JOURNAL OF MAGNETIC RESONANCE IMAGING 16:716 –720 (2002)
Original Research
Development of an Intravascular Heating Source
Using an MR Imaging Guidewire
Bensheng Qiu, PhD, Christopher J. Yeung, BS, Xiangying Du, MD, Ergin Atalar, PhD,
and Xiaoming Yang, MD, PhD*
Purpose: To develop a novel endovascular heating source
using a magnetic resonance (MR) imaging guidewire
(MRIG) to deliver controlled microwave energy into the target vessel for thermal enhancement of vascular gene transfection.
Materials and Methods: A 0.032-inch MRIG was connected to a 2.45-GHz microwave generator. We 1) calculated the microwave power loss along the MRIG, 2) simulated the power distribution around the MRIG, 3) measured
the temperature increase vs. input power with the MRIG,
and 4) evaluated the thermal effect on the balloon-compressed/microwave-heated aorta of six living rabbits. In
addition, during balloon inflation, we also simultaneously
generated high-resolution MR images of the aortic wall.
Results: The power loss was calculated to be 3.9 dB along
the MRIG. The simulation-predicted power distribution
pattern was cylindrically symmetric, analogous to the geometry of vessels. Under balloon compression, the vessel
wall could be locally heated at 41°C with no thermal damage apparent on histology.
Conclusion: This study demonstrates the possibility of using the MRIG as a multifunctional device, not only as a
receiver antenna to generate intravascular high-resolution
MR images of atherosclerotic plaques and as a conventional
guidewire to guide endovascular interventions during MR
imaging, but also as a potential intravascular heating
source to produce local heat for thermal enhancement of
vascular gene transfection.
Key Words: vascular gene therapy; microwave heating; MR
imaging; MR imaging guidewire; cardiovascular intervention
J. Magn. Reson. Imaging 2002;16:716 –720.
© 2002 Wiley-Liss, Inc.
GENE THERAPY IS A rapidly expanding field with great
potential for the treatment of cardiovascular diseases
Department of Radiology, Johns Hopkins University, School of Medicine, Baltimore, Maryland.
Contract grant sponsor: National Institutes of Health; Contract grant
numbers: R01 HL67195 and R01 HL66187.
A preliminary form of this paper was accepted as an oral presentation at
the ISMRM annual meeting, Honolulu, Hawaii, May 18 –24, 2002.
*Address reprint requests to: X.Y., Department of Radiology, Johns
Hopkins University, School of Medicine, Traylor Bldg., Rm 330, 720
Rutland Ave., Baltimore, MD 21205. E-mail: [email protected]
Received March 22, 2002; Accepted August 13, 2002.
DOI 10.1002/jmri.10213
Published online in Wiley InterScience (www.interscience.wiley.com).
© 2002 Wiley-Liss, Inc.
(1). There are two primary challenging tasks for successful vascular gene therapy. One is the precise delivery of genes into targeted atherosclerotic plaques, and
the other is the efficient transfer of genes into the endothelial cells/smooth muscle cells of the target vessels. Catheter-based delivery of genes offers a promising therapeutic approach to localize a high dose of the
transgene at the target site while minimizing any undesirable systemic transfection (2). Previous in vitro studies from other groups and our group have confirmed
that controlled heating can enhance gene transfection
(3–5). However, in clinical practice, it is not feasible to
heat the entire body of a patient. Thus, we need to
generate local heat only at the target site. One of the
strategies to address this is to develop an internal heating source, which should be small enough to be easily
placed into a local target via naturally existing anatomic
channels, such as vessels.
We have developed a magnetic resonance (MR) imaging guidewire (MRIG), a coaxial antenna with an extended inner conductor (6), that enables generation of
high-resolution MR images of the arterial wall (7) and
guidance of endovascular interventions during MR imaging (8). Using this MRIG, we have also recently developed a new technique of in vivo MR imaging of catheterbased vascular gene delivery, which enabled us to
precisely monitor gene/vector delivery into target vessel
walls (9). In the current study, we attempted to increase
the function of the MRIG by using it as an intravascular
heating source to deliver external thermal energy into
the target vessel wall during MR imaging of vascular
gene delivery, and, thereby, enhance vascular gene
transfection. This “three-in-one” MRIG design should
benefit vascular gene therapy because of its multiple
functions as: 1) a receiver antenna to generate intravascular high-resolution MR imaging of atherosclerotic
plaques of the vessel wall, 2) a conventional guidewire
to guide endovascular interventions during MR imaging, and 3) an intravascular heating source to produce
local heat in the target vessel to enhance vascular gene
transfection “on-line” during the same therapeutic interventional procedure in the same MR scanner.
