Vascular disease

Handbook of Clinical Neurology, Vol. 136 (3rd series) Neuroimaging, Part II J.C. Masdeu and R.G. Gonza´lez, Editors © 2016 Elsevier B.V. All rights reserved

Chapter 60

Vascular disease CATHERINE AMLIE-LEFOND1* AND DENNIS SHAW2 Department of Neurology, Seattle Children’s Hospital, Seattle, WA, USA

1

2

Department of Radiology, Seattle Children’s Hospital, Seattle, WA, USA

Abstract The child presenting with possible sentinel transient ischemic event or stroke requires prompt diagnosis so that strategies to limit injury and prevent recurrent stroke can be instituted. Cerebral arteriopathy is a potent risk factor for arterial ischemic stroke in childhood. Though acute imaging study in the setting of possible stroke is often a head computed tomography, when possible magnetic resonance imaging (MRI) is recommended as the first-line study as confirmation and imaging evaluation of ischemic stroke will typically require MRI. The MRI scanning approach should include diffusion-weighted imaging (DWI) early in the sequence order, since normal DWI excludes acute infarct with rare exception. In most cases, arterial imaging with time-of-flight (TOF) magnetic resonance angiography (MRA) is warranted. Dedicated MRA may not be possible in the acute setting, but should be pursued as promptly as possible, particularly in the child with findings and history suggestive of arteriopathy, given the high risk of recurrent stroke in these children. MRA can overestimate the degree of arterial compromise due to complex/ turbulent flow, and be insensitive to subtle vessel irregularity due to resolution and complex flow. In cases with high imaging suspicion for dissection despite normal MRA findings, catheter angiogram is indicated. A thoughtful, stepwise approach to arterial neuroimaging is critical to optimize diagnosis, treatment, and primary and secondary prevention of childhood stroke.

INTRODUCTION Population-based studies have reported an annual incidence of childhood arterial ischemic stroke (AIS) ranging from 2.3 to 13 per 100 000 children (Schoenberg et al., 1978; Lynch et al., 2002; Fullerton et al., 2003). Ten percent of children die following AIS, and over 75% will suffer long-term neurologic deficits (De Schryver et al., 2000; deVeber et al., 2000; Ganesan et al., 2000; Fullerton et al., 2002). Despite the burden of AIS in childhood, timely diagnosis remains a challenge in large part due to the wide differential diagnostic considerations in children relative to adults (Shellhaas et al., 2006). Beyond the ischemic stroke of arterial etiology focus of this chapter, unappreciated trauma and venous etiologies are the primary stroke differential considerations of AIS. In addition, underlying metabolic/mitochondrial

diseases causing stroke-like events are more frequently seen in children then adults. Furthermore, stroke mimics such as seizure are common in children presenting with possible acute stroke (Shellhaas et al., 2006), hence confirmation of stroke as well as exclusion of other disorders is necessary to direct the evaluation and treatment of presumed AIS in children. In adults, acute stroke will be the most likely cause of sudden onset of a focal neurologic deficit. In contrast, with children a minority of cases with acute onset of focal weakness will be due to acute cerebral ischemia, and the differential, including intracranial hemorrhage, seizure, migraine headache, infection, psychogenic diagnosis, demyelinating disease, amongst other disorders, will be of significant consideration (Shellhaas et al., 2006; Mackay et al., 2014).

*Correspondence to: Catherine Amlie-Lefond, MD, Seattle Children’s Hospital, M/S MB 7.420, 4800 Sand Point Way NE, Seattle WA 98105, USA. Tel: +1-206-987-2078, Fax: +1-206-987-2649, E-mail: [email protected]

1160 C. AMLIE-LEFOND AND D. SHAW Of significant clinical relevance, following an by the existence of potential urgent interventions, such initial AIS, one-fifth of children will have another stroke, as thrombolysis in AIS, neurointervention for thrombosometimes in the immediate poststroke period (Fullerton sis or unruptured aneurysm, and cerebral edema with et al., 2007). Hence urgent evaluation for risk factors for resultant elevated intracranial pressure. In addition, recurrent stroke is critical. In particular, cerebral arterioidentification of underlying etiologies may inform pathy is a potent risk factor for initial and recurrent stroke treatment or its secondary prevention. AIS (Adams et al., 1997; Chabrier et al., 2000; Dobson The paradigms for urgent neuroimaging in adults et al., 2002; Lanthier et al., 2005; Danchaivijitr et al., presenting with possible acute stroke are well developed, 2006; Fullerton et al., 2007; Amlie-Lefond et al., 2009; and predicated on the availability of hyperacute thromDarteyre et al., 2012), and cerebral arteriopathy itself bolytic or endovascular therapy (Jauch et al., 2013). is a risk factor for poor outcome following childhood Newer algorithms employing more advanced imaging AIS (Goldenberg et al., 2013). Many children have a seek to identify patients most likely to benefit from sentinel transient ischemic event prior to an ultimately these therapies. At the first level this imaging evaluation larger completed stroke, particularly when associated includes confirming the absence of intracranial hemorwith arteriopathy, hence recognition of this earliest event rhage before any thrombolytic or endovascular intervenis vitally important in these children, as urgent stroke tion. In adults, head computed tomography (CT), which prevention may be possible. Sources of emboli, including can be combined with CT angiography and CT perfucardiac and emboli from spontaneous or posttraumatic sion, can elucidate the presence of intracranial hemorcraniocervical vessel injury, also need to be considered rhage and is most often used in adults presenting with in the setting of AIS. clinical features of acute stroke. Although head CT While considerable research related to neuroimagand head magnetic resonance imaging (MRI) will both ing and acute stroke has been conducted in adults, evaluate for acute hemorrhage in children, commonly comparatively little has been performed in children. the need for further information, including the considerNevertheless, increasing recognition of the importance ation of a wider differential and a greater desire to limit of cerebral arteriopathy in adults with cerebrovascular radiation exposure, more often than not leads to the disease has resulted in neurovascular imaging advances preference for MRI. The relative rarity of stroke in that are proving useful when applied to the investigation childhood, the more common need for use of MRI of children. In particular, vessel wall imaging techniques in the workup of stroke, and the variable ability to coopfor evaluation of arteriopathy have particular relevance erate with imaging, however, present challenges in to children, where cerebral arteriopathy is a potent risk providing timely imaging. Ideally, implementation of factor for stroke and stroke recurrence, and in whom rapid stroke protocols, obtaining structural, diffusionarterial imaging is critical to understanding the pathoweighted imaging (DWI) as well as magnetic resonance physiology of childhood stroke, optimizing treatment, angiography (MRA) of the head and neck as indicated and reducing the incidence of recurrent stroke. are achieved in 20–30 minutes, with stroke confirmation and the immediately needed imaging data. Protocols for thrombolytic therapy in childhood stroke are in developGENERAL APPROACH TO ment and are dependent on rapid and accurate neuroimNEUROIMAGING IN CHILDHOOD aging evaluation, which will also need to be refined. One ARTERIAL ISCHEMIC STROKE factor prevalent in the management of childhood stroke Multiple neuroimaging modalities of cerebral imaging is that many children cannot reliably report onset of are available for the child presenting with possible acute symptoms which are typically unwitnessed, hence necesstroke and choosing the appropriate studies in children sitating the development of imaging criteria for estimatcan be more challenging then in adults as the study risk ing the time duration of a stroke, similar to those being versus possible yield balance includes the broader differconsidered for “wake-up” stroke in adults (Silva et al., ential considerations often in play, the desire to avoid 2010; Ma et al., 2012). unnecessary radiation in children (Brenner and Hall, Criteria for hospital qualification as an adult acute 2007) with greater lifetime risks of radiation, and often stroke center are well established, and considerable lower level of cooperation in children where the young, experience has been gathered with the diagnostic and frightened, or developmentally delayed child will be prognostic aspects of neuroimaging of acute stroke in unable to cooperate with a lengthy exam. As stroke is reladults (Alberts et al., 2005). In contrast, guidelines for atively rare in childhood diagnostic delay is frequent pediatric stroke centers are still being established since many even pediatric emergency rooms are not (Bernard et al., 2014), and considerably less work has accustomed to emergent neuroimaging in children. been performed on the associated neuroimaging. Several Use and timing of any diagnostic study are modulated considerations regarding the efficient evaluation of

