SONOlysis in Prevention of Brain InfaRctions During Internal Carotid Endarterectomy

Overview

SONOBIRDIE Trial is a randomized, single-blind, sham-controlled study designed for a demonstration of the safety and effectiveness of sonolysis (continual transcranial Doppler (TCD) monitoring) in reduction of risk of stroke or transient ischemic attack (TIA), brain infarctions and cognitive decline using a 2-MHz diagnostic probe with a maximal diagnostic energy on the reduction of risk of brain infarctions by the activation of endogenous fibrinolytic system during carotid endarterectomy (CEA) in patients with ≥ 70% symptomatic or asymptomatic internal carotid artery stenosis. The sample size is based on an expected 2.5% reduction of ischemic stroke, TIA, and death during the 30-day postoperative period in the sonolysis group (estimated prevalence, 1.5 %) compared to the control group (estimated prevalence, 4 %). Pre-study calculations showed that a minimum of 746 patients in each group is needed to reach a significant difference with an alpha value of 0.05 (two-tailed) and a beta value of 0.8 assuming that 10 % would be lost to follow-up or refuse to participate in the study. Consecutive patients will be assigned to the sonolysis or control group by a computer-generated 1:1 randomization. In patients randomized into sonolysis group, middle cerebral artery segment in a depth of 55 mm will be continuously monitored during intervention using a diagnostic 2-MHz TCD probe with a maximal diagnostic energy. In patients randomized into control group, the TCD probe will be fixed in a required position using a special helmet as in sonolysis group patients, but middle cerebral artery segment in a depth of 55 mm will be only localized using a diagnostic 2-MHz TCD probe with a maximal diagnostic energy and the TCD monitoring will be stopped afterwards. Confirmation of the investigators hypothesis that sonothrombolysis is able to activate endogenous fibrinolytic system during CEA with consecutive reduction of ischemic stroke, TIA or death, and the number and volume of brain infarcts, can lead to the increase of the safety of CEA in patients with internal carotid artery stenosis. The investigators can presume that up to 50% of patients indicated for CEA can be treated using these methods in the future.

Results of the NASCET, ECST and ACAS studies showed that carotid endarterectomy (CEA) was a beneficial therapy for patients with a symptomatic internal carotid artery (ICA) stenosis \>50% and asymptomatic ICA stenosis \>60%, resp. Surgical risk of CEA varied between 2 and 15 % and even clinically silent microembolism can cause microinfarctions presenting with postoperative cognitive deficit. Transcranial Doppler (TCD) monitoring during CEA is a common diagnostic method being used for the detection of microemboli and changes of blood flow in intracranial arteries. Reduction of periprocedural complications of CEA with TCD monitoring referred in some studies should be not only due to sophisticated indication of shunt implementation, optimization of surgery and anesthesia but also due to TCD activation of endogenous fibrinolytic system (equal to sonothrombolysis in acute stroke studies). Recently published SONOBUSTER Trial showed that intraoperative sonolysis reduced both the incidence and the volume of new brain infarctions following CEA. There are 2 possible effects of ultrasound on thrombus - mechanical destruction due to vibration of thrombus with acceleration of penetration of fibrinolytics into thrombus and elevation of temperature and stimulation of endothelium with local activation of fibrinolytic system. Grant project NR/9487-3/2007 showed that TCD monitoring has a significant effect on activation of fibrinolytic system in healthy volunteers. STUDY OBJECTIVES The objective of the randomized, double-blind, sham-controlled study is to demonstrate the safety and effectiveness of sonolysis (continual TCD monitoring) using a 2-MHz diagnostic probe with a maximal diagnostic energy on the reduction of stroke, transient ischemic attack (TIA) and brain infarction by the activation of endogenous fibrinolytic system during CEA in patients with ≥70% symptomatic or asymptomatic ICA stenosis. The substudy aims to compare the risk of brain infarction detected using magnetic resonance between sonolysis and control group. STUDY DESIGN Overview: SONOBIRDIE Trial is a randomized, single-blind, sham-controlled study designed for a demonstration of the safety and effectiveness of sonolysis (continual TCD monitoring) in reduction of risk of ischemic stroke, TIA, death, brain infarctions and cognitive decline by the sonolysis during CEA in patients with ≥70% ICA stenosis. Expected sample size: The sample size is based on an expected 2.5% absolute risk