CME JOURNAL OF MAGNETIC RESONANCE IMAGING 00:00–00 (2014)

CME Article

Arterial Spin Labeling MRI: Clinical Applications in the Brain Nicholas A. Telischak, MD,1 John A. Detre, MD,2 and Greg Zaharchuk, MD, PhD1* This article is accredited as a journal-based CME activity. If you wish to receive credit for this activity, please refer to the website: www.wileyhealthlearning.com/jmri

Tim Leiner, MD, PhD, discloses speaker fees and grant funding from Phillips, and grant funding from Bracco and Bayer. Bonnie Joe, MD, PhD, has no relevant financial relationships to disclose.

ACCREDITATION AND DESIGNATION STATEMENT Blackwell Futura Media Services designates this journalbased CME activity for a maximum of 1 AMA PRA Category 1 CreditTM. Physicians should only claim credit commensurate with the extent of their participation in the activity.

Authors: Nicholas A. Telischak, MD, John A. Detre, MD, have no relevant financial relationships to disclose. Greg Zaharchuk, MD, PhD discloses research support from GE Healthcare.

Blackwell Futura Media Services is accredited by the Accreditation Council for Continuing Medical Education to provide continuing medical education for physicians.

This manuscript underwent peer review in line with the standards of editorial integrity and publication ethics maintained by Journal of Magnetic Resonance Imaging. The peer reviewers have no relevant financial relationships. The peer review process for Journal of Magnetic Resonance Imaging is double-blinded. As such, the identities of the reviewers are not disclosed in line with the standard accepted practices of medical journal peer review.

EDUCATIONAL OBJECTIVES Upon completion of this educational activity, participants will be better able to: 1. Identify the clinical use of arterial spin labeling for perfusion in the brain. 2. Recognize common patterns of abnormal cerebral perfusion and formulate a differential diagnosis.

Conflicts of interest have been identified and resolved in accordance with Blackwell Futura Media Services’ Policy on Activity Disclosure and Conflict of Interest. INSTRUCTIONS ON RECEIVING CREDIT For information on applicability and acceptance of CME credit for this activity, please consult your professional licensing board.

ACTIVITY DISCLOSURES No commercial support has been accepted related to the development or publication of this activity.

This activity is designed to be completed within an hour; physicians should claim only those credits that reflect the time actually spent in the activity. To successfully earn credit, participants must complete the activity during the valid credit period.

Faculty Disclosures: Editor-in-Chief: Mark E. Schweitzer, MD, discloses DSMB consulting fees from Paradigm Spine and HistoGeneX, and consulting fees from MMI.

Follow these steps to earn credit:

CME Editor: Scott B. Reeder, MD, PhD discloses research support to his institution from GE Healthcare and Bracco Diagnostics; and personal stock in Novelos, Inc, and NeuWave, Inc.

• Log on to www.wileyhealthlearning.com

CME Committee: Shreyas Vasanawala, MD, PhD, discloses research collaboration with GE Healthcare and co-founder equity in Morpheus Medical.

• Read the article in print or online format.

• Read the target audience, educational objectives, and activity disclosures. • Reflect on the article. • Access the CME Exam, and choose the best answer to each question.

Frank Korosec, PhD, discloses research support to his institution from GE Healthcare.

• Complete the required evaluation component of the activity. This activity will be available for CME credit for twelve months following its publication date. At that time, it will be reviewed and potentially updated and extended for an additional period.

Scott K. Nagle, MD, PhD, discloses personal stock in GE Healthcare. Mustafa R. Bashir, MD, discloses research support and consulting fees from Siemens Healthcare and Bayer Healthcare.

1

Department of Radiology, Stanford University Medical Center, Stanford, California, USA. Department of Neurology and Radiology, University of Pennsylvania, Philadelphia, Pennsylvania, USA. *Address reprint requests to: G.Z., Stanford University 1201 Welch Rd., Mailcode 5488 Stanford, CA 94305-5488 E-mail: [email protected] Received May 6, 2014; Accepted August 5, 2014. DOI 10.1002/jmri.24751 View this article online at wileyonlinelibrary.com. 2

C 2014 Wiley Periodicals, Inc. V

1

2

Telischak et al.

Visualization of cerebral blood flow (CBF) has become an important part of neuroimaging for a wide range of diseases. Arterial spin labeling (ASL) perfusion magnetic resonance imaging (MRI) sequences are increasingly being used to provide MR-based CBF quantification without the need for contrast administration, and can be obtained in conjunction with a structural MRI study. ASL MRI is useful for evaluating cerebrovascular disease including arterio-occlusive disease, vascular shunts, for assessing primary and secondary malignancy, and as a biomarker for neuronal metabolism in other disorders such as seizures and neurodegeneration. In this review we briefly outline the various ASL techniques including advantages and disadvantages of each, methodology for clinical interpretation, and clinical applications with specific examples. Key Words: Arterial spin labeling; cerebral blood flow; MRI J. Magn. Reson. Imaging 2014;00:000–000. C 2014 Wiley Periodicals, Inc. V

ated with decreased CBF include stroke, vasospasm, hydrocephalus, tumor treatment change, and neurodegeneration. Some disorders, such as seizure and migraine, can variably show increased or decreased CBF depending on the phase of the disease (eg, interictal vs. ictal). Arterial spin labeling (ASL) is a completely noninvasive magnetic resonance imaging (MRI) method that uses magnetically labeled blood water as a flow tracer, providing CBF images of the brain. Moreover, if certain conditions are met it can potentially also provide an absolute, quantifiable CBF measurement on a voxel-by-voxel basis (2). Knowledge of the perfusion characteristics of brain and brain lesions may provide a specific diagnosis, or key physiological information critical to patient management that may be occult or difficult to discern on anatomic imaging. We outline the state-of-the-art in the clinical application of ASL in brain imaging and provide specific examples of the value of the information it provides (3–10). ASL TECHNIQUES

INTRODUCTION THE BRAIN is a unique organ, housed in a rigid calvarium and dependent on high rates of blood flow (average 50 mL/100 g/min). Its blood supply is uniquely susceptible to changes in pressure and thus highly regulated. The relationship between changes in cerebral blood volume and intracranial pressure was described by Monro and Kellie two centuries ago (1). Many disorders of the brain are associated with alterations in cerebral blood flow (CBF). Disorders associated with increased CBF include aggressive tumors, vascular shunts, and hypercapnea. Disorders associ-

