Investigation of Brain Blood Flow Changes in Acute Ischemic Stroke Patients After Successful Endovascular Thrombectomy

Introduction:

Acute ischemic stroke (AIS) is a significant global health concern representing the second-leading cause of death and the third-leading cause of disability. In up to 46% of the cases, AIS is caused by the acute occlusion of a large cerebral vessel, leading to loss of blood flow to a specific part of the brain (i.e., tissue ischemia) and resulting in severe neurological deficits. The burden of these deficits and ischemic tissue damage may be significantly reduced or even completely reversed with endovascular thrombectomy (EVT), a procedure aimed at reopening the blocked vessel by removing the blood clot (recanalization) and restoring brain blood flow. However, in clinical practice, nearly half of the patients do not experience adequate neurological improvement despite successful recanalization - a phenomenon termed reperfusion failure or clinically ineffective reperfusion. This condition is characterized by progression of brain damage, hemorrhagic transformation of the infarcted area, and poor clinical scores in the short- and long-term. Given its clinical relevance, efforts have been made to better understand this phenomenon. Several studies in animal models have identified various possible pathophysiological mechanisms that may contribute to reperfusion failure, including microvascular damage, inflammatory response, microvascular occlusion and neurovascular dysfunction. Impaired brain tissue autoregulation appears to be one hemodynamic condition that characterizes reperfusion failure and links part of these potential mechanisms. Recently, numerous imaging techniques have been proposed to investigate reperfusion failure in clinical patient populations, but, so far, the number of clinical studies remains limited and none imaging modality has emerged as superior. Identifying the best imaging biomarker to study these patients is of utmost importance because a deeper understanding of hemodynamic tissue changes following EVT and their association with clinical outcomes could help identify patients at higher risk for developing reperfusion failure. Such insights may potentially lead to personalized neuroprotective treatment strategies, ultimately improving clinical outcome after EVT.

BOLD-CVR is a magnetic resonance imaging (MRI) technique, suitable for patients with AIS, capable of measuring brain tissue's ability to regulate blood flow changes (i.e., autoregulation). This technique involves the administration of a standardized, reproducible, and safe carbon dioxide (CO2) respiratory challenge through the RespirActTM gas blender system during BOLD functional MRI (fMRI) sequence acquisition. CO2 administration acts as vasodilatory stimulus, inducing a blood flow increase in brain regions with preserved autoregulation. Conversely, in regions with impaired autoregulation, insufficient blood flow augmentation or even a paradoxical decrease may occur. This phenomenon, known as "steal phenomenon", represents an indicator of severe tissue hemodynamic suffering. Changes in cerebral blood flow induced by the CO2 stimulus are detected through alterations in the BOLD fMRI signal, which reflect variations in the oxy-/deoxyhemoglobin ratio at tissue level. Deoxyhemoglobin molecules exhibit paramagnetic properties, affecting the T2/T2* relaxation time in 3 Tesla MRI, while oxyhemoglobin molecules have no such effect. Consequently, shifts in the oxy-/deoxyhemoglobin ratio caused by blood flow changes lead to alterations in the BOLD fMRI signal. In this way, the combination of BOLD fMRI sequence with a standardized CO2 stimulus allows for the precise measurement of brain tissue hemodynamic changes with high temporal and spatial resolution, and high reproducibility.

In this project, the investigators aim to study the role of blood oxygenation-level dependent cerebrovascular reactivity (BOLD-CVR) as novel imaging biomarker to investigate reperfusion failure. The investigators thereby integrate the BOLD-CVR technique into the standard post-treatment imaging assessment of patients with LVO-AIS who have undergone EVT at our institution and examine the relationship between BOLD-CVR signals and clinical outcomes.

Project objectives and design:

The investigators hypothesize that the presence of brain tissue areas exhibiting impaired BOLD-CVR values in patients with LVO-AIS following endovascular thrombectomy is associated with higher risk of developing reperfusion failure (i.e., impaired neurological recovery).

Primary objective:

The investigators aim to characterize blood flow changes in reperfused tissue during the early post-EVT period using BOLD fMRI and examine their relationship to clinical outcomes 90 days post-EVT.

Secondary objective:

Our secondary objective is to compare the BOLD-CVR signal with co-acquired perfusion measures, including DSC MR perfusion imaging and intraprocedural angiographic signs (e.g., early venous filling, capillary blush); the investigators will also evaluate the association between BOLD-CVR signal and 7-day clinical outcomes and in-hospital clinical deterioration.

