(Radiology. 2000;216:298-303.)
© RSNA, 2000
Partial Fat-saturated Contrast-enhanced Three-dimensional MR Angiography Compared with Non-Fat-saturated and Conventional Fat-saturated MR Angiography1
Paul R. Hilfiker, MD,
Robert J. Herfkens, MD,
Steve G. Heiss, MD,
Marcus T. Alley, PhD,
Dominik Fleischmann, MD and
Norbert J. Pelc, ScD
1 From the Department of Radiology, Richard M. Lucas Center for MR Spectroscopy and Imaging, Stanford University, Stanford, Calif. From the 1999 RSNA scientific assembly. Received August 17, 1999; revision requested October 7; revision received November 5; accepted November 15. P.R.H. supported in part by a Swiss National Science Foundation Fellowship, the Holderbank Foundation, and the Novartis Foundation. M.T.A. and N.J.P. supported in part by National Institutes of Health grant P41 RR09784. D.F. supported in part by an Austrian Science Foundation Fellowship. Address correspondence to P.R.H., Institute of Diagnostic Radiology, University Hospital Zurich, Raemistr 100, CH-8091 Zurich, Switzerland (e-mail: paul.hilfiker@dmr.usz.ch).
 |
ABSTRACT
|
|---|
Abdominal three-dimensional magnetic resonance angiography was performed in 35 patients in the equilibrium phase without fat saturation, with conventional fat saturation, and with fast partial fat saturation. Qualitative and quantitative evaluation demonstrated significantly better vessel visualization with both fat-saturated techniques. The partial fat-saturated technique provided water-specific images within a breath hold, reducing motion artifacts significantly.
Index terms: Magnetic Resonance (MR), contrast enhancement, 9*.129422 • Magnetic Resonance (MR), fat suppression, 9*.129422 • Magnetic Resonance (MR), pulse sequences, 9*.129422 • Magnetic Resonance (MR), vascular studies, 9*.129422
|
INTRODUCTION
|
|---|
The application of contrast agents for magnetic resonance (MR) angiography has proved useful for obtaining high-contrast MR angiograms in relatively short examination times by using fast gradient-echo techniques (1). However, gadolinium-enhanced MR angiograms usually contain residual signal intensity in the tissues adjacent to the vessels. Fatty tissue with its short T1 can have unfavorably high signal intensity on the images and can obscure the visibility of vascular structures with low enhancement. One strategy to improve image contrast is the use of subtraction techniques (2–5). However, there must be no change in the patient position between the nonenhanced and dynamic gadolinium-enhanced imaging. This requirement is easily met in the lower extremities, whereas it is more difficult to achieve in the abdomen, because of effects of respiration and peristalsis. Another approach is the use of a fat-saturated technique. However, the conventional fat-saturated technique results in a considerable increase in imaging time and is therefore impracticable for abdominal fast three-dimensional (3D) MR angiography (6).
Our approach has been to incorporate a spectrally selective suppression pulse into a fast 3D contrast material–enhanced MR angiographic sequence. This concept of using fat suppression in an acquisition with variable repetition time was developed previously for time-of-flight intracranial imaging (7). The increase in imaging time is minimized by using the fat-suppression pulse in a limited number of repetition times (partial fat saturation). However, the increasing acceptance of contrast-enhanced 3D MR angiography offers new opportunities for optimizing fat suppression in fast MR imaging techniques.
The purpose of this study was to evaluate a partial fat-saturated technique designed for fast 3D contrast-enhanced MR angiography and to compare it with a nonfat-saturated technique and to a fat-saturated technique in abdominal 3D MR angiography.
|
Materials and Methods
|
|---|
Patient Population
Thirty-seven MR studies were acquired in 35 consecutive patients (18 male and 17 female patients; age range, 7 months to 88 years; mean age, 40.5 years). All patients were referred to our institution for abdominal MR imaging that included 3D MR angiography of the abdominal vessels. Patients were recruited into the study between January and June 1999. Patients were included only if they were willing to participate and if the examination time slot allowed the additional time necessary to conduct imaging with the three sequences. The study was approved by the institutional review board, and written informed consent was obtained from all patients or their guardians prior to the procedure.
