Abstract
New therapeutic strategies under development aim to improve recanalization rates and clinical outcomes after ischemic stroke. One such approach is ultrasound (US)-enhanced thrombolysis, or sonothrombolysis, which can improve thrombolytic drug actions and even intrinsic fibrinolysis. Although the mechanisms are not fully understood, it is postulated that thrombolysis enhancement is related to nonthermal mechanical effects of US. Recent results indicate that US with or without microbubbles may be effective in clot lysis of ischemic stroke even without additional thrombolytic drugs. Sonothrombolysis is a promising tool for treating acute ischemic stroke, but its efficacy, safety, and technical details have not been elucidated and proved yet in stroke treatment.
Acute cerebral artery occlusion leads to ischemic brain tissue damage, which is time dependent. If the artery is not opened quickly, the ischemic process worsens, leading to tissue death and cerebral infarction. These factors are the most important determinants of the quality of life and life expectancy of stroke survivors.1
Currently, the only approved treatment for acute ischemic stroke is intravenous (IV) recombinant tissue plasminogen activator (tPA) administered during the first 3 hours after symptom onset.2,3 In Europe, this time window has been extended to 4.5 hours according to recommendations from the European Cooperative Acute Stroke Study III.4
When given IV, tPA can initiate intracranial thrombus dissolution within minutes. This early and very often partial recanalization can lead to ischemic tissue rescue and subsequent recovery. However, less than 5% of patients with ischemic stroke receive IV tPA. Furthermore, only 30% to 40% of treated patients achieve early recanalization, and the recanalization is complete and sustained in only 18%.5–7
Carotid and transcranial sonography in acute stroke allows early recognition of the stroke subtype, etiology, and clinical prognosis at the bedside.8,9 Furthermore, during tPA treatment, continuous transcranial Doppler monitoring can be performed easily with a fixed head frame directed to the occluded cerebral artery. With this procedure, information about vessel patency can be obtained in real time, allowing better selection of patients who could benefit from more aggressive endovascular reperfusion therapies.10
Sonothrombolysis in Stroke Treatment
New therapeutic strategies under development aim to improve recanalization rates and clinical outcomes after ischemic stroke. One such approach is ultrasound (US)-enhanced thrombolysis, or sonothrombolysis, which can improve drug actions and even intrinsic fibrinolysis. The ability of US energy to increase enzymatic thrombolysis was first described in 1976,11 and several experimental studies have confirmed this finding.12–14 Although the mechanisms are not fully understood, it is postulated that thrombolysis enhancement is related to nonthermal mechanical effects of US.15
Negative-pressure US waves inside blood vessels create fluid motion, or microstreaming, and radiation forces, which can promote tPA circulation and increase the thrombus surface in contact with the enzyme.16 Furthermore, in vitro studies have shown that US insonation on a clot leads to reversible disaggregation of cross-linked fibrin fibers.17 Ultrasound can also increase the exposed plasmin-binding sites and the penetration of fibrinolytic enzymes into the clot.18 Acoustic cavitation, which is the ability of US to create microbubbles from gases dissolved in a liquid medium, may also play a role in sonothrombolysis by direct harm to the clot surface or an increase in tPA permeation inside the thrombus.19
Several US frequencies and intensities have been tested in vitro and in animal models.13,18,20–22 The negative US pressure is directly related to the US intensity and inversely related to the frequency.22 Higher US frequencies lead to greater energy attenuation through the skull (up to 90% of energy is lost at 1–2 MHz). On the other hand, an increase in the thermal effects of US is associated with higher intensities.23,24 Therefore, most initial in vitro and animal models for sonothrombolysis were developed with low US frequencies (kilohertz range) and low intensities (0.2–2 W/cm2).24 However, a clinical trial of low-frequency US showed harm from symptomatic bleeding into the brain.25 For that reason, clinical trials at the present time are restricted to higher-frequency, low-intensity US, similar to that used for diagnostic purposes. Frequencies of 1 to 2 MHz have been shown to be safe and effective in experimental models as well as phase 1 and 2 clinical trials. These frequencies also provide information about vessel patency in real time.7
One of the first clinical trials on sonothrombolysis in acute ischemic stroke used low frequencies delivered by pulsed US at 300 ± 1.5 kHz with a spatial-peak temporal-average intensity of 700 mW/cm2.15 A special device was designed for the study, which emitted US waves from 4 diamond-shaped transducers exposing the middle cerebral artery from the contralateral temporal bone surface.25 The Transcranial Low-Frequency Ultrasound-Mediated Thrombolysis in Brain Ischemia trial included acute stroke patients presenting within a 6-hour time window from stroke onset with a baseline stroke severity score of greater than 4 points on the National Institutes of Health (NIH) Stroke Scale at admission and with evidence of proximal intracranial occlusion on brain magnetic resonance angiography.
