(Radiology. 1999;213:185-191.)
© RSNA, 1999
Vascular and Interventional Radiology |
Thoracic Aorta: Rapid Black-Blood MR Imaging with Half-Fourier Rapid Acquisition with Relaxation Enhancement with or without Electrocardiographic Triggering1
David H. Stemerman, MD,
Glenn A. Krinsky, MD,
Vivian S. Lee, MD, PhD,
Glyn Johnson, PhD,
Ben M. Yang, MD and
Neil M. Rofsky, MD
1 From the Department of Radiology, New York University Medical Center, MRI Dept, 530 First Ave, New York, NY 10016. Received October 1, 1998; revision requested November 23; revision received December 22; accepted April 8, 1999. Address reprint requests to G.A.K.
 |
Abstract
|
|---|
PURPOSE: To evaluate and compare findings for thoracic aortic disease with three black-blood magnetic resonance (MR) pulse sequences: half-Fourier rapid acquisition with relaxation enhancement (RARE), with and without electrocardiographic (ECG) triggering, and ECG-triggered turbo spin echo (SE).
MATERIALS AND METHODS: Axial black-blood MR images of the chest acquired at 1.5 T with a phased-array coil were obtained in 38 consecutive patients referred for evaluation of thoracic aortic disease. ECG-triggered and nontriggered half-Fourier RARE images were compared with T1-weighted ECG-triggered turbo SE images. Two readers independently scored images for each of the following parameters: ghosting artifacts; clarity of the mediastinum, cardiac chambers, and aortic wall; conspicuity of abnormality; intraluminal signal void uniformity; and overall image quality.
RESULTS: Both half-Fourier RARE sequences outperformed the turbo SE sequence for all measured parameters. Scores for the ECG-triggered half-Fourier RARE sequence were significantly (P < .05) higher than those for the nontriggered version for clarity of the mediastinum and aortic wall, conspicuity of any abnormality other than aortic dissection, and overall image quality. Mean acquisition times for the ECG-triggered (48 seconds) and nontriggered (30 seconds) sequences were significantly shorter than that for the turbo SE sequence (2 minutes 20 seconds).
CONCLUSION: Rapid black-blood half-Fourier RARE sequences, with or without ECG triggering, can replace ECG-triggered turbo SE sequences for evaluation of thoracic aortic disease.
Index terms: Aneurysm, aortic, 563.73, 94.73 • Aorta, dissection, 94.74 • Aorta, MR, 94.129419 • Magnetic resonance (MR), rapid imaging, 94.129419 • Magnetic resonance (MR), technology, 94.129419
|
Introduction
|
|---|
Conventional spin-echo (SE) black-blood magnetic resonance (MR) imaging of the chest is routinely used to evaluate congenital and acquired diseases of the thoracic aorta (1–5). These pulse sequences require electrocardiographic (ECG) triggering to minimize cardiac motion and pulsatility artifacts. However, a reliance on ECG triggering may result in long examinations and poor image quality in patients with irregular cardiac rhythms. Also, slow flow may cause nonuniform vascular lumen signal intensities. To minimize respiratory and other motion-related artifacts, multiple signals (typically three or four) are averaged, which further lengthens imaging times. Shorter acquisition times can be achieved with turbo SE pulse sequences, which have replaced conventional SE pulse sequences for many applications, including T2-weighted imaging of the chest (6,7). T1-weighted black-blood turbo SE sequences result in substantial time savings when compared with conventional SE black-blood sequences (8), but they still require ECG triggering and the averaging of multiple signals to reduce motion artifacts. Acquisition times are typically 2–4 minutes.
Recently, a faster turbo SE sequence, a half-Fourier rapid acquisition with relaxation enhancement (RARE) sequence, has been introduced that provides black-blood imaging with T2-weighted contrast that is less sensitive to cardiac and respiratory motion (9–11). In this sequence, a long echo train is coupled with half-Fourier reconstruction so that all data required to form the image of a single section can be acquired in less than 500 msec. In the black-blood implementation, a double-inversion pulse is applied to null the signal of blood before the start of the half-Fourier RARE readout (12). As a single-shot technique, half-Fourier RARE is less susceptible to motion artifacts and, hence, the averaging of multiple signals is unnecessary.