Imaging Technique
All MR imaging was performed on a 1.5-T MR imager (Signa EchoSpeed; GE Medical Systems, Milwaukee, Wis) equipped with a gradient system capable of an amplitude of 22 mT/m and a slew rate of 120 mT/m/msec. The body coil was used for radio-frequency excitation and reception. For the clinically indicated study, a multiphase 3D MR angiographic data set was acquired after administration of 0.2 mmol per kilogram of body weight of gadopentetate dimeglumine (Magnevist; Berlex Laboratories, Wayne, NJ) (8). For the purposes of the present study, three additional data acquisitions were performed in the equilibrium phase. The order of acquisition was intentionally varied by directing the MR technologist to change the order of the sequences after each patient. To document equal distribution, we timed the delay between contrast material administration and image acquisition for each sequence.
The three additional data sets consisted of 3D fast spoiled gradient-recalled acquisitions without fat saturation, an acquisition with conventional chemical fat saturation in every sequence repetition, and an acquisition with partial fat saturation. All imaging was performed in coronal sections. For the nonfat-saturated and partial fat-saturated techniques, the 3D angiographic data were acquired in apnea (end inspiration). For the full fat-saturated technique (hereafter, fat-saturated), imaging was performed during shallow breathing. Section thickness (range, 2.0–3.5 mm), field of view (range, 27–48 cm), and number of images (range, 10–32 images) were optimized depending on patient size and remained the same throughout the three sequences performed with each patient. For all three sequences, interpolation in the section-select direction was performed with zero filling to produce 20–64 reconstructed sections.
The 3D fast spoiled gradient-recalled acquisition with partial fat-saturated technique was recently described by Alley et al (9). Briefly, a particular phase-encoding scheme is used in the technique in conjunction with infrequent application of a dual-band excitation pulse to produce effective fat suppression. The in-plane and section-direction phase encodes are initially arranged in an elliptic centric order (10). The resultant table is then divided into N equal segments. During sequence execution, the phase-encode values for consecutive repetition times are taken from consecutive segments. In this fashion, every Nth repetition time consists of a phase-encode pair from the most central portion of k space. Partial fat saturation is performed by inserting a spectrally selective pulse into the sequence repetitions that contain phase-encode pairs from the most central k-space segments. Therefore, sequence repetition with the frequency-selective pulse inserted has a longer repetition time than do the others. Disruption of the steady state caused by this additional pulse would normally cause the water in subsequent sequence repetitions to have higher than normal signal. Combined with the centric view ordering, higher signal-to-noise ratio (SNR) but loss of edge definition would be expected. The steady state was modeled numerically by assuming a T1 of 100 msec for enhanced blood (1).
On the basis of this model, the longitudinal steady-state condition can be maintained by incorporating a small water excitation into the suppression pulse. A Parks-McClellan finite-impulse-response filter is used to design a radio-frequency pulse that produces a 180° flip angle for fat (11) and a 33° flip angle for water. In the present study, fat suppression with the partial fat-saturated technique was performed with this dual-band pulse every N = 8 repetitions, increasing the execution time of those repetitions by 7.9 msec. For imaging with the fat-saturated technique, a conventional fat-saturated pulse and crusher gradients were applied in every repetition (12,13). On the basis of a sampling bandwidth of plus or minus 62.5 kHz, the following parameters were used: repetition time msec/echo time msec of 4.6/1 in the non–fat-saturated sequence or 13.9/1 in the fat-saturated sequence, flip angle of 25°, one signal acquired, and matrix of 512 = 192. In the partial fat-saturated sequence, 4.9/1 or 12.8/1 were used for every N = 8 repetitions. The acquisition times—which varied between patients depending on the section thickness, number of sections, and field of view—were recorded.
Image Analysis
For purposes of analysis, the vascular morphology was divided into the following segments: aorta (suprarenal or infrarenal), inferior vena cava (suprarenal or infrarenal), common iliac arteries and veins, and portal vein. Signal intensity was measured on the source images by one radiologist (P.R.H.) who defined regions of interest within these segments, in the retroperitoneal fat, and in the SD of noise outside the patient. Within the aorta and inferior vena cava, the regions of interest were placed 2 cm proximal to and 2 cm distal from the renal arteries or veins, respectively. Within the common iliac arteries and veins, the regions of interest were placed 1 cm distal from the iliac bifurcation, and within the portal vein, 1 cm proximal to the portal bifurcation. The size of the regions of interest was adapted to the size of the vessel lumen and remained unchanged for the same patient in all three image sets (14). All measurements were performed on a workstation (Advantage Windows; GE Medical). Relative SNR and contrast-to-noise ratio (CNR) for each of these segments were calculated for the three techniques as follows: SNR = SIvessel/SDnoise; CNR = (SIvessel – SIfat)/SDnoise, where SI is signal intensity.
