Transcranial Threshold of Inertial Cavitation Induced by Diagnostic Ultrasound and Microbubbles

Transcription

1 Transcranial Threshold of Inertial Cavitation Induced by Diagnostic Ultrasound and Microbubbles Liu, Jinjin a ; Gao, Shunji a ; Porter, Thomas a ; Everbach, Carr d ; Shi, William b ; Francois, Vignon b ; Powers, Jeffry c ; Lof, John a ; Turner, Joseph e ; Xie, Feng a a University of Nebraska, Nebraska Medical Center, Omaha, NE b Philips Research North America, 345 Scarborough Road, Briarcliff Manor, NY 151 c Philips Healthcare, 221 Bothell Everett Highway, Bothell, WA 9821 d Swarthmore College, 5 College Avenue, Swarthmore, PA 1981 e University of Nebraska-Lincoln, Lincoln, NE Abstract: Inertial cavitation may cause hazardous bioeffects during microbubbles mediated sonothrombolysis. The purpose of this study was to investigate the influence of ultrasound pulse length and temporal bone on inertial cavitation thresholds within the brain utilizing transtemporal imaging transducers. A pig temporal bone overlaid with muscle tissue was placed over silastic tubing containing a dilute microbubble infusion (1% Definity) within Phosphate Buffered Saline at 37 C. A 1.6 MHz Philips ie33 two-dimensional probe (S5-1) imaged at incremental peak negative pressures. Broadband noise signals were recorded to characterize inertial cavitation using two 2 MHz passive cavitation detectors. Peak-negative-pressure thresholds of inertial cavitation were approximately 263 kpa and 235 kpa with and without bone for 5 microsecond pulse length, and 217 kpa and 215 kpa with and without bone for 2 microsecond pulse length, respectively. The threshold of inertial cavitation is associated with ultrasound pulse length and temporal bone. Keywords: Inertial cavitation; Transcranial diagnostic ultrasound; Microbubbles; Thrombolysis INTRODUCTION Stroke has been the second most common cause of death and disability in the world 1. Intravenous recombinant tissue plasminogen activator (t-pa) was approved to treat ischemic stroke in the first three hours after the start of symptoms in However, a higher rate of intracranial hemorrhagic complications associated with t-pa and potential neurotoxicity limit its application 2. Thus improved thrombolysis methods have been explored 3-5. Ultrasound and microbubbles have been used in dissolving thrombi in both in vitro and in vivo experiments. Datta et al. 4 evaluated low frequency ultrasound-enhanced thrombolysis in combination of t-pa and Definity microbubbles at a range of ultrasound power output. They revealed that stable cavitation was correlated with enhanced t-pa thrombolysis. Alexamdrov and coworkers 5 performed a pilot randomized clinical safety study of sonothrombolysis augmentation with ultrasoundactivated microbubbles for acute ischemic stroke. Their research suggested ultrasound-activated microbubbles with systemic t-pa did not increase the risk of intracranial hemorrhage. Even in the absence of thrombolysis substances, cavitation induced by the interaction of ultrasound and microbubbles still shows to be effective in thrombolysis.

2 Xie and coworkers 6 reported the effectiveness of lipid-encapsulated microbubbles mediated sonothrombolysis in vitro using canine thrombus model at two different 1 MHz ultrasound intensities. Microbubbles mediated ultrasound treatments showed higher recanalization rates than ultrasound treatment alone. Culp and colleagues 7 studied microbubble potentiated low-frequency ultrasound as a method for stroke therapy in a swine model. They achieved significant higher recanalization rate in microbubbles potentiated ultrasound group than that in the control group. However, strategies to study the microbubble activity, optimize the ultrasound power output and avoid potential hazardous bioeffects have not been identified. Inertial and stable cavitations, which are usually induced by high and intermediate mechanical index ultrasound, respectively, are important mechanism of sonothrombolysis 8, 9. Some studies indicated that stable cavitation played an important role in microbubble-enhanced sonothrombolysis and is as effective for acute thrombus dissolution as inertial cavitation 1, 11. For safety consideration, stable cavitation is usually preferred, because inertial cavitation may cause tissue damage with micro-jetting and pitting on tissue surface. The objective of this study was to investigate the transcranial threshold of inertial cavitation under the influence of the skull bone and ultrasound pulse length. MATERIALS AND METHODS Experimental Setup The overall system includes an ultrasound machine (ie33; Philips Healthcare, Andover, MA) with a sector imaging probe S5-1, a large acrylic water tank, a flow system, function generator (Agilent, 3325A), oscilloscope (Waver Runner 64 MXi-A; Lecroy), and a pair of 2 MHz passive cavitation detectors (UTX IX-116; UTX Inc, Hartford, CT) with a 3-dimentional positioning system. A pig temporal bone overlaid with muscle tissue was placed over silastic tubing. Phosphate Buffered Saline (PBS) in large beaker was continuously pumped to the tube system at a constant flow rate of 19 ml/min using a Master-flex flow pump (Cole Parmer Instrument Co., Vemon Hills, IL). The temperature of PBS was maintained at 37 C using an Isotemp digital stirring hotplate (Fisher Scientific, Dubuque, IA). A 1% infusion of microbubbles (Definity; Lantheus Medical Imaging, N. Billerica, MA) at 1mL/min was continuously administered into the tube system containing flowing PBS so that microbubbles are always present in the tubing. Probe S5-1 was placed on top of the pig temporal bone. PCDs were placed at 45 o from the longitudinal axis of the silastic tubing as shown in Figure 1. For comparison, the threshold of inertial cavitation in the absence of the pig temporal bone was also studied. The pig temporal bone was removed and all other experimental setup was exactly same as the case of the presence of pig temporal bone. Probe S5-1 of ie33, tubing and PCDs were kept in the same position.