ECG-triggered sequences have been the cornerstone of black-blood MR imaging for aortic disease. Magnetization-prepared spoiled gradient-echo (turbo fast low-angle shot, or turbo FLASH) sequences have been used without ECG triggering but have proved unreliable in the diagnosis of aortic dissection (13). Our preliminary unpublished experience suggested that the black-blood half-Fourier RARE sequence, without either ECG triggering or breath holding, could be used in patients referred for MR evaluation of aortic disease. To test this hypothesis, we compared results with nontriggered black-blood half-Fourier RARE imaging to those with both ECG-triggered half-Fourier RARE and the more conventional turbo SE imaging in patients referred for evaluation of thoracic aortic disease.
|
MATERIALS AND METHODS
|
|---|
Patient Population
Over a 4-month period (November 1996 through February 1997), 38 consecutive patients (27 men and 11 women, aged 43–87 years) were referred for MR imaging of the thoracic aorta. Indications for examinations included suspected or known aneurysm and/or postoperative aneurysm (n = 23), acute or chronic aortic dissection (n = 13), coarctation (n = 1), or arteritis (n = 1). The study was approved by the institutional review board, and informed consent was obtained for administration of contrast material.
Imaging
All patients underwent imaging with a 1.5-T MR imaging system (Vision; Siemens Medical Systems, Iselin, NJ) capable of a maximum gradient of 25 mT/m and a 600-µsec rise time with a phased-array coil. All patients had ECG leads placed on the back in a standard fashion, and ECG triggering was performed with the standard software.
All patients underwent a non–breath-hold black-blood MR examination with three pulse sequences (described in detail later) followed by breath-hold gadolinium-enhanced three-dimensional MR angiography. The prospective diagnosis was established by the attending radiologist on the basis of all the available data and served as the standard of reference in the nine patients without surgical proof of disease or correlative images. Ten patients underwent surgery with intraoperative transesophageal echocardiography; 14, follow-up MR examination; and five, routine transesophageal echocardiography.
After the patient was positioned with the middle of the descending thoracic aorta in the center of the magnet, three-plane magnetization-prepared, gradient-echo scout images were acquired. Three black-blood sequences were then performed in the axial plane through the thoracic aorta. Subsequent gadolinium-enhanced three-dimensional MR angiography was performed in the oblique sagittal plane.
Section thickness, intersection gap, and field of view were kept constant for the three black-blood pulse sequences. With use of a rectangular field of view with a maximum dimension of 27.5–37.5 cm, 8–10-mm-thick sections were obtained with a 2-mm intersection gap. Superoinferior spatial presaturation bands were used with all three sequences. No special techniques (gradient moment refocusing, dephasing gradients, respiratory-ordered phase encoding, or respiratory gating) were used.
The half-Fourier RARE (HASTE; Siemens Medical Systems) sequence is shown in Figure 1. Data acquisition was gated in diastole to minimize artifacts due to cardiac motion. The sequence was preceded by a combination of a nonselective 180° inversion pulse followed immediately by a section-selective 180° pulse. The second pulse returned only those spins within the imaging section to their original, equilibrium position. In the inversion time, which was between this double-inversion pulse and application of the 90° excitation pulse, spins outside the section, including those in flowing blood, returned toward equilibrium under the influence of T1. Inversion time was chosen so that blood spins were passing the null point when the 90° pulse was applied and thus gave no signal (14,15). In this respect, the sequence was similar to other inversion-recovery sequences (short inversion time inversion recovery, or STIR, and fluid-attenuated inversion recovery, or FLAIR) used to null the signal from particular tissues by means of appropriate choice of inversion time.
View larger version (19K):
[in this window]
[in a new window]
|
Figure 1. Timing diagram of the black-blood half-Fourier RARE sequence. The acquisition is triggered by the ECG R wave, at which time spins are inverted by means of a nonselective 180° pulse. Immediately afterward, spins within the imaging section are returned to their equilibrium position by means of a section-selective 180° pulse. After a delay (inversion time [TI], chosen such that the recovering blood magnetization is passing through zero), the half-Fourier RARE acquisition is started.
|
|
The following sequences and parameters were used. (a) ECG-triggered turbo SE sequence: repetition time msec/echo time msec (effective) of 461–1,224/12; echo train length, three; matrix, 114– 180 x 256 (phase encoding x frequency encoding); and three signals acquired. ECG triggering commenced at the first R wave. Imaging time for nine to 17 sections (function of R-R interval) ranged from 1 minute 15 seconds to 5 minutes 15 seconds (mean, 2 minutes 20 seconds). (b) ECG-triggered half-Fourier RARE: 
/43/700 (inversion time msec); refocusing flip angle, 150°; echo train length, 72; matrix, 128–160 x 256 after half-Fourier processing; 21 sections; imaging time, range of 31–79 seconds and mean of 48 seconds; and one signal acquired. Two ECG periods were allowed between section acquisitions to allow inverted blood to recover. (c) Nontriggered half-Fourier RARE: parameters were the same as those for the ECG-triggered RARE sequence except for the imaging time, with a range of 25–31 seconds and a mean of 30 seconds.
