Deep Local Hyperthermia for Cancer Therapy: External Electromagnetic and Ultrasound Techniques

Deep Local Hyperthermia for Cancer Therapy: External
Electromagnetic and Ultrasound Techniques
Augustine Y. Cheung and Ali Neyzari
Cancer Res 1984;44:4736s-4744s.
Updated version
Access the most recent version of this article at:
E-mail alerts
Sign up to receive free email-alerts related to this article or journal.
Reprints and
To order reprints of this article or to subscribe to the journal, contact the AACR Publications
Department at [email protected]
To request permission to re-use all or part of this article, contact the AACR Publications
Department at [email protected]
Downloaded from on September 9, 2014. © 1984 American Association for Cancer
[CANCER RESEARCH (SUPPL.) 44, 4736s-4744s,
October 1984]
Deep Local Hyperthermia for Cancer Therapy: External
Electromagnetic and Ultrasound Techniques1
Augustine Y. Cheung and Ali Neyzari
Department of Electrical Engineering and Computer Science, George Washington University, Washington, DC 20052, and Cheung Associates, Inc.,
Beltsville, Maryland 20705
External heating techniques for delivery of localized hyperthermia in patients are reviewed. This paper covers microwaves,
radiofrequency, and ultrasound methods. Fundamental principles
governing tissue absorption, guidelines for applicator selection
and design, and restrictions of each heating approach are dis
cussed. Innovative techniques utilizing multiple applicators to
achieve better heating uniformity are also presented. The advan
tages and disadvantages of electromagnetic versus ultrasound
heating techniques are compared as a conclusion to this review.
Elevated tumor temperature, or tumor hyperthermia, is a
method used in the treatment of cancer based on a considerable
amount of good experimental data. In the early part of this
century, diseases such as arthritis, asthma, and multiple scle
rosis and infectious diseases such as syphilis and gonorrhea
were treated by hyperthermia (24).
At low-temperature hyperthermia (between 37° and 41.5°),
heat enhances cell growth and also may well enhance the growth
and proliferation of tumors. At high-temperature hyperthermia
(above 45°),heat begins to indiscriminately damage both normal
and cancer cells. Thus, to avoid both enhancement of the active
growing edge of the tumor and damage to normal cells, we are
limited to a narrow therapeutic range.
This paper describes the 2 methods of external heating (EM2
and ultrasound) that have been or can be used locally to induce
temperature elevation for the treatment of cancer. As we shall
see, each method has advantages and disadvantages.
As we go deeper inside the tissue, the number of human
tumors that can be treated with hyperthermia increases. There
fore, depth of penetration of the heating beam is an important
consideration in hyperthermia systems. Another important factor
is the noninvasiveness of the technique. Metastasis, caused by
delivering heat invasively, might increase with disruption of blood
vessels and mechanical probing of the tumor. EM and ultrasound
are the 2 main methods that are potentially useful for noninvasive
heating (17).
Localization is also a factor of consideration in hyperthermia.
In treating known or suspected multiple tumors with whole-body
hyperthermia, temperatures above 42° are hazardous due to
difficulty in quick and precise control and physiological stress
(24). Consequently, producing localized deep heating without
excessive surface heating by means of external EM and ultra
sound techniques is the primary subject of this paper.
1Presented at the Workshop Conference on Hyperthermia in Cancer Treatment,
March 19 to 21,1984, Tucson, AZ.
2 The abbreviation used is: EM, electromagnetic.
Heat-producing Modalities
Most of the heat-producing methods are divided into 2 major
modalities: (a) ohmic heating, which is produced by electrical
currents generated from radiofrequency sources and by electrical
waves generated from microwave sources; and (b) mechanical
friction, which is caused by an ultrasound wave shaking the
EM and ultrasound beams follow the general laws of waves
as they propagate through the body (14).
Because each heat-producing modality has its own physical
properties and because the anatomical site of the lesion and the
size and depth of the tumor vary, one or several methods may
have specific applications or limitations in a given topographical
area (22).
EM Techniques
Heat can be generated in tissue by different kinds of interaction
between EM fields and biological systems. One such way is by
rotating polar molecules; the friction associated with the rotation
of the atoms and molecules causes heat generation when timevarying EM fields are applied. Another kind of interaction is
oscillation of free electrons and ions. In this way, collisions
between electrons and ions with immobile atoms and molecules
within the tissues produce heat. At microwave frequencies and
radiofrequencies, the internal electric field E is primarily respon
sible for transferring energy into tissue as heat. At microwave
frequencies (300 MHz to 30 GHz), the rotation of water mole
cules dominates all interactions; therefore, water-containing tis
sues like skin and muscle are usually good microwave absorbers
In general, materials that interact with an EM field via the
interactions mentioned above are classified as lossy dielectrics
and are described by a property of material called permittivity,
designated by «.Permittivity involves a complex number for
sinusoidal steady state fields and can be expressed as
«= eo(i' - Je")
where (0 ¡sthe permittivity of free space (F/m) and («'- Ji") is
the relative permittivity, with «'as the real part and «"as the
imaginary part, both of which are unitless. From Equation A, we
see that the relative permittivity is (/<0 = <r; it is called the
dielectric constant. Tissue can be characterized by e' and a, the
conductivity (Siemens m~1) that is given by a = we0e", where w
is the angular frequency. Note that the permittivity of tissue ¡sa
strong function of frequency.