VASCULAR DISEASE 1161 acute stroke exist in children. First, it is important to conbright lesion, which is then also bright on ADC maps, will sider stroke in the child who presents with acute onset of also be bright on the trace image. Cytotoxic edema, in lateralizing or focal neurologic symptoms and signs. contrast, will show decreased ADC values consistent Second, small children and those unable to cooperate will with restricted diffusion. It must be remembered that require sedation for acquired neuroimaging. Consediffusion restriction can also be seen in non-AIS ischequently, established and close relationships with both mia as well as nonischemic diseases, some mentioned neuroradiology and anesthesiology as well as child life earlier, including acute seizures, metabolite and demyespecialists will be necessary for a functional acute stroke linating diseases. T2-weighted and T2 FLAIR images protocol in the evaluation and management of these are also reviewed for hyperintensity seen in subacute children. Complexity in the setting of pediatric stroke infarct and hyperintensity and atrophy sequelae of more necessitating general anesthesia for neuroimaging or remote injury. FLAIR is less useful in children under other management of the patient includes performing 2 years of age due to the immature state of cerebral myesedation or anesthesia such that blood pressure is mainlination. Depending on the age, hemorrhage can be dark, tained so as to optimize brain perfusion, particularly in isointense, or bright on T1- and T2-weighted images the setting of flow-limiting arteriopathy. Despite the (Kidwell and Wintermark, 2008). Susceptibilitycomplexity, the greater diagnostically relevant data weighted imaging (SWI) and gradient echo (GE) are obtained have resulted generally in application of MR added to increase the sensitivity for detection of hemortechniques over use of CT in children in the potential rhage. In most prepared centers, this series of MR AIS clinical scenario. sequences can be performed in 20–25 minutes. MRA is usually the first-line study to assess susMODALITIES RELEVANT TO pected cerebral arteriopathy and should be included in CRANIOCERVICAL IMAGING IN an acute MRI stroke protocol. In time-of-flight (TOF) CHILDHOOD ARTERIAL ISCHEMIC MRA the signal of stationary tissue is suppressed while STROKE the signal of moving blood is preserved, from which the flow in vessels is reconstructed. MRA will show Magnetic resonance imaging and magnetic evidence of arteriopathy in one-half to three-quarters resonance angiography of children with acute stroke (Danchaivijitr et al., With the spectrum of differential considerations and the 2006; Fullerton et al., 2007; Amlie-Lefond et al., 2009; desire to limit radiation, the optimal study for assessing Buerki et al., 2010); however, MRA can overestimate possible acute stroke in childhood is usually MRI, length and degree of stenosis, and will not detect lesions including DWI, commonly with the addition of head in smaller-diameter, distal vessels. Childhood AIS most and neck MRA. often involves the middle cerebral artery (MCA), and The age and condition of the child will determine the isolated anterior cerebral artery (ACA) involvement is ability to obtain an MR exam without sedation. In most rare (Buerki et al., 2010). Approximately half of strokes cases, due to the differential needing to be evaluated, involve the small-vessel territory, i.e., lenticulostriate MR will still be pursued despite requiring sedation, even or thalamoperforating arteries (Buerki et al., 2010) if a noncontrast CT might be obtainable without seda(Fig. 60.1). Quantifying stenosis is less important in chiltion. It is then paramount to have available the skill set dren than adults, however, as this information is rarely for such sedation, and a rapid, efficient MR paradigm helpful in the management of cerebral arteriopathy in outlined. children (Husson and Lasjaunias, 2004), in contrast to A rapid stroke neuroimaging MR protocol should adults, in whom cervical artery disease treatment may include DWI/apparent diffusion coefficient (ADC), depend on degree of stenosis (Brott et al., 2011). T1-weighted, T2-weighted (or T2 fluid-attenuation inverRelated imaging of the cerebral venous system is sion recovery (FLAIR)). DWI detects cytotoxic edema often also relevant in the setting of stroke evaluation. present as early as minutes following AIS, whereas T2 Though not included in this chapter, diagnosis of cerechanges in stroke may take up to 2 days to appear bral venous thrombosis can often be aided by MR veno(Buerki et al., 2010). DWI, viewed on the trace image gram (MRV). MRV can be performed also as a TOF where acute cytotoxic lesions will be bright, is extremely sequence, accentuating the venous sinus blood flow sensitive for cerebral ischemia, although DWI-negative rather than arterial, by appropriated manipulation of satinfarcts have been reported in adults in the brainstem uration pulses. Further characterization can be aided also or with lacunar locations (Oppenheim et al., 2000; be contrast MRV, with a volume T1-weighted sequence Sylaja et al., 2008). DWI assessment on the trace image being obtained following gadolinium injection. MRV requires reviewing of ADC maps to exclude the phenomis often included as an optional sequence, performed enon of “T2 shinethrough,” where an extremely T2 if there is detection of venous thrombosis on the

1162

C. AMLIE-LEFOND AND D. SHAW

Fig. 60.1. Basal ganglia infarct axial T2 fluid-attenuated inversion recovery with hyperintensity (A) and restricted diffusion on apparent diffusion coefficient (B) in distribution of the right lenticulostriate arteries in the right basal ganglia of a 3½-year-old boy who presented with acute left hemiparesis. (C) Maximum-intensity projection (MIP) of frontal view of magnetic resonance angiography demonstrates mild narrowing in the M1 segment.

standard sequences or the pattern of ischemic injury is suspicious for a venous infarct. Vessel wall imaging is increasingly used in stroke, as MRA, CT angiography (CTA), and catheter angiograms demonstrate the lumen of the vessel rather than the arterial wall which is the site of pathology. High-resolution T2 and specifically tailored pre- and postcontrast T1 FLAIR have been used extracranially in adults evaluating in particular arteriosclerotic plaque, and more recently investigating intracranial vessels. Though not currently routinely employed yet in most centers, the use of these techniques in distinguishing cerebral arteriopathies in childhood is showing some promise (Swartz et al., 2009). MRI evaluation for dissection involving the brachiocephalic vessels includes MRA to visualize the flowing vessel contour, as well as fat-saturated T1 or proton density sequences for the detection of blood in a false channel. Blood in the false channel is almost invariably bright on T1-weighted imaging between 1 and 8 weeks, but can be isointense and hence not discernable earlier then 7 days. MRA will still demonstrate contour alterations associated with dissections, within the limits of resolutions, and limitations due to normal complexity of flow in certain areas. Consideration is given for automatic incorporation of these neck sequences in a stroke imaging protocol, particularly for those children who have required sedation for scanning to avoid needing to resedate to add these sequences later.

