Author + information
- Received October 7, 2002
- Revision received February 4, 2003
- Accepted February 25, 2003
- Published online May 7, 2003.
- Takashi Miyamoto, MD*,
- Yoram Neuman, MD*,
- Huai Luo, MD*,
- Doo-Soo Jeon, MD*,
- Sergio Kobal, MD*,
- Fumiaki Ikeno, MD*,
- Michael Horzewski, BS*,
- Yasuhiro Honda, MD*,
- James M Mirocha, MS*,
- Takahiro Iwami, MD*,
- Debra Echt, MD, FACC*,
- Michael C Fishbein, MD, FACC† and
- Robert J Siegel, MD, FACC*,* ()
- ↵*Reprint requests and correspondence:
Dr. Robert J. Siegel, Division of Cardiology, Room #5335, Cedars-Sinai Medical Center, 8700 Beverly Boulevard, Los Angeles, California 90048.
Objectives We evaluated the coronary vasodilatory effects of transcutaneous low-frequency (27-kHz) ultrasound (USD).
Background Ultrasound has been shown to affect vascular function.
Methods Ultrasound energy was administered transcutaneously to 12 dogs. Coronary arterial dimensions were assessed using intravascular coronary ultrasound (IVUS) and quantitative coronary angiography (QCA).
Results The IVUS mid-left anterior descending (LAD) luminal area was 6.77 ± 1.27 mm2at baseline. After 30 s of ultrasound, this area increased by 9% (7.40 ± 1.44 mm2, p < 0.05), after 3 min by 19% (8.05 ± 1.72 mm2, p < 0.05) and after 5 min increased by 21% (8.16 ± 1.29 mm2, p < 0.05). The mean coronary diameter (2.69 ± 0.33 mm) at baseline (QCA of three segments of LAD and three segments of left circumflex coronary artery) increased by 19.3% (3.21 ± 0.28 mm) after 5 min of USD exposure. After a 90-min observation period there was a return to baseline values (p = NS). Intracoronary nitroglycerin (NTG) administered to five dogs revealed a similar magnitude of vasodilation as USD.
Conclusions Noninvasive, transthoracic low-frequency USD energy results in coronary artery vasodilation within seconds of exposure. The vasodilation is reversible and is similar in magnitude to that induced by NTG. Further evaluation is needed to assess its potential applications in humans.
Invasive, catheter-delivered, low-frequency (20 kHz) high-intensity ultrasound (USD) has been shown to cause vasodilation in peripheral and coronary arteries in animals and in humans (1–3). However, there are no data on the vasomotor effects of transcutaneous USD. In this study, we evaluated the potential vasodilatory effects of transcutaneous low-frequency USD in canine coronary arteries, as such a noninvasive approach could represent a novel method for treatment of coronary syndromes.
Twelve dogs (20 to 30 kg) were anesthetized with thiopental and maintained by isoflurane inhalation. The dogs received heparin 1,000 U subcutaneously. Studies were performed according to the position of the American Heart Association on research animal use.
Transcutaneous USD device
The device consists of a generator/amplifier and controller delivering 27-kHz pulsed-wave USD through a transducer placed over the mid-sternum of the thorax (Timi3 Systems, Inc., Santa Clara, California). The transducer-radiating surface faces the skin and it is coupled to the skin with USD gel. Ultrasound intensity is approximately 1.4 W/cm2and the pulsed wave duty is cycle 30%. The beam shape of the USD field is diverging and covers the entire heart. Because of the diverging USD field, the acoustic pressure further away from the transducer face is lower than at 5 mm from the transducer. At 5 mm, the peak rarefactional (i.e., negative) pressure is 230 kPa, at 4 cm about 170 kPa, and at 10 cm about 100 kPa. In short, between 4 and 10 cm from the transducer face the peak rarefactional pressure decreases from about 170 kPa to about 100 kPa. The output of the transducer was measured with a calibrated hydrophone (model 8103 Brüel & Kæjr, Nærum, Denmark) in accordance with the existing guidelines and standards (4–6).