From the first description of ASL, the technique has evolved and there are currently many varieties of ASL pulse sequences, which we briefly summarize. In general terms, CBF refers to perfusion per unit of tissue, and is optimally measured with a diffusible tracer that can exchange between the blood and the brain. The first measurements of CBF in humans were made by measuring arteriovenous differences in nitrous oxide (11). CBF imaging has been carried out successfully using radioactive tracers in conjunction with single-photon emission computed tomography (SPECT), positron emission tomography (PET), and more recently with stable xenon-

Table 1 Pros and Cons of ASL Compared With DSC Perfusion Imaging Quantitative

Availability

Susceptibility effects

ASL

No, IV access is unnecessary

Contrast

Infinitely

Repeatable

Yes, values in mL/100g/min

Yes, but generally not yet standard on commercial systems

DSC

Yes, patient must have an adequate IV for bolus injection

Limited by maximum contrast dose

Semiquantitative

Can be performed on any MR scanner, though dedicated post-processing software is required

Relatively robust to adjacent susceptibility (eg, blood products), due to spin echo acquisition Distortion and signal dropout due to gradient echo acquisition

Table 2 Summary of Causes of Decreased, Increased, and Mixed Disorders of Cerebral Blood Flow. Disorders of decreased cerebral blood flow

Disorders of increased cerebral blood flow

Acute ischemic stroke/TIA Moyamoya disease Acute hydrocephalus Sturge-Weber

Tumor (primary brain tumors, some metastases) Cerebritis (autoimmune, infectious, etc.) Vascular shunts (dAVF, AVM, CCF) Hypercapnea

Neurodegenerative diseases

Loss of autoregulation (e.g. hypoxic-ischemic injury,) Sickle cell disease

Disorders of mixed cerebral blood flow Seizure Migraine Hypoxic ischemic injury Posterior reversible encephalopathy syndrome (PRES)

Arterial Spin Labeling MRI

3

Figure 1. An 80-year-old female with left hemiparesis. Diffusion-weighted MRI shows acute right middle cerebral artery (MCA) territory infarct (a). ASL perfusion MRI shows right MCA territory hypoperfusion, without arterial transit abnormality, suggesting poor collateral flow (b). For reference, the Tmax map from bolus perfusion imaging is concordant with ASL findings (c).

enhanced CT with proven interreliability of each of these methods. All of these techniques entail some amount of ionizing radiation and each has significant drawbacks such that none of them can be applied routinely to or repeated frequently in a large number of patients. In ASL, the diffusible tracer is a magnetic label applied to blood water molecules, produced by saturating or inverting the longitudinal component of the MR signal. If all of the label arrives at the capillary bed or tissue at the time of imaging, this results in a T1-weighted signal reduction proportional to CBF, called the tagged image, which is compared with a control image, in which the blood water molecules are not perturbed. There are, therefore, intrinsic limita-

Figure 2. A 22-month-old boy with tuberculous meningitis and severe vasospasm of the posterior circulation resulting in infarct as shown on axial FLAIR MRI (a) and DWI (b). Upon resolution of vasospasm there is marked luxury perfusion on ASL MRI (c).

tions on signal-to-noise ratio (SNR) per unit time, as well as the potential for systematic measurement errors which include: 1) prolonged transit delay between the tagging region and the imaging slice, resulting in intravascular tagged blood that has not yet perfused the tissue of interest; 2) the dependence of measured relaxation on exchange of water between the intra- and extravascular compartments; and 3) clearance of unextracted water by outflow in the venous system. In some cases, these "errors" even yield helpful clues to diagnosis, such as in regions served by collateral flow and in arteriovenous shunt lesions. However, for applications that demand CBF quantification, several strategies have been applied to overcome these inherent limitations, which can be

Figure 3. A 41-year-old female with transient right-sided weakness. Axial FLAIR MRI at 1.5T shows signal abnormality in the left centrum semiovale, likely reflecting sequelae of borderzone ischemia (a). Axial DWI shows no acute infarct (b). ASL difference map shows high sulcal signal, called arterial transit abnormality, suggesting delayed arrival of tagged blood to the territory in question (c). Maximum intensity projection (MIP) image from source 3D time-of-flight (TOF) MRA shows severe proximal left MCA stenosis (d), which was confirmed on the follow-up catheter digital subtraction angiogram (e). The patient was treated with medical management for intracranial atherosclerosis.

Figure 4. A 6-year-old girl with neurofibromatosis-1 and headaches. FLAIR MRI images at 3T show subtle loss of MCA flow void with presence of collaterals (arrow, a), as well as high signal in cortical sulci suggestive of slow flow within collaterals (arrow, b). There is no acute infarct (c). ASL perfusion MRI shows normal posterior circulation perfusion (arrow, d), with marked decreased perfusion to the bilateral anterior circulation (e). These findings prompted an MRA evaluation which confirmed bilateral supraclinoid ICA occlusion consistent with Moyamoya syndrome (f).

Arterial Spin Labeling MRI

Figure 5. Axial FLAIR images at 3T through the cerebellum (a) and basal ganglia (b) show typical unidentified bright objects (UBOs), which are commonly seen in NF-1. ASL difference image at the time of the FLAIR imaging (c), and in 1-year followup (d) show interval decrease in perfusion to the right occipital lobe, a progression typical of NF-1 vasculopathy in this patient without Moyamoya syndrome.

