Relative Energy Index of Microembolic Signal Can Predict
Malignant Microemboli
Youngbin Choi, MD; Maher Saqqur, MD; Eileen Stewart, RN; Caroline Stephenson, RN;
Jayanta Roy, MD; Jean-Martin Boulanger, MD; Shelagh Coutts, MD; Andrew M. Demchuk, MD, FRCPC
Downloaded from http://stroke.ahajournals.org/ by guest on June 16, 2017
Background and Purpose—Microembolic signals (MES) found on transcranial Doppler range from harmless air bubbles
to large, solid, particulate emboli from the heart and large vessels. The presence of MES is not always associated with
poor clinical outcome. The purpose of our study was to determine whether the relative energy index of MES measured
by power M-mode Doppler can distinguish malignant from benign MES and to identify patients with worse prognosis.
Methods—We prospectively collected transcranial Doppler emboli monitoring data from patients with symptomatic carotid
stenosis presenting with TIA or ischemic stroke. For each patient, we calculated the relative energy index of MES and
categorized those ⬎1.0 as malignant MES. We compared the clinical characteristics, number, and volume of
diffusion-weighted imaging lesions, and degree of stenosis and plaque characteristics on CT angiogram of patients with
malignant and benign MES.
Results—We enrolled 92 patients, 29 with TIA and 63 with stroke, within 48 hours of symptom onset. Twenty-six patients
had a total of 319 MES; of these, 82.4% were benign and were 17.6% malignant. Malignant MES traveled further within
intracranial vessels than benign MES. The 9 patients with ⬎1 malignant MES had significantly larger baseline
diffusion-weighted imaging lesion volume, had a higher prevalence of intraluminal thrombus on CT angiogram of the
neck and plaque ulceration, and were more likely to have a poor clinical outcome (modified Rankin Score ⱖ2) than
those with benign MES (OR, 6.5; 95% CI, 1.47–28.68). The presence of malignant MES led to the institution of more
aggressive secondary prevention measures.
Conclusions—Power M-mode transcranial Doppler identifies a subgroup of patients with malignant MES. These signals
are more frequent in longer arterial trajectory. Patients with malignant MES have larger baseline infarcts, a higher
prevalence of intraluminal thrombus or ulcerated plaque in the carotid artery, and worse clinical outcome. (Stroke.
2010;41:700-706.)
Key Words: malignant emboli 䡲 microemboli 䡲 transcranial Doppler
T
ranscranial Doppler (TCD) is gradually gaining acceptance as a monitoring tool to detect embolic signals often
referred to as microemboli signals (MES).1 These MES
correlate well with true emboli in animal models2 and are
most frequently identified in the setting of large vessel
atherosclerotic disease, such as carotid stenosis.3 There have
been reports that the presence of MES is an independent
predictor of future stroke in patients with symptomatic4 and
asymptomatic carotid stenosis.5,6 However, not all studies
have demonstrated the anticipated association between MES
and clinical outcome.7,8
TCD MES span a broad spectrum, ranging from harmless
air bubbles produced by artificial valves to the opposite
extreme in which large, solid, particulate emboli from the
heart and large vessels traverse the intracranial vasculature. If
TCD can distinguish the benign from the malignant MES,
then it would be useful for predicting outcome and may
influence therapy. Ackerstaff et al9 reported that TCD emboli
variables called “macroembolus” and “massive air embolism” were associated with poor outcome within 7 days after
stent deployment. They applied the definition of macroembolus as an embolus that partially or completely obstructed
the middle cerebral artery (MCA) main stem for a period of
several seconds to minutes.10 That study was the first to
define the concept of malignant embolus. However, just
applying this definition to MES may not be enough to define
the size/prognosis of an embolus. This goal has been difficult
to reach to date, even though MES have been shown to be
associated with the morphology of plaque surface.11,12 Considerable work has been performed trying to distinguish
between solid emboli vs air composition.13,14 In addition,
several studies have attempted to measure the size of emboli
by their specific characteristics.15 These efforts have been
limited by the type of technology used in emboli detection
Received November 20, 2009; accepted December 14, 2009.
From Department of Neurology and Psychiatry (Y.C., J.R.), St. Louis University, St. Louis, Mo; Department of Clinical Neurosciences (Y.C., E.S.,
C.S., J.-M.B., S.C., A.M.D.), University of Calgary, Calgary, Canada; Department of Neurology (M.S.), University of Alberta, Alberta, Canada.
Correspondence to Youngbin Choi, MD, Neurology Resident, Department of Neurology and Psychiatry, St. Louis University Hospital, 3635 Vista
Avenue, St. Louis, MO 63110. E-mail [email protected]
© 2010 American Heart Association, Inc.
Stroke is available at http://stroke.ahajournals.org
DOI: 10.1161/STROKEAHA.109.573733
700
Choi et al
Energy Index and Malignant Microemboli
701
Figure 1. The PMD variable measurements are shown on the PMD display.
