Neuroimaging advances in pediatric stroke

Introduction

When a child has stroke-like symptoms, critical issues must be addressed including confirmation of stroke rather than a stroke mimic, stroke etiology, pathophysiology of the injury, and evaluation of recovery potential. Over the next decade technological advances will enable improved monitoring and personalized therapy in multiple stages of cerebrovascular disease, including stroke prevention, acute stroke, and post-stroke rehabilitation. Neuroimaging will likely play a central role in each area, and the emphasis of new neuroimaging avenues will not only focus on mapping the topography of prior tissue damage, but expand to include molecular and hemodynamic biomarkers that may be used to triage patients for preventative therapies and optimize these therapies by quantifying their impact on tissue function and health.

Computed tomography (CT) perfusion, positron emission tomography (PET), and magnetic resonance imaging (MRI) are relevant for characterizing functional tissue profiles, however MRI is particularly appealing in pediatric patients due to the ability to sequentially visualize soft tissue, vasculature, and quantitative hemo-metabolic markers of ischemia without ionizing radiation. This review focuses on clinical questions that may be answered with advanced neuroimaging in children with suspected stroke and highlights how quantitative, non-invasive imaging may be used to expand the pediatric stroke imaging landscape.

Neuroimaging in acute stroke

When a child has stroke-like symptoms, emergent MRI of the brain is frequently used to confirm that a stroke, versus stroke mimic, has occurred. Unlike in adults where CT/CTA is often used for first-line imaging, in children additional radiation exposure, CTA requirement of a rapid properly-timed contrast injection through a small intravenous line, and prevalence of stroke mimics indistinguishable on CT make MRI the preferred first approach when rapidly available and feasible without sedation. Differential diagnosis for new focal neurological symptoms most commonly includes seizure with post-ictal paralysis or hemiplegic migraine and frequently cannot be clarified with head CT¹, and CT has been reported to miss as many as 47% of acute ischemic strokes later confirmed by MRI in children². Diffusion weighted imaging (DWI)-MRI, which can be performed in approximately 30s, quickly assesses whether an ischemic stroke has occurred. MRA is also common, however requires more time (4–6 min) relative to CTA (1–2 min). If ischemic stroke is confirmed, urgent clinical decisions must be made regarding off-label interventions for cerebral reperfusion while supportive care is provided.

To evaluate intervention appropriateness, imaging can be used to triage patients outside conventional intravenous thrombolysis windows (e.g., 4.5 hours) for recanalization therapies based on the extent of ischemic tissue that remains salvageable. Due to delays in presentation and need to exclude stroke mimics, few children are candidates for intravenous thrombolysis. For instance, in a recent study of 209 children with ischemic stroke, median interval from symptom onset to diagnosis was 22.7 hours (interquartile range=7.1–57.4 hours)³. However, as the time window for mechanical thrombectomy lengthens, children will more often be candidates for off-label reperfusion therapies. Accurately identifying viable tissue at risk for infarction is especially relevant in light of recent DAWN⁴ and DEFUSE-3⁵ results that showed that adult patients can benefit from thrombectomy 6–24 hours post-symptom onset. However in children, patient size and thrombus location are also considerations when using catheters designed for adults. Even if mechanical thrombectomy is feasible, reduced institutional preparedness, better outcomes without intervention (as children often have excellent arterial collaterals), and increased plasticity of the immature brain for recovery after injury collectively suggests that the risk-benefit ratio of clot extraction in children may differ from that in older adults.

The critical question pertains to discriminating tissue that has or will progress to infarction (core) from ischemic tissue that is at risk for infarction (penumbra) or is ischemic but will not progress to infarction (benign oligemia). The traditional strategy utilizes DWI and perfusion-weighted imaging, with the expectation that hyperintense regions on DWI (hypointense apparent diffusion constant, ADC) are consistent with cellular damage and infarct core, whereas tissue beyond this region but exhibiting hypoperfusion describes ischemic penumbra. However, DWI lesion subregions may not progress to infarction⁶ and hypoperfused regions on perfusion-weighted imaging can include core, penumbra, and benign oligemia in which tissue survives spontaneously without recanalization⁷. The natural history of penumbra evolution remains less well characterized in pediatric relative to adult patients.