Intravascular coronary ultrasound (IVUS)
As shown in Figure 1, after coronary angiography for baseline measurements, IVUS was performed in the mid-left anterior descending coronary artery (LAD) in 11 of 12 dogs. Luminal areas were continuously monitored during 5 min of USD exposure using a CVIS imaging system and coronary catheters (3 F, 40 MHz Atlantis SR and 3.2 F, 30 MHz, Ultracross, Boston Scientific, Massachusetts) (Fig. 2). In five dogs, intracoronary nitroglycerin (NTG) (100 μg) was administered 90 min after stopping USD. The IVUS recordings were analyzed by an independent core laboratory (7)blinded to the status of the dog, whether control period or during the NTG or USD exposure phase of the study (Cardiovascular Core Analysis Laboratory, Stanford University, Stanford, California).
Coronary angiography protocol
As noted in Figure 1,coronary angiograms were performed at baseline in the anterior-posterior and selected oblique projections after 5, 30, 60, and 90 min of USD, as well as 30, 60, and 90 min after stopping USD. In five dogs after the 90-min observation period, quantitative coronary angiography (QCA) was also performed 5 min after intracoronary NTG administration. Six segments (three LAD and three left circumflex [LCx]) were serially measured by QCA (Fig. 2). Arterial diameters were measured using a computer edge detection program (Advantx DXC Digital System, General Electric Medical Systems, Milwaukee, Wisconsin).
During transcutaneous USD exposure, skin temperature below the transducer was measured by a thermocouple sensor (MedSource Inc., Santa Clara, California). Blood temperature within the LAD was measured with a thermocouple sensor (Timi3 Systems Inc.) mounted on a coronary guidewire placed in the mid-portion of the LAD.
Dogs were euthanized; hearts were excised and fixed in formalin for 24 to 72 h and then cut transversely in 1-cm intervals from apex to base. Representative sections from the LAD and LCx arteries, as well as the skin, soft tissues, myocardium, and lung were submitted for microscopic examination.
Continuous variables are expressed as mean ± standard deviation or percent change. Intraobserver and interobserver agreements were assessed using regression analysis and the Bland-Altman method. To assess changes from baseline to various time points, repeated measures analysis of variance was employed. Post hoc comparisons of all time points with baseline were performed using Dunnett’s procedure, which controls the experiment-wise type I error rate. A value of p < 0.05 was considered statistically significant.
Figure 3demonstrates mean mid-LAD luminal area at baseline, after 30 s, and each minute from 1 to 5 min as well as mean percent increase area after USD exposure. As shown, after 30 s of USD exposure, mean luminal area increased by 9% (from 6.77 ± 1.27 mm2to 7.40 ± 1.44 mm2, p < 0.05), after 3 min by 19% (from 6.77 ± 1.27 mm2to 8.05 ± 1.72 mm2, p < 0.05), and after 5 min by 21% (from 6.77 ± 1.27 mm2to 8.16 ± 1.29 mm2, p < 0.05). Dunnett’s procedure indicated that the luminal areas after exposure to USD at all time points (30 s, 1 min, 2 min, 3 min, 4 min, 5 min) were significantly greater than baseline luminal area.
Figure 4shows the data on QCA mean luminal diameter of the LAD and LCx. There was a 19.3% mean increase in all segments from 2.69 ± 0.33 mm to 3.21 ± 0.28 mm (p < 0.05) after 5 min of USD. After a 60-min observation period, mean diameter resolved to 5.6% above baseline values (p = NS). After the 60-min observation period, USD treatment was initiated again for 90 min and angiograms as shown in Figure 5were performed at 30-, 60-, and 90-min intervals during USD exposure. The mean increase in QCA coronary arterial diameters was 12.6% (from 2.69 ± 0.33 mm to 3.03 ± 0.30 mm, p < 0.05) at 30 min, 12.3 % (to 3.02 ± 0.28 mm, p < 0.05) at 60 min, and 8.2% (to 2.91 ± 0.28 mm, p = NS) at 90 min. During a subsequent 90-min observation period (no USD), angiographic evaluation of luminal diameter in each of the segments studied was repeated every 30 min. There was a decrease in the coronary artery luminal diameter towards baseline during the observation period after USD exposure.
