The success rate of low-intensity pulsed ultrasound (LIPUS) therapy depends on the bone site. However, the initial mechanism of physical stimulation by ultrasound and bone cellular response remains unclear. One possible physical stimulation is the induced electrical potentials due to the piezoelectricity. In this study, the output electrical potentials of ultrasound transducers made from bovine bones were investigated. Transducers made from the radius bone showed the largest electric potentials, followed by tibia, femur, and humerus. There was clear site dependence of the induced electric potentials of bone, in good accordance with the success rate of LIPUS therapy.

Low-intensity pulsed ultrasound (LIPUS) has attracted attention as a technique for healing bone fractures.1,2 This technique typically uses pulsed ultrasound from 1.5 to 3.0 MHz and shortens the healing time of bone fractures. The overall success rate of LIPUS therapy for intractable bone fractures varies depending on the bone site.3 Although many clinical studies of LIPUS have been reported, the initial mechanism of physical stimulation by ultrasound and bone cellular response has not yet been elucidated.4,5

Fukada and Yasuda6 reported that electrical potentials were generated inside bones when low-frequency mechanical stress in the kHz range was applied. In addition, Yasuda7 reported the formation of callus by applying a very small current of 1 μA using a 1.5 V battery to the femur of a rabbit. These induced electrical potentials promote bone growth and regeneration.6–10 However, there are few studies about piezoelectricity in bone in the MHz range. The piezoelectricity in the MHz range may concern the initial mechanism of bone fracture healing with ultrasound. To measure very small-magnitude piezoelectricity in bone, we have fabricated ultrasound transducers using bones as piezoelectric materials have been fabricated, and their piezoelectric properties were confirmed in the MHz range.11–15 Moreover, Ikushima et al.16 investigated stress-induced polarization in soft biological tissues by detecting the first harmonic component of the acoustically induced electric fields.

Collagen and hydroxyapatite, the main components of bone, are strongly oriented in the direction of the bone axis.17–19 Therefore, the elasticity of cortical bone shows almost uniaxial anisotropy and seems isotropic in the plane normal to the bone axis. In addition, the crystallographic symmetry of collagen, which contributes largely to the piezoelectric effect in bone, is supposed to be quasi-hexagonal.20,21 For these reasons, the piezoelectricity in the cortical bone shows anisotropy, and that the polarity and amplitude of the induced electrical potentials in bone change depending on the direction of ultrasound irradiation.13 However, the site dependence of piezoelectricity has not yet been extensively studied. Assuming that the piezoelectricity of bone contributes to the healing mechanism of LIPUS therapy, the amplitude of ultrasonically induced electrical potentials may be a factor for the success rates of LIPUS therapy.

In this work, we have experimentally observed the effects of bone anisotropy on the electrical potentials induced by ultrasonic irradiation to study the electromechanical characteristics of bone, which seems to be a factor in understanding bone healing by the low-intensity pulsed ultrasound stimulation. The study focused specifically on the site dependence of ultrasonically induced electrical potentials in bone using the following cortical bovine bones: femur, tibia, radius, and humerus to understand the relationship between ultrasound irradiation and induced electrical potentials in bone.

Figure 1(a) shows the preparation procedure of bone samples. We prepared 12 types of cortical cylindrical bone plates (diameter: 10.5–11.0 mm; thickness: 3.00 ± 0.01 mm; normal direction of circular plates; radial, tangential, or bone axis directions) from the mid shafts of left bovine femur, tibia, radius, and humerus. Then, electrodes were deposited on the plates' surfaces. Using these plates as piezoelectric materials, we fabricated bone transducers as ultrasound receivers.22 In this experiment, the femoral circular plates cut normal to the radial direction, normal to the tangential direction, and normal to the bone axis were named FR, FT, and FA, respectively; similarly, the tibial samples cut out in those three directions were named TR, TT, and TA, respectively; the radial samples were named RR, RT, and RA, respectively; and the humeral samples were named HR, HT, and HA, respectively.

Fig. 1.

(a) Preparation of bone transducer. (b) Radiation direction of ultrasound.

Fig. 1.

(a) Preparation of bone transducer. (b) Radiation direction of ultrasound.