All three MR image sets were independently evaluated by two radiologists (S.G.H., R.J.H.) on a workstation (Cemax-Icon, AUTORAD version 2.0; Cemax, Fremont, Calif). The following segments—aorta (suprarenal or infrarenal), inferior vena cava (suprarenal or infrarenal), common iliac arteries and veins, celiac and superior mesenteric arteries (origins and first degree branches), renal arteries and veins, portal and superior mesenteric veins—were rated for confidence of visualization on a five-point scale: score of 1, unsatisfactory; score of 2, fair; score of 3, average; score of 4, good; score of 5, excellent. In addition, the two readers rated edge sharpness, immunity from motion artifacts, and overall diagnostic image quality on a five-point scale for each of the three studies: score of 1, unsatisfactory; score of 2, fair; score of 3, average; score of 4, good; score of 5, excellent. Note that motion artifacts were rated on the basis of immunity from motion artifacts, to allow use of the same grading system for all qualitative evaluations: score of 1, unsatisfactory to score of 5, excellent.
The SNRs and CNRs in each patient were compared with a paired two-tailed Student t test. Qualitative data were compared with a Wilcoxon matched pairs signed rank sum test. In all cases, differences were considered statistically significant if a P value less than .05 was calculated.
|
Results
|
|---|
All patients tolerated the MR examination well and were able to suspend respiration to permit breath-hold acquisition of the 3D data sets without fat saturation and with partial fat saturation. Imaging times were 14–30 seconds (mean, 25 seconds) for non–fat-saturated, 19–39 seconds (mean, 32 seconds) for partial fat-saturated, and 41–88 seconds (mean, 74 seconds) for fat-saturated acquisitions. The delays between contrast material administration and acquisition were similar: non–fat-saturated images, 565 seconds = 428 (mean = SD); partial fat-saturated images, 580 seconds = 434; fat-saturated images, 574 seconds = 435.
The overall mean SNR in the various segments revealed a slightly higher signal intensity on the non–fat-saturated images (16.87 = 6.49) compared with that on the partial fat-saturated images (15.74 = 5.42), but the difference was not statistically significant (P > .05). The difference was probably a result of the T1 of enhanced blood being longer than 100 msec (as used in this model). Mean SNR was significantly higher (P < .001) for fat-saturated images (26.88 = 7.73) compared with either non– or partial fat-saturated images owing to the longer repetition time in the fat-saturated sequence. Analysis of the individual SNR values in the vessels revealed the same statistically significant results and is summarized in Figure 1.
View larger version (25K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 1. Graph shows SNR values (y axis) for individual vessels for acquisitions with non-fat-saturated (black bars), partial fat-saturated (white bars), and fat-saturated (gray bars) techniques. SNR values for the non-fat-saturated and partial fat-saturated techniques are similar (P < .05), but those for the fat-saturated technique are significantly higher for all vessels (P < .001). Error bars indicate SDs. IAO = infrarenal aorta, IIVC = infrarenal inferior vena cava, LIA = left common iliac artery, LIV = left common iliac vein, MPV = main portal vein, RIA = right common iliac artery, RIV = right common iliac vein, SAO = suprarenal aorta, SIVC = suprarenal inferior vena cava.
|
|
Mean CNR values between vessels and retroperitoneal fat were significantly higher (P < .001) for both fat-saturated techniques compared with the nonfat-saturated technique: nonfat saturated, 5.96 = 4.90; partial fat saturated, 11.66 = 5.42; and fat saturated, 20.05 = 6.67. Differences in CNR between the fat-saturated and partial fat-saturated techniques can be explained on the basis of the difference in vascular SNR, indicating that the techniques were comparable with respect to fat suppression. SNR values for fat were the following: nonfat saturated, 12.04 = 3.49; partial fat saturated, 4.07 = 1.13; and fat saturated, 6.63 = 2.57. CNR values for the vessel segments are summarized in Figure 2.