Maximum duration of microembolus in a
3-mm gate (MaxD), maximum intensity
with adjustment of the intensity of background blood flow (MaxI), multigate
travel time (MGTT), travel distance of
microembolus (DIST), and relative energy
index of microembolus (REIM).
Downloaded from http://stroke.ahajournals.org/ by guest on June 16, 2017
because most studies relied on single-gate spectrogram TCD,
which has limited sample volume that covers only a small
part of the travel path of an embolus.16 Transcranial power
M-mode Doppler (PMD) is now available with digital TCD
that simultaneously sample 33 gates over 60 mm of intracranial space.17 The wide range of sampling signals of PMD
could provide information such as arterial trajectory, multigate travel time, maximal duration, travel length, and velocity
of embolus, which would not be measured by the spectrogram
alone.18
Quantity of the microemboli, such as the presence or
frequency detected by transcranial Doppler, has been considered to be the only variable being correlated with the risk of
recurrent stroke. Our aim is to determine if the quality of
microemboli measured by power M-mode Doppler represents
a malignant type of MES in terms of volume of diffusionweighted imaging (DWI) lesion, 3-month clinical outcome,
pattern of carotid plaque based on CT angiogram for neck,
and recurrent stroke.
Materials and Methods
We prospectively collected TCD emboli monitoring data from
patients who participated in a prospective cohort studies, either
TASC (Transcranial Doppler and Symptomatic Carotid Disease
study)37 or VISION (Vascular Imaging of Acute Stroke for Identifying Predictors of Clinical Outcome and Recurrent Ischemic
Events)38 trials, at a single academic center between May 1, 2002
and December 31, 2006. Among them, patients who had carotid
stenosis (⬎50% stenosis), occlusion, or unstable carotid plaque were
investigated. The patients underwent bilateral hemisphere TCD
emboli monitoring for 1 hour within 48 hours of their clinical
presentation. Two registered sonographers had been practicing during that period; also, 2 experienced stroke neurologists were involved. Ten (10/102; 9.8%) patients were excluded from study
because of inadequate temporal bone window. The research protocol
was approved by our institutional ethics committee and all participants provided written informed consent.
PMD (100 mol/L; Spencer Technologies) was used to acquire
2-MHz spectral single-gate TCD information at a specific depth and
power M-mode information from 33 sample volumes placed at 2-mm
intervals from 24 to 88 mm depth of insonation.17 Probe fixation
using a head frame (Marc 500; Spencer Technologies) was used for
monitoring. One proximal MCA segment was insonated and, when
possible, ipsilateral anterior cerebral artery or internal carotid artery
were used for PMD display. The insonation depth for spectrogram
recording was between 45 and 65 mm for the ipsilateral MCA.
Semiautomated algorithm saved every possible MES for a 1-hour
period. The MES were reviewed by an experienced neurosonologist
blinded to all clinical and imaging information.
We measured several parameters of MES: maximum duration
(sec) of MES in a 3-mm gate, MES maximum intensity with
adjustment of the intensity of the background blood flow (MaxI),
multigate travel time, distance of MES travel, and the velocity
(cm/sec) of MES. In addition, we listed the arterial trajectory of MES
on PMD display.18 Arterial trajectory of MES means the pathway
MES travel around in the intracranial arterial system. For instance, it
can be M1–M2, or M1 to branch, or internal carotid artery to branch.
All information from the M-mode 33 gates provide the MES travel
pathway, which are not necessarily correlated with the duration of
MES, because the duration of MES can be affected by velocity of
MES, hemodynamic status of patient, and size of MES. To describe
the MES quality that represents malignant MES, we derived the
relative energy index of MES (REIM) as REIM⫽MaxI (maximal
intensity)⫻MaxD (maximal duration) (Figure 1). To estimate the
diameter of the embolus and define the cut-off value of REIM for
malignant microembolus, we used the following relationship:19
m2c4⫽E2⫺p2c2
Where E is the energy, m is the mass, c is the speed of light, and
p is the 3-dimensional momentums of the MES. If the velocity of the
particle is much smaller than c, then E can be approximated as
E⫽mc2.20 If the embolus is considered to be a sphere, then m can be
calculated as m⫽4/3 ␲r3␳. If ␳ is constant, then the energy can be
considered proportionate to the cubic radius of an embolus (ie, E ⬀ r3),
assuming that the minimal diameter of embolus that can be detected
by TCD is 80 ␮m3 and the maximal diameter of embolus that can
enter the MCA main trunk is 3000 ␮m,21,22 We compared REIM
with the log radius of embolus ranging from 40 ␮m to 1500 ␮m. The
embolus, with REIM 1.0 that represents 500 ␮m of diameter of
embolus, was considered to be the cut-off value for malignant
microembolus because the mean diameter of the penetrating artery of
MCA branch is ⬍500 ␮m.23 Figure 2 shows that the distribution of
702
Stroke
April 2010
Figure 2. The relation between radii of
embolus and log REIM. REIM 1.0 represented the 500-␮m diameter of embolus.