5

conveniently separated into labeling and imaging methods. Continuous ASL (CASL) is a labeling method that uses off-resonance radiofrequency (RF) power and gradients to adiabatically invert flowing water in a proximal tagging plane in the neck to maximize SNR, but requires hardware that is no longer available on modern MRI equipment and results in off-resonance effects that complicate volumetric imaging (12). Pulsed ASL (PASL), in which short (5–20 msec) RF pulses are used to saturate a proximal slab of tissue (including the blood water), is more compatible with modern MRI equipment and leads to high inversion efficiency and lower RF power deposition, but generally has lower SNR than CASL methods (13). The shorter temporal width of the PASL bolus is an additional source of systematic error in CBF quantification. Pseudocontinuous ASL (pCASL) has more recently been introduced as a method to achieve high tagging efficiency, uniting the need for a longer temporal tagged water bolus with lower RF energy deposition and compatibility with modern scanners (14). pCASL has become the labeling method of choice for clinical ASL, based on a recent consensus white paper on this topic (15). Finally, velocity selective ASL (VSASL) selectively saturates flowing spins with no spatial selectivity (16); for instance, a typical cutoff velocity of 2 cm/sec will dephase spins in arterioles that are 100 microns or more in diameter and thus results, in principle, in labeling much closer to the

Figure 6. A 6-year-old female with diffuse intrapontine glioma initially treated with chemo-irradiation who presents with nausea and vomiting. Baseline (a) ASL perfusion and (b) T2-weighted imaging prior to symptoms shows normal brain perfusion without ventricular enlargement. When acutely symptomatic with hydrocephalus, ASL perfusion shows decreased CBF especially in the anterior (arrow) and posterior (arrowhead) borderzones (c); on T2-weighted images (d) there is new ventricular enlargement and periventricular edema. After shunting, the patient’s symptoms improved, and a follow-up ASL perfusion image shows improved CBF, even though the ventricles have not yet returned to their baseline size and there is persistent periventricular signal abnormality.

6

Figure 7. A 51-year-old male with dementia. FSPGR MRI at 1.5T (a) shows cortical volume loss affecting the biparietal regions. ASL CBF map shows decreased perfusion to the bilateral parietal lobes (b), which are hypometabolic on simultaneous FDG PET-MRI (c), findings most consistent with Alzheimer’s disease.

site of capillary exchange. VS-ASL, however, suffers from lower SNR than pCASL, and is also sensitive to other flowing fluids, such as cerebrospinal fluid. We generally use a 1.5-second label and 2-second delay for routine ASL; however, certain diagnoses such as Moyamoya syndrome require longer labeling and a longer delay. In the evaluation of these patients we perform long label (3 sec) and long delay (3 sec) ASL imaging to allow tagged spins additional time to reach the capillary bed for a more accurate CBF estimation by minimizing arterial transit abnormality (ATA). Similarly, in sickle cell disease the use of a long postlabel delay (2.1 sec) gives a more accurate estimation of CBF (17). One topic worth discussing and covered in the consensus white paper is the use of crusher gradients. Crusher gradients can be applied to eliminate coherently flowing ASL signal that is within arteries with the advantage being that image quality is improved and ATA is suppressed. A disadvantage of using a crusher gradient is that signal is also attenuated more generally, reducing SNR. We further find that the presence of ATA is a useful sign that adds diagnostic value when correctly recognized and interpreted, so we do not use crusher gradients in any of

Telischak et al.

our ASL protocols (15). If scanning time is not limited, an alternative strategy would be to acquire ASL both with and without crusher gradients. Because the measured ASL effect is small, imaging sequences with high SNR are required for readout. Originally, echo-planar imaging (EPI) was used, as it offers high SNR and fast acquisition time, which also helps to reduce motion artifact between label and control images. However, this introduces artifact in regions of high susceptibility (such as the skull base or near metallic hardware), and introduces a variable delay between imaging slices. Alternative 3D sequences have thus been developed such that it is now possible to acquire the whole brain volume with either a multi- or single-shot excitation, and with reduced susceptibility weighting. This further allows for efficient background suppression of static tissue, which greatly improves temporal SNR, as well as reduced susceptibility artifacts, and has become the preferred readout method for clinical ASL (15). ASL also benefits greatly from higher field strength owing to the dual benefits of both increased sensitivity and T1 prolongation, which allows for a greater accumulation of tagged blood water. All of the ASL examples shown in this article (unless otherwise indicated), were acquired in 4 minutes at 3T using pseudocontinuous labeling with backgroundsuppressed 3D fast spin echo readout (18–20). Table 1 summarizes the distinct advantages and disadvantages of ASL compared with dynamic susceptibility contrast (DSC) (also known as bolus perfusionweighted imaging [PWI]). ASL provides quantitative values without the use of contrast, so it can be performed repeatedly (eg, before and after vasodilators such as Diamox) without concern for cumulative contrast dosing. With respect to disadvantages, DSC is a robust technique that is in widespread use, while ASL is only just starting to be used outside of major research centers. DSC reliably provides both CBF and time-to-maximum of the residue function (Tmax) values; ASL typically only allows the measurement of CBF, but newer ASL techniques (with longer acquisition times) may enable similar transit time measurements, although these are not routinely performed. Note that ASL cannot be performed once gadoliniumbased contrast (a T1-shortening agent) has been administered, as all signal will decay before reaching brain tissue, a distinct disadvantage of this technique.

ASL INTERPRETATION ASL is a quantitative technique and has been a useful tool in the study of normal brain function as well as neuropsychiatric disorders (for review, see Detre et al (20)). Although ASL MRI is potentially a quantitative method, accurate CBF quantification depends on a model that includes additional assumptions and parameters. In routine radiological practice we do not use any threshold interpretation of quantitative values and find that simple ASL difference images, reflecting the difference between the tag and control images or a map of where the labeled spins are at the time of

Arterial Spin Labeling MRI

7

Figure 8. A 53-year-old female with headache and visual disturbance, found to have a hemangiopericytoma. Contrastenhanced T1-weighted images at 3T show a large right temporal extraaxial mass extending into the infratemporal fossa (a,d). T2-weighted (b) and FLAIR (c) MR images show isointensity to gray matter and internal flow voids with minimal surrounding edema. ASL difference image (e) shows increased tumor perfusion, confirmed on the bolus perfusion-weighted CBV map (f). [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

imaging, can provide sufficient qualitative information to aid in diagnosis. For instance, in the setting of stroke, studies have shown good correlation between ASL and bolus PWI Tmax lesions on a qualitative basis or by comparing using standard deviation analysis to the contralateral normal hemisphere (21,22). Such images can direct the radiologist to an abnormality not otherwise seen or difficult to identify prospectively on routine MR sequences or narrow the differential diagnosis of a known abnormality. In children it is important to bear in mind normal changes in cerebral perfusion over time. A large review of healthy children has not been published to date, and is much needed, as the available literature on normal blood flow in development shows mixed results. It is believed that CBF is low at birth and rapidly increases until 4 years of age, where it peaks until the early teenage years, at which point CBF falls to a plateau of normal adult range by the early 20s. Even among adults, "normal" CBF values vary widely; the reasons for this variation are not known (23–26). It is convenient to categorize radiological ASL MRI abnormalities into disorders of decreased CBF, increased CBF, and mixed increased and decreased CBF (Table 2). We will use this as a framework to show specific examples of how ASL imaging in the brain informs daily clinical practice.