CT imaging Head CT is frequently the imaging modality of choice for a child presenting with acute neurologic symptoms or signs due to speed and availability, and sensitivity for most surgical emergencies, including detection of acute intracranial hemorrhage, both intraparenchymal cerebral hemorrhage and most subarachnoid hemorrhages. As such, CT is a first-line test in a patient with suspected hemorrhage or trauma if the patient is unstable or MR is not available or contraindicated (such as

with most cardiac pacemakers). However, CT is much less sensitive to early ischemic injury than MRI using DWI in particular. Early CT changes with ischemia include blurring of the gray–white-matter junction and swelling of the sulci. Only later is regional tissue hypodensity consistent with stroke seen. Consequently it is not surprising that in a recent case series CT scan failed to detect acute ischemic stroke in 62 of 74 (84%) children (Srinivasan et al., 2009). CTA of the head and neck can be used to diagnose arteriopathy, including stenosis, occlusion, and dissection (Schievink, 2001). CTA requires sedation in the younger child and significant radiation exposure, and despite slightly better resolution, may not add clinically significant information to MRA. Consequently, if MRI/MRA can be obtained, CTA is often avoided in children. If there is still strong suspicion of cerebral arteriopathy following an uninformative MRA, a catheter angiogram may be preferable to CTA, as it is more likely to yield diagnostic information about small vessels and the presence of subtle dissection in larger arteries.

Perfusion studies CT perfusion, commonly employed in adults presenting with stroke, is of little use in pediatrics largely due to the significant radiation dose, and has been replaced by MR perfusion. MR perfusion can be helpful to define ischemic brain and tissue at risk for ischemia, particularly in vasculopathies such as moyamoya (Yun et al., 2009); however, its role in directing patient management is still emerging. Several parameters can be derived from perfusionweighted MRI using a gadolinium bolus, including cerebral blood flow, cerebral blood volume, and mean transit time. These measures require postprocessing of imaging, and more commonly the cerebral blood flow and cerebral blood volume results are given as relative measures, as obtaining absolute blood flow and volume is technically challenging. As such, the relative measure is most useful in comparing an area of compromised

VASCULAR DISEASE

1163

brain to a normally perfused area in the same brain. This works for a typical stroke involving a single distribution or side of the brain, but is less informative in situations of global compromise. Brain perfusion can also be assessed on MR without contrast using arterial spin labeling (ASL), where inversion pulses are used to label blood in the neck, and measure the resulting signal drop in the brain relative to cerebral blood flow (Chen et al., 2009). ASL techniques have the advantage of not requiring contrast, but to date suffer from low signal-to-noise ratio, and longer acquisition time though employment of newer ASL techniques at 3 T will overcome some of these limitations (Duyn et al., 2005; Essig et al., 2013).

which has been used to assess the need for transfusion (Adams et al., 1998a). A significant challenge with TCD is the experience and skill required of the operator, though these are becoming more widely disseminated.

Cerebral catheter angiogram

Focal cerebral arteriopathy/transient cerebral arteriopathy

APPROACH TO SPECIFIC CLINICAL ENTITIES Despite the importance of cerebral arteriopathy, classification is still evolving (Sebire et al., 2004), and etiologic, pathologic, and imaging features often overlap. Nevertheless, well-recognized arteriopathic syndromes associated with childhood stroke have been defined.

Cerebral catheter angiogram is the most sensitive imaging method of diagnosing cerebral arteriopathy (Husson and Lasjaunias, 2004). It is frequently used in adults to determine if a lesion is occlusive or preocclusive, as this impacts treatment; however in children it is used more often to characterize arteriopathy. It requires significant radiation and anesthesia, but periprocedural complications are rare with experienced angiographers (Burger et al., 2006). If dissection is strongly suspected but not seen on noninvasive imaging, cerebral catheter angiogram is indicated (Husson and Lasjaunias, 2004). It is also more sensitve that MRA and CTA for medium and small-vessel arteriopathy, when this is a relevant consideration. In most cases of central nervous system (CNS) vasculitis, there is abnormality on standard MRI; hence, if MRI and MRA are normal, vasculitis is unlikely to be demonstrated on catheter angiogram. Potential risks of catheter angiogram include local complications such as groin hematoma, femoral artery thrombosis, or pseudoaneurysm, and CNS complications, most notably stroke. Local complications are most problematic in the youngest children, with greater concern for femoral artery thrombosis, though still uncommon, in children under 1–2 years of age. Catheter angiogram-associated femoral artery thrombosis is unlikely to lead to acute injury; there is some risk of later leg length discrepancy. Overall in children, catheter angiogram has minimal risk of serious complication (Berger et al., 2000); however there is higher risk in children with connective tissue disorders such as collagen disorders (Zilocchi et al., 2007).

Focal cerebral arteriopathy has been used to describe a focal intracranial arterial stenosis, and accounts for approximately one-quarter of arteriopathies in childhood AIS (Amlie-Lefond et al., 2009). Transient cerebral arteriopathy is a monophasic subset of focal cerebral arteriopathy where lack of progression is seen at 6 months (Chabrier et al., 1998). A viral or inflammatory trigger has been proposed (Chabrier et al., 1998; AmlieLefond et al., 2009). Focal cerebral arteriopathy is usually well visualized on MRA, and, in most cases, the radiologic progression can be followed with this modality. In transient cerebral arteriopathy arterial narrowing will most commonly be seen unilaterally in the proximal MCA (Fig. 60.2). Symptoms most often result from edema associated with a basal ganglia infarct, visualized on T2 and T2 FLAIR sequences, secondary to occlusion of lenticulostriate arteries (Braun et al., 2009). As the motor symptoms are more often the result of vasogenic edema secondary to the initial infarct, the MRI appearance will often be consistent with a subacute infarct. Less commonly, there will also be an infarct involving the ipsilateral MCA, attributed to thrombosis formation at the point of vessel narrowing in the proximal MCA, in which case motor and potentially other cortical symptoms will be more acute. In the setting of transient cerebral arteriopathy there is expected lack of evidence to suggest a longer-term ischemic process (old infarcts) or collateral development more typical of moyamoya.

