Download Noninvasive Contrast-Enhanced Ultrasound Molecular Imaging

Molecular Imaging
Noninvasive Contrast-Enhanced Ultrasound Molecular
Imaging Detects Myocardial Inflammatory Response
in Autoimmune Myocarditis
David C. Steinl, MSc; Lifen Xu, PhD; Elham Khanicheh, MD, PhD; Elin Ellertsdottir, PhD;
Amanda Ochoa-Espinosa, PhD; Martina Mitterhuber, MSc; Katharina Glatz, MD;
Gabriela M. Kuster, MD; Beat A. Kaufmann, MD, PhD
Downloaded from http://circimaging.ahajournals.org/ by guest on May 12, 2017
Background—Cardiac tests for diagnosing myocarditis lack sensitivity or specificity. We hypothesized that contrastenhanced ultrasound molecular imaging could detect myocardial inflammation and the recruitment of specific cellular
subsets of the inflammatory response in murine myocarditis.
Methods and Results—Microbubbles (MB) bearing antibodies targeting lymphocyte CD4 (MBCD4), endothelial P-selectin
(MBPSel), or isotype control antibody (MBIso) and MB with a negative electric charge for targeting of leukocytes (MBLc)
were prepared. Attachment of MBCD4 was validated in vitro using murine spleen CD4+ T cells. Twenty-eight mice were
studied after the induction of autoimmune myocarditis by immunization with α-myosin-peptide; 20 mice served as
controls. Contrast-enhanced ultrasound molecular imaging of the heart was performed. Left ventricular function was
assessed by conventional and deformation echocardiography, and myocarditis severity graded on histology. Animals were
grouped into no myocarditis, moderate myocarditis, and severe myocarditis. In vitro, attachment of MBCD4 to CD4+ T
cells was significantly greater than of MBIso. Of the left ventricular ejection fraction or strain and strain rate readouts, only
longitudinal strain was significantly different from control animals in severe myocarditis. In contrast, contrast-enhanced
ultrasound molecular imaging showed increased signals for all targeted MB versus MBIso both in moderate and severe
myocarditis, and MBCD4 signal correlated with CD4+ T-lymphocyte infiltration in the myocardium.
Conclusions—Contrast-enhanced ultrasound molecular imaging can detect endothelial inflammation and leukocyte
infiltration in myocarditis in the absence of a detectable decline in left ventricular performance by functional imaging.
In particular, imaging of CD4+ T cells involved in autoimmune responses could be helpful in diagnosing myocarditis. (Circ Cardiovasc Imaging. 2016;9:e004720. DOI: 10.1161/CIRCIMAGING.116.004720.)
Key Words: cardiomyopathy ◼ contrast media ◼ echocardiography ◼ leukocyte ◼ myocarditis
M
imaging methods for the diagnosis and characterization of
myocarditis. Echocardiography, radionuclide imaging, and
cardiac magnetic resonance imaging have all been used for
the detection of myocarditis in clinical and preclinical studies. Echocardiographic features of myocarditis including left
ventricular (LV) dilatation, segmental wall motion abnormalities, increased LV sphericity, and transient increases in LV
wall thickness7–9 are limited in specificity and hence diagnostic accuracy. For nuclear imaging, Indium-111 radiolabeled
antimyosin antibodies have been used for the detection of
yocarditis is characterized by an inflammatory infiltration of the myocardium with myocyte degeneration
and necrosis in the absence of ischemia.1 Infectious agents,
systemic diseases, drugs, and toxins cause myocarditis.2
Myocarditis has been found in 1% to 9% of autopsies and
in a higher proportion (5% to 12%) in young individuals
with sudden unexplained cardiac death.3,4 This is in contrast to the perceived low frequency of a diagnosis of myocarditis in patients. The clinical presentation of myocarditis
is variable, and routine cardiac tests either lack sensitivity
(ECG) or specificity (troponins) for diagnosing myocarditis.
Endomyocardial biopsy is the gold standard for securing the
diagnosis of a suspected myocarditis.5 However, endomyocardial biopsy is invasive, susceptible to sampling error and
suffers from low sensitivity and a high interobserver variability in interpretation.6 Thus, there is a need for noninvasive
See Editorial by Porter
See Clinical Perspective
myocardial necrosis in murine models of myocarditis and in
clinical studies.10,11 Whereas Indium-111 antimyosin antibody
Received February 25, 2016; accepted June 7, 2016.
From the Department of Biomedicine (D.C.S., L.X., E.K., E.E., A.O.-E., M.M., G.M.K., B.A.K.), Institute for Pathology University Hospital (K.G.), and
Division of Cardiology, University Hospital (G.M.K., B.A.K.), University of Basel, Switzerland.
The Data Supplement is available at http://circimaging.ahajournals.org/lookup/suppl/doi:10.1161/CIRCIMAGING.116.004720/-/DC1.
Correspondence to Beat A. Kaufmann, MD, PhD, Division of Cardiology, University Hospital Basel, Petersgraben 4, 4031 Basel, Switzerland. E-mail
[email protected]
© 2016 American Heart Association, Inc.
Circ Cardiovasc Imaging is available at http://circimaging.ahajournals.org
1
DOI: 10.1161/CIRCIMAGING.116.004720
2 Steinl et al Ultrasound Molecular Imaging of Myocarditis
Downloaded from http://circimaging.ahajournals.org/ by guest on May 12, 2017
imaging has a high sensitivity and negative predictive value,
its clinical use is hampered by limited tracer availability, radiation exposure, and delayed imaging to prevent blood pool signal.11 Cardiac magnetic resonance imaging allows detection of
myocardial edema or irreversible tissue damage inflicted by
the inflammatory infiltrate,12,13 but sensitivity is variable and
may be reduced in the absence of tissue necrosis.14 Animal
models of both virus-induced and autoimmune myocarditis suggest that heart-specific autoaggressive inflammatory
responses are important in the development of chronic cardiomyopathy, and that CD4+ T cells are instrumental in driving the transition to cardiomyopathy.15–17 Recently, fluorine-19
cardiac magnetic resonance has been used for detecting the
inflammatory infiltrate in murine autoimmune myocarditis.18
Thus, with the exception of fluorine-19 cardiac magnetic resonance, a common weakness of the aforementioned techniques
is that they detect tissue injury (edema, necrosis), which may
be absent in lymphocytic myocarditis or which can also be
because of, for example, myocardial ischemia, rather than the
inflammatory infiltrate. In the past years, microbubble (MB)
ultrasound contrast agents have been modified for the purpose
of molecular imaging. Either modifications in the shell properties or the conjugation of specific ligands to the MB shell
surface have been used to promote adhesion of MB to diseasespecific intravascular events. The feasibility of this strategy
for molecular imaging in the myocardium has been proven in
vivo.19,20 In this study, we hypothesized that contrast-enhanced
ultrasound (CEU) molecular imaging can be used to detect (1)
the inflammatory cell recruitment and endothelial inflammatory activation and (2) the recruitment of CD4+ T cells in a
murine model of myocarditis.