Downloaded from http://stroke.ahajournals.org/ by guest on June 16, 2017
emboli based on the diameter and shows that the diameter of emboli
ranged from 80 ␮m to 400 ␮m (233/319; 73.0%) when REIM 1.0
represented a 500-␮m diameter embolus. Therefore, the cut-off value
of REIM 1.0 was arbitrarily used to identify the malignant
microemboli.
The patients were categorized into 2 subgroups based on the
presence of malignant MES: patients with benign MES only (benign
MES group) vs patients with ⬎1 malignant MES (malignant MES
group). Clinical characteristics, initial NIHSS, 90-day NIHSS, modified Rankin Scale score at 90 days, DWI lesion volume and number,
the characteristics of the plaque on baseline CTA of the neck, and
recurrent stroke/TIA in 90 days were compared between malignant
MES and benign MES groups. Also, comparison of early anticoagulation or early carotid endarterectomy/stent between 2 groups had
been performed. The DWI lesion volumes were measured by
computer-assisted volumetric analysis using custom software that
incorporated a neighborhood-connected region growing threshold
segmentation method (ITK; National Library of Medicine). CTA of
the neck was performed on 64-row multidetector CT (Siemens). The
CTA results were reviewed by 2 stroke neurologists and 2 neuroradiologists by consensus on a remote research PACS station. Reviewers were blinded to the clinical data and any imaging of the head. If
there was a plaque, then the degree of stenosis was measured by the
NASCET criteria.36 The plaque length was measured electronically
as the maximum longitudinal dimension, whereas the thickness was
measured at the maximum stenosis in the axial sections. The density
was qualitatively assessed as hypodense, isodense, or heterogeneous
compared to that of neck muscles. The degree of calcification was
classified as mild, moderate, or extensive if ⬍30%, 30% to 70%, or
⬎70% of the plaque was calcified, respectively. The plaque surface
was classified as smooth, irregular, and ulcerated. The ulcer was
defined as focal extension of the contrast beyond the lumen by at
least 1 mm. Intraluminal thrombus was defined as a filling defect
within the lumen completely surrounded by contrast on at least 2
contiguous axial source images.
Demographics of study population and MES counts were reported
using descriptive statistics. Student t test was used to compare the
parameters between benign MES and malignant MES. The ␹2 test
was applied to compare the arterial trajectory, the patterns of carotid
plaque on CT angiogram, and recurrent vascular events. Student t
test was used to evaluate the statistical difference of emboli count
and DWI lesion volume. Mann-Whitney U tests was applied to
compare the NIHSS scores and modified Rankin Scale scores.
Statistical significance thresholds were established for all test, with P
ⱕ0.05 indicating difference. All statistical tests were performed
using SPSS (SPSS version 11.5; SPSS).
Results
Ninety-two patients with carotid disease had TIA/stroke and
underwent TCD emboli monitoring within 48 hours of onset
of symptoms. Mean age was 67.7⫾13.1 years and 69 patients
(75.0%) were men. Ten patients (10.9%) were excluded
because of inadequate temporal bone window. Median baseline NIHSS was 2 (interquartile range [IQR], 0 – 4). Twentynine patients (31.5%) had TIA and 63 (68.5%) patients had
stroke. The mean time from symptoms onset to clinical
evaluation was 12.8⫾6.4 hours. The TCD monitoring for
MES was performed at median time of 16.4 hours (range,
2.5– 47 hours) from onset of symptoms. Twenty-six patients
(26/92; 28.3%; mean age, 63.4⫾17.3 years) had positive
MES on PMD, and a total of 319 MES were detected by PMD
(mean MES, 3.55; ⫾SD 10.6; median, 0; IQR, 0 –1).
Fifty-six MES (56/319; 17.6%) with REIM ⬎1.0 were
considered to be malignant MES. The characteristics of
malignant MES are summarized in Table 1. Malignant MES
had greater REIM, MaxI (dB), maximum duration (ms),
multigate travel time (ms), and distance of MES travel (mm)
than benign emboli (REIM, 2.1⫾1.0 vs 0.4⫾0.2; MaxI,
19.3⫾5.5 vs 8.5⫾3.2; maximum duration, 104.8⫾26.3 vs
52.9⫾12.7; multigate travel time, 168.3⫾64.0 vs 76.4⫾36.4;
distance of MES travel, 37.5⫾8.6 vs 17.8⫾9.1; P⬍0.001).