The concept of plane wave propagation in a lossy dielectric is
often used to describe wave phenomena in tissues. Therefore,
although this concept does not actually occur physically, it is
Downloaded from on September 9, 2014. © 1984 American Association for Cancer
VOL. 44
Deep Local Hyperthermia by External Techniques
nevertheless an important tool in understanding the basic salient
characteristics of EM waves in tissue (9).
Frequency and Depth of Penetration
The use of higher frequencies results in a decreased depth of
penetration. As Chart 1 shows, as the frequency decreases, the
depth of penetration of the plane wave in muscle tissue in
creases. By drawing a vertical line at any point on the depth
axes, we would see that, for the same depth of penetration, the
use of a lower frequency results in higher power absorption. By
drawing a horizontal line at any point on the power axes, we
would see that, for the same power absorbed, use of a lower
frequency results in a higher depth of penetration. Power ab
sorbed P is given by
= Vï
where £ is the magnitude of electric field (V/m) and <r, the
conductivity, is in (S/m). P is the same as the specific absorption
rate. Penetration depth D is defined by
where a, the absorption coefficient, is given by
Chart 2. Frequency dependence of D (plane-wave depth of penetration) and <r
(conductivity) for EM waves in muscle and fat tissues (5).
necessarily desirable since as given by Equation B and Chart 2,
a, the primary factor governing absorption, decreases with de
creasing frequency (8).
and reduced wave length X««
is given by
Frequency Selection
where X0 is the free-space wave length which is always greater
than A««
Chart 2 shows penetration depth D and power absorption P
as a function of frequency for muscle and fat tissue. At any
frequency, this graph shows penetration depth in fat is higher
than that in muscle, but conductivity a of muscle is higher than
that of fat. Despite better penetration, the lower frequency is not
Because depth of penetration is a function of frequency, then
to heat tumors at various depths, it is more desirable to have a
generator covering the entire range of frequencies. However,
this is not practical because of the radiation hazards as well as
restraint on decreased absorption with lowering frequency.
Therefore, EM generators other than those of the officially des
ignated industrial, scientific, and medical band are generally
prohibited for operating on a patient in a regular hospital room;
a special shielding room is required for any frequencies other
than the above mentioned. For hyperthermia, the EM generators
that are commercially available operate at the ISM band frequen
cies of 13.56, 27.12, 40.68, 915, and 2450 MHz. A frequency of
433 MHz is also authorized in Europe. For hyperthermia, the
power range also varies normally within the range of 10 to 500
watts for a single applicator at microwave frequencies (915 and
2450 MHz).
EM Applicators
Chart 1. Power absorption in muscle by plane wave versus depth of penetration
at different frequencies (9).
Experimental studies strongly suggest that hyperthermia is
useful in the treatment of cancer. One of the most important and
difficult parts of this treatment is the delivery of well-controlled
heat into the body, a complex biological system.
One of the most significant problems facing application of EM
energy is the proper design and selection of the applicators that
direct deep penetration of EM energy into the patient. Indeed,
the success of hyperthermic treatment appears to be strongly
Downloaded from on September 9, 2014. © 1984 American Association for Cancer
A. Y. Cheung and A. Neyzari
related to the capability of the applicator(s) to focus energy
effectively into a tumor (6). Factors like reliable, simple, and safe
equipment; power output (reproducibility) and surface cooling;
localization of the treatment area; knowledge of minimum tumor
and maximum normal tissue temperature to avoid temperature
rise in the surrounding healthy tissue; and acceptable heating
duration are critical in the applicator selection and design proc
esses (3). They provide uniform, reliable, and safe heating of the
tumor volume.
In microwave diathermy (915 and 2450 MHz), spaced appli
cators are often used. However, because of the danger of
scattered radiation to the operators and the patient's body,
problems such as near-field coupling (1). If an applicator that is
short compared to the wave length is used, since the near field
is strong near the radiator but decays rapidly as its distance
from the radiator increases and since the EM field produced in
the tissue is dominated by the near fields, then greatly increased
surface heating occurs (9).
The following section includes a description of the different
kinds of EM direct-contact applicators (external), consisting of
capacitive, inductive, and radiative aperture applicators and also
multiple applicators.