Methods
Study Design and Animal Model
All experiments were performed in accordance with Swiss
Legislation and were approved by the Animal Care Committee of
Basel. A total of 52 female BALB/c mice were used. In 4 mice, specific targeting of CD4 with targeted MB was validated in vitro. In
34 mice, experimental autoimmune myocarditis (EAM) was induced
by intraperitoneal injection of 400-ng pertussistoxin (List Biological
Laboratories) followed 1 hour later by subcutaneous injection of
120 µg of α-myosin heavy chain peptide (α-myosin heavy chain,
Ac-RSLKLMATLFSTYASADR-OH, GeneCust) emulsified with
complete freund adjuvant (Sigma, 1 mg/mL) on day 0. On day 7, a
repeat injection of α-myosin heavy chain only was given. Seven mice
injected with saline/complete freund adjuvant served as controls. An
additional 7 mice were studied after induction of myocarditis to verify specific targeting of CD4 using blocking antibodies. The immunization with α-myosin heavy chain results in the development of EAM
in 60% of injected animals with inflammation peaking around day
21.21 Assessments at that time point included high-resolution echocardiography; strain imaging; measurement of myocardial perfusion;
and CEU molecular imaging for leukocytes, P-selectin, and CD4+
lymphocytes. Myocarditis severity was assessed histologically, CD4+
lymphocytes were quantified with immunohistology.
MB Preparation
Perfluorocarbon-filled, lipid-shelled MB were prepared by sonification of a gas-saturated aqueous suspension of distearoyl phosphatidylcholine (2 mg/mL; Avanti Polar Lipids, Alabaster, AL) and
polyoxyethylene-(40)-stearate (1 mg/mL; Sigma). These nontargeted
MB were used for the assessment of myocardial perfusion. MB for
targeting of leukocytes (MBLc) were prepared by adding distearoylphosphatidylserine (0.3 mg/mL; Avanti Polar Lipids) to the MB
shell.22 For targeting of P-selectin and CD4, distearoyl-phosphatidylethanolamine-PEG(3400)-biotin (0.14 mg/mL; Creative PEG Works)
was added. MB targeted to P-selectin (MBPSel) and to CD4 (MBCD4)
were then prepared by conjugation of biotinylated anti–P-selectin antibody (RB40.34) or anti-CD4 antibody (H129.19) to the MB surface
using biotin–streptavidin–biotin linking as previously described.23
Control MB (MBIso) bearing a nonspecific isotype control antibody
(R3-34) were also prepared. For in vitro experiments, MB were labeled by incorporation of dioctadecyloxacarbocyanine-perchlorate
(DiO, Molecular Probes Inc) into the shell. MB concentration and
size were measured by electrozone sensing (Multisizer III, BeckmanCoulter). MB mean (±SD) size was not statistically different for the
4 MB (2.7±0.2 μm for MBIso, 2.5±0.2 μm for MBLc, 2.7±0.2 μm for
MBPSel, and 2.7±0.2 μm for MBCD4; P=ns).
In Vitro Assessment of MB–CD4 Interaction
Flow-chamber experiments were used to assess the specific attachment of MBCD4 to CD4 protein under flow conditions. CD4 protein
was adsorbed to culture dishes (35 mm, Corning) at a concentration
of 5 ng/µL by overnight incubation. The dishes were then washed
with PBS containing 0.05% Tween 20 and blocked with 1% albumin in PBS/0.05% Tween 20. The culture plates were mounted on
an inverted parallel plate flow chamber (Glycotech). The flow chamber was placed in an inverted position on a microscope (Olympus
BX51W). A suspension of either MBIso or MBCD4 at a concentration
of 3×106 mL−1 was drawn through the flow chamber at a shear rate of
2 dynes/cm2. After 5 minutes of continuous flow, the number of attached MB was counted in 20 optical fields (×20 magnification, total
area 2.94 mm2). Experiments were performed in triplicate.
Specific attachment of MBCD4 to CD4+ lymphocytes was assessed
on cell-based adhesion assays. Single-cell suspensions of murine
spleen cells from 4 mice were generated using a cell strainer. CD4+
T cells were separated from the cell suspension by depletion of magnetically labeled nontarget cells using a cocktail of biotin-conjugated
monoclonal antibodies and antibiotin monoclonal antibodies conjugated to MicroBeads (CD4+ T Cell Isolation Kit II, Miltenyi Biotec).
Both the separated nontarget cells and the CD4+ cells were resuspended at 2×107 cells per 100 µL. 1×106 leukocytes (CD4+ or CD4−)
were then incubated with 1×107 MB (MBIso or MBCD4) for 10 minutes
followed by washing with PBS. Microscopy at ×20 magnification
(Zeiss LSM 710) was used to assess MB attachment to the cells with
the investigator (E.E.) blinded to all image information. The results
are presented as the average number of MB attached per cell.
Animal Instrumentation
On day 21 after immunization, the mice were anesthetized with 5%
isoflurane, which was subsequently reduced to 2% and adapted to stabilize the heart rate between 350 and 450 bpm. The animal core temperature was maintained at 37°C. The chest and neck were depilated.
For MB injections, the right internal jugular vein was cannulated and
the animals were transferred onto a temperature controlled imaging
stage (Vevo Imaging Station).
Echocardiography
High-frequency (40 MHz, MS 550D transducer) ultrasound imaging (Vevo 2100, VisualSonics Inc) was performed for the assessment
of cardiac structure and function. M-Mode images of the LV at the
height of the papillary muscles were used to measure anterior and
posterior wall thickness, LV end-diastolic diameter, and LV end-systolic diameter. LV ejection fraction and LV mass were derived from
these measurements.24 E-wave and A-wave velocities were measured
using pulsed Doppler from a modified apical 4-chamber view.
