Perspectives in Medicine (2012) 1, 3033

Bartels E, Bartels S, Poppert H (Editors):

New Trends in Neurosonology and Cerebral Hemodynamics an Update.
Perspectives in Medicine (2012) 1, 3033

journal homepage: www.elsevier.com/locate/permed

Update on ultrasound brain perfusion imaging

Gnter Seidel

Department of Neurology, Asklepios Klinik Hamburg Nord Heidberg, Tangstedter Landstrasse 400, D-22417 Hamburg, Germany

KEYWORDS Summary Several studies have demonstrated the value of ultrasound perfusion imaging to
visualize the area of perfusion decit in patients with acute ischemic stroke.
Ultrasound contrast;
Triggered high mechanical index (MI) imaging, which uses contrast microbubble destruction
Brain perfusion;
to analyze bolus contrast kinetics in the brain parenchyma, was used in these studies. Recently
Ischemic stroke
high sensitive, low MI imaging was introduced. With this new technology real-time bolus kinetics
as well as rell kinetics could be analyzed without triggering. In the early phase of ischemic
stroke, ultrasound perfusion imaging is useful in detecting the area of perfusion decit and to
assess outcome prognosis of the patient. This bedside technology is available for use in the
stroke unit when patients with acute ischemic stroke undergo a color-coded duplex work-up to
evaluate their vascular status.
2012 Elsevier GmbH. Open access under CC BY-NC-ND license.

Technical background
Several studies using trigger high mechanical index (MI)
techniques for visualization of cerebral perfusion after ultra-
sound contrast agent (UCA) injection have been published in
the last 13 years [16]. The studies were mostly performed
with triggered harmonic gray scale imaging techniques (con-
ventional, power modulation or pulse-inversion) analyzing
Abbreviations: A, fractional vascular volume related to the
plateau echo enhancement (rell kinetics); AUC, area under the
curve; , rate constant (slope of the wash in curve of rell kinetics);
FWHM, full width at half the maximum intensity; IU, intensity units
(linear scale); MI, mechanical index; MTT, mean transit time; PG,
positive gradient; PI, peak intensity; PPI, pixelwise peak intensity;
PSI, peak signal increase; PW, peak width; ROI, region of interest;
TPI, time to peak intensity; TTP, time between start of contrast
signal and peak intensity; UCA, ultrasound contrast agent.
Tel.: +49 040 18 18 87 3076; fax: +49 040 18 18 87 3069.
E-mail address: g.seidel@asklepios.com
the bolus kinetics in healthy subjects to nd out the best way
for the detection of UCA in the cerebral microcirculation.
Recently low mechanical index gray scale imaging was
introduced. With this new real-time technology bolus kinet-
ics as well as rell kinetics could be analyzed. Rell kinetics
is based on the reappearance of echo contrast in tissue
after complete microbubble destruction using a high MI
pulse. After destruction of the contrast agent within the
scanning plane new microbubbles enter the volume with
a certain velocity, thus allowing calculation of regional
blood ow (Fig. 1). Rell kinetics to measure regional cere-
bral blood ow was rst studied in dogs after craniectomy
[7]. Recent technological advances in ultrasound equipment
with improved sensitivity for detection of microbubbles in
the cerebral microcirculation through the acoustic bone
window in humans now enable real-time ultrasound perfu-
sion imaging [8,9]. This new real-time rell technology has
several advantages over the triggered high MI techniques.
First rell kinetics could be recorded and analyzed within
seconds (Fig. 2); therefore, several insonation planes could
be evaluated with one contrast bolus injection. Second
software tools like microvascular imaging (display of the

2211-968X 2012 Elsevier GmbH. Open access under CC BY-NC-ND license.