Patients were randomized to either IV tPA alone (control treatment arm) or IV tPA in conjunction with 90 minutes of low-frequency US insonation (active treatment arm). The study was prematurely stopped because of a significant increase in intracerebral hemorrhage rates in the target group treated with IV tPA and low-frequency US (36% in the active treatment arm versus 0% in the control treatment arm). There were no differences in the recanalization rates or long-term outcomes between the groups. Some of the intracranial hemorrhages in the target group occurred in atypical areas for patients with stroke (subarachnoid, intraventricular, and contralateral hemorrhages). Subsequent experiments suggested that low-frequency US may cause “hot spots” due to summation of US waves when the pulse repetition frequency remains relatively high. In addition, beam propagation and reflections would be different in the human skull than in the small craniums of animal models,25 and a large brain area insonated with the 4-element transducer would increase the amount of potential tissue exposed to US, where low-frequency US could cause disruptions of small arterioles or the blood-brain barrier due to longer wavelengths.26
The safety and efficacy of higher frequencies (2 MHz) was studied at low-intensity (<700 cm="" mw="" span="">2) diagnostic US settings in the Combined Lysis of Thrombus in Brain Ischemia Using Transcranial Ultrasound and Systemic tPA trial.7 This randomized international study tested the combination of 0.9-mg/kg IV tPA administered during a 3-hour window including 2 hours of continuous monitoring with conventional diagnostic transcranial Doppler imaging. After the occlusion was detected on transcranial Doppler imaging, a head frame was placed, which allowed hand-free continuous insonation during 2 hours in the target group with placebo monitoring in controls (1:1 randomization). The prespecified safety end point was symptomatic intracranial hemorrhage, causing worsening of the neurologic deficit by 4 points or more on the NIH Stroke Scale. The primary combined activity end point was complete recanalization on transcranial Doppler imaging or a dramatic clinical recovery, defined as a total NIH Stroke Scale score of 3 points or less or improvement in the score by at least 10 points within 2 hours after the tPA bolus. The clinical investigators were blinded to the group assignment (active or sham monitoring) made by the sonographers. One hundred twenty-six patients were included in the study, and after adjusting for potential confounders, 49% of the target group achieved the combined efficacy end point, compared to 30% of the controls (P = .03). The symptomatic intracranial hemorrhage rate was 4.8% in both groups, and there was a trend toward a better functional outcome in the target group.27
Recently, the results of a small randomized trial with transcranial color-coded US have been reported.28 Transcranial color-coded US transducers have a wider footprint than the 1-cm diameter of transcranial Doppler probes. Transcranial color-coded US generates multiple small beams at dual emission frequencies: ie, one for Doppler imaging (1.8 MHz) and one for grayscale imaging (4 MHz). Patients with less than 3 hours of acute proximal middle cerebral artery occlusion treated with conventional IV tPA were randomized by a coin toss to 1-hour handheld US monitoring of the occluded artery or diagnostic transcranial color-coded US every 20 minutes for 1 hour with a “refresh” mode (grayscale and color-coded imaging of the vessel) every 7 seconds to confirm the location of the pulsed Doppler insonation on the sphenoid segment of the middle cerebral artery. Thirty-seven patients were included, 19 of whom were in the target group. Partial or complete recanalization was achieved in 57% of the patients in the target group (22.2% in controls; P = .045), with better early clinical evolution and a significantly higher rate of Barthel Index values of 95 or greater at 3 months (8 of 19; P = .003). However, the rate of symptomatic intracranial hemorrhage in the patients who received continuous transcranial color-coded US monitoring tended to be higher than in the controls (15.8% versus 5.6%, respectively; P = .60). The small numbers of the study preclude definitive conclusions, but the higher hemorrhage rate compared to the transcranial Doppler studies could be related to increased brain tissue insonation (as in the Transcranial Low-Frequency Ultrasound-Mediated Thrombolysis in Brain Ischemia trial). Furthermore, the dual frequencies of the transcranial color-coded US transducers and their higher mechanical indices could increase the hemorrhagic risk. Another problem with transcranial color-coded sonothrombolysis is the absence of head frames, which forces a handheld approach and probably would prevent generalization of the treatment.29
Use of Microbubbles in Sonothrombolysis
Experimental and in vivo studies have shown that US by itself and, especially, enhanced by microbubbles can induce clot lysis without a fibrinolytic drug.30–32 This effect would probably be related to an increase in the intrinsic fibrinolytic activity, although mechanical thrombus damage by US energy can be detected in experimental models.