Capacitive Applicators
shielding is required (6). For a safe treatment with minimum
leakage of radiation, an external direct contact applicator can be
Capacitive applicators have been used widely in hyperthermia
used where shielding is not required (6, 16). Based on official
for cancer patients. They are simple devices that operate at low
safety performance standards formulated by the United States
frequencies (13.56 and 27.12 MHz). This type of applicator
Bureau of Radiological Health, direct-contact applicators should
consists of 2 plates producing an electric field (£)that is perpen
be able to induce hyperthermia in tissue at a rate exceeding 1°/
dicular to the plates and causes deep heating (Chart 3). Parallel
min, thus raising the tissue temperature from 37°to 42°in less
to the direction of the conduction current, electric field (E) is
than 5 min. At the same time, leakage exceeding the safety level
basically perpendicular to the interfaces between the tissue
of 10 milliwatts/sq cm should not be found at 5 cm from the
layers, such as fat and muscle. Due to differences in permittivity
outer edge of the applicator in use (6, 23).
of different tissues, interfaces between different tissues (e.g.,
In designing EM applicators, the size of the applicator (radiator)
fat and muscle) in wave heating is a major concern (9). For an
must be an appreciable fraction of a wave length to be efficient.
idealized geometry (parallel plate capacitor), the E field in the fat
The wave length is given by
(E,) and muscle (£m)is constant. The boundary condition at
junction between fat and muscle requires that
where f is the frequency and C is the speed of the waves in the
body, which is given by
i,£,= tm£m
where a and tm are the permittivities of fat and muscle. From
Equation B, power absorbed P (W/m3) at any point for fat and
muscle are given by (9,15)
P, =
where Ci is the speed of the waves in free space.
A complication that must be considered for the waves is the
impedance mismatch between the source, the body, and the
structures in the body. The reflections between interfaces are
related to the characteristic impedance Z,. In EM waves, this
value is Z, = 377 ii for air and Z2 = 50 ÃŽÃŽ
in the body. The
reflection power R at normal incidence for planar waves is given
Thus, the ratio of absorbed power in fat to muscle is given by
P, _ a, |£,|2
+ z,
and the transmitted power (7) is given by (14)
| i, |2
The following is a simple example that shows how excessive
s.c. fat heating occurs when the electric field is perpendicular to
7 = 1 -fl
We know that the lower frequency (/) results in an increased
depth of penetration (D), so that to have an applicator deliver
deep heating in muscle, the length of the radiator must be at
least one-half of the wave length (X) (13), and since wave length
is related to the frequency by Equation C, a lower frequency
results in a longer wave length, which leads to a large radiator.
Using too long a radiator, however, is not a practical means of
producing EM waves in regions of the patient's body. On the
\E I2
Therefore, from Equations D and E
z, - z
other hand, it is usually desirable for one to use higher-frequency
microwaves, because it is easier to localize the radiator at high
frequencies; yet deep penetration cannot be achieved.
The design of microwave heating involves solutions to EM
Charts. Capacitive applicator arrangement showing idealized parallel plate
capacitor geometry (9).
Downloaded from on September 9, 2014. © 1984 American Association for Cancer
VOL. 44
Deep Local Hyperthermia by External Techniques
the fat-muscle interface and how this condition may be pre
vented. The values needed for this calculation for frequency
27.12 MHz are given as
a, = 0.012
am = 0.61
<', = 20
t", = 7.23
t"m = 405.82
— a 0.02
From Equation A,
«m= KÕf'm —Jt'm)
- J("m\2
From Equation F,
= (0.02) (392) = 7.84
which means that power absorbed in fat is greater than power
absorbed in muscle (P, > Pm). From Equation D,
portion of a patient. These simple applicators, which are now
commercially available with the name Magnetrode, operate at a
fixed frequency (13.56 MHz). No coupling medium is necessary.
With Magnetrodes, high temperature has been achieved at
depths of 8 cm or more.
In the special case of a homogeneous dielectric placed coaxially in a thin coil, the magnetic field concentrates in the edge
vicinity of the coil, thus producing a null at the center, even in
the case of a lossless material. In the case of an inhomogeneous
medium (e.g., the human body), induced eddy currents do not
flow symmetrically around the geometric center. Instead, many
smaller locally induced loops may be found in regions of different
conductivity (5). These local eddy-current loops may cause more
uniformity and deeper heating results (21).
Three configurations of magnetic fields generated by induction
coils are illustrated in Chart 4 as follows: (a) pancake coil, where
the coil is placed on the surface of the body and may consist of
one or more turns of a conductor in a planar or axial distribution
and produces a magnetic field predominantly perpendicular to
the skin surface; (b) coaxial pair of coils, in which 2 single-turn
coils on a common axis can be placed on the anterior and
posterior sides of the body region to be heated. The arrows
indicate the magnetic field lines that pass through the body. The
dotted circles indicate the path of representative eddy currents
in coronal planes of the body; (c) concentric coil. When one or
multiple-turn coils surround a portion of the patient's body,
magnetic field lines approximately parallel to the axis of the
cylindrical volume are produced. Eddy currents associated with
the induced £field are also shown (20).