Transcranial Doppler

Moyamoya arteriopathy

Transcranial Doppler (TCD) has been used for monitoring for microemboli to determine response to antithrombotic treatment in adults. In pediatrics, TCD is most commonly used in sickle cell disease (SCD) to determine the rheology of flow in the proximal intracranial vessels,

Moyamoya is a progressive steno-occlusive disease typically involving the distal internal carotid artery (ICA) and proximal MCA bilaterally and often the ACAs with development of “moyamoya” collaterals at the base of the brain. Suzuki and Takaku (1969) describe six stages

1164

C. AMLIE-LEFOND AND D. SHAW

A

B

C

Fig. 60.2. Transient cerebral arteriopathy. A 9-year-old presenting with right hemiplegia and aphasia; left middle cerebral artery stroke evident on diffusion-weighted trace image (A) and narrow M1 segment (arrow) on magnetic resonance angiography (MRA) (B). Catheter angiogram confirmed mild narrowing of the left supraclinoid internal carotid artery and left M1 segment, with focal near-complete stenosis at the origin of the left A-1 segment. (C) Follow-up head MRA 4 years later is stable, consistent with focal cerebral arteriopathy/transient cerebral arteriopathy.

of Moyamoya: (1) bilateral ICA narrowing; (2) early moyamoya collateral formation; (3) progression of stages 1 and 2 with prominence of collaterals; (4) severe stenosis/occlusion of ICAs with involvement of circle of Willis; (5) reduction of moyamoya collaterals with increase in extracranial collaterals; and (6) disappearance of moyamoya collaterals and complete ICA occlusion. The moyamoya pattern of disease may be idiopathic (moyamoya disease) or occur in association with other syndromes (moyamoya syndrome), such as SCD and trisomy 21 (Sebire et al., 2004) or as the sequelae of radiation. The term “moyamoya disease” however is often used more loosely to include both (Scott et al., 2004). Moyamoya usually presents with bland ischemic stroke in childhood, whereas adults frequently present with hemorrhagic stroke. MRI will typically reveal white-matter lesions, encephalomalacia, atrophy, and hemorrhage (as noted above, more typically in adults) as a result of silent or overt strokes, in contrast to transient cerebral arteriopathy. Careful review of structural images will often demonstrate alteration of the normal flow voids associated with the distal ICAs and proximal MCAs, as well as the presence of flow voids through the basal ganglia and sometimes on the pial surface as a result of enlarged lenticulostriate and pial collaterals. The vascular changes are more definitely demonstrated on MRA, which can show the steno-occlusive changes in distal ICAs, proximal MCAs and ACAs, as well as proliferation of lenticulostriate vessels consistent with moyamoya (Hasuo et al., 1998). Leptomeningeal enhancement on contrast-enhanced T1 images and sulcal hyperintensity on FLAIR (“ivy sign”) can also be seen in moyamoya (Maeda and Tsuchida, 1999; Yoon et al., 2002). Contrast-enhanced T1 imaging is more sensitive than FLAIR imaging for this “ivy” sign (Yoon et al., 2002), which likely reflects the collateral vascular network which forms over the pial surface and appears to

correlate with the degree of ischemia (Mori et al., 2009), frequently receding after revascularization surgery. The angiographic findings on MRA using a modified Houkin scale have been found to correlate well with findings on catheter angiogram using the Suzuki grading system (Jin et al., 2011). Catheter angiogram with selective injections of the internal and external carotid arteries, and also a vertebral artery injection, is used to definitely evaluate the vasculature in preparation for revascularization surgery (Fig. 60.3). MR perfusion imaging using bolus tracking has been used in patients with moyamoya in an attempt to obtain quantitative information of regions at risk for ischemia (Calamante et al., 2001). Currently, however, technical limitations confound this method of perfusion quantification and it is not clear how to optimally use the information obtained to augment clinical decision making. The analysis is further complicated by large delays in arrival of bolus of contrast due to collateral supply, which limits the validity of results (Calamante et al., 2001). ASL MRI has also been used to assess arterial perfusion in patients with moyamoya treated with indirect and direct bypass. ASL has better sensitivity than MRA for demonstrating indirect revascularization, as TOF imaging can be negative due to slow flow. It is also more sensitive than MRA for evaluation of collateral circulation after direct bypass (Saida et al., 2012). As mentioned earlier, however, ASL suffers from limited signal-to-noise ratio, though newer 3 T implementations may improve this and thereby increase application.

Arteriopathy of sickle cell disease Approximately 25% of patients with SCD have cerebrovasculopathy that can precede stroke and many patients who have a stroke will have evidence of

VASCULAR DISEASE

A

B

D

1165

C

E

Fig. 60.3. Moyamoya disease in a 10-year-old boy with developmental delay presenting with expressive and receptive aphasia. (A) Diffusion-weighted imaging showing an acute stroke in the left posterior temporal lobe. (B) Axial fluid-attenuated inversion recovery (FLAIR) image shows periventricular hyperintensities not associated with restricted diffusion consistent with prior insult. (C) Frontal view of a right common carotid artery (RCCA) injection from a catheter angiogram shows occlusion at the carotid siphon (arrow) with central collateral associated with the ophthalmic artery and pial collaterals arising from external carotid artery branches. (D) Townes view of the left vertebral (LT VERT) injection from the catheter angiogram shows occlusion of the right P2 with right posterior cerebral artery filling through collaterals. (E) Middle cerebral arteries fill through posterior communication arteries and central collaterals, as well as pial collaterals evident on the lateral view of the vertebral projection.

cerebrovasculopathy (Fig. 60.4). Patients with SCD and moyamoya vasculopathy are more than twice as likely to have recurrent stroke, clinically silent stroke, or transient ischemic attack (TIA) than those without moyamoya (Dobson et al., 2002; Hulbert et al., 2011). In addition, SCD vasculopathy often progresses despite red blood cell transfusions (Bishop et al., 2011). TCD is sensitive and specific for intracranial occlusive arteriopathy associated with SCD, and can be used to monitor vascular stenosis or detect emboli. Adams and colleagues (1992) demonstrated that, in 34 TCD exams correlated with cerebral catheter angiograms in 33 patients (mean age 12 years, range 2–30 years), TCD had a sensitivity of 90% and specificity of 100%. The ability of TCD to predict stenosis was confirmed by Verlhac and colleagues (1995). Children greater than 2 years of age with SCD who have velocities above 200 cm/s are at increased risk of stroke and chronic

transfusions are indicated to decrease the risk of stroke (Adams et al., 1998b). Even among well-trained TCD examiners, however, large intrasubject variability can at times be seen, and a minimum of two exams prior to embarking on chronic blood transfusions is recommended (Brambilla et al., 2007). The American Academy of Neurology recommends TCD screening of children with SCD between the ages of 2 and 16 years (Sloan et al., 2004), although the optimal frequency is unknown. Although children with hemoglobin SC also have increased risk of stroke, although less so than children with HbgSS, a cut-off for TCD velocity in these patients has not been established. With chronic transfusion, the TCD velocities may normalize in some children, particularly younger children and those with lower TCD values, with approximately two-thirds of children reverting to normal TCD values (Kwiatkowski et al., 2011). Discontinuation of chronic transfusions was associated