DISORDERS OF DECREASED CBF Perfusion imaging in acute ischemic stroke aids in several aspects of diagnosis and treatment. It can confirm the presence of hypoperfusion in the acute setting and also aid in the distinction between regions of irreversibly damaged tissue ("core") from still viable tissue at risk of infarction (Fig. 1). It furthermore can clearly depict tissue reperfusion, either spontaneous or after therapy, often with better conspicuity than the more widely used method of bolus perfusionweighted imaging (PWI) with dynamic susceptibility contrast (DSC) (Fig. 2) (22,27). The most sensitive markers for hypoperfused tissue with DSC are the time-to-peak contrast (TTP) and Tmax, both of which are markers of blood arrival time. While not widely used yet, it is possible to quantify arrival time with ASL by using sequences that image at multiple postlabeling delay times. This latter method, dubbed multidelay ASL, may also provide better measurements of quantitative CBF. However, even routine ASL can provide useful information about delayed blood flow; ATA is often seen in regions of delayed perfusion, likely indicating the presence of collateral blood supply (27). Perhaps because of this, it has been shown to be a marker portending good prognosis in acute stroke (28). When ASL is compared with DSC in acute

8

Telischak et al.

Figure 9. A 59-year-old male with renal cell cancer. Sagittal T1 (a), axial T2-weighted (b), and axial T1-weighted postcontrast MRI (c) show a mixed intensity enhancing right orbital mass (arrows). ASL difference image (d) confirms high blood flow consistent with renal cell carcinoma metastasis.

stroke, ASL abnormality has been shown to correlate well with the Tmax >6/sec abnormality and is prognostic for identifying salvageable tissue and the chance of good clinical recovery (21,22,29). Originally defined as neurologic symptoms lasting less than 24 hours, the American Heart Association and American Stroke Association now endorse the tissue-based definition of transient ischemic attack (TIA) as brain, spinal cord, or retinal ischemia without acute infarct, as defined by DWI imaging (30). In the setting of TIA, altered blood flow to the affected hemisphere is detected by ASL imaging 50% of the time, one of the most common abnormalities being the bilateral loss of ASL signal in the vascular borderzone regions, although focal perfusion abnormalities can also be seen (31,32). The washout hypothesis of Caplan and Hennerici (33), which correlates worse observed outcomes in patients with a vulnerable plaque predisposing to embolic events combined with decreased cerebral perfusion, suggests that small infarcts in the borderzone regions are related to an inability of the vasculature to wash out small emboli.

Another finding on ASL in patients with TIA is the already-mentioned arterial transit artifact, perhaps related to persistent mild delays in the affected region (Fig. 3). This was also identified in patients with TIA and minor stroke using ASL methods that map transit time (34). Moyamoya disease is a chronic and progressive steno-occlusive arteriopathy that affects the internal carotid artery (ICA) terminus and the proximal branches of the circle of Willis, and may be spontaneous or associated with conditions such as neurofibromatosis I (NF1), the latter being dubbed Moyamoya syndrome. Perfusion imaging is commonly performed in Moyamoya disease to determine the severity of disease and sometimes also to evaluate cerebrovascular reserve for treatment planning and monitoring of patients under consideration for bypass surgery. Cerebrovascular reserve studies are routinely performed for preoperative evaluation in our institution, replacing the more difficult stable xenon CT examination. These examinations are specifically designed to evaluate cerebrovascular reserve, and we include both

Arterial Spin Labeling MRI

9

Figure 10. A 66-year-old male with history of treated esthesioneuroblastoma. Axial FLAIR (a), T2-weighted (b), and postcontrast T1-weighted (c) MRI show clear tumor recurrence involving the right orbit, greater wing of the sphenoid, and dura of the anterior right temporal horn (arrows). More difficult to appreciate on anatomic images are the dural metastases at the left temporal horn (arrowheads), clearly seen on ASL (d).

long label, long delay, and multidelay ASL before and after acetazolamide administration. Decreased CBF, detectable by ASL, may be the first sign of developing arteriopathy, before stenosis is evident by angiography (Fig. 4). Yeom et al (35) have shown decreased CBF in a cohort of NF1 patients without stenoocclusion or Moyamoya disease, particularly in the posterior circulation and borderzones, which they propose might be due to vasculopathy in cerebral microvasculature or alteration in cerebral metabolic demand (Fig. 5). Increased intracranial pressure may be caused by cerebrospinal fluid (CSF) obstruction, mass lesion, or dysautoregulation, and the measurement of CBF may be very informative in differentiating hydrocephalus from ventricular enlargement. Yeom et al (36) correlated decreased ASL perfusion in children presenting with ventricular enlargement with elevated CSF pressure measurements and clinical symptoms, successfully differentiating children with hydrocephalus from those with "benign" ventricular enlargement (Fig. 6).

While 18F-flurodeoxyglucose (FDG) PET is the most widely studied marker of metabolism in the setting of neurodegenerative disease, many studies have shown that CBF changes parallel similar changes in glucose metabolism and therefore offers similar information with respect to local synaptic activity (37). In Alzheimer’s disease (AD), studies have shown decreased perfusion in the posterior cingulate, precuneus, inferior parietal, and lateral prefrontal cortices (38–46), distinct from patterns seen in frontotemporal lobar dementia (FTLD) (44). Additionally, perfusion changes have been shown to precede the structural changes in AD, and can be incorporated along with volumetric and fractional anisotropy measurements to predict development of disease (Fig. 7) (47,48). DISORDERS OF INCREASED CBF Meningioma is the most common extracerebral neoplasm in the brain, and while CBF is variable, often shows "light bulb bright" high blood flow with ASL imaging. This can be valuable for differentiating

10

Telischak et al.

