Advertisement

Pathways for Neuroimaging of Childhood Stroke

  • Author Footnotes
    1 Co-first authors, the authors contributed equally to this work.
    David M. Mirsky
    Correspondence
    Communications should be addressed to: Dr. Mirsky; Children's Hospital Colorado; 13123 East 16th Avenue; Box B125; Aurora, CO 80045.
    Footnotes
    1 Co-first authors, the authors contributed equally to this work.
    Affiliations
    Department of Radiology, Children's Hospital Colorado, University of Colorado School of Medicine, Aurora, Colorado
    Search for articles by this author
  • Author Footnotes
    1 Co-first authors, the authors contributed equally to this work.
    Lauren A. Beslow
    Footnotes
    1 Co-first authors, the authors contributed equally to this work.
    Affiliations
    Department of Neurology, Perelman School of Medicine at the University of Pennsylvania, Division of Neurology, Children's Hospital of Philadelphia, Philadelphia, Pennsylvania
    Search for articles by this author
  • Catherine Amlie-Lefond
    Affiliations
    Department of Neurology, Seattle Children's Hospital, University of Washington, Seattle, Washington
    Search for articles by this author
  • Pradeep Krishnan
    Affiliations
    Division of Neuroradiology, Department of Diagnostic Imaging, The Hospital for Sick Children, Toronto, Ontario, Canada

    Department of Medical Imaging, University of Toronto, Toronto, Ontario, Canada
    Search for articles by this author
  • Suzanne Laughlin
    Affiliations
    Division of Neuroradiology, Department of Diagnostic Imaging, The Hospital for Sick Children, Toronto, Ontario, Canada

    Department of Medical Imaging, University of Toronto, Toronto, Ontario, Canada
    Search for articles by this author
  • Sarah Lee
    Affiliations
    Department of Neurology & Neurological Sciences, Stanford University School of Medicine, Stanford, California
    Search for articles by this author
  • Laura Lehman
    Affiliations
    Department of Neurology, Boston Children's Hospital, Harvard Medical School, Boston, Massachusetts
    Search for articles by this author
  • Mubeen Rafay
    Affiliations
    Section of Neurology, Children's Hospital, Children's Hospital Research Institute of Manitoba, University of Manitoba, Winnipeg, Manitoba, Canada
    Search for articles by this author
  • Dennis Shaw
    Affiliations
    Department of Radiology, University of Washington School of Medicine, Seattle, Washington
    Search for articles by this author
  • Author Footnotes
    2 Co-senior authors.
    Michael J. Rivkin
    Footnotes
    2 Co-senior authors.
    Affiliations
    Department of Neurology, Boston Children's Hospital, Harvard Medical School, Boston, Massachusetts

    Department of Radiology, Boston Children's Hospital, Harvard Medical School, Boston, Massachusetts

    Department of Psychiatry, Boston Children's Hospital, Harvard Medical School, Boston, Massachusetts
    Search for articles by this author
  • Author Footnotes
    2 Co-senior authors.
    Max Wintermark
    Footnotes
    2 Co-senior authors.
    Affiliations
    Division of Neuroradiology, Department of Radiology, Stanford University School of Medicine, Stanford, California
    Search for articles by this author
  • for theInternational Paediatric Stroke Study Neuroimaging Consortium and the Paediatric Stroke Neuroimaging Consortium
  • Author Footnotes
    1 Co-first authors, the authors contributed equally to this work.
    2 Co-senior authors.

      Abstract

      Background

      The purpose of this article is to aid practitioners in choosing appropriate neuroimaging for children who present with symptoms that could be caused by stroke.

      Methods

      The Writing Group members participated in one or more pediatric stroke neuroimaging symposiums hosted by the Stroke Imaging Laboratory for Children housed at the Hospital for Sick Children in Toronto, Ontario, Canada. Through collaboration, literature review, and discussion among child neurologists with expertise diagnosing and treating childhood stroke and pediatric neuroradiologists and neuroradiologists with expertise in pediatric neurovascular disease, suggested imaging protocols are presented for children with suspected stroke syndromes including arterial ischemic stroke, cerebral sinovenous thrombosis, and hemorrhagic stroke.

      Results

      This article presents information about the epidemiology and classification of childhood stroke with definitions based on the National Institutes of Health Common Data Elements. The role of imaging for the diagnosis of childhood stroke is examined in depth, with separate sections for arterial ischemic stroke, cerebral sinovenous thrombosis, and hemorrhagic stroke. Abbreviated neuroimaging protocols for rapid diagnosis are discussed. The Writing Group provides suggestions for optimal neuroimaging investigation of various stroke types in the acute setting and suggestions for follow-up neuroimaging. Advanced sequences such as diffusion tensor imaging, perfusion imaging, and vessel wall imaging are also discussed.