1166

C. AMLIE-LEFOND AND D. SHAW

A

B

Fig. 60.4. Moyamoya associated with sickle cell disease. (A) Axial fluid-attenuated inversion recovery showing scattered central white-matter hyperintensities in a 7-year-old girl with sickle cell disease. (B) Frontal view of maximum-intensity projection showing marked narrowing of the terminal carotids and proximal middle cerebral arteries bilaterally, right greater than left.

with reversion to abnormal TCD velocities and increased stroke risk, however (Adams and Brambilla, 2005). The age distribution, pathophysiology, and type of stroke in moyamoya associated with SCD closely resemble those of moyamoya disease, and imaging recommendations are as above. When catheter angiogram is used to evaluate children with SCD, hematology involvement for preparation of the child is critical. Transfusion to ensure a sickle cell hemoglobin percent of less than 30%, with some centers recommending 20% (Jones et al., 2010), and hydration are recommended. In addition, attention to underlying inflammation and associated thrombophilias such as decreased protein C and protein S activity is important (Piccin et al., 2012).

Cervicocephalic arterial dissection Cervicocephalic arterial dissection (CCAD) accounts for 7.5–20% of childhood AIS (Chabrier et al., 2000; Rafay et al., 2006; Amlie-Lefond et al., 2009), and involves the anterior circulation more often than the posterior

A

circulation (Chabrier et al., 2000; Fullerton et al., 2001). Spontaneous dissection is thought to be more common in children than adults, although difficulty obtaining a history of trauma and attribution of more minor trauma can confuse the etiology. Nonetheless, detection of dissection is critical, as these patients are at high risk for recurrent stroke, subarachnoid hemorrhage due to aneurysmal dilation, and recurrent dissection. Most dissections of the cervical carotid arteries occur 2–3 cm above the carotid bulb (Schievink, 2001). Dissections of the vertebral arteries most commonly involve the artery as it passes around the C1–C2 lateral masses and through the transverse foramina (Houser et al., 1984; Schievink, 2001) (Fig. 60.5). In the anterior circulation, intracranial dissection is more common than extracranial dissection (Chabrier et al., 2000; Fullerton et al., 2001), and in the posterior circulation extracranial dissection predominates (Songsaeng et al., 2010). Subarachnoid hemorrhage can occur due to vessel leakage or rupture of pseudoaneurysm formation. The risk of arterial rupture is higher following dissection of

B

Fig. 60.5. Subtle vertebral dissection. (A) Apparent diffusion coefficient from a magnetic resonance imaging demonstrates acute superior cerebellar artery distribution infarcts bilaterally in a 1-year-old; magnetic resonance angiography was normal (not shown). (B) Catheter angiogram of the right vertebral artery demonstrates subtle irregularity (arrow) consistent with dissection.

VASCULAR DISEASE intracranial as opposed to extracranial arteries, likely due to the thinner walls and absence of an external elastic membrane in intracranial arteries. For this reason, anticoagulant use is often avoided in intracranial dissection, although it may be safe in nonaneurysmal intracranial dissections without subarachnoid hemorrhage (Rafay et al., 2006; Metso et al., 2007). The diagnosis of arterial dissection is often made on MRA, evaluation for which should include the head and neck extending to the aortic arch, in conjunction with the use of T1W fat saturation or proton density sequences which allow for visualization of intramural hematomas (methemoglobin) in a false lumen, seen in most arterial dissections in the neck (Oelerich et al., 1999; Arnold et al., 2006; Vertinsky et al., 2008) (Fig. 60.6). More recently, use of contrast and pulse sequences designed to evaluate the vessel wall has provided novel information about wall anatomy (Srinivasan et al., 2009). Narrowing of the true lumen, analogous to that demonstrated on catheter angiogram, is classically seen on MRA, although subtle dissections, particularly in areas

A

D

B

E

1167

of turbulent flow, such as the C1–C2 level of the vertebral arteries, may be missed (Fig. 60.5). A pseudoaneurysm may be present if blood has dissected through the media, more commonly seen as a later sequela. Intimal flaps and dissecting aneurysm are diagnostic for dissection; however, as mentioned above, subtle lesions can be missed on MRA (Rafay et al., 2006; Tan et al., 2009). In a report of 13 children with CCAD subsequently diagnosed with CCAD on catheter angiogram, the diagnosis was missed on initial imaging with 1.5 T MRI and MRA in eight children: two due to absence of neck vessel imaging, three due to suboptimal technique, and three due to diagnostic error (Tan et al., 2009). Intracranial dissection has been mistaken for transient cerebral arteriopathy on MRA in a small series of four children reported to the International Pediatric Stroke Study (Dlamini et al., 2011). Three of these children died of malignant infarct edema and dissection was found at autopsy. In the fourth, catheter angiogram suggested possible dissection, which was confirmed on

C

F

Fig. 60.6. Internal carotid artery dissection in a 15-year-old boy who developed left hemiparesis shortly after falling on the right side of his head. Diffusion-weighted trace image shows hyperintensity in the posterior limb of the right internal capsule (A) and right posterior parietal lobe (B) consistent with acute strokes. (C) Axial fluid-attenuated inversion recovery image shows decreased flow and increased signal in the right supraclinoid internal carotid artery. (D) Lateral view of a right common carotid artery (RCCA) injection of a catheter angiogram shows tapering of supraclinoid right internal carotid artery. (E) Frontal view of left internal carotid artery (LICA) injection during catheter angiogram shows cross-filling supplying the right anterior circulation. (F) Magnetic resonance angiography 2 months later shows recanalization of the right internal carotid.

1168

C. AMLIE-LEFOND AND D. SHAW

dedicated MR vessel wall imaging (Dlamini et al., 2011). If the MRA is negative and there is still clinical concern, cerebral catheter angiogram generally remains the most sensitive test for dissection, although with advances in vessel wall imaging, including detection of intramural hematoma, this may change. The most common findings indicating CCAD are stenosis or occlusion of the artery or aneurysm formation (Huang et al., 2009). Catheter angiogram will not directly demonstrate arterial wall hematoma. An angiographic “string sign” can be seen as well, although this finding is most helpful in the evaluation of preocclusive atherosclerotic lesions in adults (Pappas, 2002). CTA is often pursued in adults before cerebral catheter angiogram; however, frequently the catheter angiogram is still required. Therefore if there is still uncertainty with high suspicion for dissection following MRI (e.g., multiple infarcts attributable to a singlevessel source) even with a normal or equivocal MRA, moving directly to cerebral catheter angiogram rather than pursuing a CTA is usually recommended in children. CTA may however be the preferred study in a child with acute trauma in whom rapid diagnosis is needed and CT is already being pursued. In children in whom arterial dissection is strongly suspected, treatment for presumed dissection may be appropriate even if MRA is normal and cerebral catheter angiogram may not be indicated in all cases. Doppler ultrasound has limited usefulness in diagnosis of dissection. It is less sensitive, particularly in the vertebral arteries and in the carotid arteries around the skull base, and operator-dependent.

children with small-vessel primary CNS vasculitis (Elbers and Benseler, 2008). Arterial imaging will not show abnormalities in small-vessel vasculitis, but in medium- and largeartery vasculitis, irregular lumen and vascular stenoses may be seen on MRA and CTA. In some cases, gadolinium enhancement of thickened arterial walls may be seen. Cerebral catheter angiogram has greater sensitivity for luminal abnormalities but does not show vessel walls.