Image Analysis
The quality of images acquired with all three pulse sequences was assessed retrospectively. Hard-copy images without any pulse sequence information were developed and compared side by side by two radiologists (G.A.K., D.H.S.). One of the radiologists had been involved in some of the prospective clinical interpretations, and the other had not. The two readers independently evaluated the three sets of images for the following: (a) freedom from internal ghosting, (b) freedom from external (extraaortic) ghosting, (c) clarity of the mediastinal structures, (d) clarity of aortic dissection if present, (e) clarity of the heart chambers, (f) conspicuity of abnormality present (other than aortic dissection), (g) clarity of the intramural layer of the aorta, (h) signal void uniformity, and (i) overall image quality. A five-point rating scale was used, with 5 as the highest grade, indicative of the most desirable imaging features, and 1 as the lowest.
Consensus reading was not used for discordant grades. Despite elimination of the sequence information from the images, the radiologists could not be blinded. The half-Fourier RARE images were readily distinguishable from the turbo SE images by the absence of motion artifacts and the greater number of sections. However, there were no obvious characteristics to distinguish the ECG-triggered from the nontriggered half-Fourier RARE images that could bias the readers.
Statistical Analysis
Friedman nonparametric two-way analysis of variance by ranks was used to determine if there was a significant difference in image quality parameters among the three sequences. The analysis was performed once to determine the composite significance for the two readers. The Wilcoxon matched pairs signed rank test was then performed to evaluate whether there was a significant difference in image quality parameters between each pair of sequences. A P value less than .05 was considered statistically significant.
|
RESULTS
|
|---|
At MR imaging, 33 of 38 patients (87%) had abnormal aortas: 20 patients had a thoracic aortic aneurysm (seven of which were confirmed surgically), six patients had an aortic dissection (three, Stanford type A and three, Stanford type B [two of which were surgically confirmed]), six patients had protruding aortic atheromas larger than 4 mm in diameter, and one patient had aortic coarctation (surgically confirmed). Of the patients with aortic dissection and aneurysm (n = 26), six had concomitant protruding atheromas. An example of a normal aorta, an ascending aortic aneurysm, and a Stanford type B dissection are demonstrated in Figures 2–4, respectively.
View larger version (112K):
[in this window]
[in a new window]
|
Figure 2a. (a-c) Axial images obtained with three black-blood pulse sequences show a normal thoracic aorta. (a) ECG-triggered half-Fourier RARE image (acquisition time, 41 seconds). (b) Nontriggered half-Fourier RARE image (acquisition time, 31 seconds). (c) ECG-triggered turbo SE image (imaging time, 2 minutes 51 seconds). The ascending (A) and descending (D) thoracic aorta demonstrate uniform flow void in a and nonuniform intravascular signal intensity in the descending aorta in both b and c.
|
|
View larger version (108K):
[in this window]
[in a new window]
|
Figure 2b. (a-c) Axial images obtained with three black-blood pulse sequences show a normal thoracic aorta. (a) ECG-triggered half-Fourier RARE image (acquisition time, 41 seconds). (b) Nontriggered half-Fourier RARE image (acquisition time, 31 seconds). (c) ECG-triggered turbo SE image (imaging time, 2 minutes 51 seconds). The ascending (A) and descending (D) thoracic aorta demonstrate uniform flow void in a and nonuniform intravascular signal intensity in the descending aorta in both b and c.
|
|
View larger version (117K):
[in this window]
[in a new window]
|
Figure 2c. (a-c) Axial images obtained with three black-blood pulse sequences show a normal thoracic aorta. (a) ECG-triggered half-Fourier RARE image (acquisition time, 41 seconds). (b) Nontriggered half-Fourier RARE image (acquisition time, 31 seconds). (c) ECG-triggered turbo SE image (imaging time, 2 minutes 51 seconds). The ascending (A) and descending (D) thoracic aorta demonstrate uniform flow void in a and nonuniform intravascular signal intensity in the descending aorta in both b and c.
|
|
View larger version (112K):
[in this window]
[in a new window]
|
Figure 3a. (a-c) Axial images depict an aneurysm of the ascending aorta associated with a left pleural effusion. Contiguous axial (a) ECG-triggered half-Fourier RARE images (acquisition time, 43 seconds) and (b) nontriggered half-Fourier RARE images (acquisition time, 31 seconds) both demonstrate an aneurysm of the ascending aorta (straight arrows). More uniform signal void is depicted in the ascending aorta in b and in the descending thoracic aorta (arrowheads) in a. (c) ECG-triggered turbo SE images (imaging time, 2 minutes 28 seconds) are degraded by motion artifacts and nonuniform aortic signal intensity. High signal intensity is present in the left atrium (curved arrows in a-c) from slow flow, in-plane flow, or both.