View larger version (22K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2. Graph shows CNR values (y axis) for individual vessels with the non-fat-saturated (black bars), partial fat-saturated (white bars), and fat-saturated (gray bars) techniques. CNR values with either fat-saturated technique are improved significantly (P < .001). Error bars indicate SDs. IAO = infrarenal aorta, IIVC = infrarenal inferior vena cava, LIA = left common iliac artery, LIV = left common iliac vein, MPV = main portal vein, RIA = right common iliac artery, RIV = right common iliac vein, SAO = suprarenal aorta, SIVC = suprarenal inferior vena cava.
|
|
Qualitative evaluations by the two readers (A and B) are summarized in Figure 3. Both readers rated vessel visualization significantly higher (P < .001) with either fat-saturated technique compared with the non–fat-saturated technique. Both readers rated immunity from motion artifacts significantly lower (P < .001) with the fat-saturated technique compared with the other two techniques but as not significantly different (P > .05) between the partial fat-saturated and non–fat-saturated techniques (Figs 4, 5). Both readers rated edge sharpness as significantly decreased (P < .01) with either fat-saturated technique compared with the non–fat-saturated technique. With the fat-saturated technique, this decrease could have been due to respiratory motion. Both readers rated overall diagnostic image quality with either fat-saturated technique as significantly superior (P < .01) to that with the non–fat-saturated technique. They found no difference between the two fat-saturated techniques (Fig 3).
View larger version (19K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3a. Graphs depict qualitative assessment for readers (a) A and (b) B with non-fat-saturated (black bars), partial fat-saturated (white bars), and fat-saturated (gray bars) techniques. On the y axis, the grading system was score of 1, unsatisfactory; score of 2, fair; score of 3, average; score of 4, good; score of 5, excellent.
|
|
View larger version (20K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3b. Graphs depict qualitative assessment for readers (a) A and (b) B with non-fat-saturated (black bars), partial fat-saturated (white bars), and fat-saturated (gray bars) techniques. On the y axis, the grading system was score of 1, unsatisfactory; score of 2, fair; score of 3, average; score of 4, good; score of 5, excellent.
|
|
View larger version (124K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 4a. Coronal source MR images from three sequential delayed-phase 3D MR angiograms (40 = 40-cm field of view, 2.3-mm section thickness, 64 sections) in a 37-year-old patient: (a) non-fat saturated (4.6/1, 29-second acquisition time), (b) fat saturated (13.9/1, 86-second acquisition time), and (c) partial fat saturated (4.9/1 or 12.8/1 for every N = 8 repetitions, respectively; 37-second acquisition time). With no fat saturation, the aorta (solid arrow) and inferior vena cava (open arrow) are iso- to hypointense with the surrounding fat. Both fat-saturated techniques improve vessel conspicuity. Motion artifacts seen in the liver in b (arrowhead) are not present in a or c.
|
|
View larger version (125K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 4b. Coronal source MR images from three sequential delayed-phase 3D MR angiograms (40 = 40-cm field of view, 2.3-mm section thickness, 64 sections) in a 37-year-old patient: (a) non-fat saturated (4.6/1, 29-second acquisition time), (b) fat saturated (13.9/1, 86-second acquisition time), and (c) partial fat saturated (4.9/1 or 12.8/1 for every N = 8 repetitions, respectively; 37-second acquisition time). With no fat saturation, the aorta (solid arrow) and inferior vena cava (open arrow) are iso- to hypointense with the surrounding fat. Both fat-saturated techniques improve vessel conspicuity. Motion artifacts seen in the liver in b (arrowhead) are not present in a or c.
|
|
View larger version (127K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 4c. Coronal source MR images from three sequential delayed-phase 3D MR angiograms (40 = 40-cm field of view, 2.3-mm section thickness, 64 sections) in a 37-year-old patient: (a) non-fat saturated (4.6/1, 29-second acquisition time), (b) fat saturated (13.9/1, 86-second acquisition time), and (c) partial fat saturated (4.9/1 or 12.8/1 for every N = 8 repetitions, respectively; 37-second acquisition time). With no fat saturation, the aorta (solid arrow) and inferior vena cava (open arrow) are iso- to hypointense with the surrounding fat. Both fat-saturated techniques improve vessel conspicuity. Motion artifacts seen in the liver in b (arrowhead) are not present in a or c.