doi:10.1016/j.permed.2012.02.041
Update on ultrasound brain perfusion imaging 31

Figure 1 Schematic representation of different parameters of the timeintensity curve available from bolus kinetics (left) and
rell kinetics (right). (a) Bolus kinetics: ROI = region of interest, IU = intensity units (linear scale), PI = peak intensity, PG = positive
gradient, PSI = peak signal increase, PW = peak width, FWHM = full width at half the maximum intensity, AUC = area under the curve,
TPI = time to peak intensity, TTP = time between start of contrast signal and peak intensity, MTT = mean transit time. (b) Rell
kinetics: = rate constant (slope of the wash in curve), A = fractional vascular volume related to the plateau echo enhancement.
Adapted from [11].

amount of contrast signals over time [8]) help in visualiza-
tion and documentation of perfusion decits. On the other
hand there are some disadvantages like the limited maximal
insonation depth and the high rate of insonation artifacts.
As of yet, it is not evident which method is superior for
the analysis of brain perfusion, because studies with a direct
comparison are missing.
The commercially available ultrasound contrast agents
LevovistTM (Schering), OptisonTM (Amersham Health), and
SonoVueTM (Bracco) proved to have contrast enhancing prop-
erties in human brain perfusion imaging. No severe adverse
events were documented in numerous volunteer studies
published on brain perfusion analysis using these contrast
agents including more than 200 subjects.
Various curve parameters have been described for the
analysis of the different contrast kinetics (bolus and rell).
To date (12/2011), it is not evident which kinetics or
which parameter is the most valuable for the analysis of
brain perfusion in healthy subjects. Theoretically, time-
dependent parameters like time to peak intensity (bolus
kinetics) or the -value (rell kinetics) should be more
useful than amplitude-dependent parameters, because the
latter depend also on insonation depth. Parametric images
of certain properties of the timeintensity curves have been
generated, which facilitate the evaluation of regional brain
perfusion (Fig. 3) [6].
Several studies were reported on ultrasound perfusion
imaging in healthy volunteers using perfusion weighted MRI
as reference for ultrasound perfusion imaging (Contrast
Burst and Time Variance Imaging as well as high MI harmonic
imaging) [5,10]. In these studies the time to peak inten-
sity and in one study [5] the area under the timeintensity
curve of ultrasound perfusion imaging showed a good corre-
lation to the time to peak intensity as measured in perfusion
weighted MRI.

Stroke studies

In most clinical studies on ischemic stroke patients contrast

Figure 2 Microvascular imaging: captured contrast signals
over 8 s after destruction pulse using low MI real time
contrast imaging (iU22 ultrasound system) in a 41-year-old
female patient suffering from middle and anterior cere-
bral artery infarction 11.5 h after symptom onset. We used
2.4 ml SonoVueTM as a bolus. MRI scan (DWI, below) was per-
formed 20 min after the ultrasound perfusion study. Notice the
insonation plane within the yellow margins. The area of dif-
fusion disturbance in the MRI (white area) showed no contrast
signal in the ultrasound perfusion study (dark area).
bolus kinetics was analyzed using different high MI harmonic
imaging modalities (harmonic imaging, power modulation,
and pulse inversion imaging). LevovistTM , OptisonTM , and
SonoVueTM were used as contrast agents [1216]. With
new, more sensitive multi-pulse ultrasound technologies it is
possible to analyze brain perfusion not only in the ipsilateral
but also in the contralateral hemisphere within one inves-
tigation improving the geometry of the insonation plane
and overcoming near-eld artifacts [16]. When using this
approach, additional artifacts (calcication of pineal gland
32 G. Seidel

Figure 3 Ultrasound perfusion study using parametric imaging (high-MI imaging, 2.4 ml SonoVueTM bolus kinetics). Ultrasound
(SONOS 5500, Philips) was performed 6 h after symptom onset in a 70-year-old male patient suffering from MCA occlusion (M1-
segment). Upper row: Time-to-peak image (TTP) and peak signal increase image (PPIpixelwise peak intensity). Notice the area
of delayed contrast arrival in the TTP image (within the red line), which showed in the anterior section a decrease of contrast
amplitude in the PPI image. CCT follow-up in the same imaging plane as investigated by ultrasound is shown in the lower row. In the
follow-up 180 h after symptom onset a complete middle cerebral artery infarction was displayed, which ts with the initial delay
of contrast arrival in the TTP image.