Fibromuscular dysplasia Fibromuscular dysplasia is a noninflammatory arteriopathy of unknown etiology, primarily affecting women age 20–60 years, that can occur in children and is associated with both ischemic and hemorrhagic stroke in childhood. It is classically associated with irregular narrowing (“string of beads”) on angiography of the internal carotid and renal arteries; however, diagnosis requires histopathologic confirmation. Kirton and colleagues (2013) suggest that diagnosis in childhood is complicated by overlap with other arteriopathies and not that the string-of-beads sign is seen in other childhood arteriopathies as well. In addition, the authors found most patients with pathologically proven fibromuscular dysplasia had less specific angiographic abnormalities, including focal and segmental stenosis and occlusion, and only one-quarter had a string-of-beads sign (Kirton et al., 2013).

Vasculitis

IMAGING APPROACH TO THE CHILD PRESENTING WITH POSSIBLE TRANSIENT ISCHEMIC ATTACK OR ACUTE STROKE

CNS vasculitis can be primary, or idiopathic, or secondary to a systemic cause such as collagen vascular disease or infection. In CNS vasculitis both white- and graymatter lesions can be seen, more often in the anterior than posterior circulation, with frequent involvement of the basal ganglia. Acute and subacute lesions may show DWI and ADC abnormalities consistent with ischemic stroke. Aviv et al. (2006) studied 45 children with primary angiitis of the CNS of childhood diagnosed based on clinical and radiographic criteria, established by Calabrese (2003). Multifocal, unilateral parenchymal abnormalities with involvement of the supratentorial gray matter were most commonly seen. Infratentorial abnormalities in the absence of supratentorial abnormalities were not seen (Aviv et al., 2006). In patients with suggestive neuroimaging findings, brain biopsy continues to be critical to diagnosis of small-vessel CNS vasculitis, however (Kadkhodayan et al., 2004). Biopsy, even nonlesional biopsy, has a significant yield in

Children with cerebral arteriopathy may present with TIAs, often recurrent, and TIAs may precede stroke. Signs of anterior circulation ischemia, including aphasia, dysarthria, hemiparesis, and seizures, are most commonly seen, but less common presentations, such as syncope, visual changes, and chorea, can occur. In moyamoya, symptoms are often provoked by hyperventilation due to hypocarbia-induced vasoconstriction, for example, with crying or playing a wind instrument. Cerebral arterial imaging is indicated in children presenting with an unexplained history of localized neurologic dysfunction, particularly with recurrent episodes. Hemiplegic migraine is very rare in childhood, with a mean age of onset of familial hemiplegic migraine of 16 years in males and 21 years in females (Thomsen et al., 2003), and diagnosis in childhood has typically required exclusion of recurrent ischemia and postictal etiologies. Hemiplegic migraine however can have a very dramatic appearance on SWI, with a visible difference in vessel

VASCULAR DISEASE signal intensity in one cerebral hemisphere compared to the other, attributed to differences in oxygenation. The child presenting with possible stroke requires prompt diagnosis so that strategies to limit injury and prevent recurrent stroke can be instituted. Statistically most will ultimately be diagnosed with a stroke mimic. For a child with no known risk factors for ischemic stroke, the initial imaging study is often a head CT, as it can be done quickly. CT will identify acute hemorrhage but has limited sensitivity for the detection of acute stroke. Confirmation of ischemic stroke will likely require MRI, as will workup of stroke mimics, hence the earlier advocating for MRI, in the absence of contraindications such as incompatible hardware such as a pacemaker, hemodynamic instability, or inability to cooperate with an MR exam, with consideration for a sedated exam as indicated. If routine sedation for MRI is not already in existence at the institution, setting up such a protocol is of value, before the emergent need presents itself. The MRI scanning approach should include DWI early in the sequence order as this will help direct the subsequent imaging sequences chosen, and answer the question of acute stroke up front, since normal DWI excludes acute infarct, with rare exceptions. In the child who fails an attempt at an unsedated MR, if even only the DWI is obtained, subsequent imaging with sedation can proceed at a more measured pace. In most cases, arterial imaging with TOF MRA is warranted and may alter immediate management. Dedicated MRA, and arterial vessel wall imaging if available, may not always be possible in the acute setting, but should be pursued as promptly as possible, particularly in the child with findings and history suggestive of arteriopathy, given the high risk of initial and recurrent stroke in these children. MRA can both overestimate the degree of arterial compromise due to complex/turbulent flow and be insensitive to subtle vessel irregularity due to resolution. This is particularly problematic in dissection, which has a propensity for areas with complex flow, and in which MRA findings can be subtle. In cases with high imaging suspicion for dissection despite normal MRA findings, catheter angiogram is indicated. Results of the brain MRI then dictate whether imaging of the neck for evidence of dissection should be pursued, e.g., multiple infarcts attributable to a single vessel. In conclusion, a thoughtful, stepwise approach to arterial neuroimaging is critical to optimize diagnosis, treatment, and primary and secondary prevention of childhood stroke.

REFERENCES Adams RJ, Brambilla D (2005). Discontinuing prophylactic transfusions used to prevent stroke in sickle cell disease. N Engl J Med 353: 2769–2778.

