Research at the Stanford Stroke Center

The basic premise underlying acute stroke therapy is to salvage the ischemic region from evolving into infarction, thereby maintaining brain function and improving outcome.  The concept of the ischemic penumbra concept envisions not only potentially salvageable or at-risk ischemic tissue but also nonviable tissue known as the “ischemic core”. The Stanford Stroke Center has been at the forefront of developing acute imaging and image processing techniques that provide immediate and accurate visualization of both core and penumbra.  These techniques, which involve MRI with diffusion weighted imaging (DWI) and perfusion weighed imaging (PWI), have proven to identify patients who can benefit from both intravenous and intra-arterial therapies well beyond established time frames. 

A research collaboration coordinated at Stanford helped establish that a lesion detected by DWI lesion is an extremely accurate surrogate for the ischemic core.  Subsequently, this concept was translated to CT perfusion imaging with thresholded relative CBF maps.

Data from Stanford demonstrated that perfusion imaging, when optimally processed, can accurately identify critically hypoperfused penumbral tissue.

The DEFUSE Study

This Stanford study, sponsored by the NIH, demonstrated that patients with a favorable MRI profile, called Target Mismatch, have excellent outcomes following reperfusion, even when treated up to 6 hours after symptom onset with iv tPA.  Other MRI profiles, including the No mismatch and Malignant profile failed to show any evidence of benefit from reperfusion.

RAPID Software

In order to automatically process  advance stroke imaging data quickly and accurately, Stanford Stroke Center faculty members developed a unique software platform called RAPID.  Using the databases from multiple international studies, this software program has been demonstrated to identify patients who benefit from reperfusion following late window iv tPA therapy and subsequently endovascular therapy.

DEFUSE 2

This multicenter trial was designed and run by Stanford and funded by the NIH. The trial results demonstrated that using the RAPID software, selected patients can be identified who benefit for intra-arterial clot removal therapy up to 12 hours after symptom onset. These research findings demonstrated that the paradigm for acute stroke treatment could move away from arbitrary time windows. Identification of salvageable brain tissue and determining the site of vascular obstruction has become the focus of acute imaging.  DEFUSE 3 and DAWN then demonstrated that treatment strategies can then be individualized based on imaging findings and leading to dramatic clinical benefits even in late treatment windows.

DEFUSE 3

DEFUSE 3 was a 38-center NIH-funded study led by the Stanford Stroke Center that demonstrated that nearly half of all patients treated between six and 16 hours after the onset of their symptoms could be largely spared from the consequences of their stroke and the number of stroke patients who died or required confinement to nursing homes was nearly cut in half.  These results represent the largest improvements seen in any stroke-related trial to date. This clinical trial demonstrated that far more people than previously thought can benefit from thrombectomy for acute ischemic stroke.

The improved outcomes were achieved through the use of the RAPID software platform for patient selection.  This software helps identify stroke patients who continue to have salvageable brain tissue long after the therapeutic window had generally been considered helpful has closed.  Results of the trial were published in The New England Journal of Medicine and coincided with AHA new acute-stroke treatment guidelines that expanded the stoke treatment window from 6 to 24 hours.

Stanford Neurologists Play Key Role In Redefinition Of TIA, Determining Prognosis And Optimal Management

In a 2002 article in the New England Journal of Medicine, Stanford Neurologist Greg Albers, MD, and other cerebrovascular experts called for a revision in the definition of TIA — from time-based (the resolution of symptoms in 24 hours) — to the presence or absence of brain infarction, a tissue-based definition, on neuroimaging. In 2009, the American Stroke Association released a guideline endorsing this change in the definition of TIA.

Image shows a DWI positive lesion in a patient with transient left sided weakness.

The benefits of the new TIA definition have been demonstrated in a series of recent publications in Lancet Neurology, Stroke, and Neurology that were co-authored by Stanford Stroke Center neurologists. These studies have documented that that identifying a small area of brain injury on the MRI is much more powerful than clinical data alone to predicting early stroke risk; in fact early stroke risk is about 15 times higher in patients with TIA symptoms who have a small area of tissue injury on the MRI compared to those who have a normal MRI.

High risk patients should be admitted to the hospital for close observation so that tPA can be administered rapidly if a stoke occurs, and to facilitate urgent completion of a full diagnostic evaluation. Low risk TIA patients can be managed safely and cost-effectively in an outpatient TIA clinic. A recent Stanford study, TWO ACES, documented that this novel TIA strategy resulted in extremely low stroke rates and high patient satisfaction.