Figure 11. A 57-year-old female with witnessed seizure. Axial FLAIR shows cortical high signal and thickening (arrow, a). Axial postcontrast T1-weighted MRI shows nodular enhancement (arrow, b). No vascular anomaly is seen on MRA (c). ASL difference image shows focal increased perfusion (d) which is not as easily appreciated on the CBV map obtained with bolus perfusion-weighted imaging (e). This was resected and proved to be glioblastoma multiforme.

meningioma from other lesions such as nerve sheath tumors, which do not show the same robust blood flow. Additionally, ASL increases reader sensitivity in our experience for detecting one or multiple meningiomas, especially useful when gadolinium contrast is not administered. Kimura et al (49) have shown that ASL and bolus perfusion techniques are comparable in evaluating meningiomas and that CBF corresponds to vascularity as assessed at histopathological analysis. Subtypes of meningioma can also be differentiated with the angiomatous subtype showing the highest perfusion (Fig. 8). Metastases show variable blood flow, mostly lower than that of gray matter, such that the majority cannot be easily differentiated from brain parenchyma by ASL perfusion imaging. A notable exception is renal cell carcinoma metastasis, which is known to be highly vascular and is conspicuous on ASL perfusion sequences (Fig. 9). Another unusual tumor type with high CBF is esthesioneuroblastoma, which can be localized to the anterior cranial fossa and nasal cavity, or more widely metastasize along the dural surfaces. This very vascular tumor often shows more conspicuous change on ASL sequences than it does on postcontrast T1-weighted images (Fig. 10). Gliomas are a pathologically varied set of primary brain tumors in which MRI plays an important role in grading for treatment planning purposes. Tissue is

obtained in the majority of these tumors, but even with multiple biopsies the highest grade portion of tumor might not be sampled, thus undergrading the tumor. Tumor angiogenesis is a key factor in histologic tumor grading, and tumor blood flow serves as an in vivo marker of the tumor characteristic that can be measured with ASL (50,51). There is considerable literature on the ability of bolus PWI to discern lowfrom high-grade tumors; Warmuth et al (52) showed very good agreement between ASL and PWI for quantification of blood flow in brain tumors. De novo as well as recurrent glioblastoma multiforme will typically demonstrate markedly increased CBF, consistent with the higher metabolism of the tumor tissue (Fig. 11). Thus, ASL can play a role in initial tumor diagnosis as well as in follow-up after treatment, aiding differentiation of treatment-induced changes such as radionecrosis from tumor recurrence or progression, with the latter showing neovascularization and increased CBF. Weber et al (53) found ASL to predict tumor response as early as 6 weeks following treatment, with improved accuracy over tumor volume measurements. One distinct advantage of ASL in the posttreatment setting is that spin echo imaging can be performed, which helps in the detection of small regions of recurrence near the operative bed, where it is often challenging to evaluate due to distortions on typical bolus PWI images.

Arterial Spin Labeling MRI

11

Figure 12. An 18-year-old male with pre-B-cell acute lymphoblastic leukemia in remission status post-bone marrow transplant presenting with fever and mental status change. Axial FLAIR MRI (a) shows cortical high signal and expansion in the bilateral mesial temporal lobes, which show markedly increased CBF on the ASL difference image (b). There was no diffusion abnormality to suggest infarct (c) or enhancement on postcontrast T1-weighted MRI (d). Findings are consistent with infectious etiology. Acute human herpes virus type 6 (HHV-6) infection was proven by CSF analysis.

Cerebritis and other inflammatory disorders are often associated with increased CBF, which can be useful in evaluating DWI abnormalities, especially in children, in whom DWI abnormalities commonly occur outside of the setting of infarct, thereby allowing a more confident radiologic diagnosis rather than a differential (Fig. 12). No noninvasive technique has ever been as sensitive as ASL for the detection of vascular shunt lesions. Because magnetically labeled water decays with T1 and is short relative to capillary transit time, the ASL signal is rarely seen within intracerebral veins in the absence of arteriovenous shunt. Exceptions to the rule are settings in which overall CBF is very high (such as in children or in the setting of luxury reperfusion in stroke [Fig. 13]), due to the presence of a significant amount of water protons that do not have time to exchange with

the tissue water at the capillary bed. Venous ASL signal is seen most conspicuously in the setting of arteriovenous malformations, where it often gives a good outline of the venous drainage of the AVM. Venous ASL signal is also seen in developmental venous anomalies (the import of which is still poorly understood) and in dural arteriovenous fistulae (dAVF) (54). This finding has been shown to improve the detection of dAVFs and small AVMs (Fig. 14) (55). As dAVFs may recur following treatment, ASL is also a very sensitive noninvasive follow-up tool to detect early recurrence. As such, it might hold promise to improve the triage of patients to more invasive diagnostic tests, such as conventional cerebral angiography. Sickle cell disease without Moyamoya syndrome elevates resting CBF. Additionally, perfusion asymmetries are observed in these children, which are of

12

Telischak et al.

Figure 13. A 72-year-old woman 3 days following right MCA territory stroke with spontaneous reperfusion. ASL shows high signal in the region of the right sphenoparietal sinus (arrows), representing shunted blood due to luxury reperfusion (a). DWI images demonstrate the extent of the stroke (b).

Figure 14. A 36-year-old female with recent concussion and new pulsatile tinnitus. Axial T2-weighted MRI (a) and the remainder of the anatomical sequences are normal (not shown). ASL difference image shows high signal in a tentorial vein. This finding prompted catheter angiography, and lateral and frontal projections of a left external carotid artery injection show a dural arteriovenous fistula (arrow, c) with early draining tentorial vein (arrow, d).

Arterial Spin Labeling MRI

Figure 15. A 90-year-old male with altered mental status. Anatomic imaging showed no acute infarct (a,b). ASL CBF map (color scale, mL/100g/min) shows markedly increased CBF. Arterial blood gas taken upon presentation showed pCO2 of 58.2 mmHg (reference range 35–45 mmHg).

uncertain significance, but may correlate with risk of infarct (17). Hypercapnea results in cerebral vasodilatation when cerebral autoregulation pathways are intact at least in part via activation of K-ATP channels in vascular smooth muscle, with CBF increase in the range of 6% per mmHg pCO2 rise (56). Patients with acute hypercapnia may present with headache, slurred speech, confusion, hallucination, stupor, or coma prompting imaging evaluation. As ASL provides routine quantification of cerebral blood flow, the radiologist can now suggest the diagnosis of hypercapnia and suggest correlation with an arterial blood gas (Fig. 15). Cerebrovascular autoregulation is the intrinsic tendency to maintain CBF despite changes in perfusion pressure. Loss of autoregulation may occur for a number of reasons, including diffuse hypoxicischemic injury, and will result in striking increases in cerebral perfusion (57).