Parasternal long- and short-axis B-mode images were acquired
at a frame rate of >200 frames per second. Longitudinal, radial,
and circumferential global peak strain values were derived from the
B-mode images using a speckle-tracking algorithm (VevoStrain,
VisualSonics).25 Cine-loops with 2 cardiac cycles not distorted by
3 Steinl et al Ultrasound Molecular Imaging of Myocarditis
Downloaded from http://circimaging.ahajournals.org/ by guest on May 12, 2017
Contrast ultrasound imaging (Sequoia Acuson C512; Siemens
Medical Systems) was performed with a high-frequency lineararray probe (15L8) held in place by a railed gantry system. The
LV was imaged in short axis at the papillary muscle level. ECGgated triggering was used to acquire end-systolic frames. For the
detection of the contrast agent, power modulation and pulse inversion imaging at a frequency of 7 MHz and a dynamic range of 50
dB was used. The gain settings were adjusted to levels just below
visible speckle and held constant. MBIso, MBLc, MBPSel, and MBCD4
(5×106 MB per injection) were injected in each mouse in random
order. Ultrasound imaging was paused from injection until 8 minutes later when imaging was resumed at a mechanical index of 0.87.
The first acquired image frame was used to derive the total amount
of MB present within the LV myocardium. The MB in the ultrasound beam were then destroyed with several (>10) image frames.
Several image frames at a long pulsing interval (every 9 heartbeats)
were subsequently acquired to measure signal attributable to freely
circulating MB. After log-linear conversion using known dynamic
range lookup tables, frames representing freely circulating MB
were digitally subtracted from the first image to derive signal from
attached MB. Contrast intensity was measured from a region of interest encompassing the whole LV myocardium. The selection of
the region of interest was guided by fundamental frequency anatomic images of the LV acquired at 14 MHz. In 7 additional mice
with myocarditis, imaging was performed 10 minutes after blocking
of CD4 by intravenous injection of antimouse CD4 mAb (H129.19,
BD Bioscience; 5 µg). After CEU molecular imaging, microvascular perfusion was assessed using constant infusions of nontargeted
MB at an infusion rate of 3×106 per minute. After a high-energy
destruction impulse (5 frames, mechanical index of 1.9), end-systolic short-axis image frames of the LV were acquired over several
heartbeats (>10) at a low mechanical index of 0.2. On the resulting
destruction–replenishment sequences, the average signal intensity
of a region of interest placed over the septal, anterior and lateral
wall was fitted to the monoexponential function y=A(1−e−ßt), where
y is the signal intensity at a given time point, ß is the rate of increase of the curve, which reflects myocardial blood flow velocity,
and A is the plateau intensity, which is proportional to myocardial
blood volume. Myocardial blood flow was estimated by the product
of A*ß. For each animal, 2 to 3 destruction–replenishment curves
were averaged.26 CEU imaging data were analyzed blinded to all
other experimental data (D.S.).
Histology and Immunohistology
For the assessment of myocarditis severity, hematoxylin and eosin–
stained short-axis cross-sections of the hearts were analyzed by a
pathologist blinded for all other experimental data. The cross section of the LV was divided into 4 quadrants by using (1) the papillary muscles and (2) an ink mark on the posterior interventricular
sulcus as landmarks. Each of the 4 quadrants was then individually
scored on a scale from 0 to 4 with 0 representing no inflammatory
infiltrates, 1 small foci of leukocytes between cells, 2 larger foci
of >100 leukocytes, 3>10% of the quadrant cross section involved,
and 4>30% of the quadrant cross section involved. In addition, an
overall score for the entire cross section using the same criteria was
given and used for the classification of animals.17 For immunohistochemistry, snap-frozen heart sections were stained with antibodies to CD4 (MCA4635T, AbD Serotec). A fluorescent secondary
Statistical Analysis
Data were analyzed on GraphPad Prism (version 7). Data are expressed as medians and 25th–75th percentile unless stated otherwise.
Comparisons between MB agents in the flow-chamber experiments
were made using a Mann–Whitney rank sum test. Attachment of
MB to CD4+ versus CD4− cells and MB signals within and between
animal groups were compared using a Kruskal–Wallis ANOVA with
Dunn post hoc test. Correlations between MB signals and CD4+ cells
in tissue were assessed using Spearman correlation. A P value of
<0.05 (2 sided) was considered statistically significant.
Results
In Vitro Assessment of MB–CD4 Interaction
The specific targeting of MBLc and MBPSel has been validated
previously, whereas this has not been the case for MBCD4.23,27
Therefore, we performed flow chamber and in vitro cell-based
assays to characterize MBCD4 interactions with CD4. In a parallel plate flow-chamber system coated with CD4 protein,
attachment of MBCD4 was significantly greater than for MBIso
at a shear stress encountered in the myocardial microcirculation (Figure 1A). In cell-based assays, MBCD4 attached to
CD4+ lymphocytes. Attachment to CD4− cells was 6-fold
less, and similar to the attachment of MBIso to both CD4+ and
CD4− cells (Figure 1B–1D).
Baseline Characteristics and Echocardiographic
Parameters
Myocarditis developed in 21 of the injected mice (n=7 severity
degree 1, n=7 severity degree 2, n=1 severity degree 3, and n=6
A
Attached MB / visual field
CEU Molecular Imaging and Assessment of
Myocardial Perfusion
antibody (A-11006 Alexa 488, Invitrogen) was used to visualize
CD4, and the cells/mm2 of tissue staining positive for CD4 were
quantified.
B
Attached MB / cell
respiratory movements were selected, and the endocardial and epicardial borders were traced on end-diastolic frames. After automated
tracing, manual adjustments were made as necessary for good tracking throughout the cardiac cycles. The strain measurements were then
averaged over the 2 cardiac cycles without the use of smoothing filters. All echocardiographic images were acquired and analyzed by
the same investigator (L.X.) who was blinded to all other experimental data.
C
MBIso
MBCD4
D
MBIso
MBCD4
CD4-cells
CD4+ cells
Figure 1. A, Number of microbubbles targeted to CD4 (MBCD4)
vs control microbubbles (MBIso) attaching to CD4 protein in a
parallel plate flow chamber. Data depict median value (horizontal line), 25% to 75% percentiles (box), and range of values
(whiskers). *P<0.0001 vs MBIso. B, Number of MBCD4 and MBIso
attaching to murine spleen cells not expressing CD4 vs cells
expressing CD4, *P<0.0001 vs all others. Examples showing
no attachment of MBCD4 to cells not expressing CD4 (C) and
attachment of several MBCD4 per individual CD4+ cell (D), red
color indicates cell nuclei and green color indicates microbubbles. MB indicates microbubbles; and MBCD4, MB bearing
antibodies targeting lymphocyte CD4.
4 Steinl et al Ultrasound Molecular Imaging of Myocarditis
Downloaded from http://circimaging.ahajournals.org/ by guest on May 12, 2017
severity degree 4; Table I in the Data Supplement), whereas
n=13 of the injected mice did not develop myocarditis (Figure 2). For all subsequent analyses, mice with severity degrees
3 and 4 were grouped together as severe myocarditis, severity degrees 1 and 2 as moderate myocarditis, whereas injected
mice that did not develop myocarditis were grouped together
with sham-injected mice as no myocarditis. In animals with
myocarditis, the inflammatory infiltrate consisted primarily of
mononuclear cells, including macrophages and lymphocytes.
