thin-section CT (not shown); nodule appeared larger at CT (diameter, 8 mm) because of partial calcification. Pulmonary lesions such as a nodule or an area of inflammation with a diameter of 6 mm or more were detectable with this technique.

View larger version (83K): [in this window] [in a new window] [Download PPT slide]   Figure 6a. (a) High-spatial-resolution integrated-PAT saturation-recovery true FISP short-axis cardiac MR image (175.3/1.03; section thickness, 8 mm; matrix, 320 x 320) and coronary angiogram show, respectively, perfusion defect and corresponding stenosis of left anterior descending coronary artery (open arrow). (b) Real-time true FISP short-axis cardiac MR images (same parameters as a) and (c) phase-sensitive inversion-recovery images (750/1.1; section thickness, 8 mm; matrix, 192 x 192) obtained with integrated PAT in a different patient show, respectively, regional wall motion abnormality and delayed enhancement, which indicate myocardial infarct (arrows in c) due to partial occlusion of the circumflex coronary artery. By using integrated PAT with multisection sequences, the complete cardiac MR examination was performed within three breath holds.

View larger version (55K): [in this window] [in a new window] [Download PPT slide]   Figure 6b. (a) High-spatial-resolution integrated-PAT saturation-recovery true FISP short-axis cardiac MR image (175.3/1.03; section thickness, 8 mm; matrix, 320 x 320) and coronary angiogram show, respectively, perfusion defect and corresponding stenosis of left anterior descending coronary artery (open arrow). (b) Real-time true FISP short-axis cardiac MR images (same parameters as a) and (c) phase-sensitive inversion-recovery images (750/1.1; section thickness, 8 mm; matrix, 192 x 192) obtained with integrated PAT in a different patient show, respectively, regional wall motion abnormality and delayed enhancement, which indicate myocardial infarct (arrows in c) due to partial occlusion of the circumflex coronary artery. By using integrated PAT with multisection sequences, the complete cardiac MR examination was performed within three breath holds.

View larger version (63K): [in this window] [in a new window] [Download PPT slide]   Figure 6c. (a) High-spatial-resolution integrated-PAT saturation-recovery true FISP short-axis cardiac MR image (175.3/1.03; section thickness, 8 mm; matrix, 320 x 320) and coronary angiogram show, respectively, perfusion defect and corresponding stenosis of left anterior descending coronary artery (open arrow). (b) Real-time true FISP short-axis cardiac MR images (same parameters as a) and (c) phase-sensitive inversion-recovery images (750/1.1; section thickness, 8 mm; matrix, 192 x 192) obtained with integrated PAT in a different patient show, respectively, regional wall motion abnormality and delayed enhancement, which indicate myocardial infarct (arrows in c) due to partial occlusion of the circumflex coronary artery. By using integrated PAT with multisection sequences, the complete cardiac MR examination was performed within three breath holds.

View larger version (35K): [in this window] [in a new window] [Download PPT slide]   Figure 7a. Coronal high-spatial-resolution contrast-enhanced whole-body MR angiographic images obtained with integrated PAT and 3D FLASH sequence (3.4/1.14; spatial resolution, <1.6 x 1.0 x 1.5 mm). (a, c) Images obtained with standard MR system, with five stations. (b) Images obtained with whole-body MR imager with coil matrix and four steps and with higher spatial resolution. Images in c show peripheral vascular disease and occlusion of the posterior tibial artery (solid arrows), as well as reconstitution of the plantar arch from the anterior tibial artery (open arrows). Note the excellent depiction, especially in c, of arteries in the distal calf and proximal foot, without venous overlay.

View larger version (45K): [in this window] [in a new window] [Download PPT slide]   Figure 7b. Coronal high-spatial-resolution contrast-enhanced whole-body MR angiographic images obtained with integrated PAT and 3D FLASH sequence (3.4/1.14; spatial resolution, <1.6 x 1.0 x 1.5 mm). (a, c) Images obtained with standard MR system, with five stations. (b) Images obtained with whole-body MR imager with coil matrix and four steps and with higher spatial resolution. Images in c show peripheral vascular disease and occlusion of the posterior tibial artery (solid arrows), as well as reconstitution of the plantar arch from the anterior tibial artery (open arrows). Note the excellent depiction, especially in c, of arteries in the distal calf and proximal foot, without venous overlay.