Figure 3 shows an example of a typical malignant MES and
benign MES. There was no statistically significant differences found with the velocity of MES (cm/sec) between
malignant MES and benign MES (24.6⫾9.4 vs 25.2⫾16.4;
P⫽0.73). The mean diameter (␮m) of malignant MES was
significantly larger than benign MES (1009.1⫾466.7 ␮m vs
275.2⫾98.2 ␮m; P⬍0.001). Malignant MES (51/56; 91.1%)
had more frequent M1–M2 arterial trajectories than benign
MES (48/263; 18.3%; OR, 45.7; 95% CI, 17.31–120.6;
P⬍0.001). Benign MES more frequently had only M1 arterial
trajectories (123/263; 48.7%) than malignant MES (3/56;
5.4%; OR, 9.1; 95% CI, 3.024 –27.72; P⬍0.001). Benign
MES also had more frequent branch artery trajectories (perforator and only M2; 42/263; 16.0%) than malignant MES
Choi et al
Table 1. Comparisons of Parameters Between Benign MES
Group and Malignant MES Group
Counts
REIM (joule⫻10⫺16)
Benign MES
(Mean⫾SD)
Malignant MES
(Mean⫾SD)
P
263 (82.4%)
56 (17.6%)
⬍0.001
0.4⫾0.2
2.1⫾1.0
⬍0.001
8.5⫾3.2
19.3⫾5.5
⬍0.001
MaxD, msec
52.9⫾12.7
104.8⫾260.3
⬍0.001
MGTT, msec
76.4⫾36.4
168.3⫾640.0
⬍0.001
DIST, mm
17.8⫾9.1
37.5⫾8.6
⬍0.001
MaxI, dB
VM, cm/sec
25.2⫾16.4
24.6⫾9.4
0.73
275.2⫾98.2
1009.1⫾466.7
⬍0.001
48 (18.3%)
51 (91.1%)
⬍0.001
ACA
11 (4.2%)
1 (1.8%)
M1
128 (48.7)
3 (5.4%)
M2
14 (5.3%)
1 (1.8%)
Perforator
41 (15.6%)
0 (0%)
1 (0.4%)
0 (0%)
20 (7.6%)
0 (0%)
Diameter, ␮m
Arterial trajectory
M1–M2
Downloaded from http://stroke.ahajournals.org/ by guest on June 16, 2017
ICA
M1 distal
⬍0.001
⬍0.001
⬍0.001
ACA indicates anterior carotid artery; DIST, travel distance of microembolus;
ICA, internal carotid artery; MaxD, maximum duration of microembolus in a
3-mm gate; VM, velocity of microembolus.
(1/56; 1.8%; OR, 11.50; 95% CI, 1.625– 81.375; P⬍0.001;
Table 1).
Nine out of 26 MES-positive patients (34.6%) who had ⬎1
malignant MES were categorized into malignant MES group.
Seventeen patients (65.4%; 17/26) who had only benign MES
were categorized into the benign MES group. There were no
differences between benign and malignant MES groups
regarding risk factors for stroke and antiplatelet agent or
Energy Index and Malignant Microemboli
703
statin usage before index stroke (Table 1). The baseline
clinical stroke severity measured by NIHSS score tended to
be higher in the malignant MES group (median, 4; IQR,
1.5–12) than in the benign MES group (median, 2; IQR, 1–5;
P⫽0.218), which was statistically not significant. NIHSS at
90 days was significantly higher in malignant MES group
(median, 3; IQR, 0 – 8) than in the benign MES group
(median, 1; IQR, 0 –2; (P⫽0.025). Modified Rankin Scale
score ⱖ2 was a poor clinical outcome at 90 days was
significantly more common in the malignant MES group (4/9;
44.4%) than in the benign MES group (2/17; 11.8%;
P⫽0.025; OR, 6.5; 95% CI, 1.47–28.68). Total emboli count
was significantly higher in the malignant MES group
(25.1⫾22.1) than in the benign MES group (5.3⫾8.3;
P⫽0.002).
Out of 92 patients, 65 patients (70.7%) underwent MR scan
with DWI within 48 hours from symptoms onset. All MESpositive patients underwent MRI. Twenty-seven patients
(27/92; 29.3%) were not eligible for MR scan because of
metals in the body, claustrophobia, obesity, or cardiac pacemaker. The mean time from symptoms onset to MR scan was
12.8⫾6.4 hours. All patients with any type of MES had new
lesions on their DWI MR scan. No statistically significant
difference existed between benign and malignant MES
groups regarding DWI lesion number (5.6⫾2.1 vs 7.6⫾4.5;
P⫽0.482). The baseline DWI lesion volume was significantly
larger in the malignant MES group (21.6⫾23.8 mL) than in
the benign MES group (3.89⫾6.70 mL; P⫽0.008; Table 2).
Sixty-four patients (69.6%, 64/92) underwent CT angiogram of the neck vessels within 12 hours of symptoms onset
(malignant MES group, 7/9 [77.8%]; benign MES group,
11/17 [64.7%]). Twenty-eight patients (28/92; 30.4%) were
not eligible for CT angiogram because of renal insufficiency,
allergy to dye, and other reasons. Malignant MES group had
Figure 3. An example of malignant MES
and benign MES A typical malignant
MES (arrow) and benign MES (arrow
head) are seen in the PMD display. In
this example, the malignant MES is welldifferentiated from benign MES (REIM,
3.0 vs 0; MaxI, 20 dB vs 3 dB; MaxD,
150 ms vs 50 ms; MGTT, 220 ms vs 80
ms; DIST, 35 mm vs 20 mm; velocity of
MES, 15.9 cm/sec vs 20 cm/sec).