Mice with severe myocarditis had lower body weight than
both the mice with no and moderate myocarditis (Table). Both
the heart weight determined postmortem and the LV mass
determined by echocardiography were higher in mice with
severe myocarditis compared with the 2 other groups. E-wave
velocity was lower in mice with severe myocarditis compared
with the other groups. For all other echocardiographic parameters, there were no significant differences between the groups
with a trend toward reduced values in animals with severe myocarditis. Of note, both β as an estimate of myocardial blood
flow velocity and A*β as an estimate of myocardial blood flow
were not significantly different between the groups, albeit with
large differences between individual animals.
LV Strain Parameters
Speckle-tracking–based strain measurements have been
reported to be more sensitive for the detection of global
changes in LV function when compared with conventional
echocardiography.25 We, therefore, assessed global peak strain
and peak strain rate on LV long- and short-axis images. In
mice with severe myocarditis, radial and circumferential strain
showed a trend toward lower values, and longitudinal strain
was significantly reduced. However, in mice with moderate
myocarditis, strain and strain rate values were identical to the
ones in mice without myocarditis (Figure 3).
Ultrasound Molecular Imaging
Figure 4 shows data for ultrasound molecular imaging in all the
3 groups of mice, and Figure 5 represents examples of colorcoded images for all 4 MB in an animal without, with moderate
and with severe myocarditis. Signal enhancement was similarly
low for all MB in mice without myocarditis. In contrast, in mice
with myocarditis, the signals from MBCD4, MBPSel, and MBLc
were all significantly higher than the signal from MBIso. Importantly, this was not just the case in mice with severe myocarditis
but also in mice with moderate myocarditis, as well as when
comparing mice without myocarditis to mice with myocarditis
(Figure 1 in the Data Supplement). In mice with severe myocarditis, an increase in signal from MBIso could be observed,
which was, however, not significant when compared with the
corresponding signals in the 2 other groups. Whereas the absolute signal intensities for MBCD4, MBPSel, and MBLc increased in
mice with severe myocarditis compared with mice with moderate myocarditis, the ratios for the targeted versus the control
MB remained constant at a 2- to 3-fold increase because of
nonsignificantly elevated signal from MBIso. When examining
the performance of ultrasound molecular imaging for diagnosing myocarditis using receiver-operating-characteristic curves,
all 3 targeted MB performed better than ejection fraction, longitudinal strain, and MBIso (Figure 2 in the Data Supplement).
In addition to in vitro validation, we performed in vivo experiments to confirm the specific attachment of MBCD4 to CD4+ T
cells. After injection of a blocking antibody in a dosage sufficient to saturate CD4 on CD4+ T cells in circulation, signal
for MBCD4 was abolished in mice with myocarditis (Figure 6).
Figure 2. Hematoxylin and eosin–stained histological sections of the myocardium in a control mouse (A and D), a mouse with moderate
myocarditis (B and E), and a mouse with severe myocarditis (C and F). The dashed lines (B) denote 2 inflammatory foci that covered ≈8%
of the left ventricular area, and the arrowhead (E) denotes a focus of >100 inflammatory cells, (F) shows extensive inflammatory infiltration
in a mouse with severe myocarditis.
5 Steinl et al Ultrasound Molecular Imaging of Myocarditis
Table. Physiological and Echocardiographic Data
No Myocarditis (n=20)
Moderate Myocarditis (n=14)
Severe Myocarditis (n=7)
Body weight (g)
22.5 (18.6–24.8)
20.0 (19.4–22.7)
15.5 (14.9–16.4)*
Heart rate (bpm)
388 (349–490)
413 (392–473)
403 (351–461)
Heart weight (mg)
112 (105–126)
119 (113–132)
171 (147–207)†
76 (65–90)
88 (72–96)
117 (82–137)‡
LV mass (mg)
LV Vol d (µL)
52 (40–57)
46 (44–57)
38 (24–52)
LV Vol s (µL)
15 (7–22)
13 (10–18)
9 (7–27)
69 (42–76)
EF (%)
73 (61–83)
70 (69–77)
−1
E-wave velocity (mm×s )
613 (563–734)
695 (567–802)
527 (405–561)§
A wave velocity (mm×s )
415 (371–454)
400 (346–531)
354 (154–428)
E/A
1.5 (1.3–1.7)
1.7 (1.3–1.8)
1.5 (1.0–2.0)
β (s )
0.8 (0.5–1.2)
0.6 (0.4–1.0)
0.7 (0.2–0.8)
A*β
10 (2–29)
18 (8–31)
26 (6–49)
−1
−1
CEU Molecular Imaging Detects Myocarditis
Independent of Functional Impairment
Despite advances in noninvasive imaging, myocarditis continues to represent a challenging diagnosis. Conventional
echocardiography can detect structural and functional alterations in fulminant myocarditis,28 but it is limited in sensitivity
and specificity in lower-grade acute myocarditis, which has a
higher likelihood of progression to dilated cardiomyopathy.29
In our study, LV mass was increased and LV function was
tended to be decreased in the group with severe myocarditis.
However, despite the use of high-resolution ultrasound and
80
Strain [%]
25
15
40
10
20
5
0
-40
-20
-30
-15
-20
-10
-10
-5
0
-25
-15
-20
-15
-10
-5
0
Strain rate [s-1]
20
60
radial
Discussion
Myocarditis and the ensuing inflammatory autoimmune phenomena can lead to chronic dilated cardiomyopathy and heart
failure. Currently available noninvasive imaging methods
detect the functional or structural consequences of infection
or inflammation, such as tissue necrosis or edema, and are
limited in diagnostic accuracy. We, therefore, investigated
whether CEU molecular imaging can detect the inflammatory
process in myocarditis, and whether the specific detection of
cellular subsets of the inflammatory infiltrate is possible. Our
results indicate that (1) CEU ultrasound molecular imaging
detects myocardial inflammatory infiltration and endothelial
cell activation independent of functional consequences in
autoimmune myocarditis and (2) we show for the first time
that CEU molecular imaging allows for the specific detection
of CD4+ T cells, which contribute to the pathogenesis of myocarditis and dilated cardiomyopathy.