View larger version (34K): [in this window] [in a new window] [Download PPT slide]   Figure 7c. Coronal high-spatial-resolution contrast-enhanced whole-body MR angiographic images obtained with integrated PAT and 3D FLASH sequence (3.4/1.14; spatial resolution, <1.6 x 1.0 x 1.5 mm). (a, c) Images obtained with standard MR system, with five stations. (b) Images obtained with whole-body MR imager with coil matrix and four steps and with higher spatial resolution. Images in c show peripheral vascular disease and occlusion of the posterior tibial artery (solid arrows), as well as reconstitution of the plantar arch from the anterior tibial artery (open arrows). Note the excellent depiction, especially in c, of arteries in the distal calf and proximal foot, without venous overlay.

	Discussion

TOP ABSTRACT INTRODUCTION Materials and Methods Results Discussion References

The current approaches for whole-body MR imaging are aimed at resolving this conflict by means of dedicated rolling platforms that allow imaging of the different regions of the body with a series of consecutive steps or stations (2,3,6). Even with the use of fast imaging sequences, this approach is hampered by substantial limitations of spatial and temporal resolution. On the other hand, many newer clinical MR imaging protocols that allow the maximization of temporal and spatial resolution are just becoming established (23–25). Therefore, the substantial reduction in image quality casts a shadow of doubt on the true reliability of the approach to whole-body MR imaging that is generally used at present. The mean reconstructed voxel size was 4.5 mm³. While this spatial resolution may be acceptable for the detection of a severe arterial stenosis, it does not enable accurate stenosis grading. Current state-of-the art protocols for 3D gadolinium-enhanced MR angiography of the carotid and renal arteries require voxel sizes of 1 mm³ to identify the cutoff value of a 75% reduction in vessel diameter for hemodynamically significant stenosis (14,23,24). Thus, if a stenosis is identified at standard whole-body MR angiography, an additional examination with high-spatial-resolution 3D contrast-enhanced MR angiography is required.

Parallel imaging offers the possibility of substantial increases in temporal and spatial resolution with various MR imaging techniques (12,26). In this study, we showed that integrated PAT can be combined with MR angiography and assessment of cardiac function and perfusion, as well as with MR imaging of the lung. High-spatial-resolution MR angiography with use of integrated PAT allows whole-body angiographic imaging with voxel sizes of less than 1.5 mm³. In more than 85% of cases, no venous overlay occurred, and image quality was rated as good or better. These results were only possible by using a special configuration of different surface receiver coils, a matrix coil array that allowed parallel imaging in all vascular regions. Because of the built-in autocalibration of the coils, table movement between acquisitions was not a problem with parallel imaging (27). In the study of 42 subjects, one major limitation occurred: the necessity of repositioning the subject after imaging in the upper body. Whole-body MR imaging could not be accomplished in one step with the standard clinical MR imaging system. With implementation of the protocol by using the whole-body MR imager, MR angiography could be performed with parallel imaging in the entire body, without repositioning the subject. With this system, a new approach for high-resolution whole-body 3D contrast-enhanced MR angiography became feasible: First, the availability of 32 channels with up to 76 connected matrix coil elements enabled the use of higher integrated PAT acceleration factors for MR angiography and, thus, reduction of the acquisition time by more than one-half, without compromising spatial resolution. Second, the improved geometry of the matrix coil system helped to increase the signal-to-noise ratio, which is important for integrated PAT applications because the use of integrated PAT alone leads to a reduction in the signal-to-noise ratio. Third, the range of table movement (2 m) allows the imaging of different vessel territories in a deliberate order. Therefore, we decided to perform a biphasic high-resolution acquisition in the lower leg vessels directly after imaging in the carotid arteries; this allowed a sufficient time window to ensure pure arterial contrast. It also allowed us to overcome the limitations of the initial protocol as evidenced on images obtained in group 1, on which venous overlay was present in a small percentage of cases. Additional injections for a separate acquisition of vessels in the lower part of the leg after an initial moving-table acquisition with venous overlay can thereby be completely avoided. Thus, the total dose of contrast medium can be kept at a minimum. In addition, artifacts from coil inhomogeneities at the border between array coils were completely eliminated on images obtained in group 2 subjects. These artifacts occurred mainly on images of the carotid arteries in group 1 subjects.

