Stroke Imaging Research

Stroke accounts for 7% of all deaths, and is the fourth leading cause of death in Canada. Each year, 40,000 to 50,000 Canadians will succumb to stroke and only 10% of victims will recover sufficiently to return to normal living. The cost of treatment, rehabilitation, loss of productivity from disability and premature death totals $2.7 billion a year.

There are several different groups within BIRC who are investigating aspects of stroke, from providing accurate diagnosis of stroke risk and severity, guiding the selection of appropriate treatment and preventing secondary complications, predicting stroke risk, and using technologies such as CT perfusion and arterial spin labeling to measure cerebral blood flow. Our stroke research is strengthened by collaborations with other leading stroke research centres in Canada and the US.

Extending the Stroke Treatment Window

Stroke is caused by an abrupt interruption of blood flow to the brain by a blood clot. Intravenous thrombolysis is currently an approved treatment within 3-4.5 hr of symptom onset.  For the assessment of risk vs benefit associated with this treatment in acute ischemic stroke, it is important to measure the size of the infarct core.  Hyperintensity in diffusion-weighted magnetic resonance imaging (DWI) is the most widely used method to measure infarct core size.  However, a recent review of literature found that on average 24% of DWI hyperintensity lesions can reverse, directly affecting infarct core size determination with this method.

DWI hyperintensity reflects cytotoxic edema which has been accepted as a marker of neuronal death. However, newer results from molecular pathophysiology suggest that neurons can recover from such edema particularly if the ischemia is brief as in the case of ultra-early use of new clot retrieval devices with fast and often complete recanalization and reperfusion. On the other hand, there exists definite cerebral blood flow threshold below which neurons could not survive for more than 1-2 hr. Another marker of neuronal survival is functioning GABA-A receptors, which can be monitored with positron emission tomography (PET) imaging of the receptor ligand, fluoroflumazenil (FFMZ), labeled with the positron emitter F-18. 

The team postulates that brain infarct, as defined by lack of uptake of F-18 FFMZ, has distinct cerebral blood flow (CBF) threshold, which is dependent on the ischemia time. To validate the hypothesis, the research program has used:

(1)    a porcine model of acute cerebral ischemia induced by intra-parenchymal injection of endothelin-1 (ET-1), a potent vasoconstrictor.

(2)    F-18 FFMZ PET imaging to define infarct at different ischemia times and CT Perfusion, a method developed in our lab, monitor CBF contemporaneously to PET imaging and determine the CBF threshold for infarct development. All imaging is performed on a PET-CT scanner to minimize misregistration.

Preliminary results from 6 pigs show that the 3-hr ischemia time CBF threshold for infarction (minimal FFMZ uptake) is 20 ml/min/100g, which agrees with the published threshold for ischemia induced excitotoxicity in the brain. In this condition, neurons are flooded with excitotoxic neurotransmitters, for instance, glutamate, leading to successive neuronal depolarization waves, energy depletion and finally infarction. GABA is released to compete for binding to GABA-A receptors with F-18 FFMZ. The massive amount of GABA released overwhelms competition from the trace quantities of F-18 FFMZ resulting in minimal uptake in brain tissue destined to infarction.

(A) Colour scale PET image acquired 30-50 min post F-18 FFMZ injection (180-200 min post ET-1 injection) fused with grey scale average map of CT Perfusion study acquired 120 min post ET-1 injection. (B) Colour scale PET image fused with grey scale CBF map. (C) CBF map derived from the same CT Perfusion study as in (A) & (B) in color scale from 0-120 mL/min/100g. The ischemia region in the CBF map co-localizes with the low uptake region of the F-18 FFMZ PET image (red arrows).

Following the evidence-based guideline of “time is brain”, current acute ischemic stroke treatment is moving towards ultra-early thrombolysis or thrombectomy (clot removal using interventional techniques with stentrievers or other devices). The efficacy of these early treatments relies critically on determining the size of infarct relative to penumbra.  There is increasing evidence that the clinical gold standard of identification, namely hyperintensity DWI, could overestimate the infarct size, particularly in early ischemia. Once the CBF threshold method is validated, it could be a reliable and easily accessible alternative to DWI. This would greatly increase the efficacy of ultra-early treatment, in particular, thrombectomy, by withholding treatment for those patients with a large infarct, to decrease the risk of hemorrhagic complications.

Enhancing the Safety of rtPA

Although rtPA is effective in dissolving the blood clots that cause stroke, it also can damage blood vessels already affected by stroke, resulting in bleeding in some patients. A method to assess stroke damage to blood vessels is urgently required to achieve enhancement of rtPA safety. This team of imaging researchers has developed a CT scanning method to assess vessel damage in the brain, and has identified the key limits beyond which the likelihood of bleeding is much greater, which puts the patient at greater risk. The team is now validating this vessel damage assessment method in a larger sample of stroke patients to be recruited in London, Toronto, and Ottawa.

Predicting Stroke Risk - Imaging of Carotid Atherosclerosis

The pioneering research of Dr. Henry Barnett in London, Ontario identified that atherosclerosis in the carotid artery is the main source of the blood clots that cause stroke. BIRC researchers are focused on direct imaging using ultrasound and MRI of the patient's carotid artery, plaque, and the arterial wall components, to better understand the difference between patients with stable disease and those who are deemed vulnerable or at risk of imminent stroke.

These imaging methods are providing new direct measurements (as compared to blood levels of cholesterol) to help better guide patient management and treatment before stroke occurs. This team has invented a new imaging method that provides 3D ultrasound images of the carotid arteries to provide an enhanced understanding of the local changes in the carotid artery that occur over time and after treatment. Analysis of these images helps to determine blood vessel and atherosclerosis plaque changes over time. This method will elucidate how and when treatments such as those related to drug, lifestyle, or dietary interventions affect stroke prevention and risk.

Some members of the team are also studying the imaging of vulnerable plaque by using 3-dimensional histology of surgically removed carotid plaques to validate preoperative imaging of such features of vulnerable plaque as ulceration, plaque inflammation on PET/CT, plaque composition on MRI and plaque texture on 3D ultrasound. This will permit identification of high-risk plaques that would warrant surgical removal to prevent stroke. Others are using Doppler ultrasound and flow visualization methods to study blood flow, including potentially problematic levels of turbulence and shear stress in diseased artery models near atherosclerotic plaque build-up and ulcerations. These studies will lead to advanced blood flow measurements, such as using Doppler ultrasound, for improved identification of patients potentially at risk of stroke.

CT Perfusion

CT Perfusion allows existing CT scanners in hospitals to measure tissue blood flow via a software program developed in Dr. T-Y Lee’s lab. GE Healthcare has licensed the software for use on their CT scanners to study stroke, cancer and heart attack patients. Dr. Lee developed this software to diagnose stroke and guide its treatment. The application of the software has also expanded from stroke to cancer, in particular, in the detection of angiogenesis in cancers. In 2008, 948 licenses of the software were purchased by Radiology Departments worldwide for stroke and cancer applications.

Although CT perfusion provides useful information in the management of stroke, it also delivers a substantial radiation dose to the patient. We are developing new imaging processing techniques to reduce the radiation dose so that CT Perfusion studies can be more widely used without concern for the associated radiation risks. The investigators are expanding the applications of CT Perfusion into measurement of tumour blood flow for the monitoring of anti-angiogenesis therapy and of myocardial blood flow for identifying patients who would benefit from revascularization in ischemic heart disease.

Aaron Fenster, PhD, 3D Ultrasound Physics
David Holdsworth, PhD, Blood velocity measurement
Ting-Yim Lee, PhD, CT Perfusion
Grace Parraga, PhD, Vascular Imaging
Tamie Poepping, PhD, Flow Modeling
J. David Spence, MD, Stroke Prevention

Canadian Atherosclerosis Imaging Network investigators
Mingyue Ding (China)