Adult clinical imaging tools utilize perfusion quantification methods that require tracer-based approaches and exogenous contrast (e.g., CT perfusion or dynamic susceptibility contrast MRI), but these methods are less acceptable in children. Promising MRI tools are pH-weighted chemical exchange saturation transfer (CEST) and perfusion-weighted arterial spin labeling (ASL), which can be performed non-invasively in approximately five minutes each (Table 1). The premise underlying these methods is that as cerebral perfusion reduces below critical levels, selective neuronal loss begins (approximately 50% of baseline perfusion), tissue acidosis and reversible hemiparesis initiates (35% of baseline perfusion), and ultimately cell depolarization and irreversible damage results (20% of baseline perfusion). Thresholds for normal and abnormal hemodynamic parameters have been reviewed from adult studies^{8, 9}. While similar pediatric thresholds have not been established, developmental hemo-metabolic changes have been quantified in children (10 days – 16 years) using PET¹⁰ and have shown that CBF and the cerebral metabolic rate of oxygen (CMRO₂) increase from approximately 50–60% of adult values just after birth to 120–160% of adult values by age 8 years (depending on brain region); by comparison oxygen extraction fraction (OEF) is more closely aligned with adult values throughout this age range.

Table 1.

Emerging MRI methods

Method	Measurement	Relevance	Coverage (voxel dimensions)	Current limitations	Duration (minutes)
Arterial spin labeling	CBF	Acute stroke penumbra and tissue function	Regional quantitative CBF with 2D or 3D whole-brain coverage (3–5 mm)	Quantitative interpretation in patients with vasculopathy due to variable labeling and long blood arrival times	2–5
Amide proton transfer	pH / acidosis	Acute stroke penumbra	Regional qualitative molecular information provided with 3D whole-brain coverage (2–7 mm)	Long whole-brain scan times for acute stroke imaging. No consensus on quantification	4–12
Vessel wall imaging	Vessel wall, lumen thickness, enhancement	Vasculitis, arteriopathy, dissection, atherosclerosis	Major cervical and intracranial vessels. (0.6–0.8 mm)	CSF suppression; spatial resolution for intracranial vessel walls	4–11
Blood oxygenation level-dependent reactivity	CVR	Cerebrovascular reserve capacity / stroke risk	Regional qualitative information with 2D whole-brain coverage (2–4 mm)	Non-quantitative in most implementations	5–12
T2-relaxation-under-spin-tagging	OEF	Balance of oxygen consumed and delivered / stroke risk	Global quantitative OEF measurement (2–4 mm)	Global OEF measure only	1–2
Asymmetric spin echo	OEF	Regional balance of oxygen consumed and delivered	Regional quantitative OEF with whole-brain coverage (2–4 mm)	Complex post-processing required. No consensus on quantification	3–10
Edited spectroscopy	GABA	Excitation / inhibition imbalance: plasticity	Single voxel (20–40 mm)	Limited spatial coverage and macromolecule contamination	5–10

ASL is gaining attention as a non-invasive alternative to invasive perfusion imaging. ASL is analogous to contrast-enhanced bolus tracking methods, however arterial blood water is magnetically labeled using a single or combination of radiofrequency pulses (Figure 1). Following labeling, the arterial water flows into the brain, exchanges with tissue water at the capillary level, and attenuates the tissue water signal. By comparing this labeled image with an image where arterial blood water is not labeled, a perfusion-weighted image can be obtained and converted to CBF (units: ml blood/100g tissue/min) upon application of the flow-modified Bloch equation solution. Challenges pertain to shortening the scan time required to obtain full-brain coverage while not compromising quantitative accuracy and imaging in the presence of blood arrival times longer than 2s. ASL sequences can be performed in under five minutes at spatial resolutions of 2–4 mm isotropic with 3-dimensional readouts, array-receive coils, and standard body transmit coils. ASL acquisition¹¹ and analysis¹² procedures have been presented in adult and pediatric populations, including relevance for penumbra delineation.

Figure 1.

(A) In ASL-MRI, blood water is magnetically labeled using a single (pulsed or continuous) or series of (pseudo-continuous) radiofrequency pulses, after which the inverted blood water flows into the capillary exchange site of the imaging slice, exchanges with tissue water, and attenuates the extravascular water signal. By comparing this image with an image in which blood water is unlabeled, a perfusion-weighted map is obtained. (B) Time-of-flight MRA of a patient with moyamoya and left supraclinoid ICA occlusion, focal stenosis of distal right ICA, and bilateral MCA occlusion. (C) ASL-MRI before and after bilateral encephaloduroarteriosynangiosis (22 months post-left; 16 months post-right) shows improved CBF post-surgery.

