Therapeutic ultrasound in ischemic stroke treatment: experimental evidence

Abstract

Re-opening of the occluded artery is the primary therapeutic goal in hyper-acute ischemic stroke. Systemic treatment with IV rt-PA has been shown to be beneficial at least in a 3 h ‘door to needle’ window and is approved within that interval in many countries. Trials of thrombolytic therapy with rt-PA demonstrated a small, but significant improvement in neurological outcome in selected patients. As recently shown, intra-arterial application of rt-PA is effective and opens the therapeutical window to 6 h, but requires invasive intra-arterial angiographic intervention in a high number of patients, who do not finally achieve thrombolysis. Ultrasound (US) is known to have several biological effects depending on the emission characteristics. At higher energy levels US alone has a thrombolytic effect. That effect is already used for clinical purposes in interventional therapy using US catheters. Recently, there is growing evidence that US at lower energy levels (<2 W/cm²) facilitates enzymatic mediated thrombolysis, most probably by breaking molecular linkages of fibrin polymers and therefore, increasing the working surface for the thrombolytic drug. Different in-vitro and in-vivo experiments have shown increased clot lysis as well as accelerated recanalization of occluded peripheral, coronary vessels and most recently also intracerebral arteries. Sonothrombolysis at low energy levels, however, is of great interest because of the low risk for collateral tissue damage, enabling external insonation without the need for local catheterization. Whereas little or no attenuation of US can be expected through skin and chest, intensity will be significantly attenuated if penetration of bones, particularly the skull, is required. That effect, however, is frequency dependent. Whereas >90% of intracerebral US intensity is lost (of the output power) in frequencies currently used for diagnostic purposes (mostly 2 MHz and up), that ratio is nearly reversed in the lower KHz range (<300 kHz). US at these low frequencies, however, is efficient for accelerating enzymatic thrombolysis in-vitro as well as in vivo within a wide range of intensities, from 0.5 W/cm² (MI∼0.3) to several W/cm². Since the emitted US beam widens with decreasing frequency, low-frequency US can insonate the entire intracerebral vasculature. That may overcome the limitation of US in the MHz range being restricted to insonation of the MCA mainstem. There are no reports in the preclinical literature about intracerebral bleeding or relevant cerebral cellular damage (either signs of necrosis or apoptosis) for US energy levels up to 1 W/cm². Moreover, recent investigations showed no break-down of the blood brain barrier. Safety of US exposure of the brain for therapeutic purposes has to address heating. Heating depends critically on the characteristics of the US. The most significant heating of the brain tissue itself is >1 °C/h using a continuous wave (CW) 2 W/cm² probe, whereas no significant heating could be found when using an intermittent (pulsed) emission protocol. The experimental data so far help to characterize the optimal US settings for sonothrombolysis and support the hypothesis that this combined treatment is a prospective advance in optimizing thrombolytic therapy in acute stroke.