reduction of ischemic stroke, TIA and death during the 30-day postoperative period in the sonolysis group (estimated prevalence, 1.5%) compared to the control group (estimated prevalence, 4%). Pre-study calculations showed that a minimum of 746 patients in each group is needed to reach a significant difference with an alpha value of 0.05 (two-tailed) and a beta value of 0.8 assuming that 10% would be lost to follow-up or refuse to participate in the study. The sample size for substudy is based on an expected 15% reduction of new ischemic lesions on diffusion weighted imaging-MRI (DWI-MRI) in the sonolysis group (estimated prevalence, 10%) compared to the control group (estimated prevalence, 25%). Pre-study calculations showed that a minimum of 112 patients in each group was needed to reach a significant difference with an alpha value of 0.05 (two-tailed) and a beta value of 0.8 assuming that 10 % would be lost to follow-up or refuse to participate in the study. Test device: The transcranial Doppler systems (e.g. DWL Multi-Dop T1, DWL, Sipplingen, Germany) with a diagnostic 2-MHz probe will be used for sonolysis (non-diagnostic TCD Doppler monitoring). Sonolysis: In patients randomized into sonolysis group, MCA segment in a depth of 55 mm will be continuously monitored during intervention using a diagnostic 2-MHz TCD probe with a maximal diagnostic energy (TIC\~1.3), sample volume 10 mm. Sham procedure: In patients randomized into control group, the TCD probe will be fixed in a required position using a special helmet as in sonolysis group patients, but MCA segment in a depth of 55 mm will be only localized using a diagnostic 2-MHz TCD probe with a maximal diagnostic energy and the TCD monitoring will be stopped afterwards. Carotid endarterectomy: Surgery will be performed in a general or a local anesthesia (decision will be left to the discretion of the operating team) using a cut in front angle of the sternomastoid muscle. Common carotid artery (CCA) and later ICA and external carotid artery (ECA) will be cut free. Common carotid artery, ICA and ECA will be temporarily closed. Using a longitudinal cut of CCA and ICA, atherosclerotic plaque will be visualized. Plaque will be withdrawn under the microscopic control and later a suture of arteriotomy will be performed using a monophil non-absorbent fibre 6/0. Just before the end of surgery, haemostatic process will be controlled and drainage will be done. Surgery will be completed by suture of subcutis and cutis. Unfractioned heparin (100 IU/kg bodyweight) will be administered in all patients just before the arteriotomy. In the case of the insufficient collateral flow into MCA after clipping of the CCA and ICA, a temporal shunt will be used. Antiplatelet therapy (Aggrenox, clopidogrel 75 mg/day or acetylsalicylic acid 100 mg/day) will be used continuously in all patients. Surgeon will be blinded to sonolysis or sham procedure. Magnetic resonance imaging: Magnetic resonance imaging will be performed in patients enrolled to the substudy. Magnetic resonance imaging protocol consists of 4 sequences: 1. T2TSE; 2. DWI. Sequences 1 - 3 will be applied in the same level, they will have the same slice thickness and the same cut number. The slice thickness comprises its own cut thickness (5 mm) + distant factor (30 %). Standard number of slices will be 19. Standard slice level is considered to be a modified level of skull base due to the minimalization of distant artifacts. Diffusion weighted sequence shows a middle (average) diffusivity of every point of the examined brain tissue when b value is 500 and 1000. This sequence is applied in order to assess hemorrhage (T2\*EPI) and to monitor sites of the reduced diffusion (DWI, b=500 and 1000); 3. T2 star-weighted gradient-recalled echo (GRE) sequence for detection of bleeding (including microbleeds); 4. Fluid-attenuated inversion recovery (FLAIR). Presence of new infarctions will be evaluated separately in the whole brain, in the intervened ICA territory and in the contralateral ICA territory. New ischemic lesions in the brain are defined as hyperintense lesions on postintervention DWI sequences which were not present on the pretreatment MRI. The volume of new brain infarctions will be measured manually. Ischemic lesions in the brain will be evaluated by two blinded investigators by consensus. Clinical examinations and cognitive tests: Standard physical and neurologic examinations (including the National Institutes of Health Stroke Scale - NIHSS, modified Rankin scale - mRS, and Addenbrooke's Cognitive Examination Revised - ACE-R) will be performed prior to CEA, 24 hours after CEA, 30 days and 1 year after randomization. Randomization: Consecutive patients will be assigned to the sonolysis or control group by a computer-generated 