To further differentiate benign oligemia from penumbra, knowledge of the metabolic environment of the tissue is useful. The penumbra may exhibit increased OEF and anaerobic glycolysis in the effort to maintain trans-membrane ion gradients. Anaerobic metabolism leads to tissue acidosis; thus acidosis may provide a sensitive indicator of impairment. Signal intensities of amide protons exchanging with water are influenced by pH through principles of exchange-related saturation; this effect can be assessed using proton NMR spectroscopy¹³. More recently, Amide Proton Transfer (APT) CEST MRI has been identified as a promising pH-sensitive MRI method (Figure 2)¹⁴. Chemical information is obtained through saturation transfer of magnetization between amide protons (primarily on the peptide backbone; in vivo concentration ~70–100 mM) and the imaged water protons (concentration ~110M). Importantly, over a broad physiological pH (pH>3), the exchange rate is base-catalyzed and therefore the APT effect-size on the water signal reduces with decreasing pH, or tissue acidosis.

Figure 2.

(A) APT-CEST MRI exploits the fact that amide protons, which resonance at +3.5 ppm from water, are in exchange with surrounding water protons. (B) The exchange rate is base-catalyzed over a physiological range, with more acidic environments yielding lower exchange. (C) When off-resonance pre-pulses are applied at the amide resonance prior to water excitation and detection, proton exchange will lead to an attenuation of the water signal. (D) An acute stroke patient with MCA occlusion subsequently treated with mechanical thrombectomy. There is an extensive penumbra and area with decreased APT effect on the acute MRI 2–4 hours post-symptom onset. The APT map by convention shows [1-Signal after saturation] (more acidic environment dark); a positive acute APT lesion is more similar to the FLAIR lesion on 30-day follow-up than the acute ADC or TTP map. More information in Titze et al.¹⁵

APT positive lesions may be more similar to the final infarct volume seen on 30-day follow-up MRI than the DWI-positive lesion or the time-to-peak (TTP) map (Figure 2). This is because as anaerobic metabolism during ischemia leads to tissue acidosis, APT permits insight into the energy status of ischemic parenchyma by indirect pH imaging. Sun et al. have demonstrated potential for this method by monitoring the evolution of pH changes following middle cerebral artery (MCA) occlusion in anaesthetized rats¹⁵ and human studies have now been performed in adults using this approach to better characterize ischemic penumbra^{16, 17}. Interpretation of APT changes in subacute phases remains complex however. Energy failure due to vessel occlusion, proteolysis, and inflammation are likely to increase the amount of mobile amides, opposing the change induced by lower pH. Moreover, lactate washout following partial reperfusion, buffering processes, or even small temperature variations can influence the signal. A shift from acidosis to alkalosis in subacute cerebral ischemia takes place, which makes consideration of lesion age even more important when evaluating APT changes¹⁸. APT CEST in acute stroke has recently been reviewed⁹.

Neuroimaging and stroke prevention

Reducing stroke-related morbidity will additionally depend on an improved understanding of early biomarkers that predict ischemic progression and can be used for prescient identification of patients requiring aggressive, preventative therapy prior to overt stroke. Children with cerebral arteriopathy are known to have high stroke recurrence risk¹⁹. One key to optimizing care and reducing overall morbidity is to identify children at highest risk of first or recurrent stroke and provide personalized prophylactic treatments. A large variation in infarct risk persists even in children undergoing identical risk factor management and with similar arterial steno-occlusion extent. The reason for this is because tissue-level hemodynamic compensation mechanisms vary substantially between individuals, and the manner in which (i) collateral flow networks develop, (ii) autoregulatory increases in cerebral blood volume (CBV) maintain CBF, and (iii) CBF adjusts to maintain normal ranges of OEF and CMRO₂ vary across individuals and reflect insufficient hemo-metabolic tissue-level compensation²⁰.

Dissection

Arterial dissection is a mural wall hematoma in a cervical or cerebral artery secondary to an intimal tear or to bleeding within the arterial wall. The result is often embolic or hemodynamic stroke or aneurysmal dilation of the artery. Dissections can be difficult to visualize with luminal imaging, though accurate diagnosis is important to help guide both acute and long-term management.

There is a growing interest to evaluate vessel walls using vessel wall imaging (VWI)-MRI. The strategy is to suppress the luminal blood and CSF signal, thereby allowing for visualization of what is in-between (i.e., vessel wall)^{21, 22}. Blood-water can be suppressed using a long turbo-spin-echo readout or with an inversion recovery to null the longitudinal component of the blood-water magnetization. CSF-water signal can be suppressed with similar principles of saturation/inversion recovery or with pre-readout pulse trains that suppress slow flowing spin isochromats^{23, 24}.