3 Acoustic Inertial Cavitation Detection To quantify cavtiation activity within the tubing during ultrasound exposure, passive cavitation detection scheme was used. The position of passive cavitation detectors were calibrated to make sure strong signals can be obtained from the interaction of ultrasound and microbubbles. With increasing of the ultrasound power output, inertial cavitation may occur and a great number of microbubbles collapse which introduces broadband noises. The scattered broadband noises are then detected by the passive cavitation detectors and outputted to the oscilloscope. The root mean square (RMS) voltage from the passive cavitation detectors, a quantity proportional to the sum of the pressure-wave emissions from microbubble destruction occurring in response to the incident 1.6 MHz diagnostic ultrasound, is calculated and analyzed using Matlab codes. FIGURE 1. Passive cavitation detectors FIGURE 2. Microbubbles confirmed by ie33 imaging system

4 Peak Negative Pressure Measurement The pressure levels in the presence and absence of bone at different ultrasound power output were measured using a needle hydrophone with its active element diameter of mm in diameter (Precision Acoustic LTD, United Kingdom). The active element of the hydrophone was placed at the center of the tubing where cavitation happened. RESULTS AND DISCUSSION Microbubbles appeared in the tubing, which was confirmed by ie33 at a contrast imaging mode with low amplitude output to allow replenishment of microbubbles as shown in Figure 2, and then the ie33 was switched from imaging mode to one of two custom-made therapeutic ultrasound modes. Two therapeutic treatments were applied in this experiment. One is 5 microseconds short pulses (1.6 MHz, 8 cycles), and the other one 2 microseconds long pulses (1.6 MHz, 32 cycles). Ultrasound power outputs for each treatment were increased gradually to study the threshold of inertial cavitation induced by the interaction of ultrasound and Definity microbubbles. Broadband noise from inertial cavitation was characterized by two passive cavitation detectors. The RMS values of the 2 MHz component of broad-band noise originating from the collapsing microbubbles were captured. Figures 3a and b show the averaged RMS values at 5 and 2 microseconds pulse length in the presence of pig temporal bone, respectively. Figures 3c and d show the averaged RMS values at 5 and 2 microseconds pulse length in the absence of pig temporal bone, respectively. As elapsed time increases, the ultrasound power output is increased every 6 seconds. The RMS values approach to zero at low power, because the inertial cavitation does not start and few microbubbles collapse. When ultrasound power output approaches to a critical value which is defined as the threshold of inertial cavitation, RMS values go up abruptly which indicates inertial cavitation initiates and lots of microbubbles begin to collapse. It is observed in Figures 3a, b, c and d that peak negative pressure thresholds of inertial cavitation were approximately 263 and 235 kpa with and without bone for 5 microsecond pulse length, and 217 kpa and 215 kpa with and without bone for 2 microsecond pulse length, respectively. As far as we know, the skull bone is a heterogeneous and anisotropic medium. The outer layer and interior of the bone are composed of compact bone tissue and trabecular bone tissue with porosity, respectively. The complicated bone structure would attenuate and scatter the ultrasound wave, which would influence the ultrasound field that, in turn, may affect the thresholds of inertial cavitation. For 5 microseconds pulse length, peak negative pressure threshold of inertial cavitation in the presence of bone is about 11.9% higher than that without the bone. However, the difference of threshold of ineretial cavitation is very small for 2 microseconds pulse length. It is also noted in Figures 3a, b, c and d that the peak negative pressure thresholds of inertial cavitation decrease in both the absence and presence of the skull bone as ultrasound pulse length increases from 5 microseconds to 2 microseconds. The Rayleigh-Plesset equation which describes the dynamic behavior of microbubbles