Both QCA and IVUS regression analyses indicated strong intraobserver and interobserver correlations (r ≥ 0.98). The Bland-Altman method indicated good agreement. The averaged limits of agreement (mean difference) for IVUS were 0.24 (0.01) mm2and 0.58 (0.10) mm2, and for QCA it was 0.14 (0.02) mm and 0.27 (0.11) mm. The mean absolute percentage difference (SD) ranged from 1.0% (1.0%) for IVUS to 4.2% (3.6%) for QCA.
Baseline mean luminal area measured by IVUS in the mid-LAD was 6.95 ± 1.40 mm2. After 100 μg intracoronary NTG, maximal luminal area increase within 5 min was 21% (8.40 ± 1.21 mm2) over baseline. Mean luminal QCA diameter of the LAD and LCx 5 min after NTG administration increased 16.6% from baseline luminal diameter (from 2.47 ± 0.39 mm to 2.88 ± 0.45 mm, p < 0.05). Dunnett’s procedure indicated that the luminal areas at 1 min and 2 min were significantly greater than the luminal area at baseline.
Skin temperature below the transducer in four dogs was 34.9 ± 1.08°C at baseline and 34.8 ± 1.92°C (p = 0.85) after 90 min of USD. Intracoronary temperature in one dog showed no change during USD exposure (37.0°C at baseline and 36.9°C after 60 min).
No USD-mediated injury to the coronary arteries, skin, soft tissue, myocardium, or lung was detected. Specifically, there was no thermal injury, necrosis, or inflammatory changes in the tissues.
This is the first study to demonstrate that noninvasive, transcutaneous low-frequency USD causes coronary artery vasodilation. Both IVUS and QCA were used to document that transcutaneous USD exposure results in coronary artery vasodilation in a canine model.
Effect of short duration USD exposure
In this closed-chest canine model, noninvasive USD exposure caused significant vasodilation of the coronary arteries. As measured by IVUS, the vasodilation effect was apparent within 30 s of USD exposure. As shown in Figure 3, luminal area gradually increased, resulting in a 21% increase after 5 min. By QCA, the coronary arterial diameter also increased by 19.3% after 5 min of USD. Sixty minutes after stopping USD, there was resolution of vasodilation, indicating that this is a reversible phenomenon.
Effect of prolonged duration USD exposure
In the second part of the study, the effect of the duration of USD exposure was evaluated. After 30 and 60 min of USD exposure, there was an increase in the mean luminal diameter by QCA but it was less prominent than the increase after the first short exposure of 5 min. Subsequent arterial size reduction towards normal at 90 min despite continuation of USD exposure indicates that vasodilation becomes attenuated over time. The QCA luminal diameter significantly increased in response to ultrasound in all the LAD and LCx segments studied.
Tissue heating can cause vasodilation (8). High-intensity USD can heat tissue (9,10). Although USD-induced vasodilation could be mediated by heating, we found no evidence of USD tissue heating even after 90 min of USD exposure. Thus, vasodilation was not thermally mediated in this study.
Relative vasodilatory effects of USD and NTG
The magnitude of coronary artery vasodilation induced by USD exposure and NTG was similar. There was a 21% increase in luminal area by IVUS for both USD and NTG and a 19.3% and 16.6% increase, respectively, in QCA diameter. Similar mechanisms could be responsible for USD- and NTG-induced vasodilation.
Mechanism of USD-induced vasodilation
We believe that the coronary artery vasodilation is most likely a direct effect of USD on coronary vasomotor tone. Previously studies from our laboratory demonstrated that catheter-delivered, low-frequency USD (20-kHz) induced vasodilation in rabbit aortas (2), canine coronary arteries (1), and human peripheral arteries (1,3). The magnitude of increase in arterial diameter using intravascular versus external USD was comparable, i.e., a 21% increase in canine coronary artery diameter with intravascular delivery at 20 kHz (1)and 19.3% in this study using a 27-kHz, noninvasive, transcutaneous USD device. These comparable results are consistent with a similar mechanism for transcatheter and transcutaneous USD-induced vasodilation. In these studies of catheter-delivered USD, the effect was limited to the exposed artery. Recently, Suchkova et al. (11)demonstrated that externally applied low-frequency USD (40 kHz) improves tissue perfusion to muscle in ischemic limbs as well as increasing capillary size in the exposed tissue. In addition, they showed that when the animals were pretreated with the nitric oxide synthase inhibitor Nω-nitro-L-arginine methyl-ester, the USD enhancement of tissue perfusion as well as capillary dilation was completely blocked. Thus, it seems likely that the USD-induced capillary dilation and enhanced tissue perfusion are due to a nitric oxide-dependent mechanism. As USD energy may increase mechanical shear stress on the endothelial cells, it could promote local release of the potent vasodilator nitrous oxide (12). It is possible that low-frequency USD could induce an increase in cardiac contractility and stroke volume and secondarily an increase in coronary arterial size to account for the vasodilation detected in our study. However, the results of catheter-based, USD-induced vasodilation, as well as the findings of Suchkova et al. (11)showing USD enhancement of tissue perfusion and capillary size which can be blocked by inhibition of nitric oxide synthase, are consistent with transthoracic low-frequency USD inducing coronary vasodilation by a nitric oxide-dependent mechanism.