Study outcomes

Given the research question and underlying hypothesis of this project, the association between BOLD-CVR findings (exposure variable) and predefined clinical outcomes will be investigated:

Primary outcome:

• 90-day functional outcome measured through modified Ranking Score (mRS)

Secondary outcome:

• functional outcome at 7 days (hospital discharge) measured through mRS.
• neurological deterioration during hospitalization defined as any new neurological worsening that satisfies 1 or more of the following criteria: an increase in total NIHSS score ≥2, an increase in the NIHSS subscore 1a, 1b, or 1c (level of consciousness) ≥1, or an increase in the NIHSS subscore 5a, 5b, 6a, or 6b (motor) ≥1
• radiologically confirmed hemorrhagic transformation within the reperfused tissue defined according to the Heidelberg classification.
• radiologically confirmed infarct lesion progression within the reperfused tissue.

The association between BOLD-CVR signal and clinical as well as radiological outcomes will be investigated using regression analysis.

Additionally, the investigators will explore correlations between BOLD-CVR findings and various standard hemodynamic imaging parameters, including: (a) cerebral blood flow (CBF), cerebral blood volume (CBV), mean transit time (MTT), and Tmax values in the reperfused tissue area; (b) the frequency of early venous filling or capillary blush sign in post-treatment digital subtraction angiography (DSA); (c) systolic blood flow velocities and pulsatility index in the middle cerebral artery ipsilateral to the ischemic tissue side.

Baseline characteristics Baseline characteristics to collect include age, sex, occluded side (left/right), occluded vessel (internal carotid artery, ICA/middle cerebral artery, MCA/both), symptom onset time, last-seen-well time, NIHSS score at hospital admission, mRS score at hospital admission, known comorbidities (such as presence of atrial fibrillation, smoking history, arterial hypertension, dyslipidaemia, obesity, diabetes), admission CT metrics (Alberta stroke program early CT score, ASPECTS; core volume (in mL); penumbra volume (in mL); penumbra-core mismatch volume (in mL); hypoperfusion intensity ratio, HIR; collateral grading score), type of acute reperfusion therapy (IVT + EVT, or EVT alone), symptom- or last-seen-well-to-groin puncture time (in hours), mTICI grade, DWI-derived infarct lesion volume (in mL) at MRI baseline examination.

Project design The project design comprises a single-center, prospective, observational cohort study aimed at investigating BOLD-CVR imaging in patients with LVO-AIS at three predefined time points following EVT: at 72 hours, 7 days, and 90 days. Data on baseline characteristics, clinical outcomes as well as findings derived from standard hemodynamic imaging techniques will be collected prospectively during the study period.

This project will be carried out at the Clinical Neuroscience Center of the University Hospital Zurich, a leading primary stroke center in Switzerland that offers BOLD-CVR as an established technique as part of its institutional imaging infrastructure.

The investigators would like to emphasize that this study will not affect clinical decision-making or the standard-of-care treatment provided to participating patients.

Project population and study procedure

Project population, inclusion and exclusion criteria:

• Target sample size: 100 patients
• Study period: 2.5 years (June 2025 - December 2027): The investigators aim to enroll a target sample size of 100 patients over a period of 2.5 years. This translates to an inclusion of 3-4 patients per month, with the last three months allocated for the follow-up of the last included patients.

Inclusion criteria

• Male and female subjects >18 years old.
• Acute ischemic stroke related to a new unilateral occlusion of the ICA, MCA (M1- and/or M2-segment), or both.
• Eligibility and performance of endovascular thrombectomy.
• Achievement of successful recanalization, defined as mTICI score ≥2b
• Written informed consent of the patient or when the patient is not able to participate in the consenting procedure, the written authorization of an independent doctor not involved in the research project is requested to safeguard the interests of the patients (in that case, post hoc written informed consent of the patient or next of kin had to be obtained). Foreign speaking patients should be included by a person with sufficient German language proficiency to act as a translator.

Exclusion criteria

• Documented evidence or a confirmed willingness of the patient not to participate in any scientific study.
• Standard contraindications for MRI, such as pacemaker, metallic prothesis, glaucoma, metallic tattoo dyes, verbal confirmed pregnancy.
• Unwilling or unable to cooperate with breathing manoeuvres.
• Major cardiopulmonary diseases, such as severe uncontrolled asthma bronchiale, severe chronic obstructive lung disease (i.e., GOLD stage ≥III), diffuse interstitial lung disease, pulmonary embolism, acute/subacute myocardial infarction, or severe heart failure (i.e., NYHA class ≥III). • Symptomatic increased intracranial pressure.
• Presence of intracranial hemorrhage type ≥2 according to the Heidelberg classification.
• New onset of seizure.