Radiative Aperture Applicators
which means that £field in fat is greater than £field in muscle.
Thus, if E, > £m,then P,>Pm, where the fat-overheating problem
occurs. The above calculation is based on the condition that the
£field is perpendicular to the fat-muscle interface. To prevent
fat overheating near the interface, the £field should be parallel
to the fat-muscle interface. The boundary condition requires
E, = £m.Therefore, from Equation E,
This type of applicators is classified as a high-frequency appli
cator (microwave), which couples a propagating wave from the
applicator to the patient. They are well developed and can
satisfactorily heat tissue at depths of a few cm. Furthermore,
because they are excited by wave guides, they do not produce
fat overheating problems since their £field is primarily tangential
to the fat-muscle interface (9).
Since the physical size of the applicator must be at least onehalf the wave length, at frequencies below microwave, the ap
erture applicators would be practically too large to use. However,
as Chart 1 indicates, the penetration depth at microwave fre-
which means that power absorbed in fat is much less than that
in muscle, resulting in no fat burning (9).
In the capacitive heating technique, the current spread can
also cause excessive surface heating, which would require prop
erly spacing the separation between plates and the tissue. A
circulated 0.9% NaCI solution bolus is very often used to control
the surface temperature.
Inductive Applicators
In EM heating, inductive applicators are involved when, instead
of direct electric field coupling, the main source of power depo
sition is currents produced inductively in the tissue. Recently,
Storm ef a/. (27) used a large loop induction coil surrounding a
Chart 4. Three arrangements of current loops and the corresponding directions
of magnetic field lines. Eddy currents are also shown (20).
Downloaded from on September 9, 2014. © 1984 American Association for Cancer
A. Y. Cheung and A. Neyzari
quencies is insufficient for heating deep-seated tumors. Further
more, in operating a small-aperture applicator at low frequency
(13.56 and 27.12 MHz), the production of radiation into the body
is dominated by the near field, causing surface overheat. In a
dielectrically filled wave guide as the frequency is lowered, the
size of the aperture increases and a reduction of this size is
directly proportional to the square root of the relative dielectric
constant t,, where t, = —.Therefore, by filling the empty space
(air) of the wave guide applicator with commercially available
lossless dielectrics ranging from 1 to 150, the aperture size can
be reduced by a factor of up to 12 (5, 11, 13). Indeed, Sterzer
ef al. (26) have developed a large ridge-wave guide applicator
(27.12 MHz) loaded with deionized water (lossless dielectric with
ir = 81) that produces deep heating.
Chart 5. Transmission of EM radiation from an applicator to a tumor in hyperthermia system using a bolus.
44 43°-
In EM hyperthermia, a tissue-equivalent bolus is often used to
improve the coupling between the applicator and the patient
(Chart 5). Application of bolus has the following advantages.
Smooth Transmission from Applicators into Tissue. Lack of
uniformity of the deposited energy and loss in the coupling
coefficient of energy in the heated area due to the curvature of
the body surface require use of bolus.
Skin Cooling to Avoid Surface Heating. With single applica
tors, maximum heating always occurs near the surface. There
fore, unless treating very superficial skin, deionized water is
often circulated into the bolus to act as a cooling agent against
the skin.
Safer Treatment by Reducing the Amount of Leakage from
the Applicator. Deionized water bolus greatly reduces the
amount of leakage from the applicator.
Maintenance of the Body Surface at a Fixed Distance from
the Applicator for Each Session of the Treatment. Microwave
bolus can be used as a spacer to ensure proper placement of
the applicator.
With a proper frequency, a well-designed applicator, and use
of a bolus, EM hyperthermia induction systems can deposit
uniform heating into the tissue at the depth of a few cm, but the
depth of heating can be greatly increased by using 2 or more
applicators rather than a single one.
Multiple-Applicator Technique
This technique can be incorporated into hyperthermia treat
ment to improve the depth of heating in tissue. In regions of
extreme curvature (e.g., breasts, head, neck, and limbs), it is
possible to generate deep hyperthermia by superposing several
beams. With a capacitive applicator, by placing more than one
pair of capacitive plates in a "cross-fire" arrangement, heating
from all the pairs adds in the center, where deep tissue heating
is desired. Less superficial heating may be achieved with this
arrangement (11).
Phased Array. An array of radiation designed to create con
structive interference at the focus is called phased array. In a
multiple-element array arrangement (with N elements), depend
ing on whether or not the elements are excited in phase, the
heating at the focus can be A/2 or only N greater than that
expected from a single applicator. However, in reality, it is hard
Chart 6. Distribution of heat induced by means of 2 conformai applicators facing
each other across the heated area in the thigh muscles of an anesthetized dog.