speckle-tracking analysis, there were no differences in LV size
and function between the group without and the group with
moderate myocarditis. Direct assessment of the inflammatory
process responsible not only for acute myocarditis but also
for the progression to dilated cardiomyopathy may improve
the diagnostic performance of cardiac ultrasound. Our study
shows that CEU molecular imaging with MB targeted to leukocytes and to P-selectin yields increased signal not only in
the group with severe myocarditis but also in the group with
moderate myocarditis, suggesting that the direct detection of
the inflammatory infiltrate, rather than its consequences may
result in better sensitivity. In addition to relatively nonspecific
targeting of activated leukocytes using negatively charged
0
circumferential
Immunohistochemistry was used to quantify CD4+ T cells in
myocardial tissue to assess whether signal from MBCD4 tracks
with the degree of CD4+ T cell infiltration. MBCD4 signal correlated, albeit modestly, with CD4+ cells within the myocardium
on histology (Figure 7), this was not the case for MBIso in the
same animals.
longitudinal
Downloaded from http://circimaging.ahajournals.org/ by guest on May 12, 2017
Values are median (25th–75th percentile). A-wave velocity indicates late diastolic mitral inflow velocity during atrial contraction; EF,
ejection fraction; E-wave velocity, early diastolic mitral inflow velocity; LV Vol d, left ventricular volume at end diastole; and LV Vol s, left
ventricular volume at end systole.
*P=0.0014, †P=0.0008, ‡P=0.0134, §P=0.0352 vs no myocarditis.
0
*
no
Myocarditis
n=20
moderate
Myocarditis
n=14
severe
Myocarditis
n=7
-10
-5
0
Figure 3. Strain imaging data according to myocarditis severity.
Median values (horizontal line), 25% to 75% percentiles (box) and
range of values (whiskers) for peak strain (left) and peak strain
rate (right) are given. *P=0.0403 vs no myocarditis.
6 Steinl et al Ultrasound Molecular Imaging of Myocarditis
P-selectin–targeted MB was greater in severe compared with
moderate myocarditis indicates that various degrees of disease
activity can be discerned. With regard to the clinical translation of this method for detecting myocarditis, leukocyte-targeted MB perform as well as MB targeted to P-selectin with
the advantage of a greater ease of preparation of the contrast
agent.
CEU Molecular Imaging Assesses the Infiltration
With Pathogenic CD4+ Lymphocytes
MB,27 we used antibody targeting of the endothelial cell adhesion molecule P-selectin. P-selectin is expressed on the surface of the vascular endothelium on inflammatory stimulation.
In myocarditis, P-selectin plays an important role by mediating the recruitment of pathogenic T cells into myocardial tissue.30 Our results are in line with earlier reports that show that
CEU molecular imaging can detect cardiac transplant rejection by targeting activated leukocytes or intercellular adhesion
molecule-1.31,32 However, whereas the inflammatory process
in transplant rejection shares similarities with myocarditis, an
important difference that increases the diagnostic challenges
is the low grade and focal nature of the inflammatory infiltrate
in moderate myocarditis (Figure 2B and 2E). Also, the finding that CEU molecular signal for both leukocyte-targeted and
mild
myocarditis
no
myocarditis
MBIso
severe
myocarditis
Downloaded from http://circimaging.ahajournals.org/ by guest on May 12, 2017
Figure 4. Signal enhancement from targeted and isotype control
microbubbles. Data are median values (horizontal line), 25%
to 75% percentiles (box) and range of values (whiskers) of the
video intensity. *P=0.027, ‡P=0.0133, †P=0.0004, ¶P=0.0381,
¥P=0.0192, §P=0.0073 vs MBIso in the same animal group. MB
indicates microbubbles; MBCD4, MB bearing antibodies targeting lymphocyte CD4; MBPSel, MB bearing endothelial P-selectin;
MBIso, MB bearing isotype control antibody; and MBLc, MB with a
negative electric charge for targeting of leukocytes.
MBCD4
MBPSel
In animal models of both viral and autoimmune myocarditis,
CD4+ T cells represent a relatively small subset of leukocytes
infiltrating the myocardium.33 However, development of myocarditis on transfer of appropriately activated CD4+ T cells
in host animals suggests a decisive role of CD4+ T cells.17
In addition, the number of CD4+ T cells within tissue correlates with cardiac functional impairment in the chronic stage
of EAM.34 Thus, the specific detection of CD4+ T cells could
not only be used to characterize the inflammatory infiltrate in
myocarditis but could also be of value in predicting late functional consequences of myocarditis. We, therefore, developed
and validated a novel, CD4+-targeted MB using a monoclonal antibody. Flow-chamber assays and in vitro experiments
showed that this contrast agent attaches specifically to CD4+
T cells, and that attachment occurs also under flow. In vivo,
CEU molecular imaging showed increased signals both in
moderate and in severe myocarditis, and the signal correlated
with CD4+ T cells present in tissue. Abolishment of signal by
blocking of CD4+ (Figure 6) indicates selective targeting similar to in vitro observations. A previous study used imaging
of CD8+ T-cell granzyme B activity to assess immune-mediated myocarditis triggered by injection of cytotoxic T cells
with fluorescence reflectance imaging.35 However, whereas
fluorescence imaging is limited to small-animal models, CEU
molecular imaging as used in our study does not have limited
penetration and, therefore, could also be feasible in humans.
In addition, molecular imaging of specific cellular subsets of
MBLC
Figure 5. Examples of color-coded contrastenhanced ultrasound (CEU) molecular
images in 1 animal without myocarditis (top
row), in moderate myocarditis, and in severe
myocarditis (bottom row) for microbubble
bearing isotype control antibody (MBIso),
MB bearing antibodies targeting lymphocyte CD4 (MBCD4), MB bearing endothelial
P-selectin (MBPsel), and MB with a negative
electric charge for targeting of leukocytes
(MBLc). The color scale for the CEU images
is shown at the bottom of each frame.
7 Steinl et al Ultrasound Molecular Imaging of Myocarditis
Acoustic intensity (AU)
10
8
6
MBIso
4
MBCD4 + H129.19
2
MBLc
0
moderate and severe
myocarditis
n=7
A
Acoustic intensity (AU)
the inflammatory immune response in myocarditis may help
further elucidating the mechanisms of immune-mediated
myocardial disease and be of use in the assessment of treatment effects in the future.
There are several limitations of this study that deserve
attention. We studied a model of EAM, rather than virusinduced disease. However, EAM shares many pathophysiological features with virus-induced myocarditis, and it has
been used extensively to study the autoimmune processes
caused by initial viral myocyte infection.36 Regarding CEU
molecular imaging, differences in myocardial perfusion
between the groups could potentially influence the results.