|
|
View larger version (107K):
[in this window]
[in a new window]
|
Figure 3b. (a-c) Axial images depict an aneurysm of the ascending aorta associated with a left pleural effusion. Contiguous axial (a) ECG-triggered half-Fourier RARE images (acquisition time, 43 seconds) and (b) nontriggered half-Fourier RARE images (acquisition time, 31 seconds) both demonstrate an aneurysm of the ascending aorta (straight arrows). More uniform signal void is depicted in the ascending aorta in b and in the descending thoracic aorta (arrowheads) in a. (c) ECG-triggered turbo SE images (imaging time, 2 minutes 28 seconds) are degraded by motion artifacts and nonuniform aortic signal intensity. High signal intensity is present in the left atrium (curved arrows in a-c) from slow flow, in-plane flow, or both.
|
|
View larger version (91K):
[in this window]
[in a new window]
|
Figure 3c. (a-c) Axial images depict an aneurysm of the ascending aorta associated with a left pleural effusion. Contiguous axial (a) ECG-triggered half-Fourier RARE images (acquisition time, 43 seconds) and (b) nontriggered half-Fourier RARE images (acquisition time, 31 seconds) both demonstrate an aneurysm of the ascending aorta (straight arrows). More uniform signal void is depicted in the ascending aorta in b and in the descending thoracic aorta (arrowheads) in a. (c) ECG-triggered turbo SE images (imaging time, 2 minutes 28 seconds) are degraded by motion artifacts and nonuniform aortic signal intensity. High signal intensity is present in the left atrium (curved arrows in a-c) from slow flow, in-plane flow, or both.
|
|
View larger version (106K):
[in this window]
[in a new window]
|
Figure 4a. (a-c) Axial images depict aortic dissection involving the descending thoracic aorta (Stanford type B). (a) ECG-triggered half-Fourier RARE image (acquisition time, 79 seconds) and (b) nontriggered half-Fourier RARE image (acquisition time, 31 seconds) both demonstrate a normal aortic root (A) and an intimal flap (arrow) in the descending thoracic aorta, with patency of both lumina. (c) ECG-triggered turbo SE image (imaging time, 2 minutes 17 seconds) is degraded by motion artifacts and nonuniform signal intensity within both lumina. A = aortic root, arrow indicates intimal flap.
|
|
View larger version (100K):
[in this window]
[in a new window]
|
Figure 4b. (a-c) Axial images depict aortic dissection involving the descending thoracic aorta (Stanford type B). (a) ECG-triggered half-Fourier RARE image (acquisition time, 79 seconds) and (b) nontriggered half-Fourier RARE image (acquisition time, 31 seconds) both demonstrate a normal aortic root (A) and an intimal flap (arrow) in the descending thoracic aorta, with patency of both lumina. (c) ECG-triggered turbo SE image (imaging time, 2 minutes 17 seconds) is degraded by motion artifacts and nonuniform signal intensity within both lumina. A = aortic root, arrow indicates intimal flap.
|
|
View larger version (111K):
[in this window]
[in a new window]
|
Figure 4c. (a-c) Axial images depict aortic dissection involving the descending thoracic aorta (Stanford type B). (a) ECG-triggered half-Fourier RARE image (acquisition time, 79 seconds) and (b) nontriggered half-Fourier RARE image (acquisition time, 31 seconds) both demonstrate a normal aortic root (A) and an intimal flap (arrow) in the descending thoracic aorta, with patency of both lumina. (c) ECG-triggered turbo SE image (imaging time, 2 minutes 17 seconds) is degraded by motion artifacts and nonuniform signal intensity within both lumina. A = aortic root, arrow indicates intimal flap.
|
|
Results of the qualitative assessment are listed in Table 1. When the ECG-triggered and nontriggered half-Fourier RARE sequences were compared with the triggered turbo SE sequence, both half-Fourier RARE sequences scored significantly (P < .05) higher than the triggered turbo SE sequence for all evaluated parameters, except for aortic dissection (P = .25). Statistical evaluation to compare sequences for aortic dissection is limited, however, owing to small sample size (n = 6). When the ECG-triggered and nontriggered half-Fourier RARE sequences were compared, mean scores for the former were higher than those for the latter for all variables (Table 2), but only differences in mediastinal and intramural layer clarity, conspicuity of abnormality other than aortic dissection, and overall quality reached statistical significance (P < .05) (Table 2).
|
DISCUSSION
|
|---|
Findings in our study demonstrate that the half-Fourier RARE sequence can provide an accurate and time-efficient method for providing black-blood imaging of the chest. Even when performed without ECG triggering, image quality with the half-Fourier RARE sequence was better than that with the ECG-triggered turbo SE sequence for all measured parameters, whereas mean acquisition times (30 seconds vs 2 minutes 20 seconds) were substantially shorter.