Recruitment, screening and informed consent procedure:

Patients presenting at the Stroke center of the University Hospital Zurich with LVO-AIS eligible for EVT will be consecutively considered for recruitment through the project leader and his team in daily clinical practice. Recruitment will take place at the stroke intermediate care unit after EVT has been performed to avoid any impact or delay on standard acute patient treatment. An initial patient screening to assess inclusion and exclusion criteria will be conducted by the project leader and his team, along with the treating physician, through a review of the medical chart. Any presence of an exclusion criterion (e.g., severe cardiopulmonary disease) at this stage will result in immediate exclusion from the recruitment process. If a candidate is still eligible for recruitment after this initial screening, the project leader and his team will personally approach the patient on the ward to discuss inclusion in the study. Sufficient time will be given for the patient to consider participation, and all questions will be addressed. In cases where the candidate is unable to participate in the consenting procedure, written authorization will be obtained from a legal representative or family member to protect the patient's interests. In such instances, post hoc written informed consent from the patient will be obtained.

Throughout the recruitment process, candidates will be clearly informed that refusing to participate in the project will not affect their further medical treatment.

Study procedures Overall project duration: 30 months Duration of the recruitment phase: 27 months Project duration for each participant: 3 months from EVT Planned project start (FPFV): June 2025 Planned project end (LPLV): December 2027

Study procedure description

• Patient recruitment will occur during hospitalization for LVO-AIS after EVT has been performed.
• Once a new patient has been included in the study, a study identification number will be created and his/her baseline characteristics will be added to the electronic Case Report Form (eCRF) created for the data collection of this project, an Excel file named "BOLDCVR_postEVT_MasterDatabase.xlsx". This Excel file will be stored on a secure server, accessible only to members of the study team, and as a backup measure, also on a hard drive secured in the office of the study team, under lock and key.
• Included patients will undergo a total of 3 MRI examinations including BOLD-CVR imaging at 24-48 hours, 5-7 days, and 90 days after EVT. Participation in the final examination will mark the end of the subject's involvement in the study.
• Clinical outcomes will be prospectively collected as per established institutional patient management protocols (during hospitalization, at discharge, and at the cerebrovascular outpatient clinic at 3 months) and added to the eCRF "BOLDCVR_postEVT_MasterDatabase.xlsx".
• MRI examinations will be performed with a 3-Tesla Skyra MRI scanner (Siemens, Erlangen, Germany) after the patients have been enrolled in the study. Best hardware and software available will be used to maximize patient comfort and minimize acquisition time. The MRI protocol includes the following sequences:

• Scout: 15 seconds
• Diffusion-weighted imaging (DWI): 2 minutes and 30 seconds
• Susceptibility-weighted imaging (SWI): 3 minutes
• 3D Fluid-Attenuated Inversion Recovery (FLAIR): 6 minutes
• 3D Time of Flight (TOF) angiography: 4 minutes and 15 seconds
• BOLD fMRI + CO2 stimulus: 6 minutes and 30 seconds: During the BOLD fMRI sequence, a standardized carbon dioxide (CO2) stimulus was applied using the gas blender system RespirActTM (Thornhill Research Institute, Toronto, Canada) as previously reported.This involved 100 seconds at the patient specific resting PETCO2, followed by an increase to 10 mmHg for 80 seconds, and then a return to resting PETCO2 for 220 seconds, while PETO2 levels were kept constant. Vital signs (including blood pressure, oxygen saturation, and electrocardiogram) were continuously monitored during the session to promptly address any potential adverse events.
• MPRAGE 3D native T1: 5 minutes
• Gadolinium-based Dynamic Susceptibility Contrast (DSC) perfusion: 2 minutes
• Total scan time: 29 minutes and 30 seconds
• DWI, SWI, 3D FLAIR, 3D TOF, MPRAGE 3D native T1, and Gadolinium-based DSC perfusion imaging will be acquired following clinical standard acquisition protocols.
• All imaging data will be prospectively retrieved from the MRI data-room and stored both on the secure server and on the hard drive. Patients will be saved according to their study identification number.
• Each BOLD-CVR imaging dataset will be processed and analysed to extrapolate BOLD-CVR quantitative values. Image analysis will be performed only on computers in Prof. Fierstra's research lab, following a previously described analysis pipeline that utilizes MATLAB2022 (The MathWorks, Inc., Natick, United States) and SPM12 (Wellcome Trust Centre for Neuroimaging, Institute of Neurology, University College London). Only authorized members of the research team will have access to these computers.