MIXED INCREASED AND DECREASED CBF CBF in the setting of seizure is complex: CBF is elevated during ictus and suppressed in the interictal

Figure 16. A 59-year-old male with seizure. Ictal MRI shows mildly increased FLAIR MRI signal intensity in the left mesial temporal lobe (a) and diffusion restriction (b). ASL CBF map shows increased perfusion to the left temporal lobe (c). CSF PCR confirmed HSV-1 encephalitis.

13

Figure 17. A 12-year-old girl with a seizure disorder. Interictal PET (a) shows decreased FDG uptake in the right temporal lobe (arrow). MRI ASL CBF map (b) shows decreased blood flow to the same right temporal region (arrow).

state (58). Ictus results in focally or diffusely increased CBF, often limited to a single hemisphere and not respecting vascular territory boundaries (Fig. 16). This may not be suspected clinically, and particularly in critically ill patients the characteristic ASL findings may be the first clue that the patient is in status epilepticus, prompting EEG evaluation. During the ictal period, a band-like region of increased CBF may be seen along the cortex and dural surface, which may reflect either cortical activity that does not extend into the sulci or be due to dural inflammation in the setting of the seizure. Corresponding diffusion restriction may also be seen in this setting (59,60). Reduced CBF may also be seen in a seizure focus when the patient is not seizing, corresponding to interictal hypometabolism that can be seen on FDG PET (Fig. 17). The vascular theory of migraine proposes that migraine aura is caused by vasoconstriction and subsequent headache is associated with vasodilatation and hyperperfusion (61). While migraine is incompletely understood, the perfusion changes are no

14

Telischak et al.

neuroradiology, providing novel information in a wide range of disease states. REFERENCES

Figure 18. An 8-year-old female with history of bone marrow transplant who presents with hypertensive emergency, headache, and seizure. FLAIR (a) MRI shows typical subcortical pattern of edema seen in PRES (arrow). ASL difference map (b) shows increased perfusion to the occipital and parietal lobes (arrow). It is difficult in this setting to know whether perfusion abnormalities stem primarily from PRES or from associated seizures.

longer thought to be the inciting event but rather reflect a biphasic neural phenomenon such as cortical spreading depression. More severe perfusion deficits may be associated with hemiplegic migraines, in which the headache is accompanied by sensorimotor changes (62). ASL represents an ideal sequence to study transient changes in brain perfusion. Pollock et al (63) described three patients found in a retrospective review to have migraine in whom focally increased perfusion was captured during the patients symptoms with ASL imaging. Migraine can also be a mimic of stroke, and in fact strokes are more common in patients with migraine with aura, demonstrating the underlying vascular nature of the two diseases (64). In a patient with concern for stroke and negative structural and DWI imaging it is of great value to provide support for a diagnosis of migraine, such that patients may be spared the risk of stroke treatments such as tissue plasminogen activator (tPA). Posterior reversible encephalopathy syndrome (PRES), or hypertensive encephalopathy, is incompletely understood but is hypothesized to be the result of autoregulatory failure in the setting of hypertension. Both hyperperfusion and hypoperfusion have been demonstrated, depending on the stage of the disease that is imaged (65,66). The pattern of cerebral perfusion may even be mixed in different cortical regions of the same patient (Fig. 18). SUMMARY ASL perfusion imaging provides quantification of CBF and can be performed routinely and repeatedly without contrast administration or ionizing radiation. We have categorized ASL abnormalities into decreased CBF, increased CBF, and mixed CBF disease states and demonstrated specific examples that illustrate common diseases in which ASL can aid in the diagnosis. As ASL sequences improve and with increasing availability of higher field strength scanners, we predict that this already powerful technique will become a standard sequence in

1. Macintyre I. A hotbed of medical innovation: George Kellie (17701829), his colleagues at Leith and the Monro-Kellie doctrine. J Med Biogr 2013 [Epub ahead of print]. 2. Detre JA, Leigh JS, Williams DS, Koretsky AP. Perfusion imaging. Magn Reson Med 1992;23:37–45. 3. Brown GG, Clark C, Liu TT. Measurement of cerebral perfusion with arterial spin labeling. Part 2. Applications. J Int Neuropsychol Soc 2007;13:526–538. 4. Deibler AR, Pollock JM, Kraft RA, Tan H, Burdette JH, Maldjian JA. Arterial spin-labeling in routine clinical practice. Part 1: Technique and artifacts. AJNR Am J Neuroradiol 2008;29:1228– 1234. 5. Deibler AR, Pollock JM, Kraft RA, Tan H, Burdette JH, Maldjian JA. Arterial spin-labeling in routine clinical practice. Part 2: Hypoperfusion patterns. AJNR Am J Neuroradiol 2008;29:1235– 1241. 6. Deibler AR, Pollock JM, Kraft RA, Tan H, Burdette JH, Maldjian JA. Arterial spin-labeling in routine clinical practice. Part 3: Hyperperfusion patterns. AJNR Am J Neuroradiol 2008;29:1428– 1435. 7. Liu TT, Brown GG. Measurement of cerebral perfusion with arterial spin labeling. Part 1. Methods. J Int Neuropsychol Soc 2007; 13:517–525. 8. Tourdias T, Rodrigo S, Oppenheim C, et al. Pulsed arterial spin labeling applications in brain tumors: practical review. J Neuroradiol 2008;35:79–89. 9. Hendrikse J, Petersen ET, Golay X. Vascular disorders: insights from arterial spin labeling. Neuroimaging Clin N Am 22:259–269, x–xi. 10. Wolf RL, Detre JA. Clinical neuroimaging using arterial spinlabeled perfusion magnetic resonance imaging. Neurotherapeutics 2007;4:346–359. 11. Kety SS. The determination of cerebral blood flow in man by the use of nitrous oxide in low concentrations. Am J Physiol 1945; 143:53–66. 12. Alsop DC, Detre JA. Multisection cerebral blood flow MR imaging with continuous arterial spin labeling. Radiology 1998;208:410– 416. 13. Wong EC, Buxton RB, Frank LR. A theoretical and experimental comparison of continuous and pulsed arterial spin labeling techniques for quantitative perfusion imaging. Magn Reson Med 1998;40:348–355. 14. Dai W, Garcia D, de Bazelaire C, Alsop DC. Continuous flowdriven inversion for arterial spin labeling using pulsed radio frequency and gradient fields. Magn Reson Med 2008;60:1488– 1497. 15. Alsop DCD, Golay X, Gunther M, et al. Recommended implementation of arterial spin labeled perfusion MRI for clinical applications: a consensus of the ISMRM Perfusion Study Group and the European Consortium for ASL in Dementia. Magn Reson Med 2014 [Epub ahead of print]. 16. Wong EC, Cronin M, Wu WC, Inglis B, Frank LR, Liu TT. Velocityselective arterial spin labeling. Magn Reson Med 2006;55:1334– 1341. 17. Gevers S, Nederveen AJ, Fijnvandraat K, et al. Arterial spin labeling measurement of cerebral perfusion in children with sickle cell disease. J Magn Reson Imaging 2012;35:779–787. 18. Zaharchuk G. Theoretical basis of hemodynamic MR imaging techniques to measure cerebral blood volume, cerebral blood flow, and permeability. AJNR Am J Neuroradiol 2007;28:1850– 1858. 19. Detre JA, Alsop DC. Perfusion magnetic resonance imaging with continuous arterial spin labeling: methods and clinical applications in the central nervous system. Eur J Radiol 1999;30:115– 124. 20. Detre JA, Rao H, Wang DJ, Chen YF, Wang Z. Applications of arterial spin labeled MRI in the brain. J Magn Reson Imaging 2012;35:1026–1037. 21. Zaharchuk G, El Mogy IS, Fischbein NJ, Albers GW. Comparison of arterial spin labeling and bolus perfusion-weighted