This study focused on the investigation of the possibility of using a clinical size, 0.032-inch diameter MRIG
as an intravascular heating source to deliver controlled
thermal energy from an external microwave generator
into the target vessels for the enhancement of catheter-
716
Intravascular Heating Source Using a MRIG
717
based vascular gene transfection. To this end, we 1)
calculated microwave power loss along the MRIG to
determine the adequate microwave power input, 2) simulated microwave power distribution around the active
imaging/heating region of the MRIG (centered at the
conjunction of the inner and outer conductors of the
MRIG) to determine the pattern of power distribution at
the target, 3) measured the temperature increase vs.
microwave input power with the MRIG to establish a
microwave heating protocol, and 4) measured the thermal effect on the balloon-compressed/microwaveheated target vessel wall correlated with histology to
ensure the safety of the MRIG as an intravascular heating source in vivo.
MATERIALS AND METHODS
Devices
We used a clinical size, 0.032-inch diameter MRIG that
has been successfully used to generate intravascular
high-resolution MR images of the atherosclerotic vessel
wall and for guidance of vascular interventions during
MR imaging (7,8). The MRIG was a loopless antenna
that consisted of an 8-cm long conducting wire that was
an extended inner conductor from a 25-inch long coaxial cable. The inner conductor of the MRIG consisted of
a gold-plated nitinol wire with a radius of 0.1 mm. The
dielectric was polytetrafluoroethylene (⑀r ⫽ 2) and had a
radius of 0.33 mm. The outer conductor consisted of
nitinol (␴ ⫽ 1 ⫻ 106 S/m) with a radius of 0.42 mm. The
MRIG was connected either to an external 2.45-GHz
microwave generator (Opthos Instruments, Rockville,
MD) for heating, or connected to a MR scanner for
imaging.
Evaluation of Microwave Power Loss along the
MRIG
Before using the MRIG to deliver microwave energy into
a target vessel, the first concern was power losses along
the MRIG because the diameter of MRIG is small and its
electrical resistance is relatively high. We calculated
microwave power loss along the MRIG by modeling a
lossy transmission line with distribution parameters.
The formula for the attenuation factor is given in Ulaby
(10).
Evaluation of Microwave Power Distribution
around the MRIG
The power distribution was calculated from the rootmean-square (RMS) electric field (E) distribution (11).
The electric field distribution was calculated by assuming a discretized balanced cosine current on the MRIG
and integrating the electric field from each current element, as described by Simon et al (12). The localized
power distribution of the MRIG was calculated for tissue electrical conductivity (␴) ⫽ 2.5 S/m, and tissue
dielectric constant (⑀r) ⫽ 55, which are representative
values for human tissue at 2.45 GHz (13).
In order to characterize the distribution of the heating
power along the MRIG, we defined a half-power heating
Figure 1. In vivo experimental setup in a rabbit aorta. MWG ⫽
microwave generator.
length as the length over which half of the entire emission power is deposited.
Evaluation of Temperature Increase Vs. Input
Microwave Power With the MRIG
To investigate the feasibility of using a MRIG to deliver
microwave power to the vessel and establish a heating
protocol for the enhancement of vascular gene transfection, we tested the temperature increase vs. input
microwave power in vivo. We used one New Zealand
white rabbit, approximately 5 kg in weight, with an
aorta approximately 6 mm in diameter. All animals
were treated according to the “Principles of Laboratory
Animal Care” of the National Society for Medical Research and the “Guide for the Care and Use of Laboratory Animals” (NIH Publication No. 80-23, revised
1985). The Animal Care and Use Committee at our
institution approved the experimental protocol.
Through a laparotomy, we positioned both a 5-F balloon catheter with a balloon portion 6 mm in diameter
and 2 cm in length (Boston Scientific, Boston, MA), and
a 0.6-mm fiber-optic temperature sensor (FISO Technologies, Ste-Foy, Quebec, Canada) into the lower abdominal aorta at a level 2 cm below the renal arteries
(Fig. 1). The sensor portion of the fiber-optic probe was
attached side-by-side onto the balloon. Thus, inflation
of the balloon with 37°C saline propelled the fiber-optic
probe against the arterial wall. We then placed the
MRIG into the balloon catheter so that the active imaging/heating region of the MRIG was positioned in the
center of the balloon. The MRIG was connected to an
718
Qiu et al.