and choroid plexus of lateral ventricles causing shadowing
artifacts) have to be considered.
In recent low MI real time rell kinetics studies [17,18]
perfusion decits in acute ischemic stroke patients could be
visualized qualitatively with high sensitivity in the ipsilateral
hemisphere. The maximal area without detectable contrast
signal correlates with the severity of stroke symptoms [17].
Besides this, quantitative thresholds for the occurrence of
ischemia were calculated ( < 0.76 and A < 1.91 [18]).
Different parameters of the bolus kinetics curve acquired
from ischemic brain regions in the acute phase of stroke
were compared with follow-up CT to visualize the infarction.
A combination of the peak intensity increase (PI) and time-
to-peak (TTP) proved to be most helpful in detecting the
area of infarction, with a sensitivity between 75% and 86%
as well as a specicity between 96% and 100% [13,15].
In more recent studies color-coded parametric images
were evaluated [12,19]. They provide information on the
timeintensity data in all pixels under evaluation, thus
facilitating the visualization of the perfusion state [19].
Although the supplying artery was found patent by color-
coded duplex, in 1314% of acute ischemic stroke patients
a perfusion decit in the middle cerebral artery terri-
tory could be identied with parametric perfusion imaging
[13,19]. The areas of disturbed perfusion in the parametric
images (especially the PPI image) correlate with the area of
infarction in follow-up CT and the severity of stroke symp-
toms in the early phase as well as after four months [16].
Some more investigations based on contrast bolus kinet-
ics were performed on smaller populations using Xenon-CT
[20] and MRI [14] as reference methods. These case reports
demonstrate the potential of different contrast-specic
modalities for the assessment of pathologic brain perfusion
using contrast ultrasound imaging. In a small study analyz-
ing local correlations of ultrasound perfusion parameters of
bolus kinetics with the occurrence of a perfusion-diffusion
mismatch on Stroke MRI (penumbra) thresholds were calcu-
lated. Penumbra could be assumed if the relative time delay
exceeded 4 s and the relative signal amplitude exceeded
1/3 [21]. These preliminary data should be veried by a
prospective study.
Besides the high potential of ultrasound perfusion imag-
ing as a fast, semi-invasive bedside method to evaluate
supratentorial brain perfusion in acute ischemic stroke
patients, there are some drawbacks like the insonation arti-
facts, which occur in most of the patients and the inability
to scan the whole brain. Besides these technical limitations
there are potential side effects of the new contrast agents,
which restrict the employment of these substances in severe
cardiac or pulmonary disease.
Disclosure statement
Prof. Seidel is employed by Asklepios Kliniken Hamburg
GmbH and is professor of Neurology at the University of
Update on ultrasound brain perfusion imaging
Luebeck, Germany. He has previously received unrestricted
educational grants from Schering, Bracco Imaging SpA,
Philips Medical Systems, Boehringer Ingelheim, Solvay, Bayer
HealthCare, Biogen idec, Desitin, Merck Serono, Meda, MSD,
Novartis Neuroscience, Talecris, UCB, Grunenthal, Lund-
beck, Merz, Teva and Sano Aventis. He has worked together
with Bracco Imaging SpA and Philips Medical Systems in
research projects funded by the European Union.
References
[1] Seidel G, Greis C, Sonne J, Kaps M. Harmonic grey scale imaging
of the human brain. J Neuroimaging 1999;9:1714.
[2] Postert T, Hoppe P, Federlein J, Helbeck S, Ermert H, Przun-