Arterial Spin Labeling MRI

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

37.

38.

39.

40.

imaging for detecting mismatch in acute stroke. Stroke 2012;43: 1843–1848. Bivard A, Stanwell P, Levi C, Parsons M. Arterial spin labeling identifies tissue salvage and good clinical recovery after acute ischemic stroke. J Neuroimaging 2013;23:391–396. Biagi L, Abbruzzese A, Bianchi MC, Alsop DC, Del Guerra A, Tosetti M. Age dependence of cerebral perfusion assessed by magnetic resonance continuous arterial spin labeling. J Magn Reson Imaging 2007;25:696–702. Chiron C, Raynaud C, Maziere B, et al. Changes in regional cerebral blood flow during brain maturation in children and adolescents. J Nucl Med 1992;33:696–703. Wintermark M, Lepori D, Cotting J, et al. Brain perfusion in children: evolution with age assessed by quantitative perfusion computed tomography. Pediatrics 2004;113:1642–1652. Hales PW, Kawadler JM, Aylett SE, Kirkham FJ, Clark CA. Arterial spin labeling characterization of cerebral perfusion during normal maturation from late childhood into adulthood: normal "reference range" values and their use in clinical studies. J Cereb Blood Flow Metab 2014;34:776–784. Wang DJ, Alger JR, Qiao JX, et al. The value of arterial spinlabeled perfusion imaging in acute ischemic stroke: comparison with dynamic susceptibility contrast-enhanced MRI. Stroke 2012; 43:1018–1024. Chalela JA, Alsop DC, Gonzalez-Atavales JB, Maldjian JA, Kasner SE, Detre JA. Magnetic resonance perfusion imaging in acute ischemic stroke using continuous arterial spin labeling. Stroke 2000;31:680–687. Bivard A, Krishnamurthy V, Stanwell P, et al. Arterial spin labeling versus bolus-tracking perfusion in hyperacute stroke. Stroke 2014;45:127–133. Easton JD, Saver JL, Albers GW, et al. Definition and evaluation of transient ischemic attack: a scientific statement for healthcare professionals from the American Heart Association/American Stroke Association Stroke Council; Council on Cardiovascular Surgery and Anesthesia; Council on Cardiovascular Radiology and Intervention; Council on Cardiovascular Nursing; and the Interdisciplinary Council on Peripheral Vascular Disease. The American Academy of Neurology affirms the value of this statement as an educational tool for neurologists. Stroke 2009;40: 2276–2293. Zaharchuk G, Bammer R, Straka M, et al. Arterial spin-label imaging in patients with normal bolus perfusion-weighted MR imaging findings: pilot identification of the borderzone sign. Radiology 2009;252:797–807. Zaharchuk G, Olivot JM, Fischbein NJ, et al. Arterial spin labeling imaging findings in transient ischemic attack patients: comparison with diffusion- and bolus perfusion-weighted imaging. Cerebrovasc Dis 2012;34:221–228. Caplan LR, Hennerici M. Impaired clearance of emboli (washout) is an important link between hypoperfusion, embolism, and ischemic stroke. Arch Neurol 1998;55:1475–1482. MacIntosh BJ, Lindsay AC, Kylintireas I, et al. Multiple inflow pulsed arterial spin-labeling reveals delays in the arterial arrival time in minor stroke and transient ischemic attack. AJNR Am J Neuroradiol 2010;31:1892–1894. Yeom KW, Lober RM, Barnes PD, Campen CJ. Reduced cerebral arterial spin-labeled perfusion in children with neurofibromatosis type 1. AJNR Am J Neuroradiol 2012;34:1823–8. Yeom KW, Lober RM, Alexander A, Cheshier SH, Edwards MS. Hydrocephalus decreases arterial spin–labeled cerebral perfusion. AJNR Am J Neuroradiol 2014 [Epub ahead of print]. Wolk DA, Detre JA. Arterial spin labeling MRI: an emerging biomarker for Alzheimer’s disease and other neurodegenerative conditions. Curr Opin Neurol 2012;25:421–428. Alsop DC, Casement M, de Bazelaire C, Fong T, Press DZ. Hippocampal hyperperfusion in Alzheimer’s disease. Neuroimage 2008; 42:1267–1274. Alsop DC, Dai W, Grossman M, Detre JA. Arterial spin labeling blood flow MRI: its role in the early characterization of Alzheimer’s disease. J Alzheimers Dis 2010;20:871–880. Alsop DC, Detre JA, Grossman M. Assessment of cerebral blood flow in Alzheimer’s disease by spin-labeled magnetic resonance imaging. Ann Neurol 2000;47:93–100.

