The experimental investigation of ultrasonic properties for a sonicated contrast agent and its application in biomedicine

Ultrasound in Med. & Biol., Vol. 26, No. 2, pp. 347–351, 2000 Copyright © 2000 World Federation for Ultrasound in Medicine & Biology Printed in the USA. All rights reserved 0301-5629/00/$–see front matter

PII S0301-5629(99)00146-5

● Technical Note THE EXPERIMENTAL INVESTIGATION OF ULTRASONIC PROPERTIES FOR A SONICATED CONTRAST AGENT AND ITS APPLICATION IN BIOMEDICINE DONG ZHANG*, XIU-FEN GONG*, JIE-HUI LIU*, LI-ZHENG SHAO†, XIAO-RONG LI† and QING-LANG ZHANG† *State Key Lab of Modern Acoustics, Institute of Acoustics, Nanjing University, Nanjing 210093 China; and † Wuxi No.1 People’s Hospital, Wuxi, Jiangsu Province 214000 China (Received 8 July 1999; in final form 20 October 1999)

Abstract—The ultrasonic properties of a promising ultrasound (US) contrast agent, named SDA (sonicated dextrose albumin) are reported in this paper. SDA is a suspension of stable microencapsulated gas bubbles with average diameter 2.0 ␮m prepared from sonicated dextrose albumin. The ultrasonic linear and nonlinear parameters, such as acoustic velocity, sound attenuation and acoustic nonlinearity parameter B/A of SDA, as a function of its bubble concentration from 1.0 ⴛ 107 to 2.05 ⴛ 108 microbubbles/mL in the frequency range of 2– 6 MHz are measured in vitro. The sound attenuation coefficients over 2– 6 MHz are linearly proportional to the bubble concentration and frequency. It is important to point out that the acoustic nonlinearity parameter B/A for SDA has a very large value that nonlinearly increases with the increase of bubble concentration. © 2000 World Federation for Ultrasound in Medicine & Biology. Key Words: Ultrasound contrast agent, Attenuation coefficient, Acoustic nonlinearity parameter.

finite amplitude sound wave, microbubbles in the agent are able to generate nonlinear effects due to the nonlinear vibration of microbubbles in the liquid. The nonlinear effect may be significant when the transmitting frequency is near the resonance frequency f0 of the microbubbles in the agent (De Jong et al. 1994). This phenomenon has been exploited in a new contrast-specific imaging modality: harmonic imaging (Schrope and Newhouse 1993; Chang et al. 1995). The acoustic nonlinearity parameter B/A is an important parameter in nonlinear acoustics that can be used to define the nonlinearity of the medium (Beyer 1960), and may be a new parameter for characterization in US biomedicine (Zhang et al. 1996; Zhang and Gong 1999). So, it has been an interesting issue of research to study the nonlinearity parameter B/A of the contrast agent, not only in the field of the nonlinear acoustics, but also in the field of medical ultrasonics. However, very few measurements of the acoustic nonlinearity parameter B/As for contrast agents have been reported (Gong et al. 1996; Wu and Tong 1997). In this paper, a promising contrast agent, SDA (sonicated 5% human dextrose albumin), which consists of encapsulated gas microbubbles with average diameter 2

INTRODUCTION Ultrasound (US) imaging has become a very successful modality in clinical diagnosis because of its ability to provide noninvasive, real-time cross-sectional images of soft tissue structures and blood flow without ionizing radiation. However, US differentiation of healthy from diseased tissue is sometimes difficult. This may be due to the fact that, even when tissues are pathologically different, their ultrasonic properties, such as sound velocity, attenuation and acoustic impedance, may be quite similar. For enhancement of the discrimination and evaluation of healthy and diseased tissues, the US contrast agents have been of considerable interest in recent years (Ophir and Parker 1989; Goldberg et al. 1994). Most commercially manufactured US contrast agents are liquids with encapsulated gas microbubbles, such as Albunex™ and Levovist™. Increasing the power of the backscattered signal is the most important property of the US contrast agent due to the existence of microbubbles in the agent. Nevertheless, under insonating by a Address correspondence to: Professor Xiu-fen Gong, Institute of Acoustics, State Key Lab of Modern Acoustics, Nanjing University, Nanjing 210093, China. E-mail: [email protected] 347

348

Ultrasound in Medicine and Biology

Volume 26, Number 2, 2000

Table 1. The investigated 10 different SDA concentrations (C) and their corresponding contents of microbubbles C%

100

90

80

70

60

50

30

20

10

5

N ⫻ 108 2.05 1.86 1.56 1.34 1.20 0.91 0.65 0.46 0.24 0.10 N ⫻ 108 microbubbles/mL.

Fig. 1. Microbubble size distribution of undiluted SDA solution. The vertical axis represents the total number of microbubbles per mL of suspension and the horizontal axis is microbubble diameter in ␮m.

␮m can be used to increase videodensity in the echocardiography in vivo. Development of a new ultrasonic contrast agent used in medical imaging systems will require detailed and reliable knowledge of ultrasonic properties and physical behavior of this agent. Therefore, the relationship between its US properties (sound velocity and attenuation) and microbubble concentration, insonating frequency were measured and can be used to alleviate some problems of its application in clinical US imaging. Furthermore, the acoustic nonlinearity parameter B/A of SDA at various bubble concentrations is determined by measuring second harmonic wave in the finite amplitude insert substitution method (Gong et al. 1989). MATERIALS AND METHOD Material handling procedure A variety of potential US contrast agents have been or are now under development. The ultrasonic properties of a bubbly liquid developed by Wuxi No.1 People’s Hospital (Jiangsu Province, China) used as contrast agent in diagnostic US are studied in our laboratory. A sonicator ultrasonic processor (Model Jxup-300, Wuxi, China) with a 1-cm diameter horn is utilized to sonicate an undiluted sample (5% dextrose human albumin). The sonicator is tuned with an output power about 75 w and sonicating frequency 20 kHz. Continuous sonication of the sample was performed utilizing a 30-mL sample in a vial for 20 s, and allowing it to stand for about 30 s to avoid large gas bubbles. The sonicated dextrose albumin (SDA) consisted of microbubbles of encapsulated gas surrounded by a shell of dextrose human albumin with mean diameter about 2 ␮m and a microbubble concentration before dilution of 2.05 ⫻ 108 microbubbles/mL. The normalized size distribution of SDA before dilution

measured by a Coulter Counter (Coulter Corporation, Hialeah, FL; sensitive to a diameter range from 1.0 to 32 ␮m) is shown in Fig.1. The microbubbles of the agent SDA can be stable for enough time to finish all measurements; its half-life is about 120 min in vitro. For Albunex™, however, microbubbles are stable for only a few min (Marsh et al 1997, 1998). In vivo studies, by intravenous injection of the SDA agent in a dog, both the right and the left ventricles of the heart could be successfully visualized. However, myocardial enhancement was not observed. For enough microbubbles passing through the pulmonary circulation in experimental dogs, an inert gas (suflur hexafluoride SF6) is infused with the sample for 20 s before sonication. The mean diameter and microbubble distribution of SDA-SF6 are similar to those of SDA; however, microbubbles of SDA-SF6 are more stable and additional studies about SDA-SF6 are under way. In this study, only the ultrasonic properties of SDA agent were investigated in vitro. In vitro studies For quantification of observed contrast effects, the US linear and nonlinear properties of the agent have to be known, together with the concentration in the interested frequency band. Assuming the sonicated 5% dextrose albumin solution before dilution as a solution of full concentration (100%), it is diluted in 5% dextrose solution to desired concentrations. In this study, 10 different SDA concentrations, from highest to lowest, were investigated. The investigated 10 different SDA concentrations (C) and their corresponding contents of microbubbles (N ⫻ 108 microbubbles/mL) are listed in Table 1. The container used for holding the various concentrations of SDA in our experiments is a Plexiglas cylinder with two parallel windows covered by sound-permeable membranes so that the US signal can travel through the container with minimal attenuation. The thickness d of the sample container was 2 cm. The velocity of sound in the SDA agent is determined by a discrete frequency method using a 2-MHz toneburst (Wu 1996). As shown in Fig. 2, the investigated sample with thickness d is inserted between two identical PZT transducers (2 MHz). By measuring the time of sound propagation through the agent (ts) and

Sonicated contrast agent ● D. ZHANG et al.

Fig. 2. The block diagram of the experimental system for measuring acoustic nonlinearity parameter B/A and sound velocity via the finite amplitude insert substitution method.

349

quency of the microbubbles. The nonlinearly oscillating microbubbles become secondary US sources, transmitting not only the fundamental frequency component, but also its harmonics and subharmonics. The second harmonics excitation efficiency can be defined as the acoustic nonlinearity parameter B/A. A second harmonic insert-substitution method proposed by one of the authors (Gong et al. 1989) was employed in this paper to measure the nonlinearity parameter of the SDA solutions. The principle of the method is to measure the ratio of the second harmonic pressure amplitude p2x/p20; p2x and p20 are the received pressure amplitudes of the second harmonic waves with and without inserting the sample into a referred medium with known B/A value, respectively. As shown in Fig. 2, the acoustic nonlinearity parameter (B/A)x of the sample can be expressed as:

冉冊 冋 B A

⫽

x

册

p 2x L 䡠 ⫺ 共L ⫺ d兲e ⫺␣2d P 20 D OxD x0

␳ xc x3 1 1 䡠 䡠 D 0x I共d兲 ␳ 0c 03

comparing it to the time of sound propagation of a known medium (distilled water) (t0), the sound velocity of SDA can be obtained by the following equation:

冋冉 冊 册 B A

⫹ 2 ⫺ 2,

(2)

0

where: c0 cS ⫽ , 1 ⫹ 共t s ⫺ t 0兲c 0/d

(1)

where c0 and cs are the sound velocity of distilled water and SDA agent, respectively. Broadband measurement is used to determine the frequency-dependent attenuation coefficients of the SDA solutions. Two identical broadband plane PZT transducers are employed as transmitter and receiver. The 6-dB bandwidth of the transducers cover the frequency range 2– 6 MHz. A short pulse is used to excite a 3.5-MHz transducer and the frequency response is measured with an HP spectrum analyzer (HP3585B, Hewlett Packard, Santa Clara, CA). The acoustic attenuation of the contrast agent suspension as a function of frequency can be determined by subtracting the average spectrum of received signals prior to inserting the contrast agent from the spectrum after insertion. It is well known that when the frequency of applied US is near the resonance frequency of a bubble, it absorbs the US energy effectively. Therefore, by determining where the local maximum of the attenuation coefficient occurs in the frequency domain, the US broadband pulse technique can be used to determine the resonance frequency of the bubbles in the test solution. When a US contrast agent containing microbubbles is illuminated by an external US wave, microbubbles oscillate nonlinearly. The oscillation is vigorous, especially at the external frequency near the resonance fre-

D 0x ⫽

2 ␳ xc x 2 ␳ 0c 0 , D xo ⫽ , ␳ 0 c 0 ⫹ ␳ xc x ␳ 0 c 0 ⫹ ␳ xc x I共d兲 ⫽

e ⫺␣2d ⫺ e ⫺2␣1d 2 ␣ 1 ⫺ ␣ 2, (3)

L is the distance between the transmitter and the receiver, d is the thickness of the sample, ␣1 and ␣2 are, respectively, the attenuation coefficients of the fundamental wave and the second harmonics, subscript 0 denotes water and subscript x denotes sample. The experimental system for measuring the acoustic nonlinearity parameter B/A of SDA solutions at different bubble concentrations is shown in Fig. 2. A pulse generator (XC13A, Nantong, China) and a function generator (Philips PM5193) produce a sinusoidal burst signal of 2 MHz with 15 cycles in length at a pulse-repetition frequency (PRF) of 1 kHz. This signal is amplified in a broadband 50-dB RF Power amplifier (ENI 2100L, Rochester, NY) and sent to a planar PZT transducer that has a 12-mm active diameter and is operated at a fundamental frequency of 2 MHz. Received signal is detected by a 0.6-mm diameter broadband PVDF hydrophone (TNU001A, NP1000, NTR Systems, Inc., Seattle, WA) at a distance 7 cm from the transmitter. After passing through a 60-dB band amplifier (AU-4A-0150-BNC, MITEQ, Hauppauge, NY), the pressure amplitude of the second harmonic wave is recorded by a spectrum ana-

350

Ultrasound in Medicine and Biology

Fig. 3. The sound attenuation coefficient vs. frequency for SDA at full concentration. The averaged values and standard deviations of four attenuation measurements via the broadband method are plotted. Short acoustic pulses with a duration of 15 ns were transmitted at amplitude 300 V.

lyzer (HP 3585B, Hewlett Packard, Santa Clara, CA). The acoustic output of the transmitted transducer is about 110 kPa, which is calibrated in water using the broadband PVDF hydrophone (TNU001A, NP1000, NTR Systems, Inc., Seattle, WA). RESULTS AND DISCUSSION The sound velocity of SDA at different microbubble concentrations was measured and found to decrease slightly with an increase in microbubble concentrations. In practice, the difference in compressibility will sufficiently change the velocity of sound. The existence of microbubbles will significantly reduce the velocity of a suspension of liquid. As the microbubble concentrations increase from 0.10 ⫻ 108 to 2.05 ⫻ 108 microbubbles/mL (corresponding gas volume fraction 0.0013% to 0.084%), the sound velocity decreases from 1500 m/s to 1340 m/s. This variation in sound speed with gas concentration is similar to that change described by the theory of Church (Church 1995). However, the variations in sound velocity with gas concentration are only valid for frequencies well below resonance and the variations in sound velocity will decrease with the increase of the modulus of rigidity of the shell (Church 1995). The attenuation coefficient vs. frequency for SDA agent at full concentration (2.05 ⫻ 108 microbubbles/ mL) is shown in Fig. 3. The attenuation coefficient vs. microbubble concentration for SDA agent at 2 MHz and 4 MHz is shown in Fig. 4 by using the transmission method. Owing to the limited aperture of the transducer, Fig. 3 illustrates only the bandwidth over 2– 6 MHz. It is clear that acoustic attenuation coefficient of the SDA

Volume 26, Number 2, 2000

Fig. 4. The sound attenuation coefficient vs. microbubble concentration at 2 MHz and 4 MHz. The averaged values and standard deviations of five measurements via the transmission method are plotted. Tone bursts (2 MHz and 4 MHz, respectively) with a duration of 15 cycles and 30 cycles were transmitted at a PRF 1 kHz. The output acoustic pressure are 110 kPa and 100 kPa, respectively

solution in the measured frequency range increases monotonously with frequency and is nearly a linear function of the concentration. In addition, Fig. 3 indicates that the resonance frequency of the bubbly liquid SDA is above 6 MHz. This can be approximately explained by theoretical estimation. A theoretical model for encapsulated gas-filled microbubbles predicted a peak in ultrasonic scattering intensity and attenuation due to bubble resonance (De Jong et al. 1992, 1994; De Jong and Hoff 1993). The resonance frequency f0 of the encapsulated microbubbles can be estimated by the following expression: f0 ⫽

冑冉

冊

8 ␲ r 02S p 3␥ 1 2␴ 2␴ , P0 ⫹ ⫺ ⫹ 2␲r0 ␳0 r0 ␳ 0r 0 m

(4)

where ␥ denotes the adiabatic ideal gas constant, r0 is the bubble radius and m ⫽ 4␲r30␳0. P0 and ␳0 are the ambient fluid pressure and density, respectively. ␴ is the surface tension coefficient and Sp is the shell elasticity parameter. For the SDA agent, the average radius of microbubbles in liquid is 1.0 ␮m; then, from eqn (4), we can estimate the relationship between the resonance frequency f0 and the shell elasticity parameter Sp presented as Table 2. The resonance frequency is seen to increase

Table 2. The relationship between resonance frequency and shell elasticity parameter S p (N/m) f 0 (MHz)

0 4.73

0.27 6

1 8.54

2 11.1

4.2 15.3

Sonicated contrast agent ● D. ZHANG et al.

351

indicate that the SDA suspensions of stable microencapsulated gas bubbles show significant nonlinearity and high US absorption when insonified by US of frequencies commonly used in medical US. These preliminary results demonstrate the potential of contrast-enhanced imaging and may provide a physical foundation for application of contrast agents in medical diagnosis. Acknowledgements—The authors are grateful to Professor P. N. T. Wells and Professor Z. Z. Xu for their valuable comments on the manuscript. The funding was from National Natural Science Foundation of China (No. 19834040) and Natural Science Foundation of Jiang-Su Province (No. BK99024).

Fig. 5. The acoustic nonlinearity parameter B/A vs. microbubble concentration. The averaged values and standard deviations of five measurements are plotted. A 2-MHz tone burst with 15 cycles length was transmitted at a PRF of 1 kHz. The output acoustic pressure was 110 kPa.

with the increasing of shell stiffness. In our experiment, we did not find the peak of attenuation over 2– 6 MHz; therefore, it is reasonable to indicate that the resonance frequency of SDA is above 6 MHz and its corresponding Sp will be higher than 0.27 Nm⫺1. The value of acoustic nonlinearity parameters for SDA solutions vs. microbubble concentration is shown in Fig. 5. It seems that the values of acoustic nonlinearity parameter B/A nonlinearly increase with the increasing numbers of microbubbles. Noting that, at the highest concentration (2.05 ⫻ 108 microbubbles/mL), the B/A value of SDA reaches 1125, much higher than most liquids and soft tissues (about 5–11), and the B/A value of unsonicated dextrose albumin solution is only 6.7. It can be seen that the presence of microbubbles in a liquid may enhance the nonlinearity parameter B/A to an extremely large value, even though at frequencies that are not in the vicinity of the resonance frequency of the insonified microbubbles. CONCLUSION In this paper, the experimental investigation in vitro for a promising contrast agent (sonicated 5% dextrose albumin) is presented. Results in vitro present the ultrasonic linear and nonlinear properties of the SDA agent with various microbubble concentrations. The attenuation coefficient is linearly proportional to the number of microbubbles of SDA. The nonlinearity parameter B/A is found to increase with the increasing numbers of microbubbles and, especially, it has very large B/A value (1125 for SDA before dilution). The experimental results

REFERENCES Beyer RT. Parameter of nonlinearity in fluids. J Acoust Soc Am 1960;32:719 –721. Chang PH, Shung KK, Wu S, Levene HB. Second harmonic imaging and harmonic Doppler measurements with Albunex. IEEE UFFC 1995;42:1020 –1027. Church CC. The effects of an elastic solid surface layer on the radial pulsations of gas bubbles. J Acoust Soc Am 1995;97:1510 – 1521. de Jong N, Hoff L. Ultrasound scatter properties of Albunex威 microbubbles. Ultrasonics 1993;31:175–181. de Jong N, Cornet R, Lance CT. Higher harmonics of vibrating gasfilled microbubbles. Part two: simulations. Ultrasonics 1994;32: 447– 453. de Jong N, Hoff L, Skotland T, Bom N. Absorption and scatter of encapsulated gas filled microbubbles: Theoretical considerations and some measurements. Ultrasonics 1992;30:95–103. Goldberg BB, Liu JB, Forsberg F. Ultrasound contrast agents: A review. Ultrasound Med Biol 1994;20:319 –333. Gong XF, Ye SG, Zhang D. Measurement of nonlinearity parameter of contrast agents. Nonlinear acoustics in perspective. Proceeding of 14th International conference on nonlinear acoutics. Nanjing, China: Nanjing University, 1996:475– 480. Gong XF, Zhu Z, Tao S, Huang J. Determination of the acoustic nonlinearity parameter in biological media using FAIS and ITD methods. J Acoust Soc Am 1989;86:1–5. Marsh JN, Hall CS, Hughes MS, Mobley J, Miller JG, Brandenburger GH. Broadband through-transmission signal loss measurements of Albunex suspensions at concentrations approaching in vivo doses. J Acoust Soc Am 1997;101(2):1155–1161. Marsh JN, Hughes MS, Hall CS, et al. Frequency and concentration dependence of the backscatter coefficient of the ultrasound contrast agent Albunex. J Acoust Soc Am. 1998;104(3):1654 –1665. Ophir J, Parker KJ. Contrast agents in diagnostic ultrasound. Ultrasound Med Biol 1989;15.319 –325. Schrope BA, Newhouse VL. Second harmonic ultrasonic blood perfusion measurement. Ultrasound Med Biol 1993;19:567–579. Wu J. Determination of velocity and attenuation of shear waves using ultrasonic spectroscopy. J Acoust Soc Am 1996;99(5):2871–2875. Wu J, Tong J. Measurements of the nonlinearity parameter B/A of contrast agents. Ultrasound Med Biol 1997;24:153–199. Wu J, Zhu Z, Du G. Nonlinear behavior of a liquid containing uniform bubbles: comparison between theory and experiments. Ultrasound Med Biol 1995;21:545–552. Zhang D, Gong XF. Experimental investigation of the acoustic nonlinearity parameter tomography for excised pathological biological tissues. Ultrasound Med Biol 1999;25(4):593–599. Zhang D, Gong XF, Ye SG. Acoustic nonlinearity parameter tomography for biological specimens via measurement of the second harmonics. J Acoust Soc Am 1996;99(4):2397–2402.