Dual stack black blood carotid artery CMR at 3T: Application to wall thickness visualization
© Bornstedt et al; licensee BioMed Central Ltd. 2009
Received: 08 April 2009
Accepted: 10 November 2009
Published: 10 November 2009
The increasing understanding of atherosclerosis as an important risk factor for the development of acute ischemic events like ischemic stroke has stimulated increasing interest in non-invasive assessment of the structure, composition and burden of plaque depositions in the carotid artery wall. Vessel wall imaging by means of cardiovascular magnetic resonance (CMR) is conventionally done by 2D dual inversion recovery (DIR) techniques, which often fail in covering large volumes of interest as required in plaque burden assessment. Although the technique has been extended to 2D multislice imaging, its straight extension to 3D protocols is still limited by the prolonged acquisition times and incomplete blood suppression. A novel approach for rapid overview imaging of large sections of the carotid artery wall at isotropic spatial resolutions is presented, which omits excitation of the epiglottis. By the interleaved acquisition of two 3D stacks with the proposed motion sensitized segmented steady-state black-blood gradient echo technique (MSDS) the coverage of the carotid artery trees on both sides in reasonable scan times is enabled.
10 patients were investigated with the proposed technique and compared to conventional transversal DIR turbo spin and gradient echo approaches centered at the height of the carotid bifurcation. In all MSDS experiments sufficient black-blood contrast could be obtained over the entire covered volumes. The contrast to noise ratio between vessel and suppressed blood was improved by 73% applying the motion sensitizing technique. In all patients the suspicious areas of vessel wall thickening could be clearly identified and validated by the conventional local imaging approach. The average assessable vessel wall segment length was evaluated to be 18 cm. While in 50% of the cases motion artifacts could be appreciated in the conventional images, none were detected for the MSDS technique.
The proposed technique enables the time efficient coverage of large areas of the carotid arteries without compromising wall-lumen CNR to get an overview about detrimental alterations of the vessel wall. Thickening of the vessel wall can be identified and the suspicious segments can be targeted for subsequent high-resolution CMR. The exclusion of the epiglottis may further facilitate reduction of swallowing induced motion artifacts.
The well-established understanding of atherosclerosis as an important risk factor for the development of acute ischemic events has stimulated increasing interest in non-invasive assessment of the structure, composition and burden of plaque depositions in the carotid artery wall. Over the last decade, cardiovascular magnetic resonance (CMR) has proven to contribute substantially to the non-invasive assessment and characterization of atherosclerotic lesions, especially in the aorta and the carotid arteries [1–3].
For vessel wall imaging, black-blood contrast is required for ensuring sufficient contrast between the vessel wall and blood. Most commonly, suppression of the blood signal is achieved by applying a double-inversion recovery (DIR) black-blood preparation prepulse . Most published approaches combine the DIR preparation with two-dimensional (2D) imaging by means of multi spin-echo (MSE) techniques [1–3]. Without compromising spatial resolution due to severe T2 apodization, this technique allows for the acquisition of a single slice in the minute time range, while providing sufficient signal to noise ratio (SNR) and lumen-wall contrast (CNR). Different image contrasts including T2 - (T2W), proton density (PDW) and, limited by the required inversion recovery time for blood signal nulling and heart rate, T1 - weighting (T1W) can be realized by slight modifications of the sequence timing. While the DIR technique has been extended to enable multislice acquisitions , its application for three-dimensional (3D) coverage of large sections of the arteries at sufficient isotropic spatial resolution has been limited. The resulting long acquisition times often cause image-distorting motion artifacts  due to swallowing, arterial pulsation and breathing. Incomplete blood suppression [7, 8] caused by insufficient exchange of the blood in the entire volume may mimic incorrect wall thickening. Time-efficient 3D imaging of the carotid and aortic arteries has been achieved by applying segmented gradient echo techniques [9–11]. Swallowing motion artifacts have been addressed by utilization of navigators monitoring the position of the epiglottis [10, 12] and arterial pulsation artifacts can be reduced by synchronizing the data acquisition with the heart beat. Improvement of blood suppression has been obtained by phase-sensitive reconstruction techniques  and diffusion-prepared or motion sensitized preparation techniques [10, 9, 14], all relying rather on the motion of the blood than on complete blood exchange between preparation and readout.
A major limitation of arterial wall imaging by CMR still results from the long acquisition times, which do not allow covering large sections of the arteries at sufficient spatial resolution for plaque characterization. Fast three-dimensional imaging methods, providing an overview on the arterial tree with sufficient spatial resolution and contrast, appear attractive for targeting the lesion of interest and planning of the subsequent high-resolution CMR plaque characterization protocol.
We recently developed a 3D localized volume-selective sequence for arterial wall imaging . By combining a 2D pencil beam excitation with motion sensitized black-blood prepared imaging, large segments of the carotid artery tree could be covered in reasonable image acquisition times. Although the motion sensitizing prepulse causes a preweighting of the spins similar to a T2 preparation pulse of about 15 ms length, an excellent wall lumen contrast could be maintained and the reduced spatial resolution was still sufficient for assessment of the vessel wall thickness. Motion artifacts due to swallowing could not be observed, most-likely due to the local character of the excitation, which ensures that the region of the epiglottis does not contribute to the final MR image. Limitations of this technique rise from the rather long duration of the 2D excitation pulses, which is required to avoid substantial side-loops in the 2D excitation profile and the restriction to acquire a single carotid artery tree, only.
In this contribution, a new dual-stack approach is suggested, in which two parasagittally aligned 3D volumes (stacks) are acquired temporally interleaved. Black-blood contrast is achieved by motion-sensitized preparation, which is applied prior to the acquisition of each temporally interleaved stack. As in the local excitation approach, the proposed technique omits excitation of the epiglottis and is supposed to show similar advantages regarding swallowing artifacts. The efficiency of the motion-sensitizing magnetization preparation dual stack gradient echo (MSDS) sequence is demonstrated in patients for the rapid overview imaging of long segments of the carotid artery wall for both sides simultaneously. The applicability of the technique for identification of thickened vessel wall segments was investigated in direct comparison to the conventional high-spatial resolution protocols.
The method was evaluated in 10 patients with a proven vascular disease defined as arteriosclerosis confirmed by the presence of coronary artery disease, peripheral artery disease or carotid plaque (10 males, mean age 58 +/- 7 y, mean weight 83 +/- 6 kg) without any contraindications for CMR. Written informed consent was obtained from all subjects before CMR and the protocol was approved by the ethics committee of the University.
All imaging was performed on a 3T whole-body system (Achieva, Philips Medical Systems, Netherlands) equipped with a high performance gradient system (40 mT m-1, 200 T m-1 s-1). All data was acquired utilizing a dedicated two segment four-element carotid coil (Philips Research Europe, Germany). Either segment comprises two independent coil elements with spatial extent of 65 × 50 mm2 each. The two segments were located on either side of the neck and fixated with a Vac-Lok neck cushion (Medtec, USA), which also immobilized the head of the patient to a great extent.
For localization, a multislice gradient-echo technique acquiring three orthogonal slabs with transversal, sagittal, and coronal slice orientation was applied. The resulting scout images were used for planning of the subsequent scans.
Coil-Sensitivity Reference Scan
A coil-sensitivity reference scan was acquired, which was needed for the homogenization of the image signal and the vendor specific parallel imaging reconstruction, respectively.
Scan parameters for the acquisitions performed.
FOV mm (mps)
300 × 300 × 50
150 × 200 × 30
150 × 200 × 30
150 × 150 × 20
150 × 150 × 20
Resolution mm (mps)
1.17 × 2.34 × 50
1.0 × 1.0 × 1.0
0.8 × 0.8 × 0.8
0.45 × 0.45 × 2.0
0.45 × 0.45 × 2.0
Parallel Imaging Acc.
2.0 p/1.25 s
TR/TR p. Segm. ms
Acq. Duration ms
Scan duration s
Volume Imaging of the Carotid Arteries
For comparison, a high-resolution proton-density weighted 3D double-inversion recovery multi spin echo sequence (DIR-MSE) and a 3D double-inversion recovery segmented gradient echo sequence (DIR-GRE) in axial view orientation were performed centered around the bifurcation of the respective carotid artery. For suppression of swallowing motion artifacts, a pencil beam navigator gating technique, initially developed for cardiac respiratory motion suppression , was applied in which the position of the epiglottis was monitored. Sequence details are provided in Table 1.
For image analysis, the data was transferred to a dedicated CMR image processing workstation (ViewForum, R2.6, Philips Medical Systems). The multi planar reformat tool was applied to define a path along the centerline of the common, external and internal carotid arteries (ACC, ACE and ACI). The endpoints of the paths were identified either at the location where the vessel exceeds the imaging volume or by a drop of the contrast CW between vessel wall and suppressed blood signal below 3 ( ). The length of the vessels was measured individually for the ACC (starting from the carina downwards) and the ACE/ACI (starting from the central ostia upwards).
Comparison of the suggested MSDS technique and the conventional DIR-MSE technique was done after axial reformatting of the MSDS data.
The effectiveness of the blood suppression of the motion sensitizing prepulse in comparison to the standard DIR technique was evaluated using the DIR-GRE and the MSDS data sets, as both sequences are based on nearly identical gradient echo acquisitions. Contrast to noise ratios, normalized to the voxelsize, were determined as follows: CNR norm = (S wall - S lumen )/(stddev lumen * voxelsize). The ROIs were drawn in the vicinity of the carotid bifurcation at corresponding locations to ensure that the coil sensitivity and intrinsic SNR properties were identical.
The performance of the new technique for the reduction of swallowing motion induced artifacts was qualitatively assessed by direct comparison of images acquired with the dual-stack approach with images acquired by the DIR-MSE and DIR-GRE acquisition techniques.
Morphology of the ACC, ACI and ACE were displayed real-time with a high-resolution 9 MHz broadband linear array transducer (iU22 Ultrasound System, Philips Electronics, Bothell, WA, USA). Representative pictures were taken and compared to CMR.
Visual inspection of the endovascular vessel wall revealed sufficient delineation over the entire accessible course of the carotid artery, while the contrast between the epivascular wall and the surrounding tissue was less pronounced, especially in the more distal sections of the ACE and ACI. No distinct motion artifacts (e. g. ghosting) introduced by swallowing could be observed.
The average length of the vessel wall, which could clearly be depicted, was 38.8 +/- 8.1 mm for the ACC, 77.5 +/- 14.6 mm for the ACI and 62.8 +/- 9.4 mm for the ACE. The overall assessable length (ACC + ACI + ACE) resulted to 179.1 +/- 18.1 mm for each of the 24 carotid vessel trees analyzed. The stopping criteria for the path length of the ACC was the drop of contrast and SNR respectively in 100% of the cases since the placement of the coil was restricted by the musculature of the neck. ACE/ACI paths could be followed to the border of the VOI in 83% of the cases.
None of the MSDS images showed swallowing related image artifacts, where for the acquisition of the conventional double-inversion recovery multi-spin echo (DIR-MSE) and respective gradient-echo (DIR-GRE) techniques, navigators had to be used for avoiding severe motion artifacts caused by swallowing. Despite the navigator technique, in 50% of the DIR-MSE and DIR-GRE images motion artifacts were still observable in 50% of the cases.
In this work, a 3D motion-sensitized segmented dual-stack gradient echo (MSDS) technique was developed for fast black blood CMR of large sections of the carotid arteries. The motion-sensitized preparation revealed the potential for providing excellent blood signal suppression and contrast to noise ratios over large segments of the carotid arteries without obvious loss of signal intensity of stationary tissue and enabled dual stack acquisition without prolongation of the image acquisition time. Sufficient wall lumen contrast was achieved in the ACC, the ACE and the ACI enabling a clear delineation of the vessel wall over a mean length 179 +/- 18 mm. The interleaved acquisition of the right and left carotid artery tree by the suggested dual stack approach enabled rapid acquisition of an overview visualization of the carotid artery tree at isotropic spatial resolutions between 0.83 mm3 and 1 mm3 in less than 10 minutes. Compared to the alternative approach of acquiring a single coronal slab, the dual-stack approach is supposed to be less sensitive to motion artifacts caused by swallowing , since the region of the epiglottis is not included in the excited VOI and hence does not contribute to the final image. Cancellation of signal from blood by motion-sensitized preparation clearly holds the potential to generate black blood over large segments of the carotid arteries, which cannot be achieved by the competing DIR technique, since the blood is not totally exchanged over the large volumes addressed in this contribution. Even though, blood in the carotid arteries is mainly flowing in FH direction, best blood suppression was achieved by applying motion sensitizing gradients in all three spatial directions simultaneously.
The most promising application of the proposed MSDS technique is supposed to rise in the field of rapid global assessment of vessel properties and wall thickness in the carotid arteries for rapid identification of those locations along the vessel wall showing local thickening. Here, especially the isotropic spatial resolution, which enables almost lossless reformatting of the data along the centerline is supposed to be beneficial for identification of vessel wall thickening. Due to some limitations in spatial resolution, the proposed technique may not yet be suited to replace the well-established acquisition protocols for plaque characterization published earlier [1, 3] and closer inspection of suspicious locations on the vessel wall with multi-contrast high resolution imaging is required for more detailed plaque analysis.
The application of MSDS for high-resolution imaging may become feasible with the advent of dedicated coils providing improved SNR in combination with better parallel imaging capabilities. Due to the coverage of large segments of the carotid arteries by the MSDS approach, patient motion between subsequent scans will not cause a targeted lesion to leave the VOI, as may be the case in conventional axially aligned DIR-MSE imaging approaches due to the limited coverage of the vessel in feet-head (FH) direction. This property of the MSDS technique may have further advantages with respect to longitudinal monitoring of plaque progression, since the larger covered VOI is supposed to ensure that the location of the lesion is covered completely even in case the VOI is not being planned perfectly the same. Furthermore, registrations based on a large VOI are supposed to be more precise than registrations performed on a small VOI. The suggested motion-sensitized dual-stack approach is not limited to gradient echo techniques and can be combined with other approaches for generating variable image contrast
In conclusion, the 3D MSDS technique appears to be a very promising technique for rapid artifact-free overview imaging of the vessel wall of the carotid arteries with excellent black-blood contrast. Its application to high-resolution multi-contrast imaging is, in principle, possible but remains to be proven.
Supported by the Else Kröner - Fresenius Stiftung (P28/2004)
Unrestricted grant from Takeda Pharmaceutical Company Limited
- Yuan C, Mitsumori LM, Beach KW, Maravilla KR: Carotid atherosclerotic plaque: noninvasive MR characterization and identification of vulnerable lesions. Radiology. 2001, 221: 285-99. 10.1148/radiol.2212001612.View ArticlePubMedGoogle Scholar
- Saam T, Hatsukami TS, Takaya N, Chu B, Underhill H, Kerwin WS, Cai J, Ferguson MS, Yuan C: The vulnerable, or high-risk, atherosclerotic plaque: noninvasive MR imaging for characterization and assessment. Radiology. 2007, 244: 64-77. 10.1148/radiol.2441051769.View ArticlePubMedGoogle Scholar
- Watanabe Y, Nagayama M, Suga T, Yoshida K, Yamagata S, Okumura A, Amoh Y, Nakashita S, Van Cauteren S, Dodo Y: Characterization of atherosclerotic plaque of carotid arteries with histopathological correlation: vascular wall MR imaging vs. color Doppler ultrasonography (US). J Magn Reson Imaging. 2008, 28: 478-85. 10.1002/jmri.21250.View ArticlePubMedGoogle Scholar
- Edelman RR, Chien D, Kim D: Fast selective black blood MR imaging. Radiology. 1991, 181: 655-60.View ArticlePubMedGoogle Scholar
- Yarnykh VL, Yuan C: Multislice double inversion-recovery black-blood imaging with simultaneous slice reinversion. Journal of Magnetic Resonance Imaging. 2003, 17: 478-483. 10.1002/jmri.10278.View ArticlePubMedGoogle Scholar
- Boussel L, Herigault G, de la Vega A, Nonent M, Douek PC, Serfaty JM: Swallowing, arterial pulsation, and breathing induce motion artifacts in carotid artery MRI. J Magn Reson Imaging. 2006, 23: 413-5. 10.1002/jmri.20525.View ArticlePubMedGoogle Scholar
- Luk-Pat GT, Gold GE, Olcott EW, Hu BS, Nishimura DG: High-resolution three-dimensional in vivo imaging of atherosclerotic plaque. Magn Reson Med. 1999, 42: 762-71. 10.1002/(SICI)1522-2594(199910)42:4<762::AID-MRM19>3.0.CO;2-M.View ArticlePubMedGoogle Scholar
- Crowe LA, Gatehouse P, Yang GZ, Mohiaddin RH, Varghese A, Charrier C, Keegen J, Firmin DN: Volume-selective 3D turbo spin echo imaging for vascular wall imaging and distensibility measurement. J Magn Reson Imaging. 2003, 17: 572-80. 10.1002/jmri.10294.View ArticlePubMedGoogle Scholar
- Koktzoglou I, Li D: Diffusion-prepared segmented steady-state free precession: Application to 3D black-blood cardiovascular magnetic resonance of the thoracic aorta and carotid artery walls. J Cardiovasc Magn Reson. 2007, 9: 33-42. 10.1080/10976640600843413.View ArticlePubMedGoogle Scholar
- Koktzoglou I, Li D: Submillimeter isotropic resolution carotid wall MRI with swallowing compensation: imaging results and semiautomated wall morphometry. J Magn Reson Imaging. 2007, 25: 815-23. 10.1002/jmri.20849.View ArticlePubMedGoogle Scholar
- Lin H, Flask CA, Dale BM, Duerk JL: Rapid dark-blood carotid vessel-wall imaging with random bipolar gradients in a radial SSFP acquisition. J Magn Reson Imaging. 2007, 25: 1299-304. 10.1002/jmri.20821.View ArticlePubMedGoogle Scholar
- Crowe LA, Keegan J, Gatehouse PD, Mohiaddin RH, Varghese A, Symmonds K, Cannell TM, Yang GZ, Firmin DN: 3D volume-selective turbo spin echo for carotid artery wall imaging with navigator detection of swallowing. J Magn Reson Imaging. 2005, 22: 583-8. 10.1002/jmri.20424.View ArticlePubMedGoogle Scholar
- Crowe LA, Varghese A, Mohiaddin RH, Yang GZ, Firmin DN: Elimination of residual blood flow-related signal in 3D volume-selective TSE arterial wall imaging using velocity-sensitive phase reconstruction. J Magn Reson Imaging. 2006, 23: 416-21. 10.1002/jmri.20513.View ArticlePubMedGoogle Scholar
- Wang J, Yarnykh VL, Hatsukami T, Chu B, Balu N, Yuan C: Improved suppression of plaque-mimicking artifacts in black-blood carotid atherosclerosis imaging using a multislice motion-sensitized driven-equilibrium (MSDE) turbo spin-echo (TSE) sequence. Magn Reson Med. 2007, 58: 973-81. 10.1002/mrm.21385.View ArticlePubMedGoogle Scholar
- Bornstedt A, Bernhardt P, Hombach V, Kamenz J, Spiess J, Subgang A, Rasche V: Local excitation black blood imaging at 3T: application to the carotid artery wall. Magn Reson Med. 2008, 59: 1207-11. 10.1002/mrm.21590.View ArticlePubMedGoogle Scholar
- Wang Y, Rossman PJ, Grimm RC, Riederer SJ, Ehman RL: Navigator-echo-based real-time respiratory gating and triggering for reduction of respiration effects in three-dimensional coronary MR angiography. Radiology. 1996, 198: 55-60.View ArticlePubMedGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.