· Relevant clinical questions for stroke prevention, acute stroke management, and post-stroke surveillance differ in scope and should be considered when developing and implementing new imaging methods.

· Clinical trial outcomes may differ substantially if treatments are titrated based on more sensitive biomarkers of stroke risk or treatment relevance, rather than simple randomization or tissue structure.

· Routine clinical stroke imaging relies primarily on characterizing the location and extent of steno-occlusion on angiography, and also determining the impact of this steno-occlusion on tissue structure (e.g., infarct location and size). However, understanding how parenchyma compensates for progressive reductions in cerebral perfusion pressure may provide new information that could guide personalized management of patients.

Neurologists, neurosurgeons, neuroradiologists, and imaging physicists interested in expanding the stroke imaging infrastructure by incorporating novel functional measures of parenchymal health

· To understand the current, relevant clinical questions in the setting of stroke prevention, acute stroke management, and post-stroke surveillance. To achieve this, recent clinical trial outcomes in these areas will be highlighted in the context of unmet clinical needs.

· To understand novel imaging methods that can be used to complement existing lumenography and structural imaging, the sources of these contrasts, and in what setting they may be most appropriate to implement.

· To understand existing clinical trials that use these methods, the outcomes, and to be able to summarize the existing gaps in our knowledge that must be met prior to routine clinical implementation.

The overall goal of this presentation is to provide a summary of the major unmet clinical needs in stroke imaging and management from a physicist’s perspective. Stroke imaging can broadly be considered in terms of (i) characterizing hemodynamic compensation mechanisms with the goal of stratifying treatments to prevent stroke, (ii) identifying viable tissue at risk for infarction in the setting of acute stroke, and (iii) evaluating chronic, post-stroke hemodynamic and neurochemical processes that may portend functional recovery.

While several stroke prevention trials have been seminal to guiding treatment decisions and reducing stroke incidence (1-3), many prevention trials have had discouraging results, or results that suggest minimal added benefit of new therapies over current standard of care. One reason for this may be due to suboptimal patient stratification. Although imaging is considered routinely for the clinical management of patients, it is less frequently incorporated into research trials as a prognostic indicator or outcome measure owing to cost and differences in signal quality of new methods between sites and scanner vendors. However, trial outcomes may differ substantially if treatments are titrated based on more sensitive biomarkers of stroke risk or treatment relevance, rather than randomization. To achieve this, more sensitive, cost-effective imaging approaches that can be performed routinely are required.

The physicist’s perspective to solving this fundamental problem can largely be thought of as expanding the diagnostic imaging infrastructure to more thoroughly characterize the functional sequelae that underlie stroke.

Imaging methods for characterizing tissue health in the setting of ischemic cerebrovascular disease will be summarized according to their relevance in the following categories of clinical stroke management.

i. Stroke prevention

Stroke is the leading cause of adult disability and the third leading cause of death in the United States, affecting more than 700,000 individuals annually (4). Despite progress in stroke treatment, 20-30% of strokes result in death within one month, and more than 70% result in significant long-term disability (4-6). Reducing stroke-related morbidity ultimately requires an improved understanding of early markers that identify hemodynamic impairment which can be used for prescient identification of patients requiring aggressive, preventative therapy (7, 8). New imaging methods that can be applied to determine personalized stroke risk stratification algorithms are required to better personalize treatment regimens, such as surgical revascularization or aggressive medical management. Such methods include:

· High spatial resolution vessel wall imaging for determination of plaque vulnerability

· Vessel-encoded arterial spin labeling (VE-ASL) for noninvasive cerebral blood flow (CBF) and collateralization determination

· Blood oxygenation level dependent (BOLD) cerebrovascular reactivity timing and magnitude imaging for quantifying vascular reserve capacity

· Oxygen extraction fraction (OEF) mapping for evaluating the balance of oxygen consumption and oxygen delivery

ii. Acute stroke

Acute stroke imaging is potentially the most challenging area for evaluation of new methods due to the time-sensitivity of treatments and a general inability to evaluate and optimize new methodologies in this population. While acute stroke therapy has improved (9-12), limited access to therapy within the required treatment window (i.e., 4.5 hrs intravenous and up to 12 hours endovascular) frequently results in irreversible damage. The major focus in these patients is identification of tissue that is at risk for infarction and to stratify patients for acute therapies. While CT imaging remains the standard in most hospitals for making such decisions, alternative MRI methods will be discussed as well, including:

· Diffusion weighted imaging for determining infarct core

· Gadolinium perfusion and arterial spin labeling for determining perfusion abnormalities in tissue at-risk for infarction

· Chemical exchange saturation transfer for determining acidosis and infarction risk

iii. Post-stroke plasticity

Improved management of cerebrovascular disease has reduced stroke-related mortality (13), however many stroke survivors remain impaired with nearly 33% institutionalized after stroke (14-16) and fewer than 25% able to perform pre-stroke equivalent levels of physical activity six months post-stroke (17). The development and evaluation of novel rehabilitation strategies would be accelerated with an improved understanding of cortical reorganization, or cerebral plasticity. Specifically, spared cortical tissue has increased potential for cerebral plasticity, yet effective neurorehabilitative treatments that promote plasticity remain underdeveloped. Routine implementation of such treatments requires an improved understanding of how regional neurochemical, hemodynamic, and metabolic changes relate to functional recovery and adjust in response to therapy. Here, the existing methods for determining post-stroke recovery such as vessel patency and residual infarct volume will be extended to cover:

· Magnetic resonance spectroscopic imaging for metabolite determination

· J-edited spectroscopy for g-aminobutyric acid (GABA) determination

· The relationship between these and hemodynamic factors for monitoring functional reorganization and/or susceptibility to pharmacological or electromagnetic plasticity-inducing therapies

The contrast origins of the above methods, required time, remaining methodological concerns, and applications in prospective trials will be discussed. Specific examples in the context of symptomatic intracranial stenosis, sickle cell anemia, acute ischemic stroke, and functional recovery in patients with chronic middle cerebral artery infarcts will be presented. Example images of standard clinical magnetic resonance angiography and imaging (white) and a subset of these more novel methods (yellow) are shown in Figure 1.The major strengths and remaining limitations of the newer imaging methods will be summarized, with a subset of confirmatory and conflicting perspectives from separate studies presented.As preventative, acute, and chronic stroke therapies continue to improve, there will be a growing need to stratify patients for personalized treatment regimens. While existing angiography and structural imaging methods have established utility in these areas, more novel functional methods will likely be required as well to record quantifiable observables of tissue health. This talk will summarize these methods from the technical perspective.