VWI can also be performed before and after paramagnetic contrast injection^{25, 26}. The hematoma is T₁ hyperintense, and the vessel wall generally enhances outside the hematoma following contrast injection. Other arterial pathologies such as vasculitis may also enhance (Figure 3). Enhancement may reflect inflammation, increased vascular supply, or endothelial permeability in symptomatic plaques, whereas non-enhancing wall thickening may be a general marker of microvascular disease and new lesion risk. Furthermore, just as multiple hemo-metabolic contrasts may provide a comprehensive profile of tissue-level compensation strategies, different vessel wall contrasts, such as enhancement extent, concentric thickening, and proton density or T₂-weighted contrast, may provide distinct information about disease status²¹.

Figure 3.

(A) Vessel wall contrast patterns. (B) Post-varicella transient cerebral arteriopathy shows acute infarct of the left internal capsule and narrowing of the terminal left ICA, MCA, and ACA on MRA; post-contrast VWI shows concentric wall enhancement (arrow; right panel). (C) Vertebral artery pseudoaneurysm. T₂-weighted imaging at presentation shows a chronic thalamic infarct (arrow). Catheter angiography shows luminal irregularity. Vessel wall imaging shows wall thickening and concentric enhancement (arrow; right panel). (D) Takayasu Arteritis. Post-ferumoxytol (iron-based intravascular contrast agent) angiography depicts the asymmetric smaller caliber of the left CCA (arrow), secondary to vessel wall thickening. Pre-contrast vessel wall imaging demonstrates circumferential wall thickening of the left CCA (arrow). Post-contrast imaging demonstrates enhancement (arrow) of the left CCA vessel wall indicating active inflammation. (B,C) images adapted from Dlamini et al²⁶.

Moyamoya

Moyamoya disease has unknown etiology and is often characterized by steno-occlusion of the supraclinoid internal carotid arteries, middle and anterior cerebral arteries, and development of collateral vessels. Patients with moyamoya are at high risk for stroke²⁷ and there is no consensus on optimal strategies for stratifying patients for medical management versus surgical revascularization.

The goal from a neuroimaging perspective is to identify parenchyma in which collateral pathways are inadequate to maintain sufficient oxygen delivery and as a result is at risk of infarction. The extent of intracranial vascular stenosis and the presence of collaterals have traditionally been used as surrogates for stroke risk. These macrovascular changes are measured with digital subtraction angiography (DSA), which is the gold standard for moyamoya staging. Angiography is complemented with structural diffusion-, T₂-, and T₁-weighted MRI to gauge acute infarcts, chronic infarcts, and tissue structure and atrophy, respectively. More sophisticated imaging capable of recording indicators of hemodynamic impairment and gauging the evolution of functional and compensatory changes are however needed.

Tissue-level CBF, now being increasingly measured in children using ASL-MRI (Figure 1), consistently demonstrates patterns of hypoperfusion in middle and anterior cerebral arterial flow territories, frequently with preserved posterior perfusion. ASL in moyamoya is complicated due to long blood arrival times, which can manifest as either signal voids or hyperintense signal, depending on imaging parameters. More information can be obtained by mapping reserve capacity, or the ability of arterioles to dilate in response to a cerebrovascular challenge. When CBF increases in the face of negligible or small changes in CMRO₂, the fraction of paramagnetic deoxy-hemoglobin relative to diamagnetic oxy-hemoglobin will decrease in and around capillaries and veins, thereby lengthening surrounding water T₂ and T₂* and increasing the MRI signal (i.e., the blood oxygen level-dependent, BOLD, effect). Therefore, T₂*-weighted sequences can be used for cerebrovascular reactivity (CVR) mapping when applied during isometabolic challenges that influence CO₂ and pH levels in blood and tissue: using pharmacologically-induced carbonic anhydrase inhibitors (e.g., acetazolamide) or respiratory stimuli (e.g., hypercapnia). CVR may be more prognostic than basal CBF, as it indicates parenchymal ability to respond to changes in perfusion pressure. This approach has been applied in children with moyamoya vasculopathy²⁸ and reviews of cerebrovascular reserve mapping are available^{29, 30}.