Although there was a high variability in myocardial perfusion in mice with severe myocarditis, and localized perfusion abnormalities in areas with tissue necrosis cannot be
excluded in this animal group, we did not find significant
15
B
C
rs = 0.70
p < 0.001
10
5
0
0
Acoustic intensity (AU)
Downloaded from http://circimaging.ahajournals.org/ by guest on May 12, 2017
Figure 6. Blocking of CD4 abolishes signal from microbubble
bearing antibodies targeting lymphocyte CD4 (MBCD4). Data are
median values (horizontal line), 25% to 75% percentiles (box) and
range of values (whiskers) of the video intensity. *P=0.0209 versus MB bearing isotype control antibody (MBIso). MBLc indicates
MB with a negative electric charge for targeting of leukocytes.
15
50
100
150
200
CD4+ cells/mm2
50 m
250
D
rs = 0.21
p = 0.33
10
5
0
0
50
100
150
200
CD4+ cells/mm2
250
50 m
Figure 7. Correlation of contrast-enhanced ultrasound molecular
imaging data for MBCD4 (A) and MBIso (B). Examples of extensive
myocardial CD4+ T-lymphocyte infiltration (C) and low-level infiltration (D) on immunohistochemistry, green color indicates positive staining for CD4.
differences between the groups, and perfusion was identical
in mice without and mice with moderate myocarditis. In addition, the low signals for control MB in both the groups with
myocarditis argue against an influence of myocardial flow
on CEU molecular imaging results. Also, although moderate myocarditis could be detected independent of functional
defects, ultrasound molecular imaging may not be sensitive
enough to diagnose the early stages of myocarditis when
only a few inflammatory foci or minimal myocyte necrosis
are present. Finally, the limited spatial resolution of CEU at
7 MHz precluded a regional assessment of disease activity,
which would be desirable for the guidance of endomyocardial biopsies in patients.
In summary, the results of our study indicate that CEU
molecular imaging can be used to detect the inflammatory
process in myocarditis and discriminate disease severity independent of cardiac function. Importantly, specific molecular
imaging of a leukocyte subset that is driving the autoimmune
process in myocarditis is possible. As a consequence, CEU
molecular imaging can potentially be useful in diagnosing
myocarditis. Future studies will have to address whether
imaging of the acute inflammatory response and its components also allows a risk assessment about the future development of dilated cardiomyopathy.
Sources of Funding
This work was supported by grants from the Swiss National Science
Foundation (SNF 310030_149718 to Dr Kaufmann and SNF
310030_144208 to Dr Kuster), the Swiss Life Jubiläumsstiftung and
the Kardiovaskuläre Stiftung Basel.
Disclosures
None.
References
1. Caforio AL, Pankuweit S, Arbustini E, Basso C, Gimeno-Blanes J, Felix
SB, Fu M, Heliö T, Heymans S, Jahns R, Klingel K, Linhart A, Maisch B,
McKenna W, Mogensen J, Pinto YM, Ristic A, Schultheiss HP, Seggewiss
H, Tavazzi L, Thiene G, Yilmaz A, Charron P, Elliott PM; European Society
of Cardiology Working Group on Myocardial and Pericardial Diseases.
Current state of knowledge on aetiology, diagnosis, management, and
therapy of myocarditis: a position statement of the European Society of
Cardiology Working Group on Myocardial and Pericardial Diseases. Eur
Heart J. 2013;34:2636–2648, 2648a. doi: 10.1093/eurheartj/eht210.
2. Sagar S, Liu PP, Cooper LT Jr. Myocarditis. Lancet. 2012;379:738–747.
doi: 10.1016/S0140-6736(11)60648-X.
3. Theleman KP, Kuiper JJ, Roberts WC. Acute myocarditis (predominately
lymphocytic) causing sudden death without heart failure. Am J Cardiol.
2001;88:1078–1083.
4. Doolan A, Langlois N, Semsarian C. Causes of sudden cardiac death in
young Australians. Med J Aust. 2004;180:110–112.
5. Aretz HT, Billingham ME, Edwards WD, Factor SM, Fallon JT, Fenoglio
JJ Jr, Olsen EG, Schoen FJ. Myocarditis. A histopathologic definition and
classification. Am J Cardiovasc Pathol. 1987;1:3–14.
6. Chow LH, Radio SJ, Sears TD, McManus BM. Insensitivity of right ventricular endomyocardial biopsy in the diagnosis of myocarditis. J Am Coll
Cardiol. 1989;14:915–920.
7.Pinamonti B, Alberti E, Cigalotto A, Dreas L, Salvi A, Silvestri F,
Camerini F. Echocardiographic findings in myocarditis. Am J Cardiol.
1988;62:285–291.
8. Mendes LA, Picard MH, Dec GW, Hartz VL, Palacios IF, Davidoff R.
Ventricular remodeling in active myocarditis. Myocarditis Treatment
Trial. Am Heart J. 1999;138(2 pt 1):303–308.
9. Hiramitsu S, Morimoto S, Kato S, Uemura A, Kubo N, Kimura K, Sugiura
A, Itoh T, Hishida H. Transient ventricular wall thickening in acute
8 Steinl et al Ultrasound Molecular Imaging of Myocarditis
Downloaded from http://circimaging.ahajournals.org/ by guest on May 12, 2017
myocarditis: a serial echocardiographic and histopathologic study. Jpn
Circ J. 2001;65:863–866.
10. Kishimoto C, Hung GL, Ishibashi M, Khaw BA, Kolodny GM, Abelmann
WH, Yasuda T. Natural evolution of cardiac function, cardiac pathology
and antimyosin scan in a murine myocarditis model. J Am Coll Cardiol.
1991;17:821–827.
11. Narula J, Khaw BA, Dec GW, Palacios IF, Newell JB, Southern JF, Fallon
JT, Strauss HW, Haber E, Yasuda T. Diagnostic accuracy of antimyosin
scintigraphy in suspected myocarditis. J Nucl Cardiol. 1996;3:371–381.
12. Thavendiranathan P, Walls M, Giri S, Verhaert D, Rajagopalan S, Moore S,
Simonetti OP, Raman SV. Improved detection of myocardial involvement in
acute inflammatory cardiomyopathies using T2 mapping. Circ Cardiovasc
Imaging. 2012;5:102–110. doi: 10.1161/CIRCIMAGING.111.967836.
13.Mahrholdt H, Goedecke C, Wagner A, Meinhardt G, Athanasiadis A,
Vogelsberg H, Fritz P, Klingel K, Kandolf R, Sechtem U. Cardiovascular
magnetic resonance assessment of human myocarditis: a comparison to
histology and molecular pathology. Circulation. 2004;109:1250–1258.
doi: 10.1161/01.CIR.0000118493.13323.81.
