Techniques

Intracranial Arterial System

Currently, the arterial circulation is most often studied using 3D time-of-flight (TOF) MRA. This non-invasive technique allows the visualization of the major intracranial arteries and peripheral branches in a relatively short time and generally does not require the use of contrast agent [ 18]. The principal advantages of the technique are that the overall acquisition time is shorter than those of phase contrast (PC) techniques, a single slice thickness of 1 mm or less can be obtained, a matrix of 512x512 can be acquired within a reasonable time, and, finally, the arteries can be imaged selectively due to the elevated saturation of the veins [19]. Its major limitation, however, is that the distal arterial branches are often less optimally visualized on MIP reconstructions due to the progressive saturation of these vessels during image acquisition [19]. Furthermore, intracranial arteries are often tourtuos and course predominantly within the imaging volume, thereby experiencing repetitive RF pulses and saturation effects. Saturation effects can sometimes be reduced on 3D TOF MRA by optimizing the scan parameters (TR, FA, TE, acquisition volume and scan plane) for preservation of arterial signal or by reducing background noise.

Numerous refinements have been implemented during the last ten years in order to improve the 3D TOF technique. Early developments included the Multiple Overlapping Thin Slab Acquisition (MOTSA) sequence which combines the advantages of both 2D and 3D TOF acquisitions while reducing saturation effects [20, 21]. Among the benefits of MOTSA sequences are that they permit evaluation of the cerebral arterial vessels from the intrapetrous segment of the ICA and the origin of the PICA from the vertebral arteries to the distal branches of the middle, anterior and posterior cerebral arteries. However, a disadvantage of MOTSA is that the acquisition time increases proportionally with the number of slabs required to cover the anatomic region of interest. A subsequent improvement on 3D MOTSA was achieved with the development of the Sliding-INterleaved kY (SLINKY) acquisition sequence which equalizes flow-related signal intensity across the entire slab dimension and thereby eliminates slab boundary artifact (SBA), also called "Venetian" blind artifact [22].

Since the flip angle (FA) also influences the degree of saturation, there is always an optimal value for this variable which is correlated with the blood flow velocity. In the intracranial circulation the ideal FA depends on the segment to be imaged since the blood flow velocity varies greatly from the carotid siphon to the proximal portion of the cerebral arteries and the most distal branches. To overcome this problem, a technique known as Tilted Optimized Non-saturating Excitation (TONE) was introduced in 1995 and is now routinely used in 3D TOF intracranial MRA [23,24].A further im

Table 4. Suggested parameters for 3D TOF MRA of intracranial arteries

3D TOF MRA

Coil

Head

Patient Positioning

Supine - head first

Saturation band

Superior

Sequences

3D TOF GRE (FISP, GRASS, FFE)

Adjunctive techniques

MOTSA, TONE, MTC

Sequence orientation

axial - bi-commissural line

TR, TE, flip angle

20-25 ms, 2-5 ms (flow comp), 20°-30°

Matrix

256 x 512, phase direction LR

FOV, slab, slice thick

220 x 250 mm2, 80 mm, 0.75 -1.2 mm (no gap) (contiguous)

Voxel size

0.8 x 0.5 x 0.75 -1.2 mm3

Acquisition time

6-8 minutes

Landmarks for slab position

Use scouts to cover anatomy

Image subtraction

No

Evaluation of images

Source, MIP (optional SD or VRT)

provement of the SNR between mobile and stationary spins is attainable by applying Magnetization Transfer Contrast (MTC) to 3D TOF MRA sequences of the intracranial circulation [25]. The typical parameters employed for 3D TOF MRA of the intracranial arteries are listed in Table 4.

Despite the technical advantages of 3D TOF MRA, a major limitation is the saturation of slow spins in the more distal arterial branches with consequent reduction of diagnostic accuracy. 2D TOF acquisitions can partly overcome this limitation, but these sequences lack 3D spatial resolution and are thus of little practical value for the visualization of intracranial arteries.

At present there are two possibilities to overcome the limitations of TOF MRA techniques: increased magnetic fields and the uses of contrast agent - enhanced acquisitions.

Creasy et al. [26] were among the first to describe the possibility of increasing the quality of MRA images of the intracranial circulation by combining the 3D TOF MRA technique with the intravenous infusion of a gadolinium contrast agent. Although this and other early studies revealed that visualization of the more distal portions of the intracranial arterial circulation was improved on contrast-enhanced (CE) 3D MRA, the relatively long acquisition times (8 to 10 minutes) of these early studies and rapid distribution of contrast agent led to visualization of both cerebral arteries and veins [27-29]. Additionally, the presence of contrast agent in the capillary-venous compartment was shown to lead to increased signal intensity of the stationary encephalic tissue, particularly of the intracranial structures that lack a blood-brain-barrier (e.g., choroid plexus and pituitary gland) [27,29]. Because of the superimpo-

sition of arterial and venous structures, as well as non-vascular structures, arterial illustration using these early 3D CE MRA techniques was poor and image interpretation was overly complicated [21, 30]. Therefore, this technique did not replace un-enhanced 3D TOF MRA in routine practice despite the advantage of greater sensitivity for peripheral portions of the intracranial arterial vessels.

Recently, with the introduction of stronger gradients and the development of ultrafast sequences (acquisition times of just 20-50 seconds), 3D CE MRA has been reconsidered for the diagnosis of intracranial vascular pathologies [29,31]. The technique, which requires a rapid intravenous injection of paramagnetic contrast agent, has been shown to be more sensitive and more selective then previous techniques. When scan times are in the order of seconds, an almost exclusive visualization of the intracranial arteries without visualization of the venous vessels can be achieved. These results are possible due to recent developments in K space sampling in which the central portion of K space is the first to be acquired. This reduces the overall acquisition time for arterial vessels where the contrast agent is initially concentrated [32-34].

As in 3D CE MRA of other arterial territories, the underlying principle for 3D CE MRA of the intracranial arterial circulation is the "first pass" T1-shortening effect of circulating contrast agent during the filling of the central portions of K space. However, a fundamental difference between the in-tracranial arterial circulation and other arterial regions is that the arterial-venous time interval in the brain is extremely short, typically only 4 to 6 seconds [35]. Thus, ultrafast sequences and more rigorous acquisition parameters are required in

Table 5. Suggested parameters for 3D CE MRA of intracranial arteries

3D CE MRA

Coil

Head

Patient Positioning

Supine - head first

CM dose - flow rate

0.01 mMol (Kg/bw) - 2-4 ml/sec

Bolus Timing

CM dose 2 ml - flow rate 2-4 ml/sec

Saline solution flush

15-20 ml/same flow rate of CM

Sequences

3D GRE T1 weighted (FLASH, SPGR, FFE)

Sequence orientation

axial - bi-commissural plane

TR, TE, flip angle

3-5 ms, 1-2 ms, 30°-40°

Matrix

192 x 256 (512), phase direction LR

FOV, slab, slice thick

220 x 250 mm2, 60 mm, ~1 mm (no gap)

Voxel size

1.0 x 1.0 x 1.0- 1.5 mm3

Acquisition time

20-40 sec

Landmarks for slab position

Use scouts to cover anatomy

Dynamic evaluation

4 volumes each lasting ~ 10 sec, without delay

Image subtraction

Yes

Evaluation of images

Source, MIP (optional SD or VRT)

order to obtain diagnostically useful MRA images [28,29,31].

As in other vascular territories, there are various examination factors that are critical for optimization of the 3D CE MRA examination. These factors include: optimization of scan parameters, modality of contrast agent injection (timing and quantity), relaxivity of the contrast agent, and post-processing. For most 3D CE MRA examinations of the intracranial arterial circulation, the choice of which parameters to select depends on whether the application requires high contrast and good spatial resolution or "time resolved" acquisitions at lower spatial resolution [34]. Several techniques have been proposed that favor either spatial or temporal resolution.

The 3D CE MRA techniques that favor spatial resolution are similar to the T1 weighted GE sequences and acquisition parameters used for CE MRA of the thoracic aorta and arch vessels. Ellipti-cally centric encoded acquisition must be timed precisely to the arrival of contrast agent in the circle of Willis in order to minimize enhancement of the venous components. Although the overall scan time ranges from 20 to 50 seconds, with elliptical centric phase ordering the center of k-space is filled very efficiently during the initial seconds of the scan enabling the selective visualization of major intracranial arteries without significant venous contamination (Table 5). When a phased array head coil is used, parallel imaging techniques such as SENSE (SENSitivity Encoding [36]) or SMASH (SiMultaneous Acquisition of Spatial Harmonics [37]), permit both a shorter scan time and a high er spatial resolution or coverage.

When the TR is short enough and a 192x265 matrix is used, dynamic "time resolved" 3D CE MRA can be performed of most vascular territories with a time resolution of 3-6 seconds for each 3D volume acquisition [31]. However, for intracra-nial arterial studies this temporal resolution is barely sufficient because of the extremely short interval between the arterial and venous phases. When "time resolved" 3D CE MRA or Time Resolved Imaging of Contrast Kinetics (TRICKS [31]) techniques are used for intracranial imaging, the entire dose of contrast agent must be administered as a rapid bolus during the period between the beginning of the arterial phase and the beginning of the venous phase. Therefore, a preliminary bolus test must be performed to determine the arterial delay, the venous delay, and the resulting time interval even though, due to the rapidity of the intracranial circulation, it is sometimes difficult to separate arteries from veins.

To improve time resolution, a 2D thick-slice MR digital subtraction angiography (2D MR DSA) technique can be used instead of 3D CE MRA. This method combines a series of 2D CE thick slices, each frame lasting 1-2 seconds, with subtraction of pre-contrast images from subsequent contrast-enhanced images to allow greater background tissue suppression (Fig. 23a-g). 2D MR DSA requires one or, when multiple "projections" are necessary, two or more rapid intravenous bolus injections of contrast agent. When multiple injections are needed, image subtraction prevents contamination from previously enhanced vessels (Table 6).

Tablet Dislessia
Table 6. Suggested parameters for 2D CE MRA (MR DSA) of intracranial arteries and veins

2D MR DSA

Coil

Head

Patient Positioning

Supine - head first

CM dose/flow rate

0.1 mMol (Kg/bw) - 4 ml/sec

Bolus Timing

Not required

Saline solution flush

15-20 ml/same flow rate of CM

Sequences

2D GRE T1 weighted (FLASH, SPGR, FFE)

Sequence orientation

sagittal

TR, TE, flip angle

5 ms, 1-2 ms, 40°

Matrix

256 x 256, phase direction AP

FOV, slice thick

220 x 250 mm2, 60 mm

Acquisition time

1-2 sec for each frame/ 60-30 frames

Landmarks for slab position

Use scouts to cover anatomy

Image subtraction

Yes (automatically subtracted)

Intracranial Venous System

The intracranial venous system is a complex, often asymmetric, system of vessels. MRA was the first non-invasive technique that enabled depiction of the intracranial venous system without using ionizing radiation. Both 2D TOF and 2D-3D PC sequences are available to study the cerebral venous circulation [38].

PC MRA is based on the accumulated phase difference between mobile spins and stationary spins. This characteristic renders PC acquisition more sensitive to slow flow, such as occurs in veins [39]. An appropriate velocity encoding (VENC) (Tables 7 and 8) must be selected based on the velocity of blood within the blood vessel to be imaged [14]. Unlike TOF MRA, PC MRA is not affected by saturation effects. In addition, contrast agents can improve the vascular signal on PC MRA enabling depiction of smaller venous vessels by increasing the signal intensity and hence the spatial resolution achievable [40]. For this reason PC MRA is usually the last sequence acquired during a standard contrast enhanced examination of the brain.

Usually 3D PC MRA is performed in preference to 2D PC MRA for studies of cerebral veins because of the higher spatial resolution and larger coverage achievable. However, 2D PC MRA may also be used to determine blood flow direction and as a pilot study to determine the most appropriate VENC for a definitive 3D PC MRA study. The main disadvantages of 3D PC MRA are the longer acquisition times, which can be overcome in part by the use of parallel imaging technology, and the dependence of the technique on the correct choice of VENC. An inappropriate choice may result in a representation of false stenosis.

2D TOF MRA may also be used to image cerebral veins. With this approach a presaturation band must be collocated below the acquisition to saturate arterial flow and to prevent saturation effects during the TOF acquisition. Usually, a slight oblique coronal plane is chosen to cut perpendicularly the major dural sinuses. As with PC MRA acquisitions, contrast agents have been proposed to increase the signal and to reduce the saturation effect. Unfortunately, 2D TOF MRA has a lower spatial resolution and lower background suppression resulting in an overall lower image quality compared to that achievable with 3D PC MRA [38].

Recently, high resolution 3D fast GE T1 weighted sequences (e.g. MPRAGE, FSPGR, and EPI-FFE) acquired before and during slow administration of gadolinium contrast agent have been proposed.An-other suggested protocol involves triggering the sequence with the arrival of contrast agent and fixing a delay of 8-10 seconds to acquire images at the point of maximal venous contrast concentration. The advantages of this technique include a panoramic and consistent visualization of the in-tracranial venous system. Some protocols propose image subtraction to reduce the signal intensity from arteries which is already very high on acquisitions without contrast agent and does not increase significantly following contrast agent administration. The only inconvenience of this technique is that it doubles the acquisition time, thereby increasing the probability of patient movement and, in consequence, rendering image subtraction futile.

Whichever technique is chosen, the patient should be made as comfortable as possible before initiating the study. For example, ear plugs should be used because the MRA sequences (especially 3D CE MRA sequences) are notably noisier than conventional techniques. When a contrast agent is required, an intravenous line should be connected to

Table 8. Suggested parameters for 2D PC MRA of intracranial venous system

Table 7. Suggested parameters for 3D PC MRA of intracranial circulation

3D PC MRA

Coil

Head

Patient Positioning

Supine - head first

Saturation band

Inferior

Sequences

3D GRE T1 weighted (FLASH, SPGR, FFE)

VENCs

70-80 cm/sec for arteries; 10-20 cm/sec for veins

Sequence orientation

axial - bi-commissural line

TR, TE, flip angle

80 ms, 10 ms, 20°-30°

Matrix

256 x 256, phase direction LR

FOV, slab, slice thick

200 x 250 mm2, 80 mm, 1 -1.5 mm (no gap) (contiguous)

Voxel size

0.8 x 0.9 x 1 -1.5 mm3

Acquisition time

10-12 minutes

Landmarks for slab position

Use scouts to cover anatomy

Image subtraction

Yes (automatically subtracted)

Evaluation of images

Source, MIP (optional SD or VRT)

Table 8. Suggested parameters for 2D PC MRA of intracranial venous system

2D PC MRA

Coil

Head

Patient Positioning

Supine - head first

Sequences

2D GRE T1 weighted (FLASH, SPGR, FFE)

VENCs

15-20 cm/sec

Sequence orientation

variable

TR, TE, flip angle

80 ms, 10 ms, 20°-30°

Matrix

256 x 256, phase direction LR

FOV, slab, slice thick

200 x 250 mm2, 80 mm, 1 -1.5 mm (no gap) (contiguous)

Voxel size

0.8 x 0.9 x 20 mm3

Acquisition time

2-3 minutes

Landmarks for slab position

Use scouts to cover anatomy

Image subtraction

Yes (automatically subtracted)

a power injector prior to initiating the examination in order to avoid repositioning the patient. The patient's head must be securely but comfortably immobilized and the patient should be asked to remain as still as possible during the examination. The field of view should be positioned to include the entire internal portion of the cranium in order to avoid aliasing or back-folding artifacts which significantly reduce SNR.

An important point that must be kept in mind is that MRA is not a substitute for basic brain studies, but is frequently a useful corollary examination.

Was this article helpful?

0 0
Essentials of Human Physiology

Essentials of Human Physiology

This ebook provides an introductory explanation of the workings of the human body, with an effort to draw connections between the body systems and explain their interdependencies. A framework for the book is homeostasis and how the body maintains balance within each system. This is intended as a first introduction to physiology for a college-level course.

Get My Free Ebook


Post a comment