tek H, et al. Contrast agent specic imaging modes for the
ultrasonic assessment of parenchymal cerebral echo con-
trast enhancement. J Cereb Blood Flow Metab 2000;20:
170916.
[3] Eyding J, Krogias C, Wilkening W, Meves S, Ermert H, Postert T.
Parameters of cerebral perfusion in phase-inversion harmonic
imaging (PIHI) ultrasound examinations. Ultrasound Med Biol
2003;29:137985.
[4] Harrer JU, Kltzsch C. Second harmonic imaging of the human
brain: the practicability of coronal insonation planes and alter-
native perfusion parameters. Stroke 2002;33:15305.
[5] Harrer J, Kltzsch C, Stracke CP, Mller-Hartmann W. Cerebral
perfusion sonography in comparison with perfusion MRT: a study
with healthy volunteers. Ultraschall Med 2004;25:2639.
[6] Wiesmann M, Seidel G. Ultrasound perfusion imaging of the
human brain. Stroke 2000;31:24215.
[7] Rim SJ, Leong-Poi H, Lindner JR, Couture D, Ellegala D, Mason
H, et al. Quantication of cerebral perfusion with real-time [20] Meairs S, Daffertshofer M, Neff W, Eschenfelder C, Hennerici M.
contrast-enhanced ultrasound. Circulation 2001;104:25827.
[8] Powers J, Averkiou M, Bruce M. Principles of cerebral
ultrasound contrast imaging. Cerebrovasc Dis 2009;27(Suppl.
2):1424.
[9] Seidel G, Meairs S. Ultrasound contrast agents in ischemic
stroke. Cerebrovasc Dis 2009;27(Suppl. 2):2539.
33
[10] Meves SH, Wilkening W, Thies T, Eyding J, Hlscher T, Finger M,
et al. Comparison between echo contrast agent-specic imag-
ing modes and perfusion-weighted magnetic resonance imaging
for the assessment of brain perfusion. Stroke 2002;33:24337.
[11] Della Martina A, Meyer-Wiethe K, Allemann E, Seidel G. Ultra-
sound contrast agents for brain perfusion imaging and ischemic
stroke therapy. J Neuroimaging 2005;15:21732.
[12] Seidel G, Meyer-Wiethe K, Berdien G, Hollstein D, Toth D, Aach
T. Ultrasound perfusion imaging in acute middle cerebral artery
infarction predicts outcome. Stroke 2004;35:110711.
[13] Federlein J, Postert T, Meves SH, Weber S, Przuntek H, Bt-
tner T. Ultrasonic evaluation of pathological brain perfusion in
acute stroke using second harmonic imaging. J Neurol Neuro-
surg Psychiatry 2000;69:61622.
[14] Meyer K, Wiesmann M, Albers T, Seidel G. Harmonic imaging
in acute stroke: detection of a cerebral perfusion decit with
ultrasound and perfusion MRI. J Neuroimaging 2003;13:1668.
[15] Seidel G, Albers T, Meyer K, Wiesmann M. Perfusion harmonic
imaging in acute middle cerebral artery infarction. Ultrasound
Med Biol 2003;29:124551.
[16] Eyding J, Krogias C, Wilkening W, Postert T. Detection of
cerebral perfusion abnormalities in acute stroke using phase
inversion harmonic imaging (PIHI): preliminary results. J Neurol
Neurosurg Psychiatry 2004;75:9269.
[17] Kern R, Diels A, Pettenpohl J, Kablau M, Brade J, Hennerici
MG, et al. Real-time ultrasound brain perfusion imaging with
analysis of microbubble replenishment in acute MCA stroke. J
Cereb Blood Flow Metab 2011;31:171624.
[18] Seidel G, Roessler F, Al-Khaled M. Microvascular imaging in
acute ischemic stroke. J Neuroimaging; in press.
[19] Wiesmann M, Meyer K, Albers T, Seidel G. Parametric perfusion
imaging with contrast-enhanced ultrasound in acute ischemic
stroke. Stroke 2004;35:50813.
Pulse-inversion contrast harmonic imaging: ultrasonographic
assessment of cerebral perfusion. Lancet 2000;355:5501.
[21] Meyer-Wiethe K, Cangur H, Schindler A, Koch C, Seidel G. Ultra-
sound perfusion imaging: determination of thresholds for the
identication of critically disturbed perfusion in acute ischemic
strokea pilot study. Ultrasound Med Biol 2007;33:8516.