Sickle cell anemia

Sickle cell anemia (SCA) is a monogenetic disorder with a high prevalence of cerebral vasculopathy, silent cerebral infarct (SCI), and stroke^{31, 32}. A significant mechanism for stroke is hemodynamic imbalance with a reduced supply of CBF and oxygen. Transcranial Doppler ultrasound (TCD) is frequently used to assess flow velocity in children with SCA, and elevated MCA and distal ICA flow velocities provide surrogate markers for initial stroke risk. However, TCD provides no direct information on CBF, does not assess for SCI, and does not identify infarct recurrence risk.

Children with SCA typically have chronically increased CBF^33–35 in a manner that depends on the balance of oxygen carrying capacity and hemoglobin-S fraction, vasculopathy extent, and cerebrovascular reserve. Importantly, these factors influence CBF in opposite ways, with reduced oxygen carrying capacity leading to increases in CBF and cerebral vasculopathy lowering CBF. OEF, the ratio of oxygen consumed to that which is delivered, may have added discriminatory capacity for impairment relative to CBF: as CBF becomes insufficient to meet hemodynamic demands, there will be a gradient of increasing OEF for constant or reducing CMRO_2.

Venous blood oxygenation level can be quantified by evaluating water relaxation in and around large or small veins, from which CMRO₂ or OEF can be estimated with knowledge of arterial blood oxygenation and hematocrit. T₂-relaxation-under-spin-tagging (TRUST)-MRI³⁶ is a method for fast, global OEF quantification and can be combined with perfusion measurements to quantify CMRO_2. The TRUST principle is to quantify OEF by comparing the difference in the oxygenation level of blood entering the brain through the arteries to blood leaving the brain through the veins. Arterial blood oxygenation is estimated using pulse oximetry whereas venous blood oxygenation is quantified by isolating venous blood water signal using principles of spin labeling and performing T₂-weighting preparations with different effective echo times to enable absolute quantification of venous blood water T₂, which can then be converted to venous oxygenation by applying calibration models that relate blood T₂, hematocrit, and blood oxygenation. TRUST-MRI allows global OEF to be estimated quickly (approximately one minute) in a reproducible manner, however only provides a global estimate of OEF. Regional OEF estimates can be obtained by evaluating extravascular water relaxation regionally, commonly using an asymmetric spin echo approach³⁷. This approach has the advantage of providing spatial maps of blood oxygenation level and corresponding OEF, however requires more complex signal modeling, and must account for hematocrit differences between small and large vessels, macroscopic sources of field susceptibility, intravascular and extravascular effects which vary with field strength³⁸, and in some implementations knowledge of venous CBV and vessel orientation. Recent work has demonstrated that elevated OEF using asymmetric spin echo approaches co-localize with white matter lesions in children with SCA³⁹, and separate work using the TRUST method in adults with SCA has shown that elevated OEF corresponds with higher levels of clinical impairment⁴⁰ and reduces following blood transfusions, although the extent of post-transfusion OEF and CBF reduction may vary with disease chronicity and could differ in children versus adults^{41, 42}.

Neuroimaging and stroke recovery

Children often have better outcomes after stroke than adults. Neurological outcome in 484 children with ischemic stroke in the Canadian Pediatric Stroke Registry at the time of hospital discharge was normal in 30%; and in the remaining 70%, deficits were mild in 36%, moderate in 24% and severe in 10%⁴³. Developmental plasticity may aid recovery after stroke in children though in some cases younger age at the time of brain injury may result in poorer outcome⁴⁴. Imaging strategies to predict recovery and guide or target therapies are needed⁴⁵.

Neural excitability is mediated by glutamate (GLU) and γ-aminobutyric acid (GABA), the primary excitatory and inhibitory neurotransmitters in the adult brain, respectively. GABA-releasing synapses form prior to GLUergic synapses in development, and GABA becomes inhibitory by the delayed expression of a chloride exporter, which leads to a negative shift in the reversal potential of chloride ions^{46, 47}. During acute ischemic stroke, increases in extracellular GABA result in enhanced tonic inhibition and excitability thresholds⁴⁸. As inhibition persists through sub-acute and early chronic periods post-stroke, long-term potentiation and depression can be maladaptive, resulting in impaired neural plasticity and inadequate cognitive remapping. Total tissue GABA has been reported to decrease sub-acutely by almost two-fold following MCA occlusion in rats, a finding partly attributed to depression of tricarboxylic acid cycle and GLU elevation as fuel alternates⁴⁹, and suggests GABA is dynamic post-stroke. Inhibition can be altered through adjuvant modulation of excitability in both hemispheres: transcranial magnetic stimulation or transcranial direct-current stimulation may enhance motor recovery by up-regulating excitability in the lesioned hemisphere and down-regulating excitability in the contralesional hemisphere⁵⁰. Evidence has been provided for improved post-stroke recovery following pharmacological manipulation of inhibition or targeted physical therapy⁵¹.