We adhered to the protocol of two separate injections of the contrast agent, one for imaging of the carotid arteries and the other for imaging in the thorax and abdomen, since bidirectional blood flow between the carotid arteries and the descending aorta almost inherently results in venous overlay on the renal artery when successive sections of the carotid arteries and the thoracic and abdominal aorta are acquired with a single injection of the contrast agent.

In addition to the advantages of integrated PAT for high-resolution contrast-enhanced MR angiography, valuable effects have been reported for the application of integrated PAT for single-shot MR imaging with sequences such as half-Fourier RARE. In this study, half-Fourier RARE imaging was used for multiplanar assessment of the lungs with submillimeter in-plane spatial resolution and a relatively thin section thickness of 6 mm. We were able to detect pulmonary nodules at a size of more than 8 mm, and all of these were confirmed with thin-section CT. Our results are in agreement with previous data reported by Lutterbey et al (28), who detected pulmonary nodules with a diameter of more than 3 mm by using ultrashort T2-weighted spin-echo sequences. In that study, however, the use of turbo spin-echo techniques with respiratory gating resulted in acquisition times of several minutes. In our study, the combination of half-Fourier RARE and integrated PAT allowed short acquisition times for multiplanar acquisitions of the entire thorax in only six breath holds. The use of MR imaging for detection of lung cancer is still a matter of controversy. If MR imaging is ever used for that indication, the purpose would be to enable the detection of small intrapulmonary lesions. In a group of 426 patients, Ginsberg et al (29), without foreknowledge of underlying malignancy, showed that pulmonary lesions with a diameter of 5–10 mm were malignant in 30% of cases. This finding is underlined by the results reported by Henschke et al (30), who clearly showed a tumor size–dependent cure rate even for stage IA cancers.

At cardiac imaging, the use of integrated PAT in combination with steady-state free precession and phase-sensitive imaging allowed a reduction of the examination time to three breath holds. Single-breath-hold single-shot real-time electrocardiographically gated true FISP imaging has proved reliable for measurement of global and regional cardiac function. The postprocessing time for determination of functional parameters with this sequence, however, was substantially longer than with the other sequence because of poor contrast between the intraventricular cavity and the left ventricular wall. Perfusion imaging of the heart was possible in four sections with high spatial resolution. Only high-grade stenosis could be detected with perfusion imaging at rest, however; perfusion defects were found in one case and were verified with conventional angiography of the coronary arteries. By using multisection phase-sensitive true FISP imaging for detection of delayed contrast enhancement, myocardial infarction also could be visualized within a single breath hold. Preliminary data have been reported for the highly accurate depiction of myocardial infarction with this technique, independent of optimization of inversion time (20). Steady-state free precession techniques such as true FISP imaging are particularly suited for use in combination with parallel imaging techniques, since they result in signal intensity about 10 times that achieved with standard gradient-echo sequences, thus counterbalancing the decrease in the signal-to-noise ratio that accompanies the higher acceleration factors with integrated PAT.

We found a good interobserver agreement for assessment of the entire vascular system and detection of pathologic entities, with values higher than 0.60. The agreement among observers for whole-body MR imaging was good because clinically established protocols were used to evaluate vascular territories and particular organ systems. These protocols were only adapted to optimize acquisition times and sequences to achieve whole-body imaging in less than 90 minutes. The consistent achievement of high temporal and spatial resolution with integrated PAT also helps to explain the good agreement between whole-body MR imaging and the available reference examinations. In cases of pathologic findings, the results of whole-body MR imaging agreed fully with those of subsequent reference examinations. In this screening scenario, a low prevalence of pathologic findings is expected; additional reference studies are acceptable only with suspicious findings on MR images. It is a strength of this study that all subjects were referred from a health care program and had received yearly evaluations according to screening standards for cardiovascular and malignant diseases. However, a diagnostic work-up with MR imaging is not entirely comparable with one performed with reference techniques. Despite this limitation, a high intermodality agreement was found for the evaluation of the carotid arteries, the heart, the lungs, and the abdominal organs. Because of the low prevalence of disease, sensitivity and specificity values could not be calculated, but a comparison between findings at reference examinations and those at MR imaging showed good correlation. In some cases, MR imaging provided more detailed information about pathologic entities.