Figure 2. Calculated power deposition pattern of the MRIG (antenna) at 2.45-GHz microwave frequency. The power deposition
is concentrated in a small volume that matches the 2-cm balloon within the target vessel. [Color figure can be viewed in the
online issue, which is available at www.interscience.wiley.com.]
external 2.45-GHz microwave generator, and the fiberoptic sensor was connected to a digital thermometer
(FISO Technologies, Canada). Subsequently, we operated the microwave generator, at different power levels
from 2 to 24 watts (W), to deliver thermal energy to the
MRIG. The steady-state temperature increases after
several seconds were recorded using a digital thermometer. We repeated the heating experiments in triplicate.
The results were then converted to a curve of temperature increases vs. input microwave powers.
Evaluation of the Thermal Effect on a Target
Vessel Wall With the MRIG
In our experimental setting, the target vessel wall was
locally heated during mechanical compression by the
inflated balloon. The mechanical compression would
facilitate the microwave heating-induced thermal damage on the balloon-stressed target vessel wall. We evaluated, in vivo, the potential thermal damage from microwave heating/balloon compression by examining
the histological differences before and after microwave
heating at the target in living animals. In addition, during the in vivo experiments we also tested the possibility
of simultaneously generating high-resolution MR images
of the target vessels using the same heat-delivery MRIG
with the inflated balloon in the same target vessel.
We used six New Zealand white rabbits, approximately 5 kg in weight. With the same surgical method
mentioned above, we positioned the 0.032-inch MRIG,
along with the 5-F balloon and the 0.6-mm fiber-optic
temperature sensor, into the aorta. Then, while inflating the balloon with saline, we heated the targeted aorta
with the MRIG for 20 minutes by operating the microwave generator at 20 –25 W, which resulted in a temperature increase to 41°C at the target aortic wall.
MR imaging was performed in a 1.5-Tesla MR scanner (GE Medical System, Milwaukee, WI). We acquired
high-resolution axial and sagittal images with two pulse
sequences: 1) T1-weighted imaging with a spin-echo
(SE) sequence of 500/11 msec TR/TE, 4- and 8-cm field
of view (FOV), and 256 ⫻ 256 matrix; and 2) T2weighted imaging with a fast SE (FSE) sequence of
2000/100 msec TR/TE, 15.6-kHz bandwidth, 4- and
8-cm FOV, and 256 ⫻ 256 matrix. During MR imaging,
the MRIG was connected to the MR imaging preamplifier and operated in the receive-only mode.
Immediately after the heating, we harvested the heattargeted aorta and non-heated aorta (as a control) for
histological examination. The specimens were embedded in paraffin, cut into 5-␮m slices on a cross-sectional view, and stained with Masson Trichrome stain.
RESULTS
The calculated microwave power loss along the MRIG
was 6.21 dB/m. Because the MRIG was 25 inches (0.63
m) in length, the power loss along the MRIG was 3.9 dB.
Figure 3. Measurement of temperature increase vs. microwave (MW) input power using the 0.032-inch MRIG in vivo. The
rabbit aorta is heated from 37°C to 41°C by delivering external
microwave input power from 2 to 24 W.
Intravascular Heating Source Using a MRIG
719
Figure 4. High-resolution MR imaging of a rabbit aorta. The 0.032-inch MRIG is placed within an angioplasty balloon that is
positioned in the aorta. A: Axial T1-weighted image with inflated balloon (SE, TR/TE ⫽ 500/14, 4 ⫻ 4-cm FOV, 256 ⫻ 256
matrix). B: Axial T2-weighted MR image with inflated balloon (FSE, TR/TE ⫽ 2000/100, 8 ⫻ 8-cm FOV, 256 ⫻ 256 matrix). The
arrows indicate the aortic wall. C: Sagittal T2 images of the inflated balloon with saline (FSE, TR/TE ⫽ 2000/100, 6 ⫻ 6-cm FOV,
256 ⫻ 256 matrix). The arrows indicate the metal marks of the balloon.
The pattern of simulated power distribution was cylindrically symmetric, analogous to the geometry of vessels, and was localized to the target vessel area (Fig. 2).