While we did measure coronary arterial lumen size by IVUS and QCA, we did not measure the effect of USD on coronary blood flow or velocity, or myocardial oxygen consumption. Nonetheless, we believe the findings of the current study are novel. Further studies are required to delineate the time course of action and to identify the mechanism responsible for the coronary vasodilation effect of transcutaneously applied, low-frequency USD.
Noninvasive, transthoracic, low-frequency USD induces vasodilation in canine coronary arteries. The onset of coronary vasodilation is relatively rapid (≤30 s). The magnitude of the effect is similar to that found with intracoronary NTG. This phenomenon of vasodilation that is caused by externally applied USD energy might have the potential to reduce myocardial ischemia in patients with acute coronary syndromes.
The ultrasonic power and intensity emitted by the Timi3 Systems ultrasound (USD) transducer were measured using a calibrated hydrophone. The USD transducer was connected to the Timi3 USD console and placed against an acoustic window of a tank containing degassed water. The water tank was lined with anechoic material to minimize effects of standing waves. The USD field was mapped using scanning techniques, and single-point hydrophone measurements were made at the point of maximum intensity at 5 mm from the applicator face. The hydrophone used for these measurements is a B&K type 8103 (Brüel & Kæjr, Nærum, Denmark). The hydrophone diameter is 9.5 mm. Its sensitivity at around 30 kHz is −212.7 ± 0.25 dB re 1 V/μPa, and its calibration is traceable to the National Institute of Standards and Technology. The hydrophone was scanned under computer control using a stepper motor XY-positioning system. The distance of the scan from the applicator face was 5 mm, and all measurements were made in a 25 by 25 rectangular grid across to the applicator face with the step increment of 0.200 inch (5.08 mm). The USD pressure was also measured as a function of distance from the transducer face from 5 to 155 mm. The hydrophone signal was digitized, and various acoustic parameters were calculated according to International Electrotechnical Commission standards (5,6)(IEC Publication 60601-2-5), and the Food and Drug Administration’s guidance document (4).
- intravascular coronary ultrasound
- left anterior descending coronary artery
- left circumflex coronary artery
- quantitative coronary angiography
- Received October 7, 2002.
- Revision received February 4, 2003.
- Accepted February 25, 2003.
- American College of Cardiology Foundation
- Steffen W.,
- Cumberland D.,
- Gaines P.,
- et al.
- Fischell T.A.,
- Abbas M.A.,
- Grant G.W.,
- et al.
- Siegel R.J.,
- Gaines P.,
- Procter A.,
- et al.
- ↵CFR 1050.10. Food and Drug Administration, Center for Devices and Radiological Health, 4-1-87 edition
- ↵Measurement and characterization of ultrasonic fields using hydrophones in the frequency range 0.5 MHz to 5 MHz. Geneva: International Electrotechnical Commission, 1991:Publ. IEC 61102
- Ultrasonics—physiotherapy systems—performance requirements for the safety of ultrasonic physiotherapy equipment. Geneva: International Electrotechnical Commission, 1996:Publ. IEC 61689
- Luo H.,
- Nishioka T.,
- Fishbein M.C.,
- et al.
- Suchkova V.N.,
- Baggs R.B.,
- Francis C.W.
- Corson M.A.,
- James N.L.,
- Latta S.E.,
- et al.