Neuroimaging is a promising avenue for triaging patients for plasticity-inducing therapies post-stroke, although currently uncommon. Unfortunately, variation in excitation-inhibition and hemodynamic balance has been shown to give rise to similar BOLD responses with distinct underlying mechanisms⁵². Alternatively, multi-modal imaging and spectroscopy provides a more comprehensive perspective where investigations that assess only the BOLD signal fail to adequately characterize neural function. Using ¹H magnetic resonance spectroscopy (MRS) and spectral editing, multiple groups have demonstrated techniques for measuring regional tissue GABA and GLU levels^{53, 54}. In a recent trial of 17 adult ischemic stroke patients undergoing a two-week regimen of constraint-induced movement therapy 3–12 months post-stroke⁵¹, motor performance improvement was correlated with a change in GABA levels (p<0.01). Though under-developed, it is logical that neurochemical and hemo-metabolic imaging will hold potential for triaging patients for rehabilitative therapies based on functional peri-infarct tissue profiles.

An additional consideration is that of diaschisis (neurological deficits relating to lesions remote from original site of injury but connected by fiber tracts). This can be evaluated with the use of quantitative diffusion tensor imaging (DTI) and sensorimotor white matter tractography⁵⁵. Quantitative DTI metrics (i.e., fractional anisotropy and increased diffusivity) isolated from sensorimotor tractography correctly predicted neonatal stroke cases with poor motor outcome^{56, 57}. DTI metrics have also been used as imaging biomarkers to predict rehabilitation therapy-induced improvements in motor outcome⁵⁸. More sophisticated techniques based on multi-shell diffusion MRI acquisition such as the Neurite-Orientation-Dispersion-and-Density-Imaging (NODDI)⁵⁹ and the Spherical-Mean-Technique (SMT)⁶⁰ hold added potential by providing more specific white matter microstructural biomarkers than DTI.

Neuroimaging at ultra high field

The majority of MRI protocols utilize 1.5T-3.0T scanners, however FDA approval for clinical 7.0T imaging has recently been obtained, which could allow for new directions over the next decade. Signal-to-noise ratio increases linearly with magnetic field strength, and therefore 7.0T scans could be obtained theoretically in less than half the time as 3.0T scans with identical spatial resolution, or with added spatial resolution in similar times. This is appealing for non-sedated pediatric cases where anatomical structures are smaller and compliance problematic. Additional features such as increased chemical shift dispersion, reduction of intravascular relative to extravascular water T₂*, and prolonged blood water T₁ at 7.0T should improve CEST, BOLD, and ASL fidelity (Figure 4). In practice, these benefits will allow for higher spatial resolution to be performed in all methods, functional activity to be better spatially-localized in BOLD, and ASL to be performed for longer post-labeling delays. These advantages are offset by technical difficulties of 7.0T, including transmit fields required which cannot always be obtained within current safety constraints, static and transmit field heterogeneity, implant heating, limited coil coverage, and limited accessibility of 7.0T systems (currently less than 100 worldwide). 7.0T MRI clinical relevance has recently been reviewed⁶¹.

Figure 4.

(A) 7.0T vessel wall imaging shows intracranial vessel wall segments (white arrows) and a basilar artery lesion (black arrow). (B) T₁-weighted imaging acquired at 3.0T (1 mm) and 7.0T (0.7 mm) in the same volunteer for common scan-duration=5 minutes. (C) 7.0T ASL acquired at different post-labeling delays demonstrates potential for quantifying CBF at very long arterial arrival times (3000–4000ms), which are common in stroke. Images shown are from an adult.

Conclusion

Neuroimaging advances are beginning to be utilized to characterize complex metabolic and hemodynamic signatures of tissue health, which have cerebrovascular disease relevance for (i) predicting new or recurrent strokes, (ii) characterizing tissue viability in acute stroke, and (iii) characterizing recovery potential. Work over the next decade will likely utilize non-invasive neuroimaging advances to provide a more complete perspective on tissue health. This work is especially relevant in pediatric populations where exogenous contrast agents may be undesirable or contraindicated, yet long-term surveillance measurements to assess response to therapies and recovery potential may be required.