Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
J Ultrasound Med. Author manuscript; available in PMC 2010 November 1.
Published in final edited form as:
PMCID: PMC2770178

Cardiac Arrhythmia and Injury Induced in Rats by Burst and Pulsed Mode Ultrasound with Gas Body Contrast Agent

Douglas L. Miller, Ph.D.,FAIUM,1 Chunyan Dou, M.D.,1 and Benedict R. Lucchesi, M.D.2



Premature complexes (PCs) in the electrocardiogram (ECG) signal have been reported for myocardial contrast echocardiography and also for burst mode (physical therapy) ultrasound with gas body contrast agent at lower peak rarefactional pressure amplitudes (PRPAs). For contrast echocardiography, irreversibly injured cardiomyocytes have been associated with the arrhythmia. The objective was to determine if cardiomyocyte injury is associated with the PCs induced by the burst mode at lower PRPAs.


Anesthetized rats were exposed to focused 1.5 MHz ultrasound in a water bath. Evans blue dye was injected IP to stain injured cardiomyocytes and Definity ultrasound contrast agent was infused IV. Continuous burst mode simulated physical therapy ultrasound. Intermittent 2 ms bursts, or envelopes of pulses simulating diagnostic ultrasound, were triggered 1:4 at end systole. PCs were observed on ECG recordings and stained cardiomyocytes were counted in frozen sections.


The continuous burst mode produced variable PCs and stained cells above 0.3 MPa PRPA. The triggered bursts above 0.3 MPa and pulse envelopes above 1.2 MPa produced statistically significant (P<0.01) PCs and stained cardiomyocytes.


Irreversible cardiomyocyte injury was associated with the development of PCs for burst mode and occurred at substantially lower PRPAs than for pulsed ultrasound.

Keywords: Contrast agent adverse effects, Myocardial Contrast Echocardiography, Arrhythmia, Gas body activation, Ultrasonic cavitation biology

1. Introduction

Ultrasound contrast agents are approved for use in enhancing diagnostic ultrasound images. The agents consist of suspensions of gas bodies (stabilized microbubbles), which circulate in the blood pool and return strong echos relative to the weak echos from blood. The first application of gas body contrast agents was for echocardiography with approval to opacify the blood pool within the left ventricular chamber and to improve the delineation of the endocardial border. Myocardial contrast echocardiography is another application, which seeks to image perfusion in the cardiac muscle. This can be accomplished at relative high pulse peak rarefaction pressure amplitudes (PRPAs) of diagnostic ultrasound by destroying the gas bodies and then observing the rate of refill of blood with fresh gas bodies into the tissue. Safety considerations for contrast enhanced diagnostic ultrasound have received substantial research interest because the interaction between ultrasound pulses and gas bodies is an important mechanism for nonthermal bioeffects of ultrasound. This research has lead to a variety of bioeffects findings and to potential therapeutic applications [1].

An intriguing bioeffect of myocardial contrast echocardiography is cardiac arrhythmia. This effect has been reported in humans [2, 3], as well as laboratory animals including rats [4] and dogs [5]. The arrhythmia effect of contrast echocardiography occurs at relatively high PRPAs which are also associated with gas body destabilization and destruction by the ultrasound pulses. The arrhythmia shows as premature complexes in the electrocardiogram (ECG) signal which occur coincidentally with the ultrasound image-exposure and cease after the ultrasound is stopped. A typical exposure which elicits this arrhythmia in rats involves infusion of a contrast agent, intermittent imaging every fourth complex (1:4) at end systole and a PRPA above about 1 MPa at 1.7 MHz [4]. The intermittent image-exposure allows for the refill of tissue capillaries with blood containing gas bodies, which are destroyed by each image-exposure. The induction of arrhythmia with contrast agent in the circulation also has been observed with low power burst mode ultrasound, such as is used for physical therapy [3]. Continuous burst mode (2 ms bursts repeated at 10 ms intervals) exposure at 1 MHz and 0.33-0.39 MPa produced more arrhythmia than intermittent (1:4) 1.7 MHz diagnostic ultrasound at 1.0 MPa.

Another important bioeffect at relatively high PRPAs is irreversible cardiomyocyte injury. Histological evidence of cardiomyocyte contraction band necrosis immediately after exposure has been shown for diagnostic ultrasound with commercial contrast agent in rats [6]. Irreversible cardiomyocyte injury also was identified by staining with Evans blue dye, which was detected the day after exposure [7]. The incidence of the Evans blue staining was associated with the appearance of premature complexes in the ECG signal used for triggering images, suggesting that the two effects were related. Although the mean numbers of premature complexes and stained cardiomyocytes followed a very similar exposure-response trend with PRPA, there was little correlation between the two bioeffect for separate observations in individual animals. This association was also found for high PRPAs (above the range of diagnostic ultrasound)[8]. Premature complexes were seen for 10 of 20 rats exposed to 3.1 MHz ultrasound with 1.3 μs pulses at 15.9 MPa and Optison® in the circulation. Myocardial degeneration was identified by histological staining in 16 rats, which implies that the presence of myocardial degeneration alone may not be a sufficient explanation of the premature complexes.

The association of arrhythmia and cardiomyocyte injury at moderate and high PRPAs suggests that the ultrasonic activation of the gas bodies nucleates cavitation, which induces both bioeffects. Cavitation is the most effective mechanism for nonthermal bioeffects of ultrasound. Cavitation nucleation by ultrasound contrast agents is important in regard to the bioeffects potential in vivo, because there normally are few, if any, cavitation nuclei in the body suitable for direct activation by diagnostic ultrasound pulses [9]. Ultrasound contrast agents can supply such nuclei. For example, direct evidence of cavitational activity has been obtained by detection of broad band noise emissions from the myocardium during contrast echocardiography [10]. However, this picture is confused by the observation of premature complexes with lower-amplitude physical therapy ultrasound [3]. The objective of this study was to explore the association between the arrhythmia and cardiomyocyte injury by comparing these effects for continuous burst mode ultrasound, triggered single bursts and square or ramped envelopes of ultrasound pulses (simulating diagnostic ultrasound imaging). A laboratory exposure system was developed to produce all three modes of exposure under similar conditions (i. e. with the same 1.5 MHz transducer) over a range of PRPAs. Premature complexes where detected in ECG recordings and cardiomyocyte injury was detected by the Evans blue method. The results strengthen the association between the two bioeffects of cardiac exposure to ultrasound with gas body contrast agent.

Materials and Methods

Animal preparation

All in vivo animal procedures were conducted with the approval and guidance of the University Committee on Use and Care of Animals. CD hairless rats (Charles River) were anesthetized by intraperitoneal (IP) injection of a mixture of ketamine (87 mg ml-1) and xylazine (13 ml kg-1). A 24 gauge cannula was inserted into a tail vein for IV injections. Evans blue dye in saline (20 mg/ml) at a dose of 100 mg/kg was injected IP as a vital stain for cardiomyocytes. The rats then were mounted on a holding board and ECG needle electrodes placed in the forelegs and right hind leg. The holding board was then mounted in a 37° C degassed water bath for ultrasound exposures.

The ECG signal was amplified (Model ECGA amplifier, Hugo Sachs Electronik, March FRG) and sent to an oscilloscope (Model TDS 520B, Tektronix Inc., Beaverton OR USA) and to a digitizer (Powerlab 4/30, ADInstruments Inc. Colorado Springs, CO USA). The digitized ECG was analyzed with the aid of software (Chart Pro 5, v. 5.5.5, ADInstruments Inc. Colorado Springs, CO USA), which provided automated collection of data on heart rate and the numbers of normal complexes. The software also partially automated detection of premature complexes. However, visual examination of the premature complexes was performed to classify these into two categories [11]. Supraventricular premature complexes (SPC) including a P wave were seen with or without a compensatory pause. Premature ventricular complexes (PVCs) without a P wave, but always with a compensatory pause, were seen as single or followed by re-entry complexes.

Definity® (Lantheus Medical Imaging, Inc., N. Billerica, MA) ultrasound contrast agent was prepared fresh each day. For infusion, the agent first was diluted 50:1 in sterile saline in a 3 ml syringe, which was then used to fill a 30 cm extension tube for connection to the tail vein canula. The syringe was mounted in a syringe pump (model 11plus, Harvard Apparatus, Holliston MA) and the infusion rate was set to 500 μl/ kg/ min of diluted agent (10 μl/kg/min of agent). Infusions with exposure were 5 min in duration. The total agent dose of 50 μl/kg was about 2.5 times the recommended (package insert) human infusion dose for diagnostic imaging.


Ultrasound exposure was provided by a laboratory system with guidance by diagnostic ultrasound imaging, as described previously [12]. The laboratory exposure system consisted of a transducer, power amplifier (A-500, Electronic Navigation Industries, Rochester NY), function generator for generating a pulse train (model 3314A function generator, Hewlett Packard Co., Palo Alto CA) and an arbitrary waveform generator (model 33220A, Agilent Technologies, Loveland CO). The function generator was set to provide a 1.5 MHz continuous signal or a pulse train with 3 cycle pulses and 230 μs pulse repetition period. The arbitrary waveform generator was used for amplitude modulation of the continuous or pulsed signals. Four different signals were produced: continuous burst (2 ms on, 6 ms off), triggered single bursts (2 ms), triggered pulse envelope (2 ms square envelope), or triggered ramped pulse envelopes (50 ms envelope ramping either up or down). The continuous burst simulated the therapeutic ultrasound mode commonly used for physical therapy. The triggering was used to allow intermittent exposure timed from the ECG. The triggered pulse envelope was used to partly simulate intermittent diagnostic ultrasound imaging. Finally, the ramp modulated pulses were used to induce or suppress gas body activity within the tissue, as described previously [12]. Triggering was set at end systole from the ECG signal displayed on the oscilloscope by the delayed trigger function. The oscilloscope delayed trigger out signal was used to externally trigger the arbitrary waveform generator to generate the amplitude modulation envelopes for the function generator. In addition, the modulation signal was digitized and the number of triggers during exposure was obtained with the Chart Pro software.

A damped single element transducer (Panametrics A3464, Olympus NDT Inc. Waltham, MA) with 1.9 cm diameter and 3.8 cm focus was used for ultrasound exposure. The PRPA of each signal was measured with a calibrated hydrophone (model 805, Sonora Medical Systems Inc., Longmont CO USA). The 3 cycle pulse signals produced 1.8 μs ultrasound pulses. A diagnostic ultrasound machine (GE Vingmed System V, General Electric Co., Cincinnati OH USA) was used to image the heart with an 8 MHz probe prior to exposure. Once an image of the rat heart was obtained, which showed a clear path to the heart, the exposure transducer was moved into the same position to aim at the left ventricle.

For evaluation of Evans blue staining of cardiomyocytes, the hearts were removed the day after exposure and cleared of most blood with heparin-saline. The heart was trimmed and then frozen in embedding medium (Tissue-Tek O. C. T. Compound, Sakura Finetek USA Inc. Torrance CA USA) on dry ice. These samples were stored in a -80 C freezer for later sectioning on a frozen section microtome. To search for any stained cardiomyocytes, 10 μm sections were cut each 200 μm into the sample, which typically gave 20-24 slides from the exposed central portion of the heart. Each slide was examined with a microscope using fluorescence illumination in order to estimate the number of Evans blue stained (red fluorescent) cells. The slide with the most cell staining was used to obtain a detailed count of the number of stained cells over the entire section.

The study was conducted in 16 groups of 5 rats each and one with 4 rats. For continuous burst mode, one sham group was exposed to 1 MPa, and 5 groups were exposed to different PRPAs in the range 0.25 - 1.0 MPa (in 3 dB steps). For the triggered single burst mode, 6 groups were exposed to PRPAs in the range 0.25-1.4 MPa. For the triggered pulse envelope, 4 groups were exposed to PRPAs in the range 1.0-2.8 MPa. Two groups were exposed with the ramped pulse envelopes with a maximum 2.8 MPa, one with the ramp up (pulse PRPA increasing with time) and one with the ramp down (pulse PRPA decreasing with time). For all groups, the ECG was monitored and recorded for one minute to confirm stable heart function under anesthesia, and no PCs were observed during this pre-exposure period. For exposure, the infusion was started, the ultrasound switched on and the ECG was recorded for a manually timed period of 5 min. The ultrasound and infusion were then switched off. The ECG was monitored and recorded for an additional 5 min, during which no PCs were seen (i. e. any ultrasound induced arrhythmia ceased with the cessation of ultrasound exposure). Results are reported as the means plus or minus one standard deviation, or plotted with standard error bars. For statistical analysis, Student's t-tests or the Mann-Whitney rank sum test, as appropriate, were used to compare means of the measured parameters, with statistical significance assumed at P<0.05.


The different types of premature complexes are shown in Fig. 1. The number of premature complexes are shown in Fig. 2, for 1.0 MPa continuous burst and triggered single burst exposures. The supraventricular premature complexes were primarily seen for the continuous burst mode with none occurring for the triggered burst mode. Premature ventricular complexes were the predominant arrhythmia for the triggered single burst and triggered pulse envelope exposures. PVCs had both positive and negative excursions of varying magnitude. The premature complexes began at the point of the triggered exposure with a delay of about 15-20 ms to the peak of the complex.

Fig. 1
Examples of premature complexes seen with contrast enhanced ultrasound exposures in 1 s segments of ECG data (arbitrary zero time start): (a) an SPC (arrow) with out a compensatory pause, (b) an SPC (arrow) with a compensatory pause after the complex, ...
Fig. 2
The numbers of premature complexes resembling (Fig. 1a or 1b) and premature ventricular complexes (PVCs, resembling Fig. 1c, 1d, or 1e) for the continuous burst and triggered single burst modes at 1.0 MPa. PVCs were the predominant premature complex for ...

The exposure response of the two bioeffects are shown in Fig. 3 for the continuous burst mode, Fig. 4 for the triggered single bursts and in Fig. 5 for the triggered pulse envelope. For all three modes both effects were essentially zero, then rose steadily at higher PRPAs. For the continuous burst mode, the stained cell count was small and variable, with stained cells observed in 3 of 5 rat heart samples at 0.35 MPa, 2 of 5 at 0.5 MPa, and 3 of 4 at 0.7 MPa. The count achieved statistical significance only at the 1.0 MPa level. However, the premature complexes were observed for all rats exposed at and above 0.35 MPa. The PCs were statistically significant at 0.35 MPa, which indicates a threshold at 0.3 MPa (midway between the lowest level with a significant effect, and the highest level without a significant effect). The exposures triggered form the ECG, which allowed refill of blood containing contrast agent gas bodies, produced much larger counts of PCs and stained cells. For the triggered single bursts, both effects were not statistically significant at 0.25 MPa, but were at 0.35 MPa, indicating a threshold of 0.3 MPa for both bioeffects. For the triggered pulse envelope, both bioeffects were significant at 1.4 MPa, but not at 1.0 MPa, indicating thresholds of 1.2 MPa. Modes and thresholds are listed in Table 1.

Fig. 3
Means with standard error bars for the numbers of premature complexes and Evans blue stained cells found for the continuous burst mode. The data trends with PRPA are similar for both effects.
Fig. 4
Means with standard error bars for the numbers of premature complexes and Evans blue stained cells found for the triggered burst mode. The results were much greater than for the continuous burst mode (Fig. 3) but, again, the data trends with PRPA are ...
Fig. 5
Means with standard error bars for the numbers of premature complexes and Evans blue stained cells found for the triggered pulse envelope mode. The results are diminished for the pulsed mode relative to the burst mode (Fig. 4), but the data trends with ...
Table 1
Apparent thresholds (defined as the midpoint between the lowest MPa level with a significant effect, and the next lower level) for premature complexes (PC) and stained cells (SC) for the three exposure modes. Correlation coefficients (r2) are also listed ...

The trends in the data for all three modes seem to be very similar for both the premature complexes and stained cells. Results for the two effects were examined by linear regression on the individual data and on the means. Results for the three modes are shown in Figs. 6--88 with r2 values listed in Table 1. Regression for the continuous burst mode showed no correlation between the two effects for individual data, which was highly variable, but good correlation for the means. For the triggered single burst and pulse envelope modes, the correlation was good for both the individual data and the means. The results for triggered exposure strengthens the association between these two bioeffects.

Fig. 6
Linear regressions of the stained cell counts and premature complexes for the continuous burst mode. Although the data means (squares, bi-directional error bars, dashed line) have a good correlation, the individual data points (circles, solid line) do ...
Fig. 8
Linear regressions of the stained cell counts and premature complexes for the triggered pulse envelope mode. Both the data means (squares, bi-directional error bars, dashed line) and the individual points (circles, solid line) have a good correlation ...

The results for the ramp envelope exposure are shown in Fig. 9. The ramp-up and ramp-down envelopes produced a different relative occurrence of the two bioeffects. The ramp-up produced slightly less, but not significantly less, premature complexes than the ramp-down condition. The stained cell counts were reduced significantly (P<0.001) for the ramp-up relative to the ramp-down condition.

Fig. 9
A comparison of the premature complexes and stained cell counts for the ramped pulse envelopes. The numbers of premature complexes are not significantly different for the two modes but the number of stained cells was lower (P<0.001) for the ramp-up ...

Discussion and Conclusions

Premature complexes in the ECG record have been reported for ultrasound with gas body contrast agent for diagnostic B-mode above about 1.0 MPa, and also for burst mode at a substantially lower PRPAs. For B mode, injured cardiomyocytes have been associated with the premature complexes. The objective of this study was to determine if the premature complexes seen with burst mode exposure at lower PRPAs are also associated with cardiomyocyte injury. Three modes were used for characterizing the two bioeffects versus the exposure PRPA: continuous burst mode (2 ms on, 6 ms off), single bursts and pulse envelopes both triggered 1:4 heartbeats at end systole. All three modes produced similar trends for both bioeffects, which was a steady increase with PRPA above an apparent threshold (Figs. 3--5).5). The continuous burst mode produced somewhat variable results, for which the two bioeffects were not correlated (except for regression on the means). It seems likely that the variability resulted from the continuous nature of the exposure, which might have prevented refill of the capillary bed with blood containing gas bodies. The bioeffects thresholds were both 0.3 MPa for the triggered burst mode and both 1.2 MPa for the square pulse envelope (Table 1). The pulse envelope thresholds were similar to that found for diagnostic imaging-exposure [7]. For burst mode exposure, irreversible cardiomyocyte injury was induced together with arrhythmia at substantially lower PRPAs than for pulsed ultrasound. There was a good correlation, especially for the triggered modes, between the two bioeffects (Figs. 6--8,8, Table 1).

For tests with a ramped pulse envelope, the ramp-up envelope reduced the cardiomyocyte injury (P<0.001), but the arrhythmia bioeffect was not statistically different from that for the-ramp down condition. In previous work, the ramp-up envelope essentially abolished the glomerular capillary hemorrhage effect in rat kidneys, which was seen with the ramp-down envelope [12]. This mitigation of the capillary hemorrhage was thought to be due to the destabilization and loss of potentially effective gas bodies by relatively low amplitude pulses in the ramp-up, before the supra-threshold pulses arrived (for ramp-down the reverse is true). For the cardiac bioeffects observed in this study, the ramp-up envelope did not abolish either bioeffect, and stained cells were associated with the occurrence of PCs even for this bioeffects mitigation mode.

The two bioeffects, premature complexes and injured cardiomyocytes, induced by intermittent cardiac exposure by ultrasound with gas body contrast agent occurred together, with similar thresholds, exposure-response trends and a good correlation for low PRPA burst mode as well higher amplitude pulsed ultrasound. This shows that both bioeffects are caused by gas body activation, with destabilization and nucleation of cavitation, at similar PRPAs. Irreversible cardiomyocyte injury may be expected whenever premature complexes develop from intra-cardiac ultrasonic cavitation activity.

Fig. 7
Linear regressions of the stained cell counts and premature complexes for the triggered burst mode. Both the data means (squares, bi-directional error bars, dashed line) and the individual points (circles, solid line) have a good correlation (Table 1 ...


This work was supported by PHS grant EB00338 awarded by the National Institutes of Health, DHHS.


1. Miller DL, Averkiou MA, Brayman AA, Everbach EC, Holland CK, Wible JH, Jr, Wu J. Bioeffects considerations for diagnostic ultrasound contrast agents. J Ultrasound Med. 2008;27:611–632. [PubMed]
2. van Der Wouw PA, Brauns AC, Bailey SE, Powers JE, Wilde AA. Premature ventricular contractions during triggered imaging with ultrasound contrast. J Am Soc Echocardiogr. 2000;13:288–294. [PubMed]
3. Chapman S, Windle J, Xie F, McGrain A, Porter TR. Incidence of cardiac arrhythmias with therapeutic versus diagnostic ultrasound and intravenous microbubbles. J Ultrasound Med. 2005;24:1099–1107. [PubMed]
4. Li P, Cao LQ, Dou CY, Armstrong WR, Miller DL. Impact of myocardial contrast echocardiography on vascular permeability: An in vivo dose response study of delivery mode, ultrasound power and contrast dose. Ultrasound Med Biol. 2003;29:1341–1349. [PubMed]
5. Miller DL, Driscoll EM, Dou C, Armstrong WF, Lucchesi BR. Microvascular permeabilization and cardiomyocyte injury provoked by myocardial contrast echocardiography in a canine model. J Am Coll Cardiol. 2006;47:1464–1468. [PubMed]
6. Miller DL, Li P, Gordon D, Armstrong WF. Histological characterization of microlesions induced by myocardial contrast echocardiography. Echocardiography. 2005;22:25–34. [PubMed]
7. Miller DL, Li P, Dou C, Gordon D, Edwards CA, Armstrong WF. Influence of contrast agent dose and ultrasound exposure on cardiomyocyte injury induced by myocardial contrast echocardiography in rats. Radiology. 2005;237:137–143. [PubMed]
8. Zachary JF, Hartleben SA, Frizzell LA, O'Brien WD., Jr Arrhythmias in rat hearts exposed to pulsed ultrasound after intravenous injection of a contrast agent. J Ultrasound Med. 2002;21:1347–1356. [PubMed]
9. Carstensen EL, Gracewski S, Dalecki D. The search for cavitation in vivo. Ultrasound Med Biol. 2000;26:1377–1385. [PubMed]
10. Porter TR, Everbach C, Kricsfeld D, Xie F. Myocardial cavitational activity during continuous infusion and bolus intravenous injections of perfluorocarbon-containing microbubbles. J Am Soc Echocardiogr. 2001;14:618–625. [PubMed]
11. Hamlin RL. Non-drug-related electrocardiographic features in animal models in safety pharmacology. J Pharmacol Toxicol Methods. 2005;52:60–76. [PubMed]
12. Miller DL, Dou C, Wiggins RC. Simulation of diagnostic ultrasound image pulse sequences in cavitation bioeffects research. J Acoust Soc Am. 2007;122:2002–2008. [PubMed]