1:1 randomization. ANALYSIS SETS Efficacy analyses will be performed primary for the intent-to-treat population. The secondary analysis will be performed also for per-protocol population. The intent-to-treat population will consist of all randomized subjects who signed informed consent. The per-protocol population will exclude all subjects in the intent-to-treat population who did not have any major protocol deviations: * Violation of inclusion or exclusion criteria. * Withdrawal of consent within 30 days * Crossing-over to the other treatment arm. Cross-overs are defined as patients randomized to the control group who receive continuous TCD monitoring for 10 minutes or longer and patients randomized to the sonolysis group who receive less than 10 min of sonolysis. * Total time of sonolysis less than 40 min for patients randomized to the sonolysis group * Total time of TCD monitoring of 40 min or more for patients randomized to the control group * Carotid endarterectomy not within 2 days (48 hours) of randomization * Not attending at least one of visits 3 (24 hours after CEA), 4 (30 days after CEA), or 5 (1 year after CEA) * Outcomes assessed outside the specified time windows The safety population consists of all subjects in the FAS who did receive one of the study interventions. Subjects will be analyzed according to the treatment they actually received (as treated): * Patients randomized to the control group who receive continuous TCD monitoring for 10 minutes or longer will be considered to have received sonolysis. * Patients randomized to the sonolysis group who receive less than 10 min of sonolysis will be considered to have received the control intervention. STATISTICAL METHODS The primary analysis will be based on the full analysis set. Missing data will be handled according to Statistical Analysis Plan. The proportion of the composite of ischemic stroke, TIA, or death within 30 days will be calculated in both groups with a 95% Wilson score confidence interval. Groups will be compared using the chi-squared test. As an effect measure we will calculated the absolute risk difference with Newcombe hybrid score 95% CIs. Mortality will be graphically depicted by Kaplan-Meier curves for each treatment group. Groups will be compared using a log-rank test. Mortality at 30 days for both groups and the difference between groups will be calculated with 95% CI from the Kaplan-Meier estimator. A risk difference with 95% CI will be provided, calculated on the log-scale using Greenwood standard errors and a normal approximation. The proportion of any stroke and myocardial infarction within 30 days will be calculated for each group using the non-parametric cumulative incidence function estimator with death as competing event and 95% CI. Changes in cognitive functions will be compared between groups using a linear regression with baseline values and the group as covariates (usually referred to as ANCOVA). Assumptions for linear regression (homoscedasticity and normality of the errors) will be verified using residual-vs-fitted and QQ-plots. If the assumptions are not met, transformations (e.g. log), more robust methods (e.g. robust standard errors) or non-parametric methods will be considered. Binary substudy outcomes (appearance of at least one new lesion) will be compared using chi-squared tests. The effects will be presented as absolute risk differences with Newcombe hybrid score 95% CIs. The number of new lesions will be compared between groups using an exact Poisson test. The effect will be presented as incidence rate ratio with an exact 95% CI. Multivariable logistic regression analyses will be used to analyze possible predictors of the primary outcome (composite of ischemic stroke, TIA and death), cognitive decline, or a new brain infarction. Candidate predictors include age, gender, side of stenosis, symptomatic stenosis, time from symptoms to CEA, percentage of ipsilateral ICA stenosis, percentage of contralateral ICA stenosis, arterial hypertension, diabetes mellitus, hyperlipidemia, smoking, alcohol abuse, coronary heart disease, atrial fibrillation, type of anesthesia, shunt use, antithrombotic drug use, number of antihypertensive drugs, insulin use, oral antidiabetics use, perioperative use of heparin, perioperative dose of heparin, statin use, dose of statin, type of plaque in ipsilateral carotid stenosis. All models will include the treatment group. Results will be reported as odds ratio with 95% CI. Data will be analyzed using R (R Development Core Team. 2008. R: A language and environment for statistical computing. R Foundation for Statistical Computing. Vienna, Austria), Stata (StataCorp. 2017. Stata Statistical Software: Release 15. College Station, TX: StataCorp LLC) and/or SPSS (IBM, Armonk, NY, USA).