|
|
View larger version (129K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 5a. Targeted maximum-intensity-projection images from three sequential delayed-phase 3D MR angiograms (48 = 48-cm field of view, 3.0-mm section thickness, 42 sections) in an 81-year-old patient. Images were obtained 3 months after percutaneous placement of a stent graft consisting of nitinol wires with a polytetrafluoroethylene cover (Aneurex; Medtronic, Santa Rosa, Calif) in an infrarenal aortic aneurysm: (a) non-fat saturated (4.6/1, 18-second acquisition time), (b) fat saturated (13.9/1, 54-second acquisition time), and (c) partial fat saturated (4.9/1 or 12.8/1 for every N = 8 repetitions, respectively; 23-second acquisition time). In both b and c, the proximal stent graft and iliac legs (curved arrows) are clearly visible within the excluded aortic aneurysmal sac (open arrows). In a, the iliac legs are isointense to the surrounding fat. In b, motion artifacts (arrowheads) are seen in the liver. In a-c, straight solid arrows indicate the inferior vena cava.
|
|
View larger version (127K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 5b. Targeted maximum-intensity-projection images from three sequential delayed-phase 3D MR angiograms (48 = 48-cm field of view, 3.0-mm section thickness, 42 sections) in an 81-year-old patient. Images were obtained 3 months after percutaneous placement of a stent graft consisting of nitinol wires with a polytetrafluoroethylene cover (Aneurex; Medtronic, Santa Rosa, Calif) in an infrarenal aortic aneurysm: (a) non-fat saturated (4.6/1, 18-second acquisition time), (b) fat saturated (13.9/1, 54-second acquisition time), and (c) partial fat saturated (4.9/1 or 12.8/1 for every N = 8 repetitions, respectively; 23-second acquisition time). In both b and c, the proximal stent graft and iliac legs (curved arrows) are clearly visible within the excluded aortic aneurysmal sac (open arrows). In a, the iliac legs are isointense to the surrounding fat. In b, motion artifacts (arrowheads) are seen in the liver. In a-c, straight solid arrows indicate the inferior vena cava.
|
|
View larger version (130K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 5c. Targeted maximum-intensity-projection images from three sequential delayed-phase 3D MR angiograms (48 = 48-cm field of view, 3.0-mm section thickness, 42 sections) in an 81-year-old patient. Images were obtained 3 months after percutaneous placement of a stent graft consisting of nitinol wires with a polytetrafluoroethylene cover (Aneurex; Medtronic, Santa Rosa, Calif) in an infrarenal aortic aneurysm: (a) non-fat saturated (4.6/1, 18-second acquisition time), (b) fat saturated (13.9/1, 54-second acquisition time), and (c) partial fat saturated (4.9/1 or 12.8/1 for every N = 8 repetitions, respectively; 23-second acquisition time). In both b and c, the proximal stent graft and iliac legs (curved arrows) are clearly visible within the excluded aortic aneurysmal sac (open arrows). In a, the iliac legs are isointense to the surrounding fat. In b, motion artifacts (arrowheads) are seen in the liver. In a-c, straight solid arrows indicate the inferior vena cava.
|
|
|
Discussion
|
|---|
Fast contrast-enhanced 3D MR angiography enables visualization of vascular structures on an image acquired during a single breath hold (15). This technique was especially useful for depiction of abdominal vessels difficult to see on images obtained with earlier MR angiographic methods such as black-blood, time-of-flight, or phase-contrast imaging (3,8,16–20). We expect that current MR sequences will be improved in terms of faster acquisition times and vessel visualization. Shorter data acquisition times offer the possibility for use of reduced doses of contrast material and have the potential advantage of decreased venous enhancement. However, shorter injections result in shorter periods of arterial enhancement and therefore place greater demands on the imaging technique. Additionally, imaging with shortened repetition time and higher bandwidth also results in lower SNRs and CNRs. This difference is partially compensated by the absence of respiration motion artifacts with breath-hold imaging.