Aside from short acquisition times and the need for only a single acquisition when used with a phased-array coil, the half-Fourier RARE sequence offered other advantages over the conventional ECG-triggered SE and turbo SE techniques. Because the sequence relied on a series of 180° refocusing pulses to provide each line of k space, there was diminished sensitivity to local susceptibility effects when compared with the sensitivity of conventional SE imaging. A single acquisition with the black-blood half-Fourier RARE sequence provides sufficient anatomic coverage of the entire thoracic aorta. Reduced imaging times also allow time for multiplanar assessments, when clinically relevant.
The half-Fourier RARE sequence performed with ECG triggering provided better overall image quality, anatomic detail of the aortic wall and mediastinum, and conspicuity of aortic abnormalities (with the exception of aortic dissection) compared to results with the nontriggered version. Because the mean additional time needed to acquire half-Fourier RARE images with ECG triggering was only 18 seconds, it seemed prudent to use ECG triggering when possible. For patients in whom ECG triggering cannot be accomplished, however, the nontriggered half-Fourier RARE sequence provides diagnostic quality images. This sequence may therefore help reduce the dependence of MR imaging of the thorax on a regular cardiac rhythm.
While not specifically addressed in our study, the contrast on the turbo SE and half-Fourier RARE images is different, presumably because of their different effective echo times, echo train lengths, and echo spacing. We hypothesize two possible instances in which differences in image contrast may be pertinent: aortic inflammation (infectious or autoimmune aortitis) and intramural hematoma. For the former, T2 weighting of the half-Fourier RARE sequence may render the inflammation more conspicuous. For the latter, the age of the hematoma may be more difficult to determine with the half-Fourier RARE sequence than with the T1-weighted SE sequence, for which characteristic signal changes have been described (16).
The half-Fourier RARE sequence used in this study was commercially available with a fixed effective echo time of 43 msec. This echo time was the minimum that could be achieved within the constraints imposed by hardware and limits of specific absorption rate. Although the use of a pulse sequence with a long echo train in conjunction with the relatively short effective echo time (43 msec) may theoretically result in image blurring, this did not adversely effect the quality of the half-Fourier RARE images.
The widespread use of bright-blood MR angiographic techniques, especially gadolinium-enhanced three-dimensional MR angiography, has had an effect on the MR assessment of thoracic aortic disease (5). Bright-blood MR angiographic techniques provide better evaluation of intraluminal pathologic conditions such as aortic dissection by allowing differentiation of slow or disturbed flow from thrombus and better delineation of intimal flaps. These techniques are also superior to black-blood imaging in the evaluation of branch vessel disease (13,17). Hence the reliance on black-blood imaging, which often requires imaging in two or three planes, has diminished. However, black-blood pulse sequences are still an essential component of aortic imaging for evaluation of mural disease and of the mediastinum, pleura, and pericardium for hemorrhage. In particular, gadolinium-enhanced three-dimensional techniques are insensitive to intramural hematoma and extraaortic intrathoracic pathologic conditions (18). However, by supplementing the bright-blood approach with an efficient black-blood strategy, such as half-Fourier RARE, a rapid comprehensive evaluation of the aorta is possible (Fig 5).
View larger version (122K):
[in this window]
[in a new window]
|
Figure 5a. (a-e) Axial images depict an aneurysm of the entire thoracic aorta with associated thrombus. (a) ECG-triggered half-Fourier RARE image (acquisition time, 46 seconds) and (b) nontriggered half-Fourier RARE image (acquisition time, 30 seconds) both demonstrate aneurysmal dilatation of the ascending (A) and descending (D) aorta. The thrombus (arrow in a and b) in the posterior descending thoracic aorta is well demarcated from the flow void seen anteriorly in the patent lumen. (c) ECG-triggered turbo SE image (imaging time, 2 minutes 12 seconds) is degraded by motion artifacts and nonuniform aortic signal intensity. It is difficult to differentiate flowing blood from thrombus (arrow) in the descending thoracic aorta. A = ascending aorta. (d) Axial reformation image from breath-hold, gadolinium-enhanced, three-dimensional MR angiogram demonstrates flow within the ascending aorta (arrow) and anterior aspect of the descending aorta (arrowhead). The nonenhanced posterior thrombus (T) is difficult to distinguish from contiguous lung and chest wall. (e) Oblique sagittal, maximum intensity projection image from breath-hold, gadolinium-enhanced, three-dimensional MR angiogram demonstrates the aneurysmal thoracic aorta and brachiocephalic trunk (arrow).