14.Francone M, Chimenti C, Galea N, Scopelliti F, Verardo R, Galea R,
Carbone I, Catalano C, Fedele F, Frustaci A. CMR sensitivity varies with
clinical presentation and extent of cell necrosis in biopsy-proven acute
myocarditis. JACC Cardiovasc Imaging. 2014;7:254–263. doi: 10.1016/j.
jcmg.2013.10.011.
15. Woodruff JF, Woodruff JJ. Involvement of T lymphocytes in the pathogenesis of coxsackie virus B3 heart disease. J Immunol. 1974;113:1726–1734.
16. Huber SA, Sartini D, Exley M. Vgamma4(+) T cells promote autoimmune CD8(+) cytolytic T-lymphocyte activation in coxsackievirus
B3-induced myocarditis in mice: role for CD4(+) Th1 cells. J Virol.
2002;76:10785–10790.
17.Eriksson U, Ricci R, Hunziker L, Kurrer MO, Oudit GY, Watts TH,
Sonderegger I, Bachmaier K, Kopf M, Penninger JM. Dendritic cellinduced autoimmune heart failure requires cooperation between adaptive
and innate immunity. Nat Med. 2003;9:1484–1490. doi: 10.1038/nm960.
18. van Heeswijk RB, De Blois J, Kania G, Gonzales C, Blyszczuk P, Stuber
M, Eriksson U, Schwitter J. Selective in vivo visualization of immune-cell
infiltration in a mouse model of autoimmune myocarditis by fluorine-19
cardiac magnetic resonance. Circ Cardiovasc Imaging. 2013;6:277–284.
doi: 10.1161/CIRCIMAGING.112.000125.
19.Villanueva FS, Lu E, Bowry S, Kilic S, Tom E, Wang J, Gretton J,
Pacella JJ, Wagner WR. Myocardial ischemic memory imaging with molecular echocardiography. Circulation. 2007;115:345–352. doi: 10.1161/
CIRCULATIONAHA.106.633917.
20. Kaufmann BA, Lewis C, Xie A, Mirza-Mohd A, Lindner JR. Detection of
recent myocardial ischaemia by molecular imaging of P-selectin with targeted contrast echocardiography. Eur Heart J. 2007;28:2011–2017. doi:
10.1093/eurheartj/ehm176.
21. Pummerer CL, Luze K, Grässl G, Bachmaier K, Offner F, Burrell SK,
Lenz DM, Zamborelli TJ, Penninger JM, Neu N. Identification of cardiac
myosin peptides capable of inducing autoimmune myocarditis in BALB/c
mice. J Clin Invest. 1996;97:2057–2062. doi: 10.1172/JCI118642.
22. Lindner JR, Dayton PA, Coggins MP, Ley K, Song J, Ferrara K, Kaul S.
Noninvasive imaging of inflammation by ultrasound detection of phagocytosed microbubbles. Circulation. 2000;102:531–538.
23. Lindner JR, Song J, Christiansen J, Klibanov AL, Xu F, Ley K. Ultrasound
assessment of inflammation and renal tissue injury with microbubbles targeted to P-selectin. Circulation. 2001;104:2107–2112.
24. Gao S, Ho D, Vatner DE, Vatner SF. Echocardiography in mice. Curr Protoc
Mouse Biol. 2011;1:71–83. doi: 10.1002/9780470942390.mo100130.
25. Bauer M, Cheng S, Jain M, Ngoy S, Theodoropoulos C, Trujillo A, Lin
FC, Liao R. Echocardiographic speckle-tracking based strain imaging for
rapid cardiovascular phenotyping in mice. Circ Res. 2011;108:908–916.
doi: 10.1161/CIRCRESAHA.110.239574.
26.Raher MJ, Thibault H, Poh KK, Liu R, Halpern EF, Derumeaux G,
Ichinose F, Zapol WM, Bloch KD, Picard MH, Scherrer-Crosbie M. In
vivo characterization of murine myocardial perfusion with myocardial
contrast echocardiography: validation and application in nitric oxide synthase 3 deficient mice. Circulation. 2007;116:1250–1257. doi: 10.1161/
CIRCULATIONAHA.107.707737.
27. Lindner JR, Song J, Xu F, Klibanov AL, Singbartl K, Ley K, Kaul S.
Noninvasive ultrasound imaging of inflammation using microbubbles targeted to activated leukocytes. Circulation. 2000;102:2745–2750.
28.Felker GM, Boehmer JP, Hruban RH, Hutchins GM, Kasper EK,
Baughman KL, Hare JM. Echocardiographic findings in fulminant and
acute myocarditis. J Am Coll Cardiol. 2000;36:227–232.
29. McCarthy RE III, Boehmer JP, Hruban RH, Hutchins GM, Kasper EK,
Hare JM, Baughman KL. Long-term outcome of fulminant myocarditis as compared with acute (nonfulminant) myocarditis. N Engl J Med.
2000;342:690–695. doi: 10.1056/NEJM200003093421003.
30. Cai YH, Alvarez A, Alcaide P, Duramad P, Lim YC, Jarolim P, Lowe JB,
Luscinskas FW, Lichtman AH. Abrogation of functional selectin-ligand
expression reduces migration of pathogenic CD8+ T cells into heart. J
Immunol. 2006;176:6568–6575.
31. Weller GE, Lu E, Csikari MM, Klibanov AL, Fischer D, Wagner WR,
Villanueva FS. Ultrasound imaging of acute cardiac transplant rejection with
microbubbles targeted to intercellular adhesion molecule-1. Circulation.
2003;108:218–224. doi: 10.1161/01.CIR.0000080287.74762.60.
32. Kondo I, Ohmori K, Oshita A, Takeuchi H, Yoshida J, Shinomiya K, Fuke
S, Suzuki T, Mizushige K, Kohno M. Leukocyte-targeted myocardial contrast echocardiography can assess the degree of acute allograft rejection
in a rat cardiac transplantation model. Circulation. 2004;109:1056–1061.
doi: 10.1161/01.CIR.0000115586.25803.D5.
33. Afanasyeva M, Georgakopoulos D, Rose NR. Autoimmune myocarditis:
cellular mediators of cardiac dysfunction. Autoimmun Rev. 2004;3:476–
486. doi: 10.1016/j.autrev.2004.08.009.
34.Afanasyeva M, Georgakopoulos D, Belardi DF, Ramsundar AC, Barin
JG, Kass DA, Rose NR. Quantitative analysis of myocardial inflammation by flow cytometry in murine autoimmune myocarditis: correlation
with cardiac function. Am J Pathol. 2004;164:807–815. doi: 10.1016/
S0002-9440(10)63169-0.