As demonstrated in this study, visualization of the abdominal vascular structures was substantially improved by acquiring the imaging data with fat suppression. Although vascular SNRs with the partial fat-saturated and non–fat-saturated techniques were not significantly different, CNR increased substantially with the partial fat-saturated technique (Figs 4, 5). The latter difference reflected removal of the fat signal in the paravascular tissue, which allowed better visualization of the vessels. The fact that SNR was better with the fat-saturated technique than with the partial fat-saturated or non–fat-saturated techniques can be explained on the basis of the longer repetition time in the fat-saturated technique. However, fat-saturated images were acquired with a standard fat-suppression pulse during every repetition time. Therefore, the examination time was too long to permit imaging during a breath hold, which increased motion artifacts significantly. Use of a spectrally selective suppression pulse only near the center of k space (partial fat-saturated technique) made it possible to greatly reduce the signal from fat while allowing image acquisition within a breath hold. Therefore, motion artifacts were avoided. Qualitative ratings in this study revealed no differences concerning motion artifacts with the partial fat-saturated technique compared with the non–fat-saturated technique. The former improved the vessel-to-fat CNR with a minimal time penalty. Kopka et al (21) recently described the advantage of fat suppression for the breath-hold evaluation of hepatic vessels by applying a single narrow-bandwidth, frequency-selective radio-frequency pulse before the entire centrically reordered partition loop.
Image subtraction techniques ideally remove all undesired background signal intensity and improve CNR of vessels (22). Optimal subtraction is predicated on minimal motion between successive nonenhanced and contrast-enhanced acquisitions. However, misregistration with MR angiography of the abdominal vessels can be severe and substantially reduce image quality. Partial fat saturation allows abdominal imaging in one breath hold to be comparable to non–fat-saturated 3D MR angiography and therefore avoids motion artifacts. Subtraction can also reduce image quality by deleting vascular signal intensity due to in-flow effects on the nonenhanced images (21). Subtraction also increases the noise level, imaging time (owing to the need to acquire data sets in two breath holds), and postprocessing time. In some cases, such as peripheral MR angiography involving multiple locations, acquisition of a mask volume at each location can be difficult. Furthermore, the original and subtracted image data must be reviewed by a radiologist (21). Incorporation of fat suppression in the acquisition sequence eliminates potential registration artifacts between contrast-enhanced and mask image sets. Suppressing the background signal and therefore increasing the CNR to a level similar to that on subtraction images is possible with fat suppression, which increases visualization of abdominal vascular structures. Increased blurring of the vessel margins did not decrease the overall performance of the fat-saturated technique. The quantitative improvement translated into better ratings by both readers for overall image quality.
In another approach for reducing background tissue signal, a spectrally selective inversion pulse is followed by a short inversion-recovery time and acquisition of several phase-encode values (23). The technique was used to image the vessels of the lower extremity, where motion artifacts due to breathing have no influence on image quality. However, this additional inversion-recovery period lengthens the imaging time and limits the achievable temporal and spatial resolution. The partial fat-saturated technique for 3D MR angiography effectively reduces the background tissue signal with only a slight increase in imaging time, which allows water-specific imaging within a breath hold.
This study is preliminary, and several limitations exist. Randomized image acquisition during the equilibrium phase was necessary to have the same condition for all three sequences and to be able to compare their performance. However, a comparison of arterial phase acquisitions would be preferable, even though this is not practicable in a patient collective. In general, application of a fat-saturated pulse has some limitations. Field inhomogeneity (eg, surgical clips) causes the fat-saturated pulse to drift onto the water peak, and vascular signal may be suppressed. It can be difficult to design a dual-band excitation pulse with sharp transition bands between the water and fat resonances while maintaining a reasonable pulse length. If the transition between the spectral regions of the fat inversion and the small-angle water excitation is not sharp, then field inhomogeneities within the subject can lead to unintentional variations in the water excitation. For this reason, we anticipate a single-band fat excitation with a variable flip angle excitation throughout the subsequent repetition time periods will be used in future work.