5 Cavitation (mv/sample) Cavitation (mv/sample) Cavitation (mv/sample) Cavitation (mv/sample) under external excitation implies that both the acoustic pressure levels and pulse length play important roles in destruction of microbubbles 12. Our results show that long ultrasound pulse would result in a lower cavitation threshold and induce more collapsing microbubbles than short ultrasound pulse. Attenuation of the bone was investigated by comparing the acoustic pressure levels at the same ultrasound power output with and without bone using a needle hydrophone. Experimental results showed that approximately 5% of the acoustic pressure amplitude was attenuated in the presence of bone kpa kpa kpa kpa FIGURE 3. Averaged RMS values of voltage from the PCD as function of time for (a) 5 microseconds in the presence of pig temporal bone; (b) 2 microseconds in the presence of pig temporal bone; (c) 5 microseconds in the absence of pig temporal bone; and (d) 2 microseconds in the absence of pig temporal bone; SUMMARY Transcranial thresholds of inertial cavitation induced by 1.6 MHz diagnostic ultrasound and Definity microbubbles in the presence and absence of the skull bone and at different ultrasound pulse length were quantitatively analyzed using the passive cavitation detection method. Results suggest the power output of a diagnostic ultrasound imaging system should be carefully controlled in clinical application in case that strong inertial cavitation happens when microbubbles are used.

6 REFERENCES 1. Murray CJ, Lopez AD. Mortality by cause for eight regions of the world: Global Burden of Disease Study. Lancel 1997; 349: Cannon CP. Exploring the issues of appropriate dosing in the treatment of acute myocardial infarction: potential benefits of bolus fibrinolytic agents. American Heart Journal 2; 14: S154- S6. 3. Thomas R. Porter and Feng Xie. Ultrasound, microbubbles, and thrombolysis. Progress in Cardiovascular Diseases 21; 44(2): Saurabh Datta, Constantin-C. Coussios, Azzdine Y Ammi, T. Douglass Mast, Gabrielle M. de Courten-Myers, and Christy K. Holland. Ultrasound-enhanced thrombolysis using Definity as a cavitation nucleation agent. Ultrasound Med Biol 28; 34(9): Andrei V. Alexandrov, Robert Mikulik, Marc Ribo, Vijay K. Sharma, Annabelle Y. Lao, Georgios Tsivgoulis, Rebecca M. Sugg, Andrew Barreto, Paul Siezenski, Marc D. Malkoff and James C. Grotta. A pilot randomized clinical safety study of sonothrombolysis augmentation with ultrasoundactivated perflutren-lipid microspheres for acute ischemic stroke. Stroke 28; 39: Xie Feng, Tsutsui JM, Lof J, Unger EC, Johanning J, Culp WC, Matsunaga T, Porter TR. Effectiveness of lipid microbubbles and ultrasound in declotting thrombosis. Ultrasound in Medicine and Biology 25; 31(7): William C. Culp, Eren Erdem, Paula K. Roberson, Muhammad M. Husain. Microbubble potentiated ultrasound as a method of stroke therapy in a pig model: preliminary findings. J Vasc Interv Radiol 23; 14: Everbach, E. C. and C. W. Francis. Cavitational mechanisms in ultrasound-accelerated thrombolysis at 1 MHz. Ultrasound in Medicine and Biology 2; 12(3): Behrens, S., K. Spengos, et al. Transcranial ultrasound-improved thrombolysis: diagnostic vs. therapeutic ultrasound. Ultrasound in Medicine and Biology 21; 27(12): Prokop AF, Soltani A, Roy RA. Cavitational mechanisms in ultrasound-accelerated fibrinolysis. Ultrasound in Medicine and Biology 27; 33(6): William T. Shi, Shunji Gao, Vignon, Francois, Jeff E. Powers, Lucas Drvol, Ki Won Jung, Feng Xie, John Lof, E. Carr Everbach, Thomas R. Porter. Investigation of effectiveness of microbubble stable cavitation in thrombolysis. IEEE 21 Ultrasound Symposium. 12. T. G. Leighton. The Acoustic Bubble. Academic Press, First printing paperback edition, 1997, Chapter 4.