35.Konishi M, Erdem SS, Weissleder R, Lichtman AH, McCarthy
JR, Libby P. Imaging granzyme B activity assesses immune-mediated myocarditis. Circ Res. 2015;117:502–512. doi: 10.1161/
CIRCRESAHA.115.306364.
36. Eriksson U, Penninger JM. Autoimmune heart failure: new understandings
of pathogenesis. Int J Biochem Cell Biol. 2005;37:27–32. doi: 10.1016/j.
biocel.2004.06.014.
CLINICAL PERSPECTIVE
Myocarditis is an important cause for heart failure and sudden cardiac death, particularly in young individuals. Currently
used cardiac tests for diagnosing myocarditis lack sensitivity or specificity. Ultrasound molecular imaging with site-targeted
microbubble contrast agents detects inflammatory cell recruitment and endothelial activation. In this study, we examined
whether contrast-enhanced ultrasound molecular imaging is feasible for the detection of specific components of the inflammatory process in myocarditis. In a murine model of autoimmune myocarditis, ultrasound molecular imaging with contrast
agents targeted to leukocytes, to P-selectin on endothelial cells, and to CD4+ lymphocytes was performed together with
advanced high-resolution ultrasound assessment of left ventricular function and histological evaluation. Signals from targeted contrast agents were elevated not only in histologically severe, but also in moderate myocarditis. Importantly, signal
from targeted microbubbles in moderate myocarditis was not accompanied by a decrease in left ventricular function on highresolution 2-dimensional and strain imaging. Also, signal from CD4+ targeted microbubbles correlated to CD4+ lymphocytes in tissue. Our results suggest that ultrasound molecular imaging may be feasible for the detection and characterization
of inflammation in patients with myocarditis even in the absence of functional impairment of the myocardium.
Noninvasive Contrast-Enhanced Ultrasound Molecular Imaging Detects Myocardial
Inflammatory Response in Autoimmune Myocarditis
David C. Steinl, Lifen Xu, Elham Khanicheh, Elin Ellertsdottir, Amanda Ochoa-Espinosa,
Martina Mitterhuber, Katharina Glatz, Gabriela M. Kuster and Beat A. Kaufmann
Downloaded from http://circimaging.ahajournals.org/ by guest on May 12, 2017
Circ Cardiovasc Imaging. 2016;9:
doi: 10.1161/CIRCIMAGING.116.004720
Circulation: Cardiovascular Imaging is published by the American Heart Association, 7272 Greenville Avenue,
Dallas, TX 75231
Copyright © 2016 American Heart Association, Inc. All rights reserved.
Print ISSN: 1941-9651. Online ISSN: 1942-0080
The online version of this article, along with updated information and services, is located on the
World Wide Web at:
http://circimaging.ahajournals.org/content/9/8/e004720
Data Supplement (unedited) at:
http://circimaging.ahajournals.org/content/suppl/2016/08/08/CIRCIMAGING.116.004720.DC1
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published
in Circulation: Cardiovascular Imaging 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 Circulation: Cardiovascular Imaging is online at:
http://circimaging.ahajournals.org//subscriptions/
SUPPLEMENTAL MATERIAL
Supplementary table. Global and regional histology scores in animals with
moderate (grade 1 and 2) and severe (grade 3 and 4) myocarditis.
Global histology
score
Moderate myocarditis
1
2
3
4
5
6
7
8
9
10
11
12
13
14
Severe myocarditis
1
2
3
4
5
6
7
Regional histology scores
Q1
Q2
Q3
Q4
2
2
2
1
1
2
1
1
2
2
2
1
1
1
1
0
0
0
1
0
0
1
2
0
0
1
1
1
0
2
2
0
1
2
1
0
2
0
2
0
0
0
2
2
2
1
1
2
0
1
2
0
0
0
0
0
1
0
1
0
1
0
0
0
1
2
0
0
0
0
3
3
4
4
3
4
4
1
2
4
4
3
4
4
2
3
4
4
3
4
4
3
2
4
4
3
4
4
3
1
4
4
3
4
4
Radial strain [%]
80
60
40
20
0
C
Longitudinal strain [%]
B
Circumferential strain [%]
A
-40
-30
-20
-10
0
E 15
150
100
50
0
no myocarditis
(n=20)
myocarditis
(n=21)
Acoustic intensity (AU)
Ejection fraction [%]
D
-25
-20
-15
-10
-5
0
†"
10
*"
‡"
5
0
no myocarditis
myocarditis
n=20
n=21
Supplementary Figure 1. Comparison of conventional echocardiography, strain
measurements and molecular imaging between mice without and with with
myocarditis. Radial (A), circumferential (B) and longitudinal strain (C), as well as
ejection fraction (D) did not show differences between the groups. In contrast,
molecular imaging (E) showed significantly elevated signals in animals with
myocarditis for microbubbles targeted to CD4, P-selectin and to leukocytes (**p<0.01
vs MBIso. (***p<0.001 vs MBIso).
Longitudinal strain
1.0
0.8
0.8
0.8
0.6
0.4
0.0
0.0
Sensitivity
1.0
0.6
0.4
0.2
0.2
0.2
0.4
0.6
0.8
0.0
0.0
1.0
0.6
0.4
0.2
0.2
0.4
0.6
0.8
0.0
0.0
1.0
0.4
0.6
1 - Specificity
MBCD4
MBPSel
MBLc
1.0
0.8
0.8
0.8
0.6
0.4
0.2
Sensitivity
1.0
0.6
0.4
0.4
0.6
1 - Specificity
0.8
1.0
0.0
0.0
0.8
1.0
0.6
0.4
Supplementary
0.2
0.2
0.2
0.2
1 - Specificity
1.0
0.0
0.0
1 - Specificity
Sensitivity
Sensitivity
MBIso
1.0
Sensitivity
Sensitivity
Ejection fraction
0.2
0.4
0.6
1 - Specificity
0.8
1.0
0.0
0.0
0.2
0.4
0.6
0.8
1.0
1 - Specificity
Figure 2. Receiver-operating –characteristic (ROC) curves for diagnosing
myocarditis (n=20 mice without myocarditis, n=21 mice with myocarditis). Area
under the curve (95% confidence interval) was 0.55 (0.35 – 0.7) for ejection fraction,
0.60 (0.41 – 0.78) for longitudinal strain, 0.57 (0.39 – 0.75) for MBCtr, 0.86 (0.74 –
0.98) for MBCD4, 0.90 (0.81 – 0.99) for MBPSel, and 0.82 (0.68 – 0.95) for MBLc.