As described previously, the acquisition scheme of the partial fat-saturated sequence samples the center of k space throughout the entire examination, which could make acquisition of a pure arterial image difficult if the imaging time is longer than the arterial phase. This was not an issue in the present study since we performed imaging during the equilibrium phase. Image acquisition time is only slightly increased, however, and acquisition of arterial phase images should be possible. Findings at peripheral MR angiography with the partial fat-saturated technique support this statement (9). Moreover, sampling of the center of k space throughout data acquisition makes this technique well suited for a sliding-window type of reconstruction after a dynamic imaging sequence. This technique provides greater flexibility in determining the point during contrast material passage at which image reconstruction should occur.
In conclusion, use of the partial fat-saturated technique for 3D MR angiography decreases background signal intensity, thereby increasing vessel conspicuity. The slight increase in acquisition time allows water-specific imaging within a breath hold, thereby avoiding motion artifacts.
|
FOOTNOTES
|
|---|
9*. Vascular system, location unspecified 
Abbreviations: CNR = contrast-to-noise ratio,
SNR = signal-to-noise ratio,
3D = three-dimensional
Author contributions: Guarantor of integrity of entire study, R.J.H.; study concepts, P.R.H., R.J.H.; study design, P.R.H., S.G.H.; definition of intellectual content, P.R.H., M.T.A., R.J.H., N.J.P.; literature research, P.R.H.; clinical studies, P.R.H., S.G.H., M.T.A.; data acquisition, P.R.H., S.G.H., R.J.H.; data analysis, all authors; statistical analysis, P.R.H., D.F., N.J.P.; manuscript preparation, P.R.H.; manuscript editing and review, all authors
|
REFERENCES
|
|---|
-
Prince MR. Gadolinium-enhanced MR aortography. Radiology 1994; 191:155-164.[Abstract/Free Full Text]
-
Yamashita Y, Mitsuzaki K, Tang Y, Namimoto T, Takahashi M. Gadolinium-enhanced breath-hold three-dimensional time-of-flight MR angiography of the abdominal and pelvic vessels: the value of ultrafast MP-RAGE sequences. J Magn Reson Imaging 1997; 7:623-628.[Medline]
-
Meaney JF, Prince MR, Nostrant TT, Stanley JC. Gadolinium-enhanced MR angiography of visceral arteries in patients with suspected chronic mesenteric ischemia. J Magn Reson Imaging 1997; 7:171-176.[Medline]
-
Maki JH, Prince MR, Londy FJ, Chenevert TL. The effects of time varying intravascular signal intensity and k-space acquisition order on three-dimensional MR angiography image quality. J Magn Reson Imaging 1996; 6:642-651.[Medline]
-
Levy RA, Prince MR. Arterial-phase three-dimensional contrast-enhanced MR angiography of the carotid arteries. AJR Am J Roentgenol 1996; 167:211-215.[Abstract/Free Full Text]
-
Amano Y, Gemma K, Kawamata H, Kumazaki T. Fat-suppressed gadolinium-enhanced three-dimensional magnetic resonance angiography adequately depicts the status of iliac arteries following atherectomy and stent placement. Cardiovasc Intervent Radiol 1998; 21:345-347.[Medline]
-
Tkach JA, Lin W, Duda JJ, Jr, Haacke EM, Masaryk TJ. Optimizing three-dimensional time-of-flight MR angiography with variable repetition time. Radiology 1994; 191:805-811.[Abstract/Free Full Text]
-
Leung DA, McKinnon GC, Davis CP, Pfammatter T, Krestin GP, Debatin JF. Breath-hold, contrast-enhanced, three-dimensional MR angiography. Radiology 1996; 200:569-571.[Abstract/Free Full Text]
-
Alley MT, Heiss G, Pelc NJ, Herfkens RJ. Fast fat suppression for 3D angiographic imaging (abstr) In: Proceedings of the Seventh Meeting of the International Society for Magnetic Resonance in Medicine. Berkeley, Calif: International Society for Magnetic Resonance in Medicine, 1999; 1898.
-
Wilman AH, Riederer SJ. Performance of an elliptical centric view order for signal enhancement and motion artifact suppression in breath-hold three-dimensional gradient echo imaging. Magn Reson Med 1997; 38:793-802.[Medline]
-
Kuroda K, Oshio K, Mulkern RV, Jolesz FA. Optimization of chemical shift selective suppression of fat (abstr) In: Proceedings of the Sixth Meeting of the International Society for Magnetic Resonance in Medicine. Berkeley, Calif: International Society for Magnetic Resonance in Medicine, 1998; 1981.