The calculated half-power heating length was 1.7 cm at
the microwave frequency, which is close to the 2-cm
length of the 5-F catheter balloon. The measurements
of temperature increase vs. input microwave power are
shown in Figure 3. Although some microwave energy
was lost through the MRIG during microwave energy
transfer, we could achieve the desired temperature of
41°C at the target vessel wall in vivo.
During MR imaging, we were able to monitor the inflation/deflation of the balloon. On T1- and T2-weighted MR
images, we could visualize the balloon-inflated target
aortic wall at a resolution of 157 ␮m (Fig. 4).
Clinically, all rabbits survived during the experiments.
Histopathologically, in both gross and microscopic examinations, there were no findings of thermal damage, such
as vacuolization, coagulation, or carbonization (Fig. 5).
We did not find any evidence of possible mechanical injury to the vessel wall due to the insertion of the MRIG.
DISCUSSION
The present study demonstrates the potential of using
the MRIG as a multifunctional device for vascular gene
therapy, not only as a receiver antenna to generate
intravascular high-resolution MR imaging of the target
vessel wall and as a conventional guidewire to guide
endovascular interventions (7,8), but also as an intravascular local heating source to deliver controlled therapeutic heat into target vessels. The advantages of using the MRIG as an intravascular local heating source
include: 1) the thin MRIG can be easily positioned, via
any endovascular interventional device, into a target
vessel to generate local and axisymmetric heating at the
target; 2) the MRIG produces a power distribution that
is cylindrically symmetric and analogous to the geometry of vessels and localized to the target vessel area;
and 3) the MRIG can be used as a multifunctional
device to simultaneously generate imaging and heating
at the target vessels, and thermal power input through
the MRIG can be easily controlled at the external mi-
Figure 5. Histologic findings in a control (unheated) aorta (A) and the heated aorta (B) of a rabbit. Microscopy shows no abnormal changes
(200⫻, Masson Trichrome stain). [Color figure can be viewed in the online issue which is available at www.interscience.wiley.com.]
720
crowave generator. In addition, it should be greatly beneficial to combine MR thermal mapping techniques with
our current design, to monitor and control the location,
distribution, and extent of the delivered therapeutic
heat at the target vessels during MR imaging.
In our experimental setup, the resistance of the MRIG
was relatively high because the outer conductor of the
0.032-inch MRIG is made of nitinol (␴ ⫽ 106 S/m).
There is about 60% input power lost along the MRIG,
and 40% input power is used to heat the target. When
inputting 24 W of power, the heat strength at the active
imaging/heating region of the MRIG is 5.65 W/cm, and
the average power lost along the remaining region of the
MRIG is 0.23 W/cm, which is much less than that at
the active imaging/heating region, and, therefore, there
is no significant temperature increase along the MRIG.
To decrease the resistance of the MRIG, we may coat the
inner surface of the outer conductor (tube) with copper
(␴ ⫽ 5.8 ⫻ 107 S/m) or silver (␴ ⫽ 6.2 ⫻ 107 S/m), and,
thereby, its power loss will be decreased from 6.21
dB/m to 2.65 dB/m. Thus, along the 25-inch long
MRIG we used, the total power loss would decrease to
1.7 dB. The same coating method should enable the
construction of a 0.014-inch MRIG to fit the local heating requirement in the coronary artery.
The results of the present study showed that the
localized microwave power distribution radiated from
the active imaging/heating region of the MRIG is cylindrically symmetric, analogous to the geometry of vessels, and covers the heated area of the target. This
simulated power distribution indicates that 2.45 GHz
microwave power is an ideal heating source, which can
be delivered by the MRIG to locally heat the target vessel
rather than a large volume of the body. Theoretically,
the higher the microwave frequency, the more concentrated the microwave power distribution and the
smaller the half-power heating length. However, the
surface electrical resistance of the MRIG is proportional
to the square root of the working frequency. The higher
the frequency, the higher the power loss along the MRIG
will be. Thus, it is necessary to balance the benefits of
local heating and the electrical resistance of the MRIG
for the power transmission. In principle, we could use
low-frequency microwave power if it satisfies the demands of intravascular local heating. In our study, operating the microwave generator at 2.45 GHz (␴ ⫽ 2.5
S/m, ⑀r ⫽ 55) resulted in the 1.7-cm half-power heating
length deposited in the surrounding media, which is
close to the 2-cm length of the 5-F catheter balloon. If
we heat the target vessel using a one-GHz generator
(␴ ⫽ 1.5 S/m, ⑀r ⫽ 60) (13), we should achieve a halfpower heating length of 2.8 cm. For our purposes of
intravascular local heating with a 2-cm long balloon,
the 2.45-GHz microwave generator is an adequate
choice with an acceptable power loss.