Graph represents temperature readings at various points of thermocouples. Inser
tion along the distances between applicators (19).
to design a phase array radiating into a lossy inhomogeneous
dielectric (human tissue) (6).
Radiative aperture applicators have been used in arrays to
obtain improved heating patterns. Cheung ef al. (7) used 2
applicators at 2450 MHz to obtain more uniform heating. Mendecki ef al. (19) used a single conformai applicator at 2450 MHz.
The heat induced in the tissue was not uniform, and the thera
peutic temperature range was limited to 1.5 to 2 cm below the
surface level (cutaneous or s.c. heating). To improve deep heat
ing, they used 2 conformai applicators facing each other across
the heated area. As illustrated in Chart 6, perfectly uniform
heating in tissue with a thickness of 5 cm is achieved. GuerquinKern ef al. (12) used two 2450-MHz applicators perpendicular to
each other; an improved temperature field resulted from the
superposition of the 2 intersecting beams.
In microwave hyperthermia, a single dielectrically loaded openended waveguide, horn, or coaxial antenna is often used. To
avoid the disadvantage of the single applicator, phased array is
used in layered lossy media with the focal point several cm away
from the radiating aperture. Gee ef al. (11) developed a theory
for analyzing an arbitrary array designed for near-field focusing
and for testing its predictions for a 4-element linear array against
experimental data. The focused linear array of 2450 MHz con
sists of 4 titanium dioxide-loaded horn antennas with apertures
(2.0 x 1.4 cm). The experiments conducted with the 4-element
linear array have successfully demonstrated that the near-field
focusing of an array can be accomplished by appropriate phasing
of each antenna element for the desired focal point. This validates
the theoretical model. Furthermore, Gee ef al. have obtained a
reasonable beam spot size (1.3 cm) that is amenable to electric
scanning and achieved sufficient sidelobe suppression (as is
evident by the 19-element hexagonal planar array) to ensure that
most of the EM energy can be confined and directed to the
intended focal region.
Downloaded from on September 9, 2014. © 1984 American Association for Cancer
VOL. 44
Deep Local Hyperthermia by External Techniques
Ultrasound Technique
Ultrasound is another method of producing deep heating in
hyperthermia cancer therapy. This therapeutic modality has been
used for some years but, like the EM technique, has advantages
and disadvantages.
Vibration due to passage of ultrasound waves through tissues
causes the displacement of tissue molecules. Heating is pro
duced as a result of the absorption of this ultrasound vibration
in the tissue. The speed of sound in tissue is considerably lower
as compared to the velocity of EM wave propagation. This
difference in velocity and difference of the ultrasound and EM
radiation results in vast differences between ultrasonic and EM
heating. Because of the relatively low speed of sound in tissue
(1.5 x 105 cm/sec), at frequencies between 1 and 10 MHz
(ultrasonic frequencies), the acoustical wave lengths (between
1.5 to 0.15 mm) are much shorter than those in the EM range.
This frequency range is still low enough to avoid high tissue
absorption and as a result provides deep penetration in tissue.
The propagation of ultrasound in the body is similar to that of
microwave beams. The acoustic impedance z»is related to the
velocity of ultrasound V (speed of sound in region x in m/s) and
the average density P, (kg/m~3).
Because at ultrasound frequencies both speed of sound and
average density are almost constant for most tissues (e.g. , water,
brain, liver, muscle, and fat, but not bone), the acoustic imped
ance Zx is constant for different tissues. For this reason, the
internal reflections between fat and muscle are usually neglected
in ultrasound technique. However, propagation of ultrasound
waves in bone and air is quite different from that of soft tissue;
a great deal of reflection occurs at the interfaces of bone (or air)
and tissue (2, 15). This is one of the disadvantages of ultrasound
Because heating by plane-wave energy causes the intensity
and temperature to decay exponentially as the depth in tissue
increases (Chart 7, Curves A), deep heating is not achieved, and
regardless of the wave length of the plane-wave energy, surface
heating occurs and injury is possible. Therefore, for selective
deep-heat deposition in a limited region, focusing the energy is
essential (17). As Chart 7, Curves B, shows, by focusing, higher
intensity and temperature can be achieved at the desired point
of depth, and due to the small size of ultrasound energy wave
lengths ultrasound waves can be focused easily into local regions
of tissue for producing controlled localized hyperthermia to heat
deep-seated tumors (17).
Chart 7. Intensity and temperature distribution patterns, with plane wave in a
homogeneous medium (Curves A) and with a focused radiation field (Curves B)
diameters are used for deep tumors and operate at the lower
frequencies (18).
Focusing Lens. Energy from the transducer can be focused
or concentrated into the tissue with a focusing lens. Different
sizes of lenses are available for different sizes of transducers.
When selecting focusing lenses, factors such as good impedance
matching and low-attenuation loss properties should be consid
Degassed water or 0.9% NaCI solutions are used for acoust
ical coupling between the transducer and the body during insonation(18).