Close modal

For the experiments, the transmitter (a polyvinylidene difluoride, PVDF, focusing-type transducer, diameter, 20 mm; focal length, 40 mm; Toray Engineering, Tokyo, Japan) and the receiver (the bone transducer) were set to cross at a right angle in water. A function generator (33250A, Agilent Technologies, Santa Clara, CA) was used to generate a square electrical pulse (pulse width: 70 μs), which was then amplified to 70-V peak-to-peak by a bipolar power supply (HAS 4101, NF, Yokohama, Japan). An ultrasonic pulse wave was radiated from the transducer at the initial rising time of the square pulse (main frequency, 760 kHz; sound pressure, 7.4 kPa peak-to-peak). The pulse was irradiated into the side surface of the bone transducer, as shown in Fig. 1(b). The distance between the transmitter and the receiver was set to 40 mm so that the side of the bone sample was located at the focal point of the transmitter. The received signal was amplified by 40 dB using a pre-amplifier (BX-31A, NF, Yokohama, Japan) and observed using an oscilloscope (DPO3054, Tektronix, Beaverton, OR). The direction of ultrasound irradiation was changed by rotating the bone transducer. Measurements were performed at each rotation angle in steps of 10° from 0° to 350°. Here, 0° means that the ultrasound propagates in the tangential (FR, TR, RR, and HR samples) and radial (FT, TT, RT, HT, FA, TA, RA, and HA samples) directions.

We succeeded in observing the ultrasonically induced electrical potentials in all bone transducers. Figure 2 shows the observed waveforms of the electrical potentials measured by the HR, HT, and HA transducers. Since the transducer output is the sum of all responses to the propagating wave in the surface electrode, the observed wave electrical potential depends on the receiver diameter.23 Then, induced electrical potential in bone is different from the waveform of the radiated short ultrasound pulse. We confirmed that the polarity of the wave front changed depending on the ultrasound propagation direction when using the FR, TR, RR, HR, FT, TT, RT, and HT transducers, whose bone plates were normal to the radial and tangential directions. However, the polarity of the wave front could not be judged when using the FA, TA, RA, and HA transducers whose bone plates were normal to the axial direction because the amplitudes of the electrical potential were much smaller than those observed with the other bone transducers.

Fig. 2.

Waveforms of induced electrical potentials acquired from humerus bone samples.

Fig. 2.

Waveforms of induced electrical potentials acquired from humerus bone samples.

Close modal

Figure 3(a) shows the piezoelectric anisotropy results in the FR, TR, RR, and HR bone transducers (the bone plate surfaces were perpendicular to the radial direction). The output electrical potential values (peak-peak) of all these transducers showed maximum around 45°, 135°, 225°, and 315° angles between the axial and tangential directions and minimum at approximately the axial and tangential directions. Maximum induced electrical potential of RR (65 μV) was 1.2 times larger than that from TR (54 μV), 1.4 times larger than that from FR (45 μV), and 2.4 times larger than that from HR (27 μV). Figure 3(b) also shows the piezoelectric anisotropy results of the FT, TT, RT, and HT bone transducers (the bone plate surfaces that were perpendicular to the tangential direction). The output electrical potential amplitudes of all these transducers also showed maximum around 45°, 135°, 225°, and 315° angles between the axial and radial directions and minimum at approximately the axial and radial directions. Maximum induced electrical potential of RT (95 μV) was 1.5 times larger than that from TT (61 μV), 1.9 times larger than that from FT (49 μV), and 2.2 times larger than that from HT (43 μV). These data show that oblique incidence of ultrasound to bone axis causes large electrical potentials. The mechanical studies of Fukada and Yasuda6,20 also showed that the magnitude and polarity of the induced electrical potentials changed depending on the direction of pressure application. Notably, it has been reported that the magnitude of the charge that appeared in human and bovine femur showed a maximum when the angle between the pressure application direction and the bone axis was about 50°.6,13,20 Our results showed maximum electrical potentials at approximately 45°, and there was a similar tendency in relationships between the ultrasound irradiation direction and the value of the induced electrical potential. Figure 3(d) shows the enlarged piezoelectric anisotropy results in the FA, TA, RA, and HA bone transducers (the bone plate surfaces were perpendicular to the axial direction). The amplitude of the electrical potential stayed almost constant in the range of 6–12 μV peak-to-peak, and no significant change in amplitude was observed with change in the ultrasound propagation direction. The crystal structure of collagen, which contributes to the piezoelectricity of bone, is considered to be quasi-hexagonal and align in the direction of the bone axis.19,20 However, from these data, we could not judge whether the data have 60° rotational symmetry or not.

Fig. 3.

Relationships between peak-peak values of the induced electrical potentials and ultrasound irradiation directions.

Fig. 3.

Relationships between peak-peak values of the induced electrical potentials and ultrasound irradiation directions.

Close modal

Samples made from bovine radius showed the largest amplitudes of induced electrical potential, followed by tibia, femur, and humerus, in all three directions (radial, tangential, and axial). Watanabe et al.3 reported that the overall success rate of human LIPUS therapy for delayed union and non-union fractures was around 90%, 87%, 82%, and 67% for radius, tibia, femur, and humerus, respectively. Although bovine cortical bone was used in this research, we found large potential amplitudes in the sites where the success rate of human LIPUS therapy was high. Further studies are expected on the relationship between ultrasound irradiation-induced electrical potentials in human bone and the healing rate of intractable bone fracture.

In this study, the used ultrasonic wave was very short, and the pressure was comparatively small, and the output of the electric potential was the averaged values in the electrodes, which may be smaller than the actual potential in the small local area. In future studies, it will be important to examine the relationship between the optimum LIPUS sound pressure and piezoelectricity of bone. Thus, high-intensity focused ultrasound (HIFU) might be addressed as part of our future studies, but we must be careful about safety.

Regarding bone as a piezoelectric material, we examined the site dependence of ultrasonically induced electric potentials in bone. As a result, the amplitudes and polarities of the induced electrical potentials in bovine bone depended on the ultrasound propagation direction and the site from which the bone was sampled. In addition, the site dependence of induced electrical potentials in bovine bone samples was in good accordance with the success rate of LIPUS therapy of intractable human bone fracture. Although the bones used in this study were obtained from bovine, the site dependence may show a possible contribution of ultrasonically induced electrical potentials to LIPUS therapy.

Part of this study was supported by the JST A-step program.

1.
L. R.
Duarte
, “
The stimulation of bone growth by ultrasound
,”
Arch. Orthop. Trauma Surg.
101
(
3
),
153
159
(
1983
).
2.
R.
Leighton
,
J. T.
Watson
,
P.
Giannoudis
,
C.
Papakostidis
,
A.
Harrison
, and
R. G.
Steen
, “
Healing of fracture nonunions treated with low-intensity pulsed ultrasound (LIPUS): A systematic review and meta-analysis
,”
Injury
48
(
7
),
1339
1347
(
2017
).
3.
Y.
Watanabe
,
T.
Matsushita
,
M.
Bhandari
,
R.
Zdero
, and
E. H.
Schemitsch
, “
Ultrasound for fracture healing: Current evidence
,”
J. Ortho. Trauma.
24
,
S55
S61
(
2010
).
4.
F.
Padilla
,
R.
Puts
,
L.
Vico
, and
K.
Raum
, “
Simulation of bone repair with ultrasound: A review of the possible mechanic effects
,”
Ultrasonics
54
,
1125
1145
(
2014
).
5.
J. D.
Heckman
,
J. P.
Ryaby
,
J.
McCabe
,
J. J.
Frey
, and
R. F.
Kilcoyne
, “
Acceleration of tibial fracture-healing by non-invasive, low-intensity pulsed ultrasound
,”
J. Bone Joint Surg. Am.
76
,
26
34
(
1994
).
6.
E.
Fukada
and
I.
Yasuda
, “
On the piezoelectric effect of bone
,”
J. Phys. Soc. Jpn.
12
,
1158
1162
(
1957
).
7.
I.
Yasuda
, “
Electrical callus and callus formation by electret
,”
Clin. Orthop. Relat. Res.
124
,
53
56
(
1977
).
8.
M.
Mohammadkhah
,
D.
Marinkovic
,
M.
Zehn
, and
S.
Checa
, “
A review on computer modeling of bone piezoelectricity and its application to bone adaptation and regeneration
,”
Bone.
127
,
544
555
(
2019
).
9.
B.
Tandon
,
J. J.
Blaker
, and
S. H.
Cartmell
, “
Piezoelectric materials as stimulatory biomedical materials and scaffolds for bone repair
,”
Acta Biomater.
73
,
1
20
(
2018
).
10.
M. H.
Shamos
and
L. S.
Lavine
, “
Physical bases for bioelectric effects in mineralized tissues
,”
Clin. Orthop. Relat. Res.
35
,
177
188
(
1964
).
11.
M.
Okino
,
S.
Coutelou
,
K.
Mizuno
,
T.
Yanagitani
, and
M.
Matsukawa
, “
Electrical potentials in bone induced by ultrasound irradiation in the megahertz range
,”
Appl. Phys. Lett.
103
,
103701
(
2013
).
12.
H.
Tsuneda
,
S.
Matsukawa
,
S.
Takayanagi
,
K.
Mizuno
,
T.
Yanagitani
, and
M.
Matsukawa
, “
Effects of microstructure and water on the electrical potentials in bone induced by ultrasound irradiation
,”
Appl. Phys. Lett.
106
,
073704
(
2015
).
13.
S.
Matsukawa
,
T.
Makino
,
S.
Mori
,
D.
Koyama
,
S.
Takayanagi
,
K.
Mizuno
,
T.
Yanagitani
, and
M.
Matsukawa
, “
Effect of anisotropy on stress-induced electrical potentials in bovine bone using ultrasound irradiation
,”
Appl. Phys. Lett.
110
,
143701
(
2017
).
14.
S.
Mori
,
T.
Makino
,
D.
Koyama
,
S.
Takayanagi
, and
M.
Matsukawa
, “
Ultrasonically-induced electrical potentials in demineralized bovine cortical bone
,”
AIP Adv.
8
(
4
),
045007
(
2018
).
15.
T.
Makino
,
T.
Nakamura
,
L.
Bustamante
,
S.
Takayanagi
,
D.
Koyama
, and
M.
Matsukawa
, “
Piezoelectric and inversely piezoelectric responses of bone tissue plates in the MHz range
,”
IEEE Trans. Ultrason. Ferroelect. Freq. Control
67
(
8
),
1152
1532
(
2020
).
16.
K.
Ikushima
,
T.
Kumamoto
,
K.
Ito
, and
Y.
Anzai
, “
Electric polarization of soft biological tissues induced by ultrasound waves
,”
Phys. Rev. Lett.
123
(
23
),
238101
(
2019
).
17.
K.
Raum
, “
Microelastic imaging of bone
,”
IEEE Trans. Ultrason. Ferroelect. Freq. Control
55
(
7
),
1417
1431
(
2008
).
18.
N.
Sasaki
and
Y.
Sudoh
, “
X-ray pole figure analysis of apatite crystals and collagen molecules in bone
,”
Calcif. Tissue Int.
60
(
4
),
361
367
(
1997
).
19.
Y.
Yamato
,
M.
Matsukawa
,
T.
Yanagitani
,
K.
Yamazaki
,
H.
Mizukawa
, and
A.
Nagano
, “
Correlation between hydroxyapatite crystallite orientation and ultrasonic wave velocities in bovine cortical bone
,”
Calcif. Tissue Int.
82
,
162
169
(
2008
).
20.
E.
Fukada
and
I.
Yasuda
, “
Piezoelectric effects in collagen
,”
Jpn. J. Appl. Phys.
3
,
117
121
(
1964
).
21.
J. P.
Orgel
,
A.
Miller
,
T. C.
Irving
,
R. F.
Fischetti
,
A. P.
Hammersley
, and
T. J.
Wess
, “
The in situ supermolecular structure of type I collagen
,”
Structure
9
(
11
),
1061
1069
(
2001
).
22.
Y.
Nakamura
and
T.
Otani
, “
Study of the surface elastic wave induced on backing material and diffracted field of a piezoelectric polymer film hydrophone
,”
J. Acoust. Soc. Am.
94
(
3
),
1191
1199
(
1993
).
23.
See supplementary material at https://www.scitation.org/doi/suppl/10.1121/10.0003195 for describing the mechanism of wave observation using a polyvinylidene difluoride transducer for example. We have irradiated ultrasound to the transducer from side surface. The plus and minus parts of the wave may be canceled in the output from the large electrodes during the propagation of the sound wave. Therefore, the observed wave electrical potential due to the wave propagation depends on the receiver diameter.

Supplementary Material