-
Wang Y, Rossman PJ, Grimm RC, Wilman AH, Riederer SJ, Ehman RL. 3D MR angiography of pulmonary arteries using real-time navigator gating and magnetization preparation. Magn Reson Med 1996; 36:579-587.[Medline]
-
Haase A, Frahm J, Hanicke W, Matthaei D. 1H NMR chemical shift selective (CHESS) imaging. Phys Med Biol 1985; 30:341-344.[Medline]
-
Schoenberg SO, Bock M, Knopp MV, et al. Renal arteries: optimization of three-dimensional gadolinium-enhanced MR angiography with bolus-timing–independent fast multiphase acquisition in a single breath hold. Radiology 1999; 211:667-679.[Abstract/Free Full Text]
-
Prince MR, Narasimham DL, Stanley JC, et al. Breath-hold gadolinium-enhanced MR angiography of the abdominal aorta and its major branches. Radiology 1995; 197:785-792.[Abstract/Free Full Text]
-
Nghiem HV, Freeny PC, Winter TC, III, Mack LA, Yuan C. Phase-contrast MR angiography of the portal venous system: preoperative findings in liver transplant recipients. AJR Am J Roentgenol 1994; 163:445-450.[Abstract/Free Full Text]
-
Suto Y, Ohuchi Y, Kimura T, et al. Three-dimensional black blood MR angiography of the liver during breath holding: a comparison with two-dimensional time-of-flight MR angiography. Acta Radiol 1994; 35:131-134.[Medline]
-
Holland GA, Dougherty L, Carpenter JP, et al. Breath-hold ultrafast three-dimensional gadolinium-enhanced MR angiography of the aorta and the renal and other visceral abdominal arteries. AJR Am J Roentgenol 1996; 166:971-981.[Abstract/Free Full Text]
-
Prince MR, Chenevert TL, Foo TK, Londy FJ, Ward JS, Maki JH. Contrast-enhanced abdominal MR angiography: optimization of imaging delay time by automating the detection of contrast material arrival in the aorta. Radiology 1997; 203:109-114.[Abstract/Free Full Text]
-
Shirkhoda A, Konez O, Shetty AN, Bis KG, Ellwood RA, Kirsch MJ. Mesenteric circulation: three-dimensional MR angiography with a gadolinium-enhanced multiecho gradient-echo technique. Radiology 1997; 202:257-261.[Abstract/Free Full Text]
-
Kopka L, Rodenwaldt J, Vosshenrich R, et al. Hepatic blood supply: comparison of optimized dual phase contrast-enhanced three-dimensional MR angiography and digital subtraction angiography. Radiology 1999; 211:51-58.[Abstract/Free Full Text]
-
Leung DA, Pelkonen P, Hany TF, Zimmermann G, Pfammatter T, Debatin JF. Value of image subtraction in 3D gadolinium-enhanced MR angiography of the renal arteries. J Magn Reson Imaging 1998; 8:598-602.[Medline]
-
Hany TF, Pfammatter T, Ladd ME, Schneider E, Schmidt M, Debatin JF. Single injection, 2-station 3D MR angiography with a dedicated peripheral coil for the assessment of the pelvic and run-off vasculature (abstr). Proceedings of the Sixth Meeting of the International Society for Magnetic Resonance in Medicine Berkeley, Calif: International Society for Magnetic Resonance in Medicine, 1998; 65.
This article has been cited by other articles:
|
|
|
|
 
J. P. Finn, V. Baskaran, J. C. Carr, R. M. McCarthy, F. S. Pereles, R. Kroeker, and G. A. Laub
Thorax: Low-Dose Contrast-enhanced Three-dimensional MR Angiography with Subsecond Temporal Resolution—Initial Results
Radiology,
September 1, 2002;
224(3):
896 - 904.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
C. H. Coulam, F. P. Chan, and K. C. P. Li
Can a Multiphasic Contrast-Enhanced Three-Dimensional Fast Spoiled Gradient-Recalled Echo Sequence Be Sufficient for Liver MR Imaging?
Am. J. Roentgenol.,
February 1, 2002;
178(2):
335 - 341.
[Abstract]
[Full Text]
[PDF]
|
|
|
TOP
ABSTRACT
INTRODUCTION