|
|
View larger version (111K):
[in this window]
[in a new window]
|
Figure 5b. (a-e) Axial images depict an aneurysm of the entire thoracic aorta with associated thrombus. (a) ECG-triggered half-Fourier RARE image (acquisition time, 46 seconds) and (b) nontriggered half-Fourier RARE image (acquisition time, 30 seconds) both demonstrate aneurysmal dilatation of the ascending (A) and descending (D) aorta. The thrombus (arrow in a and b) in the posterior descending thoracic aorta is well demarcated from the flow void seen anteriorly in the patent lumen. (c) ECG-triggered turbo SE image (imaging time, 2 minutes 12 seconds) is degraded by motion artifacts and nonuniform aortic signal intensity. It is difficult to differentiate flowing blood from thrombus (arrow) in the descending thoracic aorta. A = ascending aorta. (d) Axial reformation image from breath-hold, gadolinium-enhanced, three-dimensional MR angiogram demonstrates flow within the ascending aorta (arrow) and anterior aspect of the descending aorta (arrowhead). The nonenhanced posterior thrombus (T) is difficult to distinguish from contiguous lung and chest wall. (e) Oblique sagittal, maximum intensity projection image from breath-hold, gadolinium-enhanced, three-dimensional MR angiogram demonstrates the aneurysmal thoracic aorta and brachiocephalic trunk (arrow).
|
|
View larger version (101K):
[in this window]
[in a new window]
|
Figure 5c. (a-e) Axial images depict an aneurysm of the entire thoracic aorta with associated thrombus. (a) ECG-triggered half-Fourier RARE image (acquisition time, 46 seconds) and (b) nontriggered half-Fourier RARE image (acquisition time, 30 seconds) both demonstrate aneurysmal dilatation of the ascending (A) and descending (D) aorta. The thrombus (arrow in a and b) in the posterior descending thoracic aorta is well demarcated from the flow void seen anteriorly in the patent lumen. (c) ECG-triggered turbo SE image (imaging time, 2 minutes 12 seconds) is degraded by motion artifacts and nonuniform aortic signal intensity. It is difficult to differentiate flowing blood from thrombus (arrow) in the descending thoracic aorta. A = ascending aorta. (d) Axial reformation image from breath-hold, gadolinium-enhanced, three-dimensional MR angiogram demonstrates flow within the ascending aorta (arrow) and anterior aspect of the descending aorta (arrowhead). The nonenhanced posterior thrombus (T) is difficult to distinguish from contiguous lung and chest wall. (e) Oblique sagittal, maximum intensity projection image from breath-hold, gadolinium-enhanced, three-dimensional MR angiogram demonstrates the aneurysmal thoracic aorta and brachiocephalic trunk (arrow).
|
|
View larger version (57K):
[in this window]
[in a new window]
|
Figure 5d. (a-e) Axial images depict an aneurysm of the entire thoracic aorta with associated thrombus. (a) ECG-triggered half-Fourier RARE image (acquisition time, 46 seconds) and (b) nontriggered half-Fourier RARE image (acquisition time, 30 seconds) both demonstrate aneurysmal dilatation of the ascending (A) and descending (D) aorta. The thrombus (arrow in a and b) in the posterior descending thoracic aorta is well demarcated from the flow void seen anteriorly in the patent lumen. (c) ECG-triggered turbo SE image (imaging time, 2 minutes 12 seconds) is degraded by motion artifacts and nonuniform aortic signal intensity. It is difficult to differentiate flowing blood from thrombus (arrow) in the descending thoracic aorta. A = ascending aorta. (d) Axial reformation image from breath-hold, gadolinium-enhanced, three-dimensional MR angiogram demonstrates flow within the ascending aorta (arrow) and anterior aspect of the descending aorta (arrowhead). The nonenhanced posterior thrombus (T) is difficult to distinguish from contiguous lung and chest wall. (e) Oblique sagittal, maximum intensity projection image from breath-hold, gadolinium-enhanced, three-dimensional MR angiogram demonstrates the aneurysmal thoracic aorta and brachiocephalic trunk (arrow).