In human stroke, a small study tested application of US without fibrinolytic drugs in patients with contraindications to or ineligibility for IV tPA treatment.33 Fifteen patients with proximal middle cerebral artery occlusion and contraindications to tPA were randomized to 1 hour of continuous transcranial color-coded US monitoring or a placebo. The target group achieved a higher recanalization rate and an earlier clinical outcome, but these promising results need to be confirmed in larger studies.
Microbubbles, or microspheres, are gas- or air-filled lipid shell bubbles in the micrometer size range that have been used for a long time as diagnostic US echo contrast agents.30 In the last few years, some experimental data have suggested that microbubbles can also increase the effect of US in sonothrombolysis.19,31 When microbubbles pass through the US energy field, they undergo translations and size oscillations (static cavitation), which generate harmonic signals that are able to increase the acoustic impedance mismatch between the blood and surrounding tissue, improving the diagnostic value of vascular US. These harmonic emissions also release energy and agitate the fluid where the spheres are dissolved, improving tPA delivery and penetration inside a clot.32 If the negative US pressure is increased, the bubble collapses (inertial cavitation), leading to intense localized stresses and microjets, which can cause mechanical fragmentation of the thrombus.34 Therefore, microbubbles act as nuclei for acoustic cavitation, lowering the US intensity threshold for this acoustic phenomenon, which probably cannot be achieved without microbubbles with the low-intensity US emissions permitted in human practice.19
In human stroke, the largest study of microbubble-enhanced sonothrombolysis published to date tested the synergic effect of 3 bolus injections of air-filled galactose-based microbubbles (Levovist; Schering AG, Berlin, Germany) associated with 2 hours of continuous high-frequency, low-intensity diagnostic transcranial Doppler monitoring and IV tPA.35 This pilot study was nonrandomized, included patients with acute middle cerebral artery occlusion within a 3-hour window, and compared the protocol described in the US + tPA group and the tPA-alone group from the Combined Lysis of Thrombus in Brain Ischemia Using Transcranial Ultrasound and Systemic tPA trial. Thirty-eight patients were treated with microbubbles, and the investigators showed that 2 hours after the tPA bolus, the tPA + transcranial Doppler US + Levovist group achieved a 55% sustained recanalization rate compared to 40% and 23% in the tPA + transcranial Doppler US and tPA-alone groups. The rate of intracranial hemorrhage in the target group was 23%, with only 3% symptomatic intracranial hemorrhage, probably reflecting a higher recanalization rate, according to the authors.