15 41. Chen Y, Wolk DA, Reddin JS, et al. Voxel-level comparison of arterial spin-labeled perfusion MRI and FDG-PET in Alzheimer disease. Neurology 2011;77:1977–1985. 42. Dai W, Lopez OL, Carmichael OT, Becker JT, Kuller LH, Gach HM. Mild cognitive impairment and Alzheimer disease: patterns of altered cerebral blood flow at MR imaging. Radiology 2009; 250:856–866. 43. Fleisher AS, Podraza KM, Bangen KJ, et al. Cerebral perfusion and oxygenation differences in Alzheimer’s disease risk. Neurobiol Aging 2009;30:1737–1748. 44. Hu WT, Wang Z, Lee VM, Trojanowski JQ, Detre JA, Grossman M. Distinct cerebral perfusion patterns in FTLD and AD. Neurology 2010;75:881–888. 45. Johnson NA, Jahng GH, Weiner MW, et al. Pattern of cerebral hypoperfusion in Alzheimer disease and mild cognitive impairment measured with arterial spin-labeling MR imaging: initial experience. Radiology 2005;234:851–859. 46. Yoshiura T, Hiwatashi A, Yamashita K, et al. Simultaneous measurement of arterial transit time, arterial blood volume, and cerebral blood flow using arterial spin-labeling in patients with Alzheimer disease. AJNR Am J Neuroradiol 2009;30:1388– 1393. 47. Chao LL, Buckley ST, Kornak J, et al. ASL perfusion MRI predicts cognitive decline and conversion from MCI to dementia. Alzheimer Dis Assoc Disord 2010;24:19–27. 48. Zhang Y, Schuff N, Ching C, et al. Joint assessment of structural, perfusion, and diffusion MRI in Alzheimer’s disease and frontotemporal dementia. Int J Alzheimer’s Dis 2011;2011: 546871. 49. Kimura H, Takeuchi H, Koshimoto Y, et al. Perfusion imaging of meningioma by using continuous arterial spin-labeling: comparison with dynamic susceptibility-weighted contrast-enhanced MR images and histopathologic features. AJNR Am J Neuroradiol 2006;27:85–93. 50. Silva AC, Kim SG, Garwood M. Imaging blood flow in brain tumors using arterial spin labeling. Magn Reson Med 2000;44: 169–173. 51. Wolf RL, Wang J, Wang S, et al. Grading of CNS neoplasms using continuous arterial spin labeled perfusion MR imaging at 3 Tesla. J Magn Reson Imaging 2005;22:475–482. 52. Warmuth C, Gunther M, Zimmer C. Quantification of blood flow in brain tumors: comparison of arterial spin labeling and dynamic susceptibility-weighted contrast-enhanced MR imaging. Radiology 2003;228:523–532. 53. Weber MA, Thilmann C, Lichy MP, et al. Assessment of irradiated brain metastases by means of arterial spin-labeling and dynamic susceptibility-weighted contrast-enhanced perfusion MRI: initial results. Invest Radiol 2004;39:277–287. 54. Iv M, Fischbein NJ, Zaharchuk G. Association of developmental venous anomalies with perfusion abnormalities on arterial spin labeling and bolus perfusion–weighted imaging. J Neuroimaging 2014 [Epub ahead of print]. 55. Le TT, Fischbein NJ, Andre JB, Wijman C, Rosenberg J, Zaharchuk G. Identification of venous signal on arterial spin labeling improves diagnosis of dural arteriovenous fistulas and small arteriovenous malformations. AJNR Am J Neuroradiol 2011;33:61–68. 56. Pollock JM, Deibler AR, Whitlow CT, et al. Hypercapnia-induced cerebral hyperperfusion: an underrecognized clinical entity. AJNR Am J Neuroradiol 2009;30:378–385. 57. Pollock JM, Whitlow CT, Deibler AR, et al. Anoxic injuryassociated cerebral hyperperfusion identified with arterial spinlabeled MR imaging. AJNR Am J Neuroradiol 2008;29:1302– 1307. 58. Wu RH, Bruening R, Noachtar S, et al. MR measurement of regional relative cerebral blood volume in epilepsy. J Magn Reson Imaging 1999;9:435–440. 59. Unrath A, Muller HP, Ludolph AC, Kassubek J. Reversible cortical diffusion restriction, hyperperfusion and T2-hyperintensity caused by two different types of epileptic seizure. Clin Neuroradiol 2012;22:239–243. 60. de Crespigny A, Rother J, van Bruggen N, Beaulieu C, Moseley ME. Magnetic resonance imaging assessment of cerebral hemodynamics during spreading depression in rats. J Cereb Blood Flow Metab 1998;18:1008–1017.

16 61. Friberg L, Olesen J, Lassen NA, Olsen TS, Karle A. Cerebral oxygen extraction, oxygen consumption, and regional cerebral blood flow during the aura phase of migraine. Stroke 1994;25:974– 979. 62. Hsu DA, Stafstrom CE, Rowley HA, Kiff JE, Dulli DA. Hemiplegic migraine: hyperperfusion and abortive therapy with intravenous verapamil. Brain Dev 2008;30:86–90. 63. Pollock JM, Deibler AR, Burdette JH, et al. Migraine associated cerebral hyperperfusion with arterial spin-labeled MR imaging. AJNR Am J Neuroradiol 2008;29:1494–1497.

Telischak et al. 64. Kurth T, Slomke MA, Kase CS, et al. Migraine, headache, and the risk of stroke in women: a prospective study. Neurology 2005;64: 1020–1026. 65. Bartynski WS, Boardman JF. Catheter angiography, MR angiography, and MR perfusion in posterior reversible encephalopathy syndrome. AJNR Am J Neuroradiol 2008;29:447–455. 66. Schwartz R, Mulkern R, Vajapeyam S, Kacher DF. Catheter angiography, MR angiography, and MR perfusion in posterior reversible encephalopathy syndrome. AJNR Am J Neuroradiol 2009;30: E19; author reply E20.