|
|
View larger version (97K):
[in this window]
[in a new window]
|
Figure 5e. (a-e) Axial images depict an aneurysm of the entire thoracic aorta with associated thrombus. (a) ECG-triggered half-Fourier RARE image (acquisition time, 46 seconds) and (b) nontriggered half-Fourier RARE image (acquisition time, 30 seconds) both demonstrate aneurysmal dilatation of the ascending (A) and descending (D) aorta. The thrombus (arrow in a and b) in the posterior descending thoracic aorta is well demarcated from the flow void seen anteriorly in the patent lumen. (c) ECG-triggered turbo SE image (imaging time, 2 minutes 12 seconds) is degraded by motion artifacts and nonuniform aortic signal intensity. It is difficult to differentiate flowing blood from thrombus (arrow) in the descending thoracic aorta. A = ascending aorta. (d) Axial reformation image from breath-hold, gadolinium-enhanced, three-dimensional MR angiogram demonstrates flow within the ascending aorta (arrow) and anterior aspect of the descending aorta (arrowhead). The nonenhanced posterior thrombus (T) is difficult to distinguish from contiguous lung and chest wall. (e) Oblique sagittal, maximum intensity projection image from breath-hold, gadolinium-enhanced, three-dimensional MR angiogram demonstrates the aneurysmal thoracic aorta and brachiocephalic trunk (arrow).
|
|
Our study has recognized limitations, including its retrospective design, the inability to blind the readers to half-Fourier RARE and turbo SE images, and the use of subjective grading criteria. The shorter effective echo time (14 msec) of the turbo SE sequence may be more vulnerable to nonuniform intraluminal signal. Some centers use longer echo times to obtain more uniform signal void within the aorta. However, such sequences provide fewer sections for a given repetition time, which results in less anatomic coverage. Also, use of a longer echo time may result in greater susceptibility artifacts and the potential for other motion artifacts. Respiratory gating or respiratory-ordered phase encoding was not used with the turbo SE sequence, and these strategies may improve image quality. Finally, when compared with T1-weighted SE imaging, our turbo SE sequence had an echo train length of three, which may, in theory, cause blurring.
In conclusion, the half-Fourier RARE sequence provided black-blood images of the chest with greatly reduced imaging times and greater anatomic coverage per acquisition time, improved lesion conspicuity, and reduced motion artifacts compared to those obtained with an ECG-triggered turbo SE sequence. Although the use of ECG triggering improved the quality of black-blood half-Fourier RARE images, nontriggered images were still considered better than the turbo SE images. The short acquisition times and freedom from the constraints of ECG triggering means that aortic imaging could be performed in patients who otherwise could not be studied with MR imaging. When coupled with gadolinium-enhanced MR angiography, black-blood half-Fourier RARE imaging allows a complete MR evaluation of the thoracic aorta in less than 15 minutes. This may extend the option of MR imaging to a greater number of patients, including those who are critically ill or have irregular cardiac rhythms.
|
Acknowledgments
|
|---|
We thank Orlando Sinonetti for his help in the preparation of the manuscript for this article.
|
Footnotes
|
|---|
Abbreviations: ECG = electrocardiographic
RARE = rapid acquisition with relaxation enhancement
SE = spin echo
Author contributions: Guarantor of integrity of entire study, G.A.K.; study concepts and design, G.A.K., N.M.R.; definition of intellectual content, all authors; literature research, G.A.K., D.H.S.; clinical studies, G.A.K., N.M.R.; data acquisition, G.A.K., N.M.R.; data analysis, all authors; statistical analysis, V.S.L.; manuscript preparation, editing, and review, all authors.
|
References
|
|---|
-
Amparo EG, Higgins CB, Hricak H, et al. Aortic dissection: magnetic resonance imaging. Radiology 1985; 155:399-406.[Abstract/Free Full Text]
-
Dinsmore RE, Liberthson RR, Wismer GL, et al. Magnetic resonance imaging of thoracic aortic aneurysm: comparison with other imaging methods. AJR 1986; 146:309-314.[Abstract/Free Full Text]
-
White RD, Dooms GC, Higgins CB. Advances in imaging thoracic aortic disease. Invest Radiol 1986; 21:761-778.[Medline]
-
Kersting-Sommerhoff BA, Higgins CB, White RD, Sommerhoff CP, Lipton MJ. Aortic dissection: sensitivity and specificity of MR imaging. Radiology 1988; 166:651-655.[Abstract/Free Full Text]
-
Flamm SD, VanDyke CW, White RD. MR imaging of the thoracic aorta. Magn Reson Imaging Clin N Am 1996; 4:217-235.[Medline]
-
Seelos KC, von Smekal A, Vahlensieck M, Gieseke J, Reiser M. Cardiac abnormalities: assessment with T2-weighted turbo spin-echo MR imaging with electrocardiogram gating at 0.5 T. Radiology 1993; 189:517-522.[Abstract/Free Full Text]
-
Haddad JL, Rofsky NM, Ambrosino MM, Naidich DP, Weinreb JC. T2-weighted MR imaging of the chest: comparison of electrocardiograph-triggered conventional and turbo spin-echo and nontriggered turbo spin-echo sequences. JMRI 1995; 5:325-329.