The same microbubble type was tested with transcranial color-coded US in patients with acute middle cerebral artery occlusion of less than 3 hours who were randomized to conventional IV tPA alone or tPA and 1 hour of handheld transcranial color-coded US monitoring and a continuous infusion of Levovist during the same period. Complete recanalization in the target group was achieved in 48% of cases, but the study was stopped with only 9 patients included because of an unexpected increase in the intracranial hemorrhage rate (78%). However, none of the hemorrhages were symptomatic, which made the decision of stopping the study controversial. It was also unclear whether the hemorrhage increase could be related to the synergistic effect of the microbubbles or the transcranial color-coded US monitoring (which has been suggested previously), and this issue may not be resolved in the absence of a non-US group.36
On the basis of these promising results, a multicenter international dose escalation phase 1 and 2 single-blinded trial (Transcranial Ultrasound in Clinical Sonothrombolysis) was designed.37 Patients with acute intracranial artery occlusion detected by transcranial Doppler imaging were randomized 2:1 to IV tPA plus continuous infusion of lipid-coated microbubbles containing C3F8 (perflutren) (MRX-801; ImaRx Therapeutics, Inc, Tucson, AZ) plus transcranial Doppler monitoring during 90 minutes or conventional IV tPA. The primary safety outcome was the presence of symptomatic intracranial hemorrhage, and activity outcomes were the complete recanalization rate and functional outcome. Thirty-five patients were included: 23 in the target group (cohort 1, 1.4 mL of microbubbles, 12 patients; cohort 2, 2.8 mL of microbubbles, 11 patients). Twelve control patients received 0.9-mg/kg IV tPA and brief transcranial Doppler assessments. There was no symptomatic hemorrhage in the first cohort and controls, but there were 3 in cohort 2 (27%), of which 2 were fatal, so the study was stopped by the sponsor. The recanalization tended to be faster and higher in both treatment groups (67% and 46% of complete recanalization in cohorts 1 and 2, respectively, compared to 33% in controls), and there was a trend toward a better functional outcome after 3 months in the patients who received the microbubbles. The authors suggested that the increased symptomatic intracranial hemorrhage in cohort 2 may have been influenced by imbalances between the groups. Cohort 2 had a higher baseline NIH Stroke Scale score, a longer interval between tPA and microbubble (MRX-801) infusion, and, probably most important, a higher systolic blood pressure with blood pressure control protocol violations. Therefore, a safe dose of 1.4 mL has been identified as having a trend toward an increased recanalization rate and better functional outcomes; thus, additional larger studies with extended enrollment and further experiments are needed to determine the mechanisms by which microbubbles and US interact with tissue, particularly ischemic stroke tissue.
Safety and Efficacy of Sonothrombolysis
A meta-analysis of randomized and nonrandomized studies on the safety and efficacy of sonothrombolysis38 identified and analyzed all studies of US-enhanced thrombolysis in acute ischemic stroke. Recanalization rates and symptomatic intracranial hemorrhage were compared between tPA, tPA + transcranial Doppler US ± microbubbles, tPA + transcranial color-coded US ± microbubbles, and tPA + low-frequency US.
A total of 6 randomized (n = 224) and 3 nonrandomized (n = 192) studies were identified. The rates of symptomatic intracranial hemorrhage in randomized studies were as follows: tPA + transcranial Doppler US, 3.8% (95% confidence interval [CI], 0%–11.2%); tPA + transcranial color-coded US, 11.1% (95% CI, 0%–28.9%); tPA + low-frequency US, 35.7% (95% CI, 16.2%–61.4%); and tPA alone, 2.9% (95% CI, 0%–8.4%). Complete recanalization rates were higher in patients receiving a combination of transcranial Doppler with tPA compared to patients treated with tPA alone: 37.2% (95% CI, 26.5%–47.9%) versus 17.2% (95% CI, 9.5%–24.9%), respectively.