      Conclusions

      This article provides protocols for the imaging of children who present with suspected stroke.

      Keywords

      Introduction and epidemiology

      Stroke is a major cause of morbidity and mortality in children worldwide. The reported annual incidence of childhood stroke ranges from 2.3 to 13 per 100,000 children per year in developed countries.
      • Giroud M.
      • Lemesle M.
      • Gouyon J.B.
      • Nivelon J.L.
      • Milan C.
      • Dumas R.
      Cerebrovascular disease in children under 16 years of age in the city of Dijon, France: a study of incidence and clinical features from 1985 to 1993.
      • Lynch J.K.
      • Hirtz D.G.
      • DeVeber G.
      • Nelson K.B.
      Report of the National Institute of Neurological Disorders and Stroke workshop on perinatal and childhood stroke.
      • Fullerton H.J.
      • Wu Y.W.
      • Zhao S.
      • Johnston S.C.
      Risk of stroke in children: ethnic and gender disparities.
       Despite increasing awareness, this condition is often overlooked by medical providers and families. In adults, presentation with sudden onset hemiparesis with or without facial weakness and language problems constitutes hallmark presenting features of stroke which raise concern for the diagnosis without unnecessary delay. In children, stroke diagnosis is not as straightforward. Despite a growth in awareness about childhood stroke, when children present with acute neurological deficits, stroke is often not the first diagnosis considered by the medical providers. Delay in diagnosis derives, in part, from clinicians' difficulty recognizing that presenting signs and symptoms such as seizure, altered mental status, headache, and lethargy can be associated with acute stroke in children. Neuroimaging is essential for diagnosis and differentiation of stroke from stroke mimics that can present similarly such as hypoglycemia, demyelinating disorders, tumors, posterior reversible leukoencephalopathy syndrome, and complex migraine. Importantly neuroimaging is essential for identification of children who may be candidates for hyperacute therapy.
      This report will briefly describe childhood stroke classification and then will discuss the imaging of each major stroke subtype individually. The objective is to provide practitioners with a guide for neuroimaging children with various stroke subtypes.

      Pediatric stroke classification

      Childhood stroke is defined as occurring in children aged 29 days after birth to 18 years. Perinatal stroke, defined as stroke occurring from birth to 28 days of life (and in some cases in utero beginning at 20 weeks' gestation), will be discussed in a separate article. Stroke is traditionally subdivided into two types: ischemic and hemorrhagic. As opposed to adults who have ischemic stroke 85% of the time, stroke in children is almost evenly divided between ischemic and hemorrhagic events.
      • Fullerton H.J.
      • Wu Y.W.
      • Zhao S.
      • Johnston S.C.
      Risk of stroke in children: ethnic and gender disparities.
       Ischemic stroke is further subclassified into arterial ischemic stroke (AIS) and cerebral sinovenous thrombosis (CSVT). Childhood AIS is defined as presentation with a focal deficit or seizure that localizes to an ischemic area of brain injury in a known arterial territory. Most children present with hemiplegia, with or without aphasia. CSVT can occur alone or in association with venous infarction or hemorrhage. Isolated cortical vein thrombosis (ICVT) is rare, accounting for less than 1% of all cerebral infarctions.
      • Singh R.
      • Cope W.P.
      • Zhou Z.
      • De Witt M.E.
      • Boockvar J.A.
      • Tsiouris A.J.
      Isolated cortical vein thrombosis: case series.
       Hemorrhagic stroke in children includes spontaneous intracerebral hemorrhage with or without intraventricular extension, intraventricular hemorrhage (IVH), and nontraumatic subarachnoid hemorrhage.

      Childhood AIS

      After childhood AIS, more than 75% of children will suffer long-term neurological deficits and 10% of children will die.
      • De Schryver E.L.
      • Kappelle L.J.
      • Jennekens-Schinkel A.
      • Boudewyn Peters A.C.
      Prognosis of ischemic stroke in childhood: a long-term follow-up study.
      • deVeber G.
      • Roach E.S.
      • Riela A.R.
      • Wiznitzer M.
      Stroke in children: recognition, treatment, and future directions.
      • Ganesan V.
      • Hogan A.
      • Shack N.
      • Gordon A.
      • Isaacs E.
      • Kirkham F.J.
      Outcome after ischaemic stroke in childhood.
      • Fullerton H.J.
      • Chetkovich D.M.
      • Wu Y.W.
      • Smith W.S.
      • Johnston S.C.
      Deaths from stroke in US children, 1979 to 1998.
      • Fullerton H.J.
      • Wu Y.W.
      • Sidney S.
      • Johnston S.C.
      Risk of recurrent childhood arterial ischemic stroke in a population-based cohort: the importance of cerebrovascular imaging.
       Recurrence risk after childhood AIS has been estimated at 12% at one year
      • Fullerton H.J.
      • Wintermark M.
      • Hills N.K.
      • et al.
      Risk of recurrent arterial ischemic stroke in childhood: a Prospective International Study.
      and 19% at five years.
      • Fullerton H.J.
      • Wu Y.W.
      • Sidney S.
      • Johnston S.C.
      Risk of recurrent childhood arterial ischemic stroke in a population-based cohort: the importance of cerebrovascular imaging.
      Approximately 30% of children with AIS encountered in academic centers have an associated cardiac disorder that presumably leads to cardioembolism (Fig 1), whereas cerebral arteriopathy is found in up to half of all children with childhood AIS.
      • Mackay M.T.
      • Wiznitzer M.
      • Benedict S.L.
      • Lee K.J.
      • Deveber G.A.
      • Ganesan V.
      Arterial ischemic stroke risk factors: the International Pediatric Stroke Study.
      • Wintermark M.
      • Hills N.K.
      • deVeber G.A.
      • et al.
      Arteriopathy diagnosis in childhood arterial ischemic stroke: results of the vascular effects of infection in pediatric stroke study.
       The presence of cerebral arteriopathy on neuroimaging also predicts recurrent stroke and stroke after transient ischemic attack (TIA).
      • Fullerton H.J.
      • Wu Y.W.
      • Sidney S.
      • Johnston S.C.
      Risk of recurrent childhood arterial ischemic stroke in a population-based cohort: the importance of cerebrovascular imaging.
      • Lehman L.L.
      • Watson C.G.
      • Kapur K.
      • Danehy A.R.
      • Rivkin M.J.
      Predictors of stroke after transient ischemic attack in children.
      Figure thumbnail gr1
      Figure 1Cardioembolic stroke. (A) Axial diffusion-weighted image (DWI) demonstrates a large left middle cerebral artery (MCA) territory infarct (black arrows) in a one-year-old boy found to have a thrombus within his left ventricle. (B) Three-dimensional time-of-flight magnetic resonance angiography maximum intensity projection image reveals lack of normal flow-related signal involving the left MCA and branch vessels (white asterisk).
      Cervicocephalic arterial dissection, one type of arteriopathy, accounts for 7.5% to 20% of childhood AIS.
      • Rafay M.F.
      • Armstrong D.
      • Deveber G.
      • Domi T.
      • Chan A.
      • MacGregor D.L.
      Craniocervical arterial dissection in children: clinical and radiographic presentation and outcome.
      • Chabrier S.
      • Husson B.
      • Lasjaunias P.
      • Landrieu P.
      • Tardieu M.
      Stroke in childhood: outcome and recurrence risk by mechanism in 59 patients.
      • Amlie-Lefond C.
      • Bernard T.J.
      • Sebire G.
      • et al.
      Predictors of cerebral arteriopathy in children with arterial ischemic stroke: results of the International Pediatric Stroke Study.
       Involvement of the anterior circulation is reported more often than the posterior circulation.
      • Chabrier S.
      • Husson B.
      • Lasjaunias P.
      • Landrieu P.
      • Tardieu M.
      Stroke in childhood: outcome and recurrence risk by mechanism in 59 patients.
      • Fullerton H.J.
      • Johnston S.C.
      • Smith W.S.
      Arterial dissection and stroke in children.
       However, vertebral artery dissection was recently identified in 50% of children with posterior circulation stroke
      • McCrea N.
      • Saunders D.
      • Bagkeris E.
      • Chitre M.
      • Ganesan V.
      Diagnosis of vertebral artery dissection in childhood posterior circulation arterial ischaemic stroke.
      (Fig 2).
      Figure thumbnail gr2
      Figure 2Arterial dissection. (A) Axial diffusion-weighted image shows an infarct of the right cerebellum (black arrows). (B) Three-dimensional time-of-flight magnetic resonance angiography image illustrates focal irregularity of the V3 segment of the right vertebral artery and, to a lesser degree, the contralateral vessel (white arrows). The appearance and location of the vessels involved favor the diagnosis arterial dissection. Other less common entities such as fibromuscular dysplasia can have a similar appearance.
      Focal cerebral arteriopathy of childhood describes a localized intracranial arterial stenosis and accounts for approximately 25% of arteriopathies in childhood AIS.
      • Amlie-Lefond C.
      • Bernard T.J.
      • Sebire G.
      • et al.
      Predictors of cerebral arteriopathy in children with arterial ischemic stroke: results of the International Pediatric Stroke Study.
       It may comprise a monophasic process, also known as “transient cerebral arteriopathy of childhood,” most commonly involving the proximal middle cerebral artery (MCA), that may resolve. Focal cerebral arteriopathy can also be a fixed vasculopathy with no improvement over time (Fig 3).
      • Chabrier S.
      • Rodesch G.
      • Lasjaunias P.
      • Tardieu M.
      • Landrieu P.
      • Sebire G.
      Transient cerebral arteriopathy: a disorder recognized by serial angiograms in children with stroke.
      In contrast, moyamoya arteriopathy represents a progressive steno-occlusive arteriopathy that typically involves the distal internal carotid artery and proximal MCAs or anterior cerebral arteries bilaterally and, much less commonly, the posterior circulation. Ultimately there is development of “moyamoya” collaterals, which create the typical appearance of “puff of smoke” on angiography that inspired the name of this condition (Fig 4).
      Figure thumbnail gr3
      Figure 3Focal cerebral arteriopathy-transient cerebral arteriopathy. (A) Axial diffusion-weighted image demonstrates patchy areas of ischemia involving the right middle cerebral artery (MCA) territory (black arrows). (B) Three-dimensional time-of-flight magnetic resonance angiography (3D TOF MRA) maximum intensity projection (MIP) image reveals focal severe narrowing of the right M1 segment (white arrow) and mild irregularity of the right A1 segment (white arrowhead). (C) Contrast-enhanced black blood vessel wall image illustrates wall enhancement of the right MCA (white arrows). (D) Four month follow-up 3D TOF MRA MIP image shows near-complete resolution of vascular irregularity (white arrow and arrowhead).
      Figure thumbnail gr4
      Figure 4Moyamoya. (A) Three-dimensional time-of-flight magnetic resonance angiography maximum intensity projection image demonstrates lack of flow-related signal involving the right middle cerebral artery with the presence of moyamoya collaterals (white asterisk). (B) Axial fluid attenuated inversion recovery (FLAIR) image illustrates a region of a chronic ischemia within the right periventricular white matter (black arrows).
      Sickle cell anemia is a risk factor for stroke. In studies that predate the STOP trial (Stroke Prevention Trial in sickle cell anemia), 11% of children with sickle cell anemia experienced a stroke before age 20 years.
      • Earley C.J.
      • Kittner S.J.
      • Feeser B.R.
      • et al.
      Stroke in children and sickle-cell disease: Baltimore-Washington Cooperative Young Stroke Study.
      • Ohene-Frempong K.
      • Weiner S.J.
      • Sleeper L.A.
      • et al.
      Cerebrovascular accidents in sickle cell disease: rates and risk factors.
       Not all children with sickle cell disease or sickle cell disease and stroke have vasculopathy (including moyamoya), although many develop steno-occlusive vasculopathy and moyamoya. In a recent study, 43% of children with recurrent strokes on chronic transfusion therapy had moyamoya.
      • Dobson S.R.
      • Holden K.R.
      • Nietert P.J.
      • et al.
      Moyamoya syndrome in childhood sickle cell disease: a predictive factor for recurrent cerebrovascular events.
       Given the robust association between progressive vasculopathy on magnetic resonance angiography (MRA) and new infarcts on magnetic resonance imaging (MRI) in children with sickle cell disease,
      • Hulbert M.L.
      • McKinstry R.C.
      • Lacey J.L.
      • et al.
      Silent cerebral infarcts occur despite regular blood transfusion therapy after first strokes in children with sickle cell disease.
       additional therapies like revascularization surgery should be considered in those with progression of arteriopathy or in whom recurrent stroke occurs despite chronic transfusion therapy.
      • Kennedy B.C.
      • McDowell M.M.
      • Yang P.H.
      • et al.
      Pial synangiosis for moyamoya syndrome in children with sickle cell anemia: a comprehensive review of reported cases.
      • Hall E.M.
      • Leonard J.
      • Smith J.L.
      • et al.
      Reduction in overt and silent stroke recurrence rate following cerebral revascularization surgery in children with sickle cell disease and severe cerebral vasculopathy.
       Unfortunately, there are no evidence-based guidelines about the appropriate timing of revascularization surgery, the degree of arteriopathy progression that should prompt consideration for surgery, or the long-term efficacy of various revascularization procedures in this population. As in other patients with moyamoya, catheter angiogram is typically necessary for the evaluation for revascularization surgery. However, the value of perfusion imaging, particularly if there is bilateral arterial disease or recurrent stroke, is unknown. In the setting of progressive arteriopathy or recurrent strokes and sickle cell anemia, some hematologists will also consider hematopoietic stem cell transplant.
      • Kassim A.A.
      • Galadanci N.A.
      • Pruthi S.
      • DeBaun M.R.
      How I treat and manage strokes in sickle cell disease.
      Central nervous system vasculitis represents a less frequently encountered cause of childhood AIS. Vasculitis can be primary or secondary to a systemic cause such as collagen vascular disease or septic meningitis.
      • Aviv R.I.
      • Benseler S.M.
      • Silverman E.D.
      • et al.
      MR imaging and angiography of primary CNS vasculitis of childhood.
      • Twilt M.
      • Benseler S.M.
      The spectrum of CNS vasculitis in children and adults.
       It is characterized by irregular vascular stenoses that result in both deep and superficial sites of ischemia. Fibromuscular dysplasia, a noninflammatory arteriopathy, rarely presents in childhood and is associated with both ischemic and hemorrhagic strokes in childhood.
      • Kirton A.
      • Crone M.
      • Benseler S.
      • et al.
      Fibromuscular dysplasia and childhood stroke.
       The cervical vasculature is most frequently involved, classically described as having alternating areas of vascular constriction and dilatation.

      Role of imaging in childhood AIS

      Timely diagnosis of acute stroke remains challenging because of the following factors: (1) the wider differential diagnosis in children relative to adults, (2) the relative rarity of stroke in children compared with adults (including a lack of knowledge that children have strokes and that treatment is time-dependent in some circumstances), and (3) challenges to acquisition of urgent diagnostic neuroimaging in children.
      • Rafay M.F.
      • Pontigon A.M.
      • Chiang J.
      • et al.
      Delay to diagnosis in acute pediatric arterial ischemic stroke.
       Twenty percent to more than half of children presenting urgently with stroke-like symptoms will have stroke mimics;
      • Shellhaas R.A.
      • Smith S.E.
      • O'Tool E.
      • Licht D.J.
      • Ichord R.N.
      Mimics of childhood stroke: characteristics of a prospective cohort.
      • Rivkin M.J.
      • deVeber G.
      • Ichord R.N.
      • et al.
      Thrombolysis in pediatric stroke study.
      • Ladner T.R.
      • Mahdi J.
      • Gindville M.C.
      • et al.
      Pediatric acute stroke protocol activation in a children's hospital emergency department.
      therefore, more than in adults, the first question in children is whether the cause of the child's symptoms is a stroke or stroke mimic. In the setting of a stroke, the vascular distribution(s), presence of arterial clot, presence of associated hemorrhage, and presence of acute or chronic arteriopathy direct management. Given that some arteriopathies in childhood affect the cervical vessels and some the intracranial vasculature, vascular imaging of the head and neck is required in most cases.
      Standardization of a stroke imaging protocol in children is challenging because various pediatric issues must be addressed. These include understanding the changing appearance of the developing brain because of ongoing myelination and cortical organization, concern for ionizing radiation from computed tomography (CT) and conventional angiography, the potential need for general anesthesia and/or sedation, presence of metallic orthodonture that can obscure MRI, and varying availability of MRI in the acute setting. Recent advances in adult stroke imaging and treatment protocols have provided an impetus to establish uniform protocols in children, particularly in the hyperacute/acute phase with the potential for thrombolytic therapy with or without endovascular intervention.
      Noncontrast head CT is often the initial study in a child presenting with possible stroke and can rule out intracranial hemorrhage. However, CT has limited sensitivity for the detection of acute childhood AIS and stroke mimics. CT imaging fails to identify the diagnosis in more than 40% of children.
      • Rafay M.F.
      • Pontigon A.M.
      • Chiang J.
      • et al.
      Delay to diagnosis in acute pediatric arterial ischemic stroke.
       Considering this limited sensitivity, the concerns for radiation, and the likelihood of needing MRI to confirm diagnosis, many centers have developed “rapid brain” or hyperacute MRI protocols for stroke that take 15 to 20 minutes. These rapid brain protocols incorporate diffusion-weighted imaging (DWI) and apparent diffusion coefficient (ADC) maps to confirm the diagnosis of stroke as well as susceptibility-weighted imaging (SