Evaluation of intracranial stenoses and aneurysms with accelerated
Thomas A. Hopea,⁎, Michael D. Hopea, Derk D. Purcella, Cornelius von Morzea,
Daniel B. Vignerona, Marcus T. Alleyb, William P. Dillona
aDepartment of Radiology, University of California San Francisco, San Francisco, CA 94143-0628, USA
bDepartment of Radiology, Stanford University, Stanford, CA 94305, USA
Received 6 January 2009; revised 7 March 2009; accepted 10 May 2009
The aim of this study was to evaluate intracranial arterial stenoses and aneurysms with accelerated time-resolved three-dimensional (3D)
phase-contrast MRI or 4D flow. The 4D flow technique was utilized to image four normal volunteers, two patients with intracranial stenoses
and two patients with intracranial aneurysms. In order to reduce scan time, parallel imaging was combined with an acquisition strategy that
eliminates the corners of k-space. In the two patients with intracranial stenoses, 4D flow velocity measurements showed that one patient had
normal velocity profiles in agreement with a previous magnetic resonance angiogram (MRA), while the second showed increased velocities
that indicated a less significant narrowing than suspected on a previous MRA, as confirmed by catheter angiography. This result may have
prevented an invasive angiogram. In the two patients with 4-mm intracranial aneurysm, one had a stable helical flow pattern with a large jet,
while the other had a temporally unstable flow pattern with a more focal jet possibly indicating that the second aneurysm may have a higher
likelihood of rupture. Accelerated 4D flow provides time-resolved 3D velocity data in an 8- to 10-min scan. In the stenosis patients, the
addition of 4D flow to a traditional MRA adds the velocity data provided from transcranial Doppler ultrasound (TCD) possibly allowing for
more accurate grading of stenoses. In the aneurysm patients, visualization of flow patterns may help to provide prognostic information about
future risk of rupture.
© 2010 Elsevier Inc. All rights reserved.
Keywords: Phase contrast; Velocity; Intracranial aneurysm; Intracranial stenosis; 4D Flow
Evaluation of intracranial arterial blood velocity data is
important for the understanding and characterization of
many pathological processes including arterial stenoses and
aneurysm formation and rupture. Time-resolved 3D phase-
contrast MRI (4D flow) allows for the acquisition of
dynamic, multidirectional blood velocity data and has
previously been used to image intracardiac flow and thoracic
aorta flow patterns [1–5]. The technique has recently been
used to visualize flow in silicone models of intracranial
aneurysms and for intracranial flow evaluation [6–10].
Currently, transcranial Doppler ultrasound (TCD) is the
primary tool for intracranial velocity measurements and has
been used in the evaluation of intracranial arterial stenoses
. Advantages of 4D flow over TCD for intracranial blood
flow evaluation include not being limited by acoustic
windows for data acquisition and allowing for 4D visualiza-
tion of complex blood flow patterns and calculation of
secondary vascular parameters such as wall shear stress
(WSS). Phase-contrast MR is the gold standard for
measuring cerebral blood flow and has been shown to be
more accurate than US measurements . For velocity
measurements, US has shown good agreement with phase-
contrast MR measurements, and studies in pulsatile models
of stenoses have shown good correlation between intravas-
cular US and phase contrast [13,14].
The main clinical limitation of 4D flow is the long scan
time required for acquiring data sets with reasonable spatial
Available online at www.sciencedirect.com
Magnetic Resonance Imaging 28 (2010) 41–46
⁎Corresponding author. Tel.: +1 415 476 8358.
E-mail address: firstname.lastname@example.org (T.A. Hope).
0730-725X/$ – see front matter © 2010 Elsevier Inc. All rights reserved.
and temporal resolution. We have used a parallel imaging
technique, generalized autocalibrating partially parallel
reconstruction (GRAPPA), in combination with a compli-
mentary approach to k-space subsampling that omits the
corners of k-space [8,15,16]. With the combination of these
techniques, we are able to keep the scan time under 10 min.
With intracranial stenoses, magnetic resonance angio-
grams (MRAs) and TCD have low positive predictive
values for determining the presence of a stenosis, although
they are useful for excluding patients without stenoses
. In order to determine whether a narrowing is
considered stenotic, TCD uses velocity cutoffs . It is
hypothesized that the combination of velocity data
provided by 4D flow with the anatomic imaging provided
by an MRA will aid in the diagnosis of these lesions
limiting the need for invasive angiography.
In regard to intracranial aneurysms, although the largest
trial to date states that aneurysms less than 6 mm in size
are at low risk to rupture, multiple studies indicate that
there is a real risk of rupture with these aneurysms [18–
21]. Aneurysms greater than 6 mm in size will often be
treated either by surgical clipping or interventional
placement of coils, while aneurysms less than 6 mm in
size are observed . As more aneurysms are found and
followed that are less than 6 mm in size with the increase
in imaging studies, methods to evaluate patient-specific
risk of rupture of these smaller aneurysms are needed. By
leveraging knowledge from computational fluid data (CFD)
findings, it is hypothesized that 4D flow imaging of flow
patterns will help to risk stratify patients with aneurysms
less than 6 mm.
In this study, we evaluated velocity fields using 4D
flow in four patients, two with intracranial stenoses and
two with aneurysms.
2. Materials and methods
All procedures were approved by our local institutional
review board, and informed consent was obtained for all
patients. Time-resolved 3D PC-MRI was used to assess
neurovascular velocities in four patients (Table 1) with
intracranial pathology at 3.0T magnetic field strengths
(Signa CV/i; GE Healthcare, Milwaukee, WI, USA; Gmax,
40 mT/m; rise time, 268 μs) . GRAPPA with an
acceleration factor of 2 was utilized . The corners of
k-space, as defined by the ellipse πkzmaxkymax, were not
acquired [16,24]. An eight-channel head coil was used. Pulse
oximetry was used for cardiac gating, and each data set was
interpolated over 20 time frames. The following parameters
were used: field of view 162×180 mm, a fractional phase
FOVof 90% (rFOV), matrix size of 162×300×30, flip angle
15° and receiver bandwidth of ±65 kHz. Avelocity encode of
125 cm/s was used along each axis. Three slice encodes were
repeated in each RR interval, resulting in a temporal
resolution (TRes) of 65 ms (TRes=12 TR). Prior to analysis
of the image data, corrections were performed for Maxwell
phase effects, eddy currents and errors due to gradient field
Female 4-mm aneurysm at the left MCA trifurcation
Female 4-mm aneurysm at the right MCA
Angulation and narrowing of the right MCA
Narrowing of the right MCA
Fig. 1. Images from a 73-year-old man with angulation and narrowing of the
right MCA just distal to the bifurcation of the internal carotid artery. (A)
Axial computed tomography angiogram (CTA) maximum intensity projec-
tion image (MIP) depicting a sharp angulation and narrowing at the location
of the ICA bifurcation. Note that there is no right A1 segment. (B) 4D flow
image of the bilateral ICA and MCAs. Streamlines show an increase in
velocity at the area of narrowing (arrow) compared to the contralateral side.
(C) Measurements of mean velocity across the vessel lumen show a minimal
increase in velocities at the stenosis in agreement with the MRA.
42T.A. Hope et al. / Magnetic Resonance Imaging 28 (2010) 41–46
To visualize complex flow patterns, streamlines and
vector fields were created using a 3D visualization software
(EnSight, CEI, Inc., Apex, NC, USA). Streamlines are
imaginary lines that are aligned with the local velocity vector
field and represent flow at a specific point in time. To analyze
the mean velocity across the vessel lumen, planes were
extracted from the data sets and segmented using a
proprietary program (Aspire2, Stanford, CA, USA).
All patients were successfully imaged using 4D flow, with
scan times ranging from 8 to 10 min per acquisition. The first
patient had a focal narrowing of the middle cerebral artery
(MCA) just distal to the bifurcation of the internal carotid
artery (Fig. 1). The mean velocity and peak mean velocity at
the stenosis were 29 and 42 cm/s, respectively. From the
streamlines, an elevation of local velocities can be seen, but
velocity measurements in the vessel agree with the computed
tomography angiogram that there was no significant
stenosis. The second patient had a narrowing of the proximal
MCA that is evident in the streamline images (Fig. 2). The
original MRA showed a severe stenosis; however, a
subsequent catheter angiogram showed that the narrowing
was less than 50%. From the 4D flow data, the calculated
mean velocity across the vessel lumen showed an increase in
peak velocity at the stenosis with a slowing of velocities
distally when compared to the normal contralateral side
(Fig. 1). The mean velocity of the flow through the stenosis
was 51 cm/s with a peak mean velocity of 76 cm/s, compared
to 29 and 43 cm/s on the contralateral side.
The third patient had a 4-mm aneurysm at the right
MCA trifurcation. Vector field images show a large jet
impacting on the lateral wall of the aneurysm creating a
helical flow pattern in the aneurysm sac that is stable
throughout the cardiac cycle (Fig. 3). The fourth patient had
a 4-mm aneurysm at the left MCA trifurcation. Streamline
images show a smaller jet impinging on the lateral wall of
the aneurysm and a helical flow pattern displaced from the
center of the aneurysm that was not stable throughout the
cardiac cycle (Fig. 4).
Time-resolved 3D phase-contrast MRI (4D flow)
provides time-resolved 3D velocity data sets that were
used in this study to evaluate intracranial stenoses and
aneurysms. The accelerated technique shorted scan time
from 25–30 min reported previously to 8–10 min in the
four patients imaged .
The two patients with stenoses illustrate the utility of 4D
flow velocity measurements. The first patient would have
been appropriately excluded based on measurements of both
velocity and vessel diameter. In the second patient, the vessel
diameter determined by MRA showed a critical stenosis and
the patient subsequently went on to catheter angiography,
which did not show a stenotic lesion. Catheter angiography
is the gold standard for determining the degree of stenosis,
but it is associated with a small but significant risk of severe
neurologic deficit as well as other vascular complications
Fig. 2. Images from a 56-year-old man with a proximal right MCA
narrowing (arrow). (A) Axial MRA MIP depicting right MCA stenosis. (B)
Subsequentangiogramdepictingthe right MCA narrowing of less than50%.
(C) Streamlines of flow through the left and right MCA. There is an increase
in velocity at the stenosis in the right MCA compared to the contraleteral
side (arrow). (D) Close-up view of the right MCA showing an abrupt
decrease in velocity just distal to the stenosis and turbulent flow at the
stenosis. (E) Measurement of mean velocity across the vessel lumen shows
that velocities are increased at the stenosis compared to the contralateral side
and drop by greater than 50% distal to the stenosis.
43T.A. Hope et al. / Magnetic Resonance Imaging 28 (2010) 41–46
including arterial dissection. In the second patient, depend-
ing on what criteria are used for 4D flow velocity data, the
4D flow measurements may have disagreed with the MRA-
determined diameter and indicated that there was not a
stenosis, preventing the need for catheter angiography. In
patients with intracranial stenoses, measurement of average
velocity across the vessel lumen, which 4D flow provides,
may provide a better noninvasive alternative to TCD,
especially when combined with a traditional MRA. None-
theless, further validation of the 4D flow measurements is
needed to determine the appropriate velocity criteria for
intracranial stenoses as TCD-based velocity criteria may not
be appropriate for 4D flow velocities.
With intracranial aneurysms, the clinical question
revolves around flow patterns as compared to velocity
measurements. Previously, there has been no technique
available for imaging three-dimensional flow patterns in
vivo. In the late 1990s, work by Imbesi and Kerber 
using slipstream analysis in silicone models showed the
existence of high-velocity jets that impacted on the wall of
aneurysms where eventual rupture occurred. CFD analysis of
patients with larger intracranial aneurysms indicated that
small velocity jets with complex unstable flow patterns are
seen in ruptured aneurysms .
In the third patient, a large jet enters the aneurysm
creating a single large helical pattern that persists through-
out the cardiac cycle. The fourth patient has a smaller jet
creating a temporally unstable helical flow pattern along the
Fig. 4. Images from a 63-year-old woman with a 4-mm aneurysm at the right
MCA trifurcation. (A) Axial MIP from a CTA through the aneurysm. The
arrow points to the MCA proximal to the trifurcation. (B) Streamlines show
a small high-velocity jet entering the aneurysm and a helical flow pattern
along the medial wall that is displaced from the center of the aneurysm
(arrowhead). (C–F) Vector fields on planes placed through the center of the
aneurysm at different points in the cardiac cycle (C — 15%, D — 35%, E —
55% and F — 75% of the cardiac cycle). The helical flow pattern is short
lived and does not encompass the entire aneurysm.
Fig. 3. Images from a 50-year-old woman with a 4-mm aneurysm at the left
MCA trifurcation. (A) Axial MIP from a CTA through the aneurysm. (B)
Representative streamlines through the MCA bifurcation in a normal
volunteer. Note the absence of helical flow. (C) In the patient, streamlines
through the left MCA trifurcation and aneurysm depicting formation of a
helical flow pattern that encompasses the entire aneurysm. (D–F) Vector
fields superimposed on axial planes placed through the aneurysm showing
helical flow in the center of the aneurysm at three different time points in the
cardiaccycle (D — 15%,E — 35%, F — 55% of the cardiac cycle).There is
a large high-speed jet entering the aneurysm from the MCA (arrow) creating
a stable helical flow pattern throughout the cardiac cycle.
44 T.A. Hope et al. / Magnetic Resonance Imaging 28 (2010) 41–46
medial wall of the aneurysm. Based on the prior CFD
results, when comparing these two patients, one might
recommend closer follow-up in the fourth patient as it likely
has a higher risk of rupture. As the use of 4D flow becomes
more common, further work needs to be done to determine
which flow patterns are correlated with aneurysm progres-
sion and rupture.
CFD techniques have also been used to calculate WSS
and intraluminal pressure. Low WSS has been associated
with aneurysmal growth and rupture, while intraluminal
pressure variation has not been associated with rupture
[31,32]. Data from 4D flow can be used to calculate these
secondary parameters, and basic WSS calculations have
already been performed on 4D flow data . With the
development of better segmentation algorithms, patient-
specific WSS analysis will be possible without the use
Limitations to the present study include the small
number of patients and lack of a gold standard comparison.
Additionally, the 4D flow technique has an inherently low
temporal and spatial resolution. The temporal resolution of
65 ms may limit the ability to visualize short-lived flow
patterns and lead to the underestimation of peak velocities.
The data is also a temporal average of multiple heartbeats
and does not provide beat-by-beat analysis of flow. The
limited spatial resolution may limit the visualization of
complex flow features due to partial volume effects. Lastly,
the high-velocity encode chosen in this study will limit the
sensitivity for low-velocity flow, although it has been
previously noted that a higher velocity encode does not
affect the recognition of helical flow patterns to a great
Accelerated 4D flow can provide accurate velocity
information about intracranial pathology, specifically ste-
noses and aneurysms, in a 10-min scan time. Future studies
are required to determine the clinical significance of 4D flow
findings in both these pathologies. There is promise that the
combination of 4D flow-acquired velocity data with an MRA
may aid in the risk stratification of patients with intracranial
stenoses and that flow profiles based on 4D flow data may be
able to help determine which intracranial aneurysms have a
higher risk of rupture.
 Markl M, Draney MT, Miller DC, Levin JM, Williamson EE, Pelc NJ,
et al. Time-resolved three-dimensional magnetic resonance velocity
mapping of aortic flow in healthy volunteers and patients after valve-
sparing aortic root replacement. J Thorac Cardiovasc Surg 2005;130
 Fyrenius A, Wigstrom L, Ebbers T, Karlsson M, Engvall J, Bolger AF.
Three dimensional flow in the human left atrium. Heart 2001;86(4):
 Kozerke S, Hasenkam JM, Pedersen EM, Boesiger P. Visualization of
flow patterns distal to aortic valve prostheses in humans using a fast
approach for cine 3D velocity mapping. J Magn Reson Imaging 2001;
 Kilner PJ, Yang GZ, Wilkes AJ, Mohiaddin RH, Firmin DN, Yacoub
MH. Asymmetric redirection of flow through the heart. Nature 2000;
 Hope TA, Markl M, Wigstrom L, Alley MT, Miller DC, Herfkens RJ.
Comparison of flow patterns in ascending aortic aneurysms and
volunteers using four-dimensional magnetic resonance velocity
mapping. J Magn Reson Imaging 2007;26(6):1471–9.
 Yamashita S, Isoda H, Hirano M, Takeda H, Inagawa S, Takehara Y,
et al. Visualizationof hemodynamicsin intracranial arteriesusingtime-
resolved three-dimensional phase-contrast MRI. J Magn Reson
 Moftakhar R, Aagaard-Kienitz B, Johnson K, Turski PA, Turk AS,
Niemann DB, et al. Noninvasive measurement of intra-aneurysmal
pressure and flow pattern using phase contrast with vastly under-
sampled isotropic projection imaging. AJNR Am J Neuroradiol 2007;
 Bammer R, Hope TA, Aksoy M, Alley MT. Time-resolved 3D
quantitative flow MRI of the major intracranial vessels: initial
experience and comparative evaluation at 1.5T and 3.0T in combina-
tion with parallel imaging. Magn Reson Med 2007;57(1):127–40.
 Isoda H, Hirano M, Takeda H, Kosugi T, Alley MT, Markl M, et al.
Visualization of hemodynamics in a silicon aneurysm model using
time-resolved, 3D, phase-contrast MRI. AJNR Am J Neuroradiol
 Hope MD, Purcell DD, Hope TA, von Morze C, Vigneron DB, Alley
MT, et al. Complete intracranial arterial and venous blood flow
evaluation with 4D flow MR imaging. AJNR Am J Neuroradiol 2009;
 Feldmann E, Wilterdink JL, Kosinski A, Lynn M, Chimowitz MI,
Sarafin J, et al. The Stroke Outcomes and Neuroimaging of
Intracranial Atherosclerosis (SONIA) trial. Neurology 2007;68(24):
 Lee VS, Spritzer CE, Carroll BA, Pool LG, Bernstein MA, Heinle SK,
et al. Flow quantification using fast cine phase-contrast MR imaging,
conventional cine phase-contrast MR imaging, and Doppler sonogra-
phy: in vitro and in vivo validation. AJR Am J Roentgenol 1997;169
 Hoppe M, Heverhagen JT, Froelich JJ, Kunisch-Hoppe M, Klose KJ,
Wagner HJ. Correlation of flow velocity measurements by magnetic
resonance phase contrast imaging and intravascular Doppler ultra-
sound. Invest Radiol 1998;33(8):427–32.
 Zananiri FV, Jackson PC, Halliwell M, Harris RA, Hayward JK,
Davies ER, et al. A comparative study of velocity measurements in
major blood vessels using magnetic resonance imaging and Doppler
ultrasound. Br J Radiol 1993;66(792):1128–33.
 Griswold MA, Jakob PM, Heidemann RM, Nittka M, Jellus V, Wang J,
et al. Generalized autocalibrating partially parallel acquisitions
(GRAPPA). Magn Reson Med 2002;47(6):1202–10.
 Bernstein MA, Fain SB, Riederer SJ. Effect of windowing and zero-
filled reconstruction of MRI data on spatial resolution and acquisition
strategy. J Magn Reson Imaging 2001;14(3):270–80.
 Stroke Outcome and Neuroimaging of Intracranial Atherosclerosis
(SONIA): design of a prospective, multicenter trial of diagnostic tests.
 Wiebers DO, Whisnant JP, Huston 3rd J, Meissner I, Brown Jr RD,
Piepgras DG, et al. Unruptured intracranial aneurysms: natural history,
clinical outcome, and risks of surgical and endovascular treatment.
 Juvela S, Porras M, Poussa K. Natural history of unruptured
intracranial aneurysms: probability of and risk factors for aneurysm
rupture. J Neurosurg 2000;93(3):379–87.
 Juvela S, Porras M, Heiskanen O. Natural history of unruptured
intracranial aneurysms: a long-term follow-up study. J Neurosurg
 Weir B, Disney L, Karrison T. Sizes of ruptured and unruptured
aneurysmsin relation to theirsites andthe agesof patients. J Neurosurg
45T.A. Hope et al. / Magnetic Resonance Imaging 28 (2010) 41–46
 Brisman JL, Song JK, Newell DW. Cerebral aneurysms. N Engl J Med
 Markl M, Chan FP, Alley MT, Wedding KL, Draney MT, Elkins CJ,
et al. Time-resolved three-dimensional phase-contrast MRI. J Magn
Reson Imaging 2003;17(4):499–506.
 Hope T, Hope M, Bammer R, Alley M. Combination of parallel
imaging and a cut-corner acquisition for neurovascular 4D-flow.
Proceedings 16th Scientific Meeting, International Society for
Magnetic Resonance in Medicine; 2008. p. 1984. Toronto.
 Walker PG, Cranney GB, Scheidegger MB, Waseleski G, Pohost GM,
Yoganathan AP. Semiautomated method for noise reduction and
background phase error correction in MR phase velocity data. J Magn
Reson Imaging 1993;3(3):521–30.
 Bernstein MA, Zhou XJ, Polzin JA, King KF, Ganin A, Pelc NJ, et al.
Concomitant gradient terms in phase contrast MR: analysis and
correction. Magn Reson Med 1998;39(2):300–8.
 Markl M, Bammer R, Alley MT, Elkins CJ, Draney MT, Barnett A,
et al. Generalized reconstruction of phase contrast MRI: analysis and
correction of the effect of gradient field distortions. Magn Reson Med
 Meckel S, Stalder AF, Santini F, Radu EW, Rufenacht DA, Markl M,
et al. In vivo visualization and analysis of 3-D hemodynamics in
cerebral aneurysms with flow-sensitized 4-D MR imaging at 3 T.
 Imbesi SG, Kerber CW. Analysis of slipstream flow in two ruptured
intracranial cerebral aneurysms. AJNR Am J Neuroradiol 1999;20(9):
 Cebral JR, Castro MA, Burgess JE, Pergolizzi RS, Sheridan MJ,
Putman CM. Characterization of cerebral aneurysms for assessing risk
of rupture by using patient-specific computational hemodynamics
models. AJNR Am J Neuroradiol 2005;26(10):2550–9.
 Jou LD, Wong G, Dispensa B, Lawton MT, Higashida RT, Young WL,
et al. Correlation between lumenal geometry changes and hemody-
namics in fusiform intracranial aneurysms. AJNR Am J Neuroradiol
 Shojima M, Oshima M, Takagi K, Torii R, Hayakawa M, Katada K,
et al. Magnitude and role of wall shear stress on cerebral aneurysm:
computational fluid dynamic study of 20 middle cerebral artery
aneurysms. Stroke 2004;35(11):2500–5.
 Kvitting JP, Ebbers T, Wigström L, Engvall J, Olin CL, Bolger AF.
Flow patterns in the aortic root and the aorta studied with time-
resolved, 3-dimensional, phase-contrast magnetic resonance imaging:
implicationsfor aorticvalve-sparingsurgery.J ThoracCardiovascSurg
46T.A. Hope et al. / Magnetic Resonance Imaging 28 (2010) 41–46