The Stanford TIA program is currently evaluating a new technology, perfusion weighted imaging (PWI), as a method of confirming an ischemic “footprint” that can verify a transient neurological episode was caused by ischemia, rather than a non-ischemic condition. In a preliminary study, Stanford Stroke Center neurologists demonstrated that about 30% of patients with symptoms suggestive of a TIA have a positive PWI scan.

Stanford has collaborated with the National Stroke Association, the American Heart Association and the American College of Chest Physicians to produce several guideline statements aimed at refining the diagnosis and management of TIA:

National Stroke Association Recommendations for TIA

Guidelines for the Prevention of Stroke in Patients with TIA

Definition and Evaluation of Transient Ischemic Attack

Antithrombotic and thrombolytic therapy for ischemic stroke

The CTP to predict Response to recanalization in Ischemic Stroke Project (CRISP) is a study funded by the National Institutes of Health (NIH) to develop a practical tool to identify acute stroke patients who are likely to benefit from endovascular therapy. 

Stroke is the number one cause of disability in the United States.  It is caused by an occlusion of a blood vessel in the brain.  In order to reduce the burden of disability caused by stroke there is a need for better stroke treatments that are available to more stroke victims.  Endovascular stroke treatment is increasingly being used and may fill this need, as it can be very effective at opening up occluded blood vessels in the brain.  It is, however, not known which patients benefit clinically.

Previous studies suggest that stroke patients with a small volume of irreversible ischemic injury (infarct core) and a large volume of reversible ischemic injury (penumbra) are most likely to benefit from restoration of blood flow.  MRI shows promise for identification of the ischemic core and penumbra but it has very limited availability in US emergency rooms. Computed Tomography Perfusion (CTP) imaging is a potential solution as it is widely available and can easily be added to a non-contrast head CT, already routinely obtained to evaluate stroke patients in the emergency room.  However, methods for processing of CTP images and criteria for interpretation of the images are still immature.  Therefore, patient selection based on CTP images is not ready for implementation in clinical trials or clinical practice.

As part of the CRISP study we have developed a fully automated system (RAPID) for processing of CT Perfusion (CTP) images that generates brain maps of the ischemic core and penumbra.  We have defined criteria, based on these CTP maps, which we believe will predict if a patient is likely to benefit from restoration of blood flow. We are currently conducting a prospective cohort study at Stanford University and several collaborating hospitals across the USA to test if physicians in the emergency setting, with the aid of RAPID, can accurately predict if a patient will benefit from an endovascular revascularization procedure.

The successful execution of this research will provide physicians with an easy, automated method to select patients who are likely to benefit from restoration of blood flow. Such a method, thanks to easy accessibility to CT technology, would be of great value for patient selection in multi-center clinical acute stroke trials and, eventually, in routine clinical practice.

The looming healthcare financial crisis in America led to the creation of the Stanford Clinical Excellence Research Center (CERC). The mission of the program is: "better health, less spending." CERC brings together individuals with backgrounds in medicine, industrial engineering, and management and social sciences in an effort to create innovative healthcare delivery models that safely lower per capita spending while maintaining or improving health outcomes and patient experience. The goal is to bend the national trend of the ever increasing portion of our national GDP being spent on healthcare. Stanford Stoke Center faculty member Amy Tai is collaborating with CERC on a novel stroke/TIA heath care delivery project.

Stroke is the leading cause of disability and eighth most expensive health condition in America. Dr. Tai spent a year working on a redesigned care delivery process to address the gaps in the current system of stroke care in the United States.  An innovative stroke care model was developed using a systematic design method that included: (1) A literature review of major guidelines, stroke care delivery methods, and cost effectiveness studies to identify the best practices and how to achieve them at the lowest cost; (2) Site visits with those identified to be best at delivering high quality care for the lowest cost; (3) Observations of patients and providers to identify their unmet needs; (4) Development of a model to address stakeholders' unmet needs using methods from design thinking and health care delivery science; (5) Estimation of the model's cost-saving potential, using national averages for risk factor prevalence, stroke incidence, and costs of conditions and interventions.

Synthesis of this design process yielded a high-value stroke care model that: (1) prevents strokes through maximal use of preventative medications; (2) stratifies care for patients with TIA and mild stroke symptoms and; (3) delivers tPA in most time-efficient way possible and delivers a strengthened transition to community program for those at high risk for readmissions.  It is estimated that this model will significantly improve patient outcomes and reduce healthcare spending in cerebrovascular disease by 11%.