The attenuation coefficient of tissue increases approximately
linearly with frequency; i.e., the shorter the wave length in tissue,
the greater is the attenuation. Therefore, when deep penetration
is needed for deep-tumor heating, a low frequency should be
selected. The size and shape of the focus are also determined
by wavelength. Therefore, a target as small as 1 mm can be
selectively heated by ultrasound.
In heating a deep-seated tumor by localized hyperthermia
using ultrasound technique, the longest wave length should be
approximately one-fifth of the dimension (thickness or diameter)
of the tumor (18). Chart 8 shows that in ultrasound technique
most of the power is concentrated in the region (heating area)
with the diameter S, where S is related to the wave length of the
energy X, the depth d (focal length of lens), and the diameter of
the transducer D and is given by
S =
1.22 \d
Insonation, or irradiation with ultrasound, elevates the temper
ature in tissue and consists mainly of the following.
Transducer. Ultrasound is generated from a transducer (x-cut
quartz crystals) which, when activated by a high-frequency volt
age, produces pressure waves that heat the tissue (4). A rea
sonable transducer size in ultrasound is several wave lengths in
diameter, such as 8, 12, or 16 cm. Transducers with larger
Results from the above equation for a transducer 9 cm in
diameter lead to the following (9). For any depth (focal length)
smaller than 12 cm and frequently greater than 0.5 MHz, the
diameter of the heating area is less than 0.5 cm, which would
not be practical for spot size. Ultrasonic power absorption per
unit volume of tissue is a function of depth d and is given by
Wa = W<,exp(-2a.„d)
Downloaded from on September 9, 2014. © 1984 American Association for Cancer
A. Y. Cheung and A. Neyzari
- Transducer
Focused Field
Chart 8. Schematic of focusing of energy from the transducer into the tumor
where W0 is initial power incident at the tissue surface and am
is the acoustic attenuation coefficient. Ultrasound intensity in
depth d is also given by
' = T = T
(watt/sq cm)
more and the tumor margin less, due to conduction and blood
perfusion. Since at low temperature heat may well enhance the
growth and proliferation of a tumor, this situation would create
serious problems. Therefore, to raise tumor temperature evenly
to the desired level in the entire tumor, translocation must be
used, allowing deposition of enough heat at the periphery of the
tumor. With a tumor larger than the heat source, stationary focus
or pulsing stationary heat source on and off for production of
effective hyperthermia is not adequate. Therefore, the heat
source (or focal region) must be moved over the entire tumor for
deposition of heat at different parts of a large tumor.
From the above thermophysical properties of normal tissue
and tumors and the length of the trajectory, which depends on
the size of the tumor, the velocity of translocation can be
determined. Since the generation of heat in tissue is a function
of both local intensity and duration of insonation, in order to
generate more heat in tumors by increasing the local intensity
without possible focal damage to the tumor at focus, the duration
of insonation needs to be decreased. This can be done by
increasing the translocation velocity (17).
Multiple Transducers
/ = /oexp(-2a«^)
(watt/sq cm)
where I0 is the initial intensity and A is the cross-section of the
focused spot, given by A = -— (sq cm) (9,14).
When deep heating is needed, the smallest practical diameter
of heating area S (focused spot) should be 0.5 cm; therefore, the
area is equal to
= 0.196
(sq cm)
To deliver deep heating to large or vascular tumors, it may not
be possible to use a single transducer. By superposition of more
than one beam entering the tissue surface at different points,
sufficient power and depth of penetration can be achieved (17).
Phase arrays of transducer elements that are being activated in
sequence can produce a good deep temperature elevation. Two
beams can interface destructively, however, if they are out of
phase where they overlap. Consequently, the heat generation
may be lower in the overlap compared to that at the beams
A good example of a combination of multiple transducers and
translocation is Lele's (18) device, shown in Chart 9. With
steered, focused ultrasound, a spatially uniform level of hyper
thermia restricted to the target volume and located at depth can
/ = 5 Woexp(-2a«<y) (watt/sq cm)
absorbed power per unit volume is given by
P = 2/0amexp(-2a«^)
(watt/cu cm)
Remark. For heating a fixed spot size (S = 0.5 cm) of homo
geneous muscle with ultrasound at different frequencies, we
should consider the following. For a depth of 2 cm or less, a
frequency of 2 MHz or higher is required. For a depth greater
than 5 cm, a frequency of 1 MHz or less is required. Therefore,
for a depth of 12 cm, the frequency of 0.5 MHz is optimal. If the
initial 3 cm of fat are followed by homogeneous muscle, then for
depths of 7 cm or more a frequency of less than 1 MHz is
required (9).
Translocation, or moving the heat source (or focal region), is
important in the production of hyperthermia by ultrasound be
cause of uniform temperature distribution. Tumors have lower
blood perfusion than do normal tissues, and the lowest appears
to be in their central regions (25). Because more heat can be
removed from a region with higher blood flow (2), depositing
energy evenly throughout the tumor would heat the central region
j i \ i l i
Diameter, cm
Chart 9. Unitomi temperature distribution in beef muscle mass in vitro using
beams focused at 6 cm depth at 0.9 MHz frequency in circular trajectories (18).
Downloaded from on September 9, 2014. © 1984 American Association for Cancer
VOL. 44
Deep Local Hyperthermia by External Techniques
be achieved. This device is based on the conduction and blood
perfusion in tissue and tumors. Two well-focused beams are
moved in circular trajectories, one in the peripheral region of the
tumor and the other close to the central region. In experimenta
tion, such a technique resulted in excellent uniform temperature
distribution at 2 to 7.5 cm of depth at a frequency of 0.9 MHz
Pounds ef al. [from Hunt (14)] used another approach that
was later followed by Fessenden ef al. (10) in which 6-planar, 7cm-diameter PZT-4 discs mounted on a 90°spherical shell sector
with a 26-cm radius of curvature were utilized at 0.35 MHz to
produce therapeutic heating up to a depth of 15 cm.
Advantages and Disadvantages of EM and Ultrasound Tech
EM Technique
Advantages. Since EM energy can propagate through air, in
this technique coupling is not required.
Due to the presence of air within and in the vicinity of areas
such as the lungs, stomach, bowel, bladder, rectum, and pelvis,
the use of EM technique is suggested for cancer therapy in
these regions.
EM energy is not hindered by bones. Therefore, this technique
can be used for treatment of cancer in the chest area and all
portions of upper and lower extremities.
The preferred approach for brain tissue heating is microwave
with single or multiple external beams.
Microwave radiation can penetrate deeply into low-watercontaining tissue, like fat, and since the breast is composed
largely of fat, deep penetration for cancer therapy is possible.
Large volumes can be heated with multiple applicators or
phase-array microwave.
Producing microwave power is relatively inexpensive as a
result of the commercialization of the microwave ovens.
Mechanisms of interactions of microwaves with biological tis
sue are reasonably understood. This allows a better design of
safe and effective hyperthermia systems.
It is relatively simple to control the power output of a micro
wave generator.
Depending on the type of treatment, there are different meth
ods of induction of hyperthermia by EM system. These are
noninvasive and invasive methods. Noninvasive methods can
also be divided into simple and multiple-applicator techniques. In
invasive methods, the objects can be either implanted in the
body or inserted into a body orifice.
Disadvantages. EM waves are absorbed by water-containing
tissues and cause excessive heat elevation due to both higher
absorption and lower heat dissipation. Thus, there are potential
hazards for the EM technique in hydrate tissues or in tissues
close to the organs containing or surrounded by fluids, such as
the heart, stomach, and spinal cord.
Depth of dose is limited to a few cm by using a single
applicator, particularly with microwaves.
The fat near the fat-muscle interface may overheat due to
large reflections. These reflections may generate standing waves
close to the fat surface.
Focusing is difficult at low frequencies.
Interaction with metal temperature-measuring devices is pos
There is a potential danger to patients using pacemakers.
Advantages. Deep penetration of controlled beams up to 12
cm is possible.
Tumors absorb ultrasound energy better than does normal
tissue, as compared to EM energy.
Excellent focusing is possible because the wave lengths are
small compared to the diameter of their source.
The acoustic impedance of most of the body fluids is close to
that of the soft tissue, and absorption in the fluids is lower than
that in the tissues. Thus, there is no possibility of excessive
There are no significant reflections at the interfaces between
fatty and muscle tissues.
The method is noninteractive with thermometry devices.
Imaging and thermometry are possible with ultrasound.
No special radio frequency-shielded room is required.
Disadvantages. There is high absorption in bone, causing
bone heating.
Reflection between bone-tissue interfaces is large.
Reflected energy cannot be refocused within the soft tissue.
Potential problems lie with cavities containing air. Acoustic
impedance mismatch between air and soft tissues is very high,
and energy is completely reflected at air-tissue interfaces, be
cause there is no transmission through air cavities.
Ultrasound is not suitable for lung, abdominal, or brain cancer
and also not recommended for deep heating in extremities.
Coupling medium is required.
1. Audet, J., Chive, M., Botomey, J. C., Pichot, C., N'Guyen, D. D., Robillard, M.,
and Leroy, Y. Applicators for medical applications. J. Microwave Power, 75:
Babbs, C. F., Oleson, J. R., and Pearce, J. A. Equipment for local hyperthermia
therapy of cancer. Med. Instrum. (Arlington), 76: 245-248,1982.
Bteehen, N. M. Hyperthermia in the treatment of cancer. Br. J. Cancer, 45
(Suppl. 5V 96-100, 1982.
Bteehen, N. M. Hyperthermia for the treatment of cancer: biological, physical,
and clinical considerations. Excerpta Med. Int. Congr. Ser., 569: 73-100,
Cheung, A. Y. Microwave and radiofrequency techniques for clinical hyper
thermia. Br. J. Cancer, 45 (Suppl. 5V 16-24,1982.
Cheung, A. Y. Microwave techniques for clinical hyperthermia. In: Physical
Aspects of Hyperthermia, AAPM Monograph No. 8, pp. 357-371. New York:
American Institute of Physics, 1983.
Cheung, A. Y., Dao, T., and Robinson, J. E. Dual beam TEM applicator for
direct-contact heating of dieiectncaiiy encapsulated malignant mouse tumor.
Radio Sci., 72: 81-85, 1977.
Cheung, A. Y., Goldmg. W. M., and Samaras, G. M. Direct contact applicators
for microwave hyperthermia. J. Microwave Power, 76:151-159,1981.
Christensen, D. A., and Dumey, C. H. Hyperthermia production for cancer
therapy: a review of fundamentals and methods. J. Microwave Power, 76:89105, 1981.
Fessenden, P., Lee, E. R., Anderson, T. L., Strohbehn, J. W., Meyer, J. L.
Samulski, T. V., and Marmor, J. B. Experience with a multitransducer ultra
sound system for localized hyperthermia of deep tissues. IEEE Trans. Biomed.
Eng., BME-31:126-135,1984.
Gee, W., Lee, S. W., Bong, N. K., Cain, C. A., Mittra, R., and Magin, R. L
Focused array hyperthermia applicator: theory and experiment. IEEE Trans.
Biomed. Eng., BME-31: 38-46, 1984.
Guerquin-Kem, J. L., Palas, L., Priou, A., and Gautherie, M. Therapeutic
purposes-experimental studies of various applicators. J. Microwave Power,
Hand, J. W. Physical techniques for delivering microwave energy to tissues.
Br. J. Cancer, 45 (Suppl. 5V 9-15,1982.
Hunt, J. W. Applications of microwave, ultrasound, and radiofrequency heating.
Nati. Cancer Inst. Monogr., 67: 447-456, 1982.
Iskander, M. F. Physical aspects and methods of hyperthermia production by
Downloaded from on September 9, 2014. © 1984 American Association for Cancer
A. Y. Cheung and A. Neyzari
RF currents and microwaves. In: Physical Aspects of Hyperthermia, AAPM
Monograph No. 8, pp. 462-481. New York: American Institute of Physics,
Kantor, G. Evaluation and survey of microwave and radiofrequency applicators.
J. Microwave Power, 16:135-150,1981.
Lele, P. P. Induction of deep, local hyperthermia by ultrasound and electromagnetic field. Radiât.Environ. Biophys., 17: 205-217,1980.
Lele, P. P. Local hyperthermia by ultrasound. In: Physical Aspects of Hyperthermia, AAPM Monograph No. 8, pp. 393-440. New York: American Institute
of Physics, 1983.
Mendecki, J., Friedenthal, E., Botstein, C., Sterzer, F., and Paglionme, R.
Therapeutic potential of conformai applicators for induction of hypertnermia.
J. Microwave Power, 14: 139-144,1979.
Oleson, J. R. A review of magnetic induction methods for hypertnermia
treatment of cancer. IEEE Trans. Biomed. Eng., BME-31: 91-97, 1984.
Oleson, J. R., Cetas, T. C., Stauffer, P. R., and Jones, R.C. Magnetic induction
heating in heterogeneous material. Third International Symposium: Cancer
Therapy by Hyperthermia, Drugs, and Radiation, Colorado State University,
Fort Collins, CO, June 22-26, 1980.
Perez.C. A. Rationale for clinical application of hyperthermia alone or combined
with irradiation or cytotoxic drugs in cancer therapy. In: Physical Aspects of
Hyperthermia, AAPM Monograph No. 8, pp. 63-89. New York: American
Institute of Physics, 1983.
Samaras, G. M., and Cheung, A. Y. Microwave hyperthermia for cancer
therapy. CRC Crit. Rev. Biomed. Eng., 5:123-184,1981.
Short, J. G., and Turner, P. F. Physical hyperthermia and cancer therapy. Proc.
IEEE, 68: 133-141,1980.
Song, C. W., Rhee, J. G., and Levitt, S. H. Blood flow in normal tissues and
tumors during hyperthermia. J. Nat). Cancer Inst, 64:119-124,1980.
Sterzer, F., Paglione, R., Mendecki, J., Friedenthal, E., and Botsteins, C. R.F.
therapy for malignancy. IEEE Spectrum, 77:32-37,1980.
Storm, F. K., Harrison, W. H., Elliot, R. S., and Morton, D. L. Normal tissue
and solid tumor effects of hyperthermia in animal models and clinical trials,
Cancer Res., 39: 2245-2251,1979.
Downloaded from on September 9, 2014. © 1984 American Association for Cancer
VOL. 44