Stroke is the leading cause of adult disability and the second leading cause of death in developed countries.1 Despite progress in stroke treatment, 20–30 % of strokes result in death within one month and 70–80 % result in significant long-term disability.2–4 Improved preventative and acute management of cerebrovascular disease has reduced stroke-related mortality;5 however, many stroke survivors remain impaired with nearly 33 % institutionalised after stroke.6–9
Neuroimaging is a critical component of patient care in all stages of cerebrovascular disease: from identification of stroke risk factors to stratifying acute and chronic stroke patients to the most effective revascularisation and rehabilitation treatments. Evaluation of management decisions and post-stroke therapy strategies would be accelerated with an improved understanding of the complex interplay between vascular, neurochemical and tissue-level haemodynamic impairment.
Characterisation of soft-tissue structure using fluid attenuated inversion recovery (FLAIR) and T1-weighted magnetic resonance imaging (MRI) is central to most standard MRI protocols and extensive work has documented the relevance of these techniques.10–12 In more specialised protocols, diffusion weighted imaging (DWI) or diffusion tensor imaging (DTI) can be applied to characterise membrane integrity and fibre tract directionality, respectively,13 and neurochemical tissue signatures can be assessed using principles of magnetic resonance spectroscopy.14 These relatively well-known approaches have been implemented in patients with cerebrovascular disease for more than two decades. However, newer approaches are continuously being developed that hold potential for further expanding our understanding of the functional and structural sequelae of tissue changes during and following ischaemia.
Here, we review the state of novel, non-invasive MRI approaches capable of complementing the existing stroke imaging infrastructure for more comprehensive evaluation of cerebrovascular disease. As each method is in a different stage of development, with varying timelines for routine clinical implementation, the purpose of this article is to review these emerging methods in the context of their strengths and limitations for evaluating cerebrovascular disease and to provide references relevant for further evaluation of their existing impact on stroke imaging (see Table 1).
Magnetic Resonance Imaging
MRI has been routinely used clinically for more than 20 years, with a primary purpose for visualising structural soft-tissue contrast. The main signal source in MRI is water and the contrast derives from how the magnetic properties (longitudinal and transverse relaxation times: T1 and T2, respectively) of water protons in different environments adjust in the presence of a static magnetic field (B0; generally 1.5–3.0 Tesla) following perturbation by additional radiofrequency (B1) and gradient (G) fields. Water protons in different environments have unique relaxation properties and therefore can exhibit exquisite contrast between different healthy tissue types and between healthy and diseased tissue. Importantly and unlike contrast derived from positron emission tomography (PET), single-photon emission computed tomography (SPECT) or computed tomography (CT), different contrasts in MRI can be generated simply by adjusting scan parameters (e.g., B1 and G). This accelerates innovation and clinical implementation in MRI, as testing or approval that may be required from new exogenous tracers is irrelevant. However, a valid concern with MRI is that our ability to validate new techniques may be outpaced by diverging technical innovations. When this occurs, the clinical feasibility of new MR approaches and their true prognostic potential may be unclear. Here, we present an overview of several cerebrovascular imaging methods that have tested clinical potential and which deserve interest for implementation into large-scale clinical trials.
Blood Oxygenation Level-dependent Cerebrovascular Reactivity
Blood oxygenation level-dependent (BOLD) MR neuroimaging with predominately T2*-weighted contrast15,16 and less commonly T2-weighted contrast,17 has been widely applied to map brain activity18–21 non-invasively, transforming the field of cognitive neuroscience. BOLD contrast arises owing to the mismatch in cerebral blood flow (CBF), cerebral blood volume (CBV) and cerebral metabolic rate of oxygen (CMRO2) increases secondary to both vascular and neuronal stimulation in healthy parenchyma.22 Specifically, CBF increases by generally 20–100 % (depending on task), whereas CMRO2 increases only marginally by 0–20 %.23,24 This leads to an increase in oxyhaemoglobin (HbO2) relative to deoxyhaemoglogin (Hb) in capillaries and veins. As HbO2 is diamagnetic whereas Hb is paramagnetic, the relative reduction of fractional Hb in the vasculature will increase surrounding water T2 and T2*, thereby increasing MR signal. Additional dephasing arises from CBV increases in vasculature containing Hb, which will reduce the surrounding water signal.25,26