Possible project biases:

The investigators acknowledge that our project may be susceptible to certain biases, including selection bias and confounding bias.

To mitigate the risk of selection bias, the investigators plan to include the maximum number of consecutive patients presenting at our institution with LVO-AIS who undergo EVT. It is crucial to emphasize that inclusion criteria should not be influenced by the patient's clinical status. Even patients who are unable to participate in the consenting procedure - vulnerable patients - should be included following the steps outlined above.

The risk of confounding bias will be addressed by identifying potential confounding variables relevant to the research question and controlling for them through appropriate statistical analyses.

Withdrawal and discontinuation

On one hand, study participants may be withdrawn from the project by any member of the study team at any time if any of the following conditions are met:

• Development of any of the exclusion criteria after participant inclusion in the study.
• Development of malignant infarct.
• Low quality of the baseline BOLD-CVR images acquired during the first examination session.

On the other hand, participants may choose to withdraw from the study at any time without prior notice and without the need for an explanation.

If a participant misses an appointment for a follow-up MRI examination (at 7 days or 90 days after EVT), efforts will be made to encourage their continued participation in the study, and a new appointment will be scheduled. Participants lost to follow-up and withdrawn patients will be accurately tracked in the master database.

If a new severe adverse event linked to the application of BOLD-CVR imaging in this specific patient cohort is identified (HRO, art. 21), the project will be terminated.

Statistics and methodology Given the observational study design and exploratory nature of this project, no sample size calculation can be performed. The provided target sample size (N = 100) has been estimated considering the inclusion of as many consecutive subjects as possible.

Statistical significance will be defined as two-sided p values < 0.05. Due to the exploratory nature of the statistical analysis, no correction for multiple testing will be implemented.

Regulatory aspects and safety

Local regulations / Declaration of Helsinki:

This research project will be conducted in accordance with the protocol, the Declaration of Helsinki, the principles of Good Clinical Practice, the Human Research Act (HRA) and the Human Research Ordinance (HRO) as well as other locally relevant regulations. The project leader acknowledges his responsibilities as both the project leader and the Sponsor.

Notification of safety and protective measures (HRA Art. 15, HRO Art. 20):

If, during the research project, circumstances arise which could jeopardise the safety or health of the participants or lead to a disproportionate relationship between the risks and burdens and the benefits, all the measures required to ensure protection are to be taken without delay.

The project leader and the Sponsor are promptly notified (within 24 hours) if immediate safety and protective measures have to be taken during the conduct of the research project. The Ethics Committee will be notified via BASEC of these measures and of the circumstances necessitating them within 7 days.

Serious events (HRO Art. 21):

If a serious event occurs, the research project will be interrupted, and the Ethics Committee notified on the circumstances via BASEC within 7 days according to HRO Art. 211.

Amendments Substantial changes to the project set-up, the protocol and relevant project documents will be submitted to the Ethics Committee for approval according to HRO Art. 18 before implementation. Exceptions are measures that have to be taken immediately in order to protect the participants.

Inclusion criteria

• Male and female subjects >18 years old.
• Acute ischemic stroke related to a new occlusion of the internal carotid artery, middle cerebral artery (M1- and/or M2-segment), or both.
• Eligibility and performance of endovascular thrombectomy.
• Achievement of successful recanalization, defined as mTICI score ≥2b
• Written informed consent of the patient or when the patient is not able to participate in the consenting procedure, the written authorization of a family member. Foreign speaking patients should be included by a person with sufficient German language proficiency to act as a translator.

Exclusion criteria

• Documented evidence or a confirmed willingness of the patient not to participate in any scientific study.
• Standard contraindications for MRI, such as pacemaker, metallic prothesis, glaucoma, metallic tattoo dyes, verbal confirmed pregnancy. Unwilling or unable to cooperate with breathing manoeuvres.
• Major cardiopulmonary diseases, such as severe uncontrolled asthma bronchiale, severe chronic obstructive lung disease (i.e., GOLD stage ≥III), diffuse interstitial lung disease, pulmonary embolism, acute/subacute myocardial infarction, or severe heart failure (i.e., NYHA class ≥III).
• Symptomatic increased intracranial pressure.
• Presence of intracranial haemorrhage type ≥2 according to the Heidelberg classification.
• New onset of seizure.