This study shows that it is possible to simultaneously
generate high-resolution MR imaging of a target vessel
wall during the inflation of a gene delivery balloon for
vascular gene transfer, which should enable simultaneous monitoring and guiding of catheter-based vascular gene delivery and distribution (9). Our histologic
results have proven that in the current experimental
setting, it is safe to create local heating in target vessels
Qiu et al.
using the 0.032-inch MRIG to deliver 2.45 GHz microwave thermal energy. Although our in vivo studies show
no abnormal manifestations in either clinical or histologic validations, we need to establish a therapeutic heating/thermal safety guideline to ensure that the MR imaging/microwave heating system can be used safely and
efficiently for MR imaging-based vascular gene therapy.
In addition, we need to design a “switch” or duplexer
between the two components of MR imaging and microwave heating, to enable simultaneous generation of
high-resolution MR imaging, MR thermal mapping, and
local heating of the gene-targeted vessel wall during the
same MR examination.
In conclusion, the present study demonstrates the
possibility of using the MRIG as a multifunctional device, not only as a receiver antenna to generate intravascular high-resolution MR imaging of atherosclerotic
plaques of the vessel wall and as a conventional guidewire to guide endovascular interventions during MR
imaging, but also as an intravascular heating source to
produce local heat for thermal enhancement of vascular gene transfection. This study has established the
groundwork to further refine a MR imaging/microwave
heating system for the enhancement of vascular gene
therapy, and for the management of cardiovascular
atherosclerotic diseases using intravascular MR imaging-based vascular gene therapy.
ACKNOWLEDGMENT
The authors thank Mary McAllister for her editorial
assistance.
REFERENCES
1. Nabel E, Leiden J. Gene transfer approaches for cardiovascular
disease. In: Chien K, editor. Molecular basis of cardiovascular disease. Philadelphia: W.B. Saunders Company; 1999. p 86 –112.
2. Kandarpa K. Catheter-based techniques for endovascular local
drug delivery. J Vasc Interv Radiol 2000;11(Suppl):419 – 423.
3. Takai T, Ohmori H. Enhancement of DNA transfection efficiency by
heat treatment of cultured mammalian cells. Biochim Biophys Acta
1992;1129:161–165.
4. Madio D, van-Gelderen P, DesPres D, et al. On the feasibility of
MRI-guided focused ultrasound for local induction of gene expression. J Magn Reson Imaging 1998;8:101–104.
5. Du X, Qiu B, Wang D, Yang X. MR imaging-guidewire as a heating
source for enhancement of gene transduction in human vascular
smooth muscle cells [Abstract]. In: Proceedings of the 10th Annual
Meeting of ISMRM, Hawaii, 2002.
6. Ocali O, Atalar E. Intravascular magnetic resonance imaging using
a loopless catheter antenna. Magn Reson Med 1997;37:112–118.
7. Yang X, Serfaty JM, Quick H, et al. Intracoronary high-resolution
MR imaging using a loopless receiver antenna: an initial in vivo
study. 2000 Scientific Program, Annual Meeting of RSNA, Chicago,
Nov 26 –Dec 1, 2000; p 286.
8. Serfaty JM, Yang X, Quick H, Aksit P, Atalar E. MRI-guided coronary artery intervention. Circulation 2000;102(Suppl):2480.
9. Yang X, Atalar E, Li D, et al. Magnetic resonance imaging permits in
vivo monitoring of catheter-based vascular gene delivery. Circulation 2001;104:1588 –1590.
10. Ulaby FT. Applied electromagnetics. New Jersey: Prentice Hall;
1999.
11. Cheng DK. Field and wave electromagnetics. New York: Addison
Wesley; 1992.
12. Simon R, Whinnery J, Duzer V. Fields and waves in communication
electronics. New York: John Wiley & Sons, Inc.; 1994.
13. Polk C, Postow E. Handbook of biological effects of electromagnetic
fields. Boca Raton, FL: CRC Press; 1986.