1169

Adams RJ, Nichols FT, Figueroa R et al. (1992). Transcranial Doppler correlation with cerebral angiography in sickle cell disease. Stroke 23: 1073–1077. Adams RJ, McKie VC, Carl EM et al. (1997). Long-term stroke risk in children with sickle cell disease screened with transcranial Doppler. Ann Neurol 42: 699–704. Adams RJ, McKie VC, Brambilla D et al. (1998a). Stroke prevention trial in sickle cell anemia. Control Clin Trials 19: 110–129. Adams RJ, McKie VC, Hsu L et al. (1998b). Prevention of a first stroke by transfusions in children with sickle cell anemia and abnormal results on transcranial Doppler ultrasonography. N Engl J Med 339: 5–11. Alberts MJ, Latchaw RE, Selman WR et al. (2005). Recommendations for comprehensive stroke centers: a consensus statement from the Brain Attack Coalition. Stroke 36: 1597–1616. Amlie-Lefond C, Bernard TJ, Sebire G et al. (2009). Predictors of cerebral arteriopathy in children with arterial ischemic stroke: results of the International Pediatric Stroke Study. Circulation 119: 1417–1423. Arnold M, Bousser MG, Fahrni G et al. (2006). Vertebral artery dissection: presenting findings and predictors of outcome. Stroke 37: 2499–2503. Aviv RI, Benseler SM, Silverman ED et al. (2006). MR imaging and angiography of primary CNS vasculitis of childhood. AJNR Am J Neuroradiol 27: 192–199. Berger TM, Caduff JH, Gebbers JO (2000). Fatal varicellazoster virus antigen-positive giant cell arteritis of the central nervous system. Pediatr Infect Dis J 19: 653–656. Bernard TJ, Rivkin MJ, Scholz K et al. (2014). Emergence of the Primary Pediatric Stroke Center: Impact of the Thrombolysis in Pediatric Stroke Trial. Stroke 45: 2018–2023. Bishop S, Matheus MG, Abboud MR et al. (2011). Effect of chronic transfusion therapy on progression of neurovascular pathology in pediatric patients with sickle cell anemia. Blood Cells Mol Dis 47: 125–128. Brambilla DJ, Miller ST, Adams RJ (2007). Intra-individual variation in blood flow velocities in cerebral arteries of children with sickle cell disease. Pediatr Blood Cancer 49: 318–322. Braun KP, Bulder MM, Chabrier S et al. (2009). The course and outcome of unilateral intracranial arteriopathy in 79 children with ischaemic stroke. Brain 132: 544–557. Brenner DJ, Hall EJ (2007). Computed tomography – an increasing source of radiation exposure. N Engl J Med 357: 2277–2284. Brott TG, Halperin JL, Abbara S et al. (2011). 2011 ASA/ ACCF/AHA/AANN/AANS/ACR/ASNR/CNS/SAIP/SCAI/ SIR/SNIS/SVM/SVS guideline on the management of patients with extracranial carotid and vertebral artery disease: executive summary. Stroke 42: e420–e463. Buerki S, Roellin K, Remonda L et al. (2010). Neuroimaging in childhood arterial ischaemic stroke: evaluation of imaging modalities and aetiologies. Dev Med Child Neurol 52: 1033–1037. Burger IM, Murphy KJ, Jordan LC et al. (2006). Safety of cerebral digital subtraction angiography in children:

1170

C. AMLIE-LEFOND AND D. SHAW

complication rate analysis in 241 consecutive diagnostic angiograms. Stroke 37: 2535–2539. Calabrese LH (2003). Clinical management issues in vasculitis. Angiographically defined angiitis of the central nervous system: diagnostic and therapeutic dilemmas. Clin Exp Rheumatol 21: S127–S130. Calamante F, Ganesan V, Kirkham FJ et al. (2001). MR perfusion imaging in moyamoya syndrome: potential implications for clinical evaluation of occlusive cerebrovascular disease. Stroke 32: 2810–2816. Chabrier S, Rodesch G, Lasjaunias P et al. (1998). Transient cerebral arteriopathy: a disorder recognized by serial angiograms in children with stroke. J Child Neurol 13: 27–32. Chabrier S, Husson B, Lasjaunias P et al. (2000). Stroke in childhood: outcome and recurrence risk by mechanism in 59 patients. J Child Neurol 15: 290–294. Chen J, Licht DJ, Smith SE et al. (2009). Arterial spin labeling perfusion MRI in pediatric arterial ischemic stroke: initial experiences. J Magn Reson Imaging 29: 282–290. Danchaivijitr N, Cox TC, Saunders DE et al. (2006). Evolution of cerebral arteriopathies in childhood arterial ischemic stroke. Ann Neurol 59: 620–626. Darteyre S, Chabrier S, Presles E et al. (2012). Lack of progressive arteriopathy and stroke recurrence among children with cryptogenic stroke. Neurology 79: 2342–2348. discussion 2346. De Schryver EL, Kappelle LJ, Jennekens-Schinkel A et al. (2000). Prognosis of ischemic stroke in childhood: a long-term follow-up study. Dev Med Child Neurol 42: 313–318. Deveber GA, Macgregor D, Curtis R et al. (2000). Neurologic outcome in survivors of childhood arterial ischemic stroke and sinovenous thrombosis. J Child Neurol 15: 316–324. Dlamini N, Freeman JL, MacKay MT et al. (2011). Intracranial dissection mimicking transient cerebral arteriopathy in childhood arterial ischemic stroke. J Child Neurol 26: 1203–1206. Dobson SR, Holden KR, Nietert PJ et al. (2002). Moyamoya syndrome in childhood sickle cell disease: a predictive factor for recurrent cerebrovascular events. Blood 99: 3144–3150. Duyn JH, Van Gelderen P, Talagala L et al. (2005). Technological advances in MRI measurement of brain perfusion. J Magn Reson Imaging 22: 751–753. Elbers J, Benseler SM (2008). Central nervous system vasculitis in children. Curr Opin Rheumatol 20: 47–54. Essig M, Nguyen TB, Shiroishi MS et al. (2013). Perfusion MRI: the five most frequently asked clinical questions. AJR Am J Roentgenol 201: W495–W510. Fullerton HJ, Johnston SC, Smith WS (2001). Arterial dissection and stroke in children. Neurology 57: 1155–1160. Fullerton HJ, Chetkovich DM, Wu YW et al. (2002). Deaths from stroke in US children, 1979 to 1998. Neurology 59: 34–39. Fullerton HJ, Wu YW, Zhao S et al. (2003). Risk of stroke in children: ethnic and gender disparities. Neurology 61: 189–194.

Fullerton HJ, Wu YW, Sidney S et al. (2007). Risk of recurrent childhood arterial ischemic stroke in a population-based cohort: the importance of cerebrovascular imaging. Pediatrics 119: 495–501. Ganesan V, Hogan A, Shack N et al. (2000). Outcome after ischaemic stroke in childhood. Dev Med Child Neurol 42: 455–461. Goldenberg NA, Jenkins S, Jack J et al. (2013). Arteriopathy, D-dimer, and risk of poor neurologic outcome in childhood-onset arterial ischemic stroke. J Pediatr 162 (1041–6): e1. Hasuo K, Mihara F, Matsushima T (1998). MRI and MR angiography in moyamoya disease. J Magn Reson Imaging 8: 762–766. Houser OW, Mokri B, Sundt Jr TM et al. (1984). Spontaneous cervical cephalic arterial dissection and its residuum: angiographic spectrum. AJNR Am J Neuroradiol 5: 27–34. Huang YC, Chen YF, Wang YH et al. (2009). Cervicocranial arterial dissection: experience of 73 patients in a single center. Surg Neurol 72 (Suppl 2): S20–S27. discussion S27. Hulbert ML, Mckinstry RC, Lacey JL et al. (2011). Silent cerebral infarcts occur despite regular blood transfusion therapy after first strokes in children with sickle cell disease. Blood 117: 772–779. Husson B, Lasjaunias P (2004). Radiological approach to disorders of arterial brain vessels associated with childhood arterial stroke-a comparison between MRA and contrast angiography. Pediatr Radiol 34: 10–15. Jauch EC, Saver JL, Adams Jr HP et al. (2013). Guidelines for the early management of patients with acute ischemic stroke: a guideline for healthcare professionals from the American Heart Association/American Stroke Association. Stroke 44: 870–947. Jin Q, Noguchi T, Irie H et al. (2011). Assessment of moyamoya disease with 3.0-T magnetic resonance angiography and magnetic resonance imaging versus conventional angiography. Neurol Med Chir (Tokyo) 51: 195–200. Jones BP, Ganesan V, Saunders DE et al. (2010). Imaging in childhood arterial ischaemic stroke. Neuroradiology 52: 577–589. Kadkhodayan Y, Alreshaid A, Moran CJ et al. (2004). Primary angiitis of the central nervous system at conventional angiography. Radiology 233: 878–882. Kidwell CS, Wintermark M (2008). Imaging of intracranial haemorrhage. Lancet Neurol 7: 256–267. Kirton A, Crone M, Benseler S et al. (2013). Fibromuscular dysplasia and childhood stroke. Brain 136: 1846–1856. Kwiatkowski JL, Yim E, Miller S et al. (2011). Effect of transfusion therapy on transcranial Doppler ultrasonography velocities in children with sickle cell disease. Pediatr Blood Cancer 56: 777–782. Lanthier S, Armstrong D, Domi T et al. (2005). Post-varicella arteriopathy of childhood: natural history of vascular stenosis. Neurology 64: 660–663. Lynch JK, Hirtz DG, Deveber G et al. (2002). Report of the National Institute of Neurological Disorders and Stroke workshop on perinatal and childhood stroke. Pediatrics 109: 116–123.

VASCULAR DISEASE Ma H, Parsons MW, Christensen S et al. (2012). A multicentre, randomized, double-blinded, placebo-controlled phase III study to investigate EXtending the time for Thrombolysis in Emergency Neurological Deficits (EXTEND). Int J Stroke 7: 74–80. MacKay MT, Lee M, Churilov L et al. (2014). Pediatric brain attacks: differentiating between stroke and mimics in the emergency room, International Stroke Conference. Maeda M, Tsuchida C (1999). “Ivy sign” on fluid-attenuated inversion-recovery images in childhood moyamoya disease. AJNR Am J Neuroradiol 20: 1836–1838. Metso TM, Metso AJ, Helenius J et al. (2007). Prognosis and safety of anticoagulation in intracranial artery dissections in adults. Stroke 38: 1837–1842. Mori N, Mugikura S, Higano S et al. (2009). The leptomeningeal “ivy sign” on fluid-attenuated inversion recovery MR imaging in moyamoya disease: a sign of decreased cerebral vascular reserve? AJNR Am J Neuroradiol 30: 930–935. Oelerich M, Stogbauer F, Kurlemann G et al. (1999). Craniocervical artery dissection: MR imaging and MR angiographic findings. Eur Radiol 9: 1385–1391. Oppenheim C, Stanescu R, Dormont D et al. (2000). Falsenegative diffusion-weighted MR findings in acute ischemic stroke. AJNR Am J Neuroradiol 21: 1434–1440. Pappas JN (2002). The angiographic string sign. Radiology 222: 237–238. Piccin A, Murphy C, Eakins E et al. (2012). Protein C and free protein S in children with sickle cell anemia. Ann Hematol 91: 1669–1671. Rafay MF, Armstrong D, Deveber G et al. (2006). Craniocervical arterial dissection in children: clinical and radiographic presentation and outcome. J Child Neurol 21: 8–16. Saida T, Masumoto T, Nakai Y et al. (2012). Moyamoya disease: evaluation of postoperative revascularization using multiphase selective arterial spin labeling MRI. J Comput Assist Tomogr 36: 143–149. Schievink WI (2001). Spontaneous dissection of the carotid and vertebral arteries. N Engl J Med 344: 898–906. Schoenberg BS, Mellinger JF, Schoenberg DG (1978). Cerebrovascular disease in infants and children: a study of incidence, clinical features, and survival. Neurology 28: 763–768. Scott RM, Smith JL, Robertson RL et al. (2004). Long-term outcome in children with moyamoya syndrome after cranial revascularization by pial synangiosis. J Neurosurg 100: 142–149. Sebire G, Fullerton H, Riou E et al. (2004). Toward the definition of cerebral arteriopathies of childhood. Curr Opin Pediatr 16: 617–622. Shellhaas RA, Smith SE, O’Tool E et al. (2006). Mimics of childhood stroke: characteristics of a prospective cohort. Pediatrics 118: 704–709.

1171

Silva GS, Lima FO, Camargo EC et al. (2010). Wake-up stroke: clinical and neuroimaging characteristics. Cerebrovasc Dis 29: 336–342. Sloan MA, Alexandrov AV, Tegeler CH et al. (2004). Assessment: transcranial Doppler ultrasonography: report of the Therapeutics and Technology Assessment Subcommittee of the American Academy of Neurology. Neurology 62: 1468–1481. Songsaeng D, Srivatanakul K, Krings T et al. (2010). Symptomatic spontaneous vertebrobasilar dissections in children: review of 29 consecutive cases. J Neurosurg Pediatr 6: 233–243. Srinivasan J, Miller SP, Phan TG et al. (2009). Delayed recognition of initial stroke in children: need for increased awareness. Pediatrics 124: e227–e234. Suzuki J, Takaku A (1969). Cerebrovascular “moyamoya” disease. Disease showing abnormal net-like vessels in base of brain. Arch Neurol 20: 288–299. Swartz RH, Bhuta SS, Farb RI et al. (2009). Intracranial arterial wall imaging using high-resolution 3-tesla contrastenhanced MRI. Neurology 72: 627–634. Sylaja PN, Coutts SB, Krol A et al. (2008). When to expect negative diffusion-weighted images in stroke and transient ischemic attack. Stroke 39: 1898–1900. Tan MA, Deveber G, Kirton A et al. (2009). Low detection rate of craniocervical arterial dissection in children using timeof-flight magnetic resonance angiography: causes and strategies to improve diagnosis. J Child Neurol 24: 1250–1257. Thomsen LL, Ostergaard E, Olesen J et al. (2003). Evidence for a separate type of migraine with aura: sporadic hemiplegic migraine. Neurology 60: 595–601. Verlhac S, Bernaudin F, Tortrat D et al. (1995). Detection of cerebrovascular disease in patients with sickle cell disease using transcranial Doppler sonography: correlation with MRI, MRA and conventional angiography. Pediatr Radiol 25 (Suppl 1): S14–S19. Vertinsky AT, Schwartz NE, Fischbein NJ et al. (2008). Comparison of multidetector CT angiography and MR imaging of cervical artery dissection. AJNR Am J Neuroradiol 29: 1753–1760. Yoon HK, Shin HJ, Chang YW (2002). “Ivy sign” in childhood moyamoya disease: depiction on FLAIR and contrastenhanced T1-weighted MR images. Radiology 223: 384–389. Yun TJ, Cheon JE, Na DG et al. (2009). Childhood moyamoya disease: quantitative evaluation of perfusion MR imaging – correlation with clinical outcome after revascularization surgery. Radiology 251: 216–223. Zilocchi M, Macedo TA, Oderich GS et al. (2007). Vascular Ehlers–Danlos syndrome: imaging findings. AJR Am J Roentgenol 189: 712–719.