-
Simonetti OP, Kiefer B, Laub G, Fluegel H, Finn P. Rapid black-blood imaging of the heart with turbo spin-echo and turbo gradient spin-echo techniques (abstr). JMRI 1994; 4(P):81.
-
Laub G, Simonetti OP, Nitz W. Single-shot imaging of the heart with HASTE (abstr) In: Proceedings of the Society of Magnetic Resonance in Medicine and the European Society of Magnetic Resonance in Medicine and Biology. Berkeley, Calif: Society of Magnetic Resonance in Medicine, 1995; 246.
-
Stehling MK, Holzknecht N, Laub G, Peller M, van Smekal A, Reiser M. Single-shot imaging of the heart with HASTE (abstr) In: Proceedings of the Fourth Meeting of the International Society for Magnetic Resonance in Medicine. Berkeley, Calif: International Society for Magnetic Resonance in Medicine, 1996; 656.
-
Hartnell GG, Charlamb MC, Cohen MD, Saouaf R, Simonetti OP, Finn JP. Breath hold cardiac MRI: image quality and motion susceptibility of turbo spin echo, turboSTIR, and HASTE compared with standard sequences (abstr) ; Proceedings of the Society of Magnetic Resonance in Medicine and the European Society of Magnetic Resonance in Medicine and Biology. Berkeley, Calif: Society of Magnetic Resonance in Medicine, 1995; 658.
-
Simonetti OP, Finn JP, White RD, Laub G, Henry DA. "Black blood" T2-weighted inversion-recovery MR imaging of the heart. Radiology 1996; 199:49-57.[Abstract/Free Full Text]
-
Hartnell GG, Finn JP, Zenni M, et al. MR imaging of the thoracic aorta: comparison of spin-echo, angiographic, and breath-hold techniques. Radiology 1994; 191:697-704.[Abstract/Free Full Text]
-
Edelman RR, Chien D, Kim D. Fast selective black blood MR imaging. Radiology 1991; 181:655-660.[Abstract/Free Full Text]
-
Liu Y, Riederer SJ, Ehman RL. Magnetization-prepared cardiac imaging using gradient echo acquisitions. Magn Reson Med 1993; 30:271-275.[Medline]
-
Murray JG, Manisali M, Flamm SD, et al. Intramural hematoma of the thoracic aorta: MR image findings and their prognostic implications. Radiology 1997; 204:349-356.[Abstract/Free Full Text]
-
White RD, Obuchowski NA, VanDyke CW, et al. Thoracic aortic disease: evaluation using a single MRA volume series. J Comput Assist Tomogr 1994; 18:843-854.[Medline]
-
Krinsky G, Rofsky NM, DeCoroto DR, et al. Thoracic aorta: comparison of gadolinium-enhanced three-dimensional MR angiography with conventional MR imaging. Radiology 1997; 202:183-193.[Abstract/Free Full Text]
This article has been cited by other articles:
|
|
|
|
 
V. Mani, V. V. Itskovich, M. Szimtenings, J. G. S. Aguinaldo, D. D. Samber, G. Mizsei, and Z. A. Fayad
Rapid Extended Coverage Simultaneous Multisection Black-Blood Vessel Wall MR Imaging
Radiology,
July 1, 2004;
232(1):
281 - 288.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
R. P. Kunz, K. Oberholzer, W. Kuroczynski, G. Horstick, F. Krummenauer, M. Thelen, and K.-F. Kreitner
Assessment of Chronic Aortic Dissection: Contribution of Different ECG-Gated Breath-Hold MRI Techniques
Am. J. Roentgenol.,
May 1, 2004;
182(5):
1319 - 1326.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
O. B. Vignaux, J. Augui, J. Coste, C. Argaud, P. Le Roux, P. G. Carlier, D. Duboc, and P. Legmann
Comparison of Single-Shot Fast Spin-Echo and Conventional Spin-Echo Sequences for MR Imaging of the Heart: Initial Experience
Radiology,
May 1, 2001;
219(2):
545 - 550.
[Abstract]
[Full Text]
|
|
|
|
|
|
|
R. M. Botnar, M. Stuber, K. V. Kissinger, W. Y. Kim, E. Spuentrup, and W. J. Manning
Noninvasive Coronary Vessel Wall and Plaque Imaging With Magnetic Resonance Imaging
Circulation,
November 21, 2000;
102(21):
2582 - 2587.
[Abstract]
[Full Text]
[PDF]
|
|
|
TOP
Abstract
Introduction