704
Stroke
April 2010
Table 2. Comparison of Baseline Characteristics, DWI Lesion
Volume, Patterns of CTA, and Recurrent Stroke/TIA of Benign
MES and Malignant MES Groups
Benign MES
Group
Malignant MES
Group
Discussion
P
N of patients (%)
17 (65.4)
Stroke (%)
13 (76.5)
8 (88.9)
0.452
Total MES count
5.3⫾8.3
25.1⫾22.1
0.002
9 (34.6)
0.079
Risk factors
Hypertension
Downloaded from http://stroke.ahajournals.org/ by guest on June 16, 2017
12 (70.6)
3 (33.3)
Hyperlipidemia
3 (17.6)
1 (11.1)
0.569
Diabetes
5 (29.4)
3 (33.3)
0.587
Coronary artery disease
4 (23.5)
3 (33.3)
0.462
Peripheral artery disease
2 (11.8)
9 (0.0)
0.418
Previous stroke
4 (23.5)
2 (22.2)
0.668
Smoking
5 (29.4)
3 (33.3)
0.587
9 (52.9)
5 (55.6)
0.613
Medications before index
stroke/TIA
Antiplatelet agents
Statin
10 (58.8)
4 (44.4)
0.387
NIHSS baseline, median (IQR)
2 (0–2)
4 (1.5–12)
0.218
NIHSS at 90 days, median
(IQR)
1 (0–2)
4 (0–8)
0.025
Modified Rankin Scale score at
90 days, median (IQR)
0 (0–1)
3 (0–4)
0.025
MRI
DWI lesion N
DWI lesion volume, mL
5.6⫾2.1
7.6⫾4.5
0.482
3.89⫾6.70
21.6⫾23.8
0.008
CTA for carotid artery
Patients undergoing CT
angiogram
Intraluminal thrombus (%)
11 (64.7)
7 (77.8)
3 (27.3)
6 (85.7)
0.025
Sessile/pedunculated (%)
4 (36.4)
3 (42.9)
0.583
Hypodensity (%)
4 (36.4)
2 (28.6)
0.572
Ulceration (%)
2 (18.2)
5 (71.4)
0.039
Massive calcification (%)
3 (27.3)
3 (42.9)
0.428
62.6⫾23.1
0.546
Degree of stenosis, ipsilateral %
68.2⫾15.7
P⫽0.002) or early carotid endarterectomy/stent (77.8% vs
17.6%; P⫽0.005; Table 2).
Interventions for secondary
prevention
Anticoagulation
4 (23.5)
8 (88.9)
0.002
Carotid endarterectomy/stent
3 (17.6)
7 (77.8)
0.005
higher incidence of intraluminal thrombus (6/7 [85.7%] vs
3/11 [27.3%]; P⫽0.025) and plaque ulceration (5/7 [71.4%]
vs 2/11 [18.2%]; P⫽0.039).in the carotid artery group compared to benign MES group. The other high- risk patterns of
carotid artery, such as sessile/pedunculated, ulceration of
surface, hypodensity, calcification, or degree of stenosis,
were not associated with presence of malignant MES (Table
2). The higher tendency of subsequent vascular events rate
was found in the malignant MES group (3/9; 33.3%) compared to the benign MES group (2/17; 11.8%); however, no
statistical significance was present (P⫽0.208). The malignant
MES group tended to receive more aggressive secondary
prevention, such as early anticoagulation (88.9% vs 23.4%;
Our study has demonstrated the capability of PMD TCD to
define the emboli signals based on quality of MES. This has
resulted in the identification of a subgroup of emboli defined
as malignant MES. These emboli have characteristics exhibiting a higher energy level signature, which represent a
greater mass than the typical MES based on quantum physics
principles. We found that this subgroup of MES is associated
with larger size of baseline infarct and more frequent intraluminal thrombus of carotid artery. In addition, patients with
malignant MES on TCD monitoring are more prone to poor
clinical outcome and receiving more aggressive treatment for
prevention of recurrent stroke. Defining this subgroup of
patients with acute TIA/stroke and malignant MES could help
the physician in preventing further stroke recurrence by
optimizing medical treatment or in justifying early
revascularization.
Our findings may be of significant clinical importance
because the focus of prevention treatment using MES as a
surrogate outcome has solely rested on the frequency or
presence of MES. All therapeutic studies to date have
measured treatment effect by the reduction in MES counts
with a variety of therapies.24 –27 The number of MES likely
has some importance in predicting risk of stroke, as best
demonstrated by Levi et al28 in the dextran study of carotid
endarterectomy candidates, which reported a 77% chance of
stroke if ⬎50 MES occurred per hour. Subsequent work
suggested detection rate ⬎10 MES per 30 minutes was
associated with high postoperative stroke rate.29 The importance of the commonly seen MES counts of (1–10 MES per
hour) is much less clear. At these levels, MES appears to have
only modest predictive value for subsequent clinical stroke
events and, therefore, only limited value in stroke prevention.
The detection of subgroup of malignant MES may define a
more malignant course that may require aggressive antithrombotic treatment and more urgent revascularization.
PMD has significant advantages over single-gate spectrogram TCD by providing the necessary information to evaluate the “size” of MES. The size of embolus could influence
the back-scatter cross-section,30 which would be estimated
indirectly by measured embolus-to-blood ratio.31 Measured
embolus-to-blood ratio can be converted into MaxI on PMD
display, which represents maximal power of MES.18 Powerbased methods using measured embolus-to-blood ratio (MaxI
on PMD display) are most promising28; however, sound
intensity for solid emboli (for example, red cell aggregate)
does not increase monotonically with power; therefore, any 1
value of MaxI on PMD display could correspond to 2 or 3
different embolus sizes.15 For this reason, the MaxI is not
considered sufficient to characterize size of emboli unless
other information is simultaneously provided. Distance of
travel of MES may have some advantages over the MaxI on
PMD display because it is not affected by amplifier overload.13 However, it can be influenced by insonation status,
such as bone window or the sonographer’s skill. MES
duration measurement is another promising approach,14
Choi et al
Downloaded from http://stroke.ahajournals.org/ by guest on June 16, 2017
which could be longer in the case of larger emboli.15
However, the duration of MES can also be affected by
velocity of MES and hemodynamic status of patients. If the
size of MES is considered to be proportionate in its mass
because the density of embolus, such as red cell aggregates,
is homogenous, then relative energy level of MES can be
calculated by product of power as measured sound intensity
(MaxI on PMD display) and duration of MES (maximum
duration on PMD display).
Even though presence of malignant MES failed to demonstrate more frequent subsequent vascular event such as
stroke/TIA in our study, the tendency toward higher risk of
recurrent stroke/TIA in the malignant MES group was distinct (3/9 [33.3%] vs 2/17 [11.8%]). Also, it needs to be
considered that finding malignant MES on TCD monitoring
led the physicians to commence more aggressive treatment,
such as early anticoagulation or carotid endarterectomy/stent
for secondary stroke prevention. There was a lack of power
attributable to low number of patients in our study. Validation
of this sizing method is required with a controlled study using
various sizes of imaging contrast nanobubbles. In addition,
large-scale clinical trials using malignant MES as a surrogate
outcome could also confirm the relationship between malignant MES and prognosis.
Our study suggested that most of the emboli produced
by carotid stenosis were smaller than the mean size of the
penetrating artery. The more severe the carotid disease is,
the more likely the MES is malignant. Judging from the
analysis of arterial trajectory of embolus, malignant MES
would preferentially enter the larger-circumferential cerebral arteries and smaller benign MES could enter any
arteries, including the small perforating arteries that arise
from the major vessels of the circle of Willis. MacDonald
et al32 reported that larger emboli had a significant
tendency to enter larger arteries and smaller emboli were
more likely to be distributed in the small perforating
arteries. Embolism as a cause of lacunar infarction had
been reported in 18% to 25% of cases.33,34
Our study was limited in that we could not completely rule
out air bubbles as the cause of the high-intensity MES seen in
the malignant MES group. However, this would seem very
unlikely because patients who had undergone mechanical
valvular disease or any type of surgery with air entry into the
vasculature were excluded from the study.35 Another limitation is the fact that MaxI on PMD display is influenced by
many factors, such as the size of MCA, shape of the
ultrasonic beam, the emboli trajectory, and the interaction of
these parameters.15Another limitation of our study is that
clinical outcome may not reflect the natural history of MES
because positive MES at the time of admission were considered clinically significant, which caused the physician to
initiate aggressive antithrombotic therapies. In addition, patients also underwent fairly urgent carotid endarterectomy or
stenting if stenosis was appropriate (⬎50% stenosis), and this
affects the long-term outcome. Therefore, this limitation
should not overestimate the poor outcome seen in the malignant MES group; however, if it had an effect, it could
underestimate that.
Energy Index and Malignant Microemboli
705
Conclusion
In conclusion, our results suggest that among patients with
symptomatic carotid disease, there is a subgroup of patients
with embolic signals defined as malignant MES that can be
detected by PMD display. These malignant MES have the
characteristics typical of a longer trajectory through the MCA
vasculature, larger baseline infarcts, and worse clinical outcome. These malignant MES cases should be cause for
concern and may require more aggressive antithrombotic
treatment and investigation for intraluminal thrombus or
plaque ulceration.
Sources of Funding
The study was supported by funding from the Alberta Heritage
Foundation for Medical Research, Canadian Institutes for Health
Research, and Heart and Stroke Foundation of Alberta, Northwest
Territories, and Nunavut.
Disclosure
None.
References
1. Ringelstein EB, Droste DW, Babikian VL, Evans DH, Grosset DG, Kaps
M, Markus HS, Russell D, Siebler M. Consensus on Microembolus
Detection by TCD. Stroke. 1998;29:725–729.
2. Russell D, Madden KP, Clark WM, Sandset PM, Zivin JA. Detection of
arterial emboli using Doppler ultrasound in rabbits. Stroke. 1991;22:
253–258.
3. Kaposzta Z, Young E, Bath PMW, Markus HS. Clinical application of
asymptomatic embolic signal detection in acute stroke. A prospective
study. Stroke. 1999;30:1814 –1818.
4. Markus HS, MacKinnon A. Asymptomatic embolization detected by
Doppler ultrasound predicts stroke risk in symptomatic carotid artery
stenosis. Stroke. 2005;36:971–975.
5. Valton L, Larrue V, le Traon AP, Massabuau P, Géraud G. Microembolic
signals and risk of early recurrence in patients with stroke or transient
ischemic attack. Stroke. 1998;29:2125–2128.
6. Molloy J, Markus HS. Asymptomatic embolization detected by Doppler
ultrasound predicts stroke risk in symptomatic carotid artery stenosis.
Stroke. 1999;30:1440 –1443.
7. Abbott AL, Chambers BR, Stork JL, Levi CR, Bladin CF, Donnan GA.
Embolic signals and prediction of ipsilateral stroke or transient ischemic
attack in asymptomatic carotid stenosis: a multicenter prospective cohort
study. Stroke. 2005;36:1128 –1133.
8. Lund C, Rygh J, Stensrod B, Sandset PM, Brucher R, Russell D. Cerebro
microembolus detection in an unselected acute ischemic stroke population. Cerebrovasc Dis. 2000;10:403– 408.
9. Ackerstaff RG, Suttorp MJ, van den Berg JC, Overtoom TT, Vos JA,
Zanen P. Prediction of early cerebral outcome by transcranial doppler
monitoring in carotid bifurcation angioplasty and stenting. J Vasc Surg.
2005;41:618 – 624.
10. Claus SP, Louwerse ES, Mause HW, van der Mee M, Moll FL, Mess
WH, Ackerstaff RG. Temporary occlusion of middle cerebral artery by
macroembolism in carotid surgery. Cerebrovasc Dis. 1999;9:261–264.
11. Valton L, Larrue V, Arrue P, Geraud G, Bes A. Asymptomatic cerebral
embolic signals in patients with carotid stenosis. Correlation with
appearance of plaque ulceration on angiography. Stroke. 1995;26:
813– 815.
12. Orlandi G, Parenti G, Landucci PL, Sartucci F, Paoli C, Puglioli M,
Collavoli P, Murri L. Plaque surface and microembolic signals in
moderate carotid stenosis. Ital J Neurol Sci. 1999;20:179 –182.
13. Smith JL, Evans DH, Bell PRF, Naylor AR. A comparison of four
methods for distinguishing doppler signals from gaseous and particulate
emboli. Stroke. 1998;29:1133–1138.
14. Droste DW, Markus HS, Nassiri D, Brown MM. The effect of velocity on
the appearance of embolic signals studied in transcranial doppler models.
Stroke. 1994;25:986 –991.
15. Evans DH. Ultrasound Detection of Cerebral Emboli. IEEE Ultrasonics
symposium. 2003;316 –326.
706
Stroke
April 2010
Downloaded from http://stroke.ahajournals.org/ by guest on June 16, 2017
16. Azarpazhooh MR, Chambers BR. Clinical application of transcranial
doppler monitoring for embolic signals. J Clin Neurosci. 2006;13:
799 – 810.
17. Moehring MA, Spencer MP. Power M-mode Doppler (PMD) for
observing cerebral blood flow and tracking emboli. Ultrasound Med Bio.
2002;28:49 –57.
18. Sharma VK, Tsivgoulis G, Lao AY, Malkoff MD, Alexandrov AW,
Alexandov AV. Quantification of microbubble appearance in brain
vessels; Implications for dose calculations and drug delivery. Stroke
2007;38:493.
19. Wheeler MT. Gravitation. San Francisco, CA: W. H. Freeman; 1973.
20. Serway R, Jewett J. Physics for scientists and engineers, 6th ed. Pacific
Grove, CA: Brooks/Cole Publishing; 2003.
21. Gibo H, Carver CC, Rhoton AL, Lenkey C, Mitchell RJ. Microsurgical
anatomy of the middle cerebral artery. J Neurosurg. 1981;54:151–169.
22. Grand W. Microsurgical anatomy of the proximal middle cerebral artery
and the internal carotid artery bifurcation. Neurosurgery. 1980;7:
215–218.
23. Fisher CM. Capsular infarct. Arch Neurol. 1979;36:65–73.
24. Markus HS, Droste DW, Kaps M, Larrue V, Lees KR, Siebler M,
Ringelstein EB. Dual antiplatelet therapy with clopidogrel and aspirin in
symptomatic carotid stenosis evaluated using Doppler embolic signal
detection: the Clopidogrel and Aspirin for Reduction of Emboli in Symptomatic Carotid Stenosis (CARESS) trial. Circulation. 2005;111:
2233–2240.
25. Junghans U, Siebler M. Cerebral microembolism is blocked by tirofiban,
a selective nonpeptide platelet glycoprotein IIb/IIIa receptor antagonist.
Circulation. 2003;107:2717–2721.
26. Levi CR, Stork JL, Chambers BR, Abbott AL, Cameron HM, Peeters A,
Royle JP, Roberts AK, Fell G, Hoare MC, Chan AT, Donnan GA.
Dextran reduces embolic signals after carotid endarterectomy. Ann
neurol. 2001;50:544 –547.
27. Molloy J, Martin JF, Baskervile PA, Fraser SC, Markus HS.
S-Nitrosoglutathione reduces the rate of embolization in humans.
Circulation. 1998;98:1372–1375.
28. Georgiadis D, Mackay TG, Kelman AW, Grosset DG, Wheatley DJ, Lees
KR. Differentiation between gaseous and formed embolic materials in
vivo: Application in prosthetic heart valve patients. Stroke. 1994;25:
1559 –1563.
29. Abbott AL, Levi CR, Stork JL, Donnan GA, Chambers BR. Timing of
Clinically Significant Microembolism after Carotid Endarterectomy.
Cerebrovasc Dis. 2007;23:362–367.
30. Rschevkin SN. A course of lectures on the theory of sound. Pergamon
Press, Oxford, 1963.
31. Moehring MA, Klepper JR. Pulse doppler ultrasound detection, characterization and size estimation of emboli in flowing blood. IEEE Trans
Biomed Eng. 1994;41:35– 44.
32. Macdonald RL, Kowalczuk A, Johns L. Emboli enter penetrating arteries
of monkey brain in relation to their size. Stroke. 1995;26:1247–1251.
33. Horowitz DDR, Tuhrim S, Weinberger JM, Rudolph SH. Mechanisms in
lacunar infarction. Stroke. 1992;23:325–327.
34. Ay H, Oliveira-Filho J, Buonanno FS, Ezzeddine M, Schaefer PW,
Rordorf G, Schwamm LH, Gonzalez RG, Koroshetz WJ. Diffusionweighted imaging identifies a subset of lacunar infarction associated with
embolic source. Stroke. 1999;30:2644 –2650.
35. Georgiadis D, Preiss M, Lindner A, Gybels Y, Zierz S, Zerkowski HR.
Doppler microembolic signals in children with prosthetic cardiac valves.
Stroke. 1997;28:1328 –1329.
36. North Am Symptomatic Carotid Endarterectomy Trial Collaborators.
Beneficial effect of carotid endarterectomy in symptomatic patients with
high-grade stenosis. N Engl J Med. 1991;325:445– 453.
37. Choi Y, Talip A, Saqqur M, Jin A, Stewart E, Stephenson C, Ibrahim M,
Roy J, Boulanger JM, Coutts SB, Khan F, Demchuk AM. A combined
Power M-mode and Spectrogram Transcranial Doppler Ultrasound
microemboli signal criteria for improving emboli detection and reliability.
J Neuroimaging. 2009. November 13. [Epub ahead of print].
38. Coutts SB, Simon JE, Eliasziw M, Sohn CH, Hill MD, Barber PA,
Palumbo V, Kennedy J, Roy J, Gagnon A, Scott JN, Buchan AM,
Demchuk AM. Triaging transient ischemic attack and minor stroke
patients using acute magnetic resonance imaging. Ann Neurol. 2005;57:
848 – 854.
Relative Energy Index of Microembolic Signal Can Predict Malignant Microemboli
Youngbin Choi, Maher Saqqur, Eileen Stewart, Caroline Stephenson, Jayanta Roy, Jean-Martin
Boulanger, Shelagh Coutts and Andrew M. Demchuk
Downloaded from http://stroke.ahajournals.org/ by guest on June 16, 2017
Stroke. 2010;41:700-706; originally published online February 25, 2010;
doi: 10.1161/STROKEAHA.109.573733
Stroke is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright © 2010 American Heart Association, Inc. All rights reserved.
Print ISSN: 0039-2499. Online ISSN: 1524-4628
The online version of this article, along with updated information and services, is located on the
World Wide Web at:
http://stroke.ahajournals.org/content/41/4/700
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published
in Stroke can be obtained via RightsLink, a service of the Copyright Clearance Center, not the Editorial Office.
Once the online version of the published article for which permission is being requested is located, click
Request Permissions in the middle column of the Web page under Services. Further information about this
process is available in the Permissions and Rights Question and Answer document.
Reprints: Information about reprints can be found online at:
http://www.lww.com/reprints
Subscriptions: Information about subscribing to Stroke is online at:
http://stroke.ahajournals.org//subscriptions/