The initial implementation of the imaging protocol by using the standard cardiovascular imager with parallel imaging was time consuming. Although the examination time was well tolerated by most subjects, it was too long to allow routine application. With use of the new whole-body MR imager, the total examination time was significantly reduced and superior image quality was achievable. The reduction in examination time was possible because of higher acceleration factors and a large range of table movement, which obviated subject repositioning between the first and second parts of the examination. A second contributing factor is the coil design (the use of matrix coils). Because of this whole-body matrix coil system, which provides complete anatomic coverage, no subject or coil repositioning was necessary during the examination.

Until now, we have not included colon cancer screening in our protocol. It is well acknowledged that colonic polyps are an important target of screening because of their high prevalence and the resectability of adenomatous polyps prior to malignant transformation. The use of the reported MR imaging techniques for such screening, however, is controversial (31,32). The new whole-body imager nevertheless holds the potential for overcoming the limitations of spatial resolution with the use of higher integrated PAT acceleration factors for fast gradient-echo imaging of the colon.

With regard to the limitations of our study, it should be noted that the data are based on findings in a small study sample. In addition, no systematic correlation between MR images and CT images was performed, and therefore we do not know the true sensitivity of MR imaging. False-negative findings may occur.

In this feasibility study, perfusion imaging during pharmacologically induced stress with adenosine was not performed. With the use of a whole-body imaging protocol and a whole-body MR imager, and with the consequent substantial reduction in total acquisition time, this test will be possible in the future for patients with risk factors for coronary artery disease (33).

Long acquisition times and low temporal and spatial resolution are factors that previously limited the use of MR imaging for whole-body screening. New acquisition techniques such as integrated PAT and multisection sequences, together with newly developed hardware, can help to solve these problems. It is now possible to achieve a comprehensive examination that includes whole-body MR angiography and a complete cardiac examination (functional imaging, perfusion imaging, delayed contrast-enhanced imaging, and MR imaging of the lungs, brain, and abdominal organs) without compromising temporal or spatial resolution and within a reasonable acquisition time.

	FOOTNOTES

 

Abbreviations: FISP = fast imaging with steady-state precession • FLASH = fast low-angle shot • PAT = parallel acquisition techniques • RARE = rapid acquisition with relaxation enhancement • 3D = three-dimensional

See Materials and Methods for pertinent disclosures.

Author contributions: Guarantors of integrity of entire study, H.K., S.O.S.; study concepts, H.K., S.O.S., K.N., O.D., B.K.; study design, H.K., S.O.S.; literature research, H.K., S.O.S.; clinical studies, H.K., E.W.; data acquisition, H.K., S.O.S., A.S., E.W.; data analysis/interpretation, H.K., S.O.S., K.N., A.H., B.J.W.; statistical analysis, H.K., K.N., S.O.S., A.H.; manuscript preparation, H.K., K.N., O.D.; manuscript definition of intellectual content, M.F.R.; manuscript editing, S.O.S., M.F.R.; manuscript revision/review, S.O.S., K.N., A.H., A.S., E.W., B.J.W., O.D., B.K., M.F.R.; manuscript final version approval, M.F.R., S.O.S., K.N., A.H., B.J.W.

	References

TOP ABSTRACT INTRODUCTION Materials and Methods Results Discussion References

 

1. Sans S, Kesteloot H, Kromhout D. The burden of cardiovascular diseases mortality in Europe. Task Force of the European Society of Cardiology on Cardiovascular Mortality and Morbidity Statistics in Europe. Eur Heart J 1997; 18:1231–1248.
2. Goyen M, Herborn CU, Kroger K, Lauenstein TC, Debatin JF, Ruehm SG. Detection of atherosclerosis: systemic imaging for systemic disease with whole-body three-dimensional MR angiography—initial experience. Radiology 2003; 227:277–282.[Abstract/Free Full Text]
3. Goyen M, Quick HH, Debatin JF, et al. Whole-body three-dimensional MR angiography with a rolling table platform: initial clinical experience. Radiology 2002; 224:270–277.[Abstract/Free Full Text]
4. Tan KT, van Beek EJ, Brown PW, van Delden OM, Tijssen J, Ramsay LE. Magnetic resonance angiography for the diagnosis of renal artery stenosis: a meta-analysis. Clin Radiol 2002; 57:617–624.[CrossRef][Medline]
5. Vasbinder GB, Nelemans PJ, Kessels AG, Kroon AA, de Leeuw PW, van Engelshoven JM. Diagnostic tests for renal artery stenosis in patients suspected of having renovascular hypertension: a meta-analysis. Ann Intern Med 2001; 135:401–411.[Abstract/Free Full Text]
6. Ruehm SG, Goyen M, Barkhausen J, et al. Rapid magnetic resonance angiography for detection of atherosclerosis. Lancet 2001; 357:1086–1091.[CrossRef][Medline]
7. Vasbinder GB, Maki JH, Nijenhuis RJ, et al. Motion of the distal renal artery during three-dimensional contrast-enhanced breath-hold MRA. J Magn Reson Imaging 2002; 16:685–696.[CrossRef][Medline]
8. Vasbinder GBC, De Haan MW, van Engelshoven JM. Accuracy of CTA and 3D contrast-enhanced MRA as compared to intra-arterial digital subtraction angiography for assessment of the number of renal arteries in 356 subjects (abstr). Radiology 2002; 225(P):400.[Abstract/Free Full Text]
9. Vasbinder GBC. Pitfalls and limitations of contrast enhanced renal magnetic resonance angiography for the detection of atherosclerotic renal artery stenosis. Presented at the 25th international meeting of the MR Angio Club, Dublin, Ireland, September 24–26, 2003.
10. Vasbinder GB, Nelemans PJ, Kessels AG, et al. Accuracy of computed tomographic angiography and magnetic resonance angiography for diagnosing renal artery stenosis. Ann Intern Med 2004;141:674–682.[Abstract/Free Full Text]
11. Grant EG, Benson CB, Moneta GL, et al. Carotid artery stenosis: grayscale and Doppler ultrasound diagnosis—Society of Radiologists in Ultrasound consensus conference. Ultrasound Q 2003; 19:190–198.[Medline]
12. Weiger M, Pruessmann KP, Kassner A, et al. Contrast-enhanced 3D MRA using SENSE. J Magn Reson Imaging 2000; 12:671–677.[CrossRef][Medline]
13. Griswold MA, Jakob PM, Heidemann RM, et al. Generalized autocalibrating partially parallel acquisitions (GRAPPA). Magn Reson Med 2002; 47:1202–1210.[CrossRef][Medline]
14. Schoenberg SO, Rieger J, Weber CH, et al. High-spatial-resolution MR angiography of renal arteries with integrated parallel acquisitions: comparison with digital subtraction angiography and US. Radiology 2005; 235(2):687–698.[Abstract/Free Full Text]
15. Biederer J, Busse I, Grimm J, et al. Sensitivity of MRI in detecting alveolar infiltrates: experimental studies [in German]. Rofo Fortschr Geb Rontgenstr Neuen Bildgeb Verfahr 2002; 174:1033–1039.[Medline]
16. Barkhausen J, Goyen M, Ruhm SG, Eggebrecht H, Debatin JF, Ladd ME. Assessment of ventricular function with single breath-hold real-time steady-state free precession cine MR imaging. AJR Am J Roentgenol 2002; 178:731–735.[Abstract/Free Full Text]
17. Kramer U, Fenchel M, Helber U, et al. Multislice TrueFISP-MR imaging for identifying stress-induced myocardial functional disturbances in coronary heart disease [in German]. Rofo Fortschr Geb Rontgenstr Neuen Bildgeb Verfahr 2003; 175:1355–1362.[Medline]
18. Wintersperger BJ, Nikolaou K, Dietrich O, et al. Single breath-hold real-time cine MR imaging: improved temporal resolution using generalized autocalibrating partially parallel acquisition (GRAPPA) algorithm. Eur Radiol 2003; 13:1931–1936.[CrossRef][Medline]
19. Schreiber WG, Schmitt M, Kalden P, Mohrs OK, Kreitner KF, Thelen M. Dynamic contrast-enhanced myocardial perfusion imaging using saturation-prepared TrueFISP. J Magn Reson Imaging 2002; 16:641–652.[CrossRef][Medline]
20. Kellman P, Arai AE, McVeigh ER, Aletras AH. Phase-sensitive inversion recovery for detecting myocardial infarction using gadolinium-delayed hyperenhancement. Magn Reson Med 2002; 47:372–383.[CrossRef][Medline]
21. Huber A, Hayes C, Wintersperger BJ, Franz W, Schoenberg SO, Reiser MF. Value of phase-sensitive inversion recovery for detection of myocardial infarction (abstr). In: Radiological Society of North America scientific assembly and annual meeting program. Oak Brook, Ill: Radiological Society of North America, 2003; 592.
22. Henschke CI, Yankelevitz DF. Screening for lung cancer. J Thorac Imaging 2000; 15:21–27.[CrossRef][Medline]
23. Fain SB, King BF, Breen JF, Kruger DG, Riederer SJ. High-spatial-resolution contrast-enhanced MR angiography of the renal arteries: a prospective comparison with digital subtraction angiography. Radiology 2001; 218:481–490.[Abstract/Free Full Text]
24. 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]
25. 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]
26. Jakob PM, Griswold MA, Edelman RR, Manning WJ, Sodickson DK. Accelerated cardiac imaging using the SMASH technique. J Cardiovasc Magn Reson 1999; 1:153–157.[Medline]
27. Jakob PM, Griswold MA, Edelman RR, Sodickson DK. AUTO-SMASH: a self-calibrating technique for SMASH imaging—simultaneous acquisition of spatial harmonics. MAGMA 1998; 7:42–54.
28. Lutterbey G, Leutner C, Gieseke J, et al. Detection of focal lung lesions with magnetic resonance tomography using T2-weighted ultrashort turbo-spin-echo-sequence in comparison with spiral computerized tomography [in German]. Rofo Fortschr Geb Rontgenstr Neuen Bildgeb Verfahr 1998; 169:365–369.[Medline]
29. Ginsberg MS, Griff SK, Go BD, Yoo HH, Schwartz LH, Panicek DM. Pulmonary nodules resected at video-assisted thoracoscopic surgery: etiology in 426 patients. Radiology 1999; 213:277–282.[Abstract/Free Full Text]
30. Henschke CI, Wisnivesky JP, Yankelevitz DF, Miettinen OS. Small stage I cancers of the lung: genuineness and curability. Lung Cancer 2003; 39:327–330.[CrossRef][Medline]
31. Lauenstein TC, Debatin JF. Magnetic resonance colonography with fecal tagging: an innovative approach without bowel cleansing. Top Magn Reson Imaging 2002; 13:435–444.[CrossRef][Medline]
32. Lauenstein TC, Goehde SC, Debatin JF. Fecal tagging: MR colonography without colonic cleansing. Abdom Imaging 2002; 27:410–417.[CrossRef][Medline]
33. Gould KL, Lipscomb K, Hamilton GW. Physiologic basis for assessing critical coronary stenosis: instantaneous flow response and regional distribution during coronary hyperemia as measures of coronary flow reserve. Am J Cardiol 1974; 33:87–94.[CrossRef][Medline]

This article has been cited by other articles:

				M. Fenchel, J. Doering, A. Seeger, U. Kramer, K. Rittig, B. Klumpp, C. D. Claussen, and S. Miller Ultrafast Whole-Body MR Angiography with Two-dimensional Parallel Imaging at 3.0 T: Feasibility Study Radiology, November 10, 2008; (2008) 2501080494. [Abstract] [Full Text]

				K. Nael, M. Fenchel, M. Krishnam, G. Laub, J. P. Finn, and S. G. Ruehm High-Spatial-Resolution Whole-Body MR Angiography with High-Acceleration Parallel Acquisition and 32-Channel 3.0-T Unit: Initial Experience Radiology, March 1, 2007; 242(3): 865 - 872. [Abstract] [Full Text] [PDF]

				T. Hansen, J. Wikstrom, L.O. Johansson, L. Lind, and H. Ahlstrom The Prevalence and Quantification of Atherosclerosis in an Elderly Population Assessed by Whole-Body Magnetic Resonance Angiography Arterioscler. Thromb. Vasc. Biol., March 1, 2007; 27(3): 649 - 654. [Abstract] [Full Text] [PDF]

				K. Nael, S. G. Ruehm, H. J. Michaely, R. Saleh, M. Lee, G. Laub, and J. P. Finn Multistation Whole-Body High-Spatial-Resolution MR Angiography Using a 32-Channel MR System Am. J. Roentgenol., February 1, 2007; 188(2): 529 - 539. [Abstract] [Full Text] [PDF]

K. Nikolaou, H. Kramer, C. Grosse, D. Clevert, O. Dietrich, M. Hartmann, P. Chamberlin, S. Assmann, M. F. Reiser, and S. O. Schoenberg High-Spatial-Resolution Multistation MR Angiography with Parallel Imaging and Blood Pool Contrast Agent: Initial Experience Radiology, December 1, 2006; 241(3): 861 - 872. [Abstract]