In 8 trials of high-frequency (transcranial Doppler/transcranial color-coded US) US-enhanced thrombolysis, tPA + transcranial Doppler/transcranial color-coded US ± microbubbles was associated with a higher likelihood of complete recanalization (pooled odds ratio, 2.99; 95% CI, 1.70–5.25; P = .0001) compared to tPA alone. High-frequency sonothrombolysis was not associated with an increased risk of symptomatic intracerebral hemorrhage (pooled odds ratio, 1.26; 95% CI, 0.44–3.60; P = .67). The conclusion of this extensive meta-analysis was that sonothrombolysis with high-frequency diagnostic US appears to be safe, leading to higher rates of complete recanalization compared to systemic thrombolysis.38
To date, no studies of the thrombolytic effects of microbubbles and US alone, without tPA, have been done in human stroke patients.39 Combined therapy with US and microbubbles without tPA was successfully used for lysis of thrombosed dialysis grafts in 22 humans. No adverse events were encountered, and the microbubble clearing of the thrombosed grafts was similar to the effect of tPA alone.40
Recent results suggest that US and microbubbles may be effective for clot lysis in ischemic stroke even without additional thrombolytic drugs. Although improved clinical thrombolysis with microbubbles in combination with US with or without thrombolytic drugs shows great promise in stroke treatment, the optimal techniques and application protocols, indications, and contraindications have to be defined. On the other hand, the optimal microbubble dosage, mode of delivery, thrombolytic drug dosage, and US characteristics (frequency, duration, and mode of insonation) all remain uncertain and need clarification.39,41
Novel Developments in Sonothrombolysis
Novel developments combine nanotechnology with microbubbles for drug delivery (like glycoprotein IIb/IIIa inhibitors). Entrapment of tPA in liposomes can also improve the efficacy of thrombolysis by clot cavitation and acoustic radiation force. Targeting of clot-dissolving therapeutic agents could potentially decrease the frequency of complications while simultaneously increasing treatment effectiveness by concentrating the available drug at the desired site, thus permitting a lower systemic dose of a thrombolytic drug.39 Recent data also suggest that US and microbubbles with or without thrombolytic drugs can also improve flow in the cerebral microcirculation, which may provide new concepts for stroke treatment.39
The complex effect of US on acceleration of thrombus lysis has not been completely elucidated yet. It is assumed that US waves accelerate enzymatic fibrinolysis primarily by nonthermal mechanisms. Other mechanical effects of US, such as temporary peripheral vasodilatation caused by increased production of dinitrogen oxide in the endothelium, radiation forces, and acoustic cavitation, have been noted in the peripheral circulation.15,24
A recently published article by Bardoň et al42 assessed the effect of continuous 1-hour insonation (sonolysis) of the middle cerebral artery in 15 healthy volunteers using a 2-MHz diagnostic transcranial probe. Measurements of blood flow parameters were performed at 2-minute intervals, and during the second session, a flow curve was recorded for 10 seconds at 2-minute intervals. Although irregular changes were recorded in the measured parameters (peak systolic velocity, end-diastolic velocity, mean flow velocity, pulsatility index, and resistive index) during both measurements, there were no significant differences between the two measurements. As opposed to sonolysis of the radial artery, sonolysis of the middle cerebral artery using a diagnostic 2-MHz frequency in healthy volunteers did not lead to significant changes in the cerebral hemodynamic parameters (peak systolic velocity, end-diastolic velocity, pulsatility index, and resistive index) or peripheral vasodilatation. This finding implied that exposure to US energy in the range used for sonolysis is unlikely to cause a primary alteration in cerebral hemodynamic parameters or vasodilatation. However, a synergistic effect during the release of bioactive compounds, enzymes, and proteins cannot be excluded.
The different results seen for the effects of US on peripheral arteries (radial arteries) and cerebral arteries (middle cerebral arteries) can at least in part be explained by the influence of autoregulation in cerebral arteries, which prevents strong vasodilatation caused by US. This factor could be one of the important explanations for the huge differences in recanalization and reocclusion rates, symptomatic intracranial hemorrhage rates, and outcomes for different US frequencies and modes of application of sonothrombolysis in acute ischemic stroke treatment published until now.
In conclusion, sonothrombolysis could become an effective, safe, standardized, and highly recommended treatment of acute ischemic stroke, but further studies are needed to elucidate its complex mechanisms of action and to identify subgroups of patients with ischemic stroke who would have the highest benefit from this kind of therapy.
Abbreviations
CI=confidence interval, IV=intravenous, NIH=National Institutes of Health, tPA=tissue plasminogen activator, US=ultrasound
© 2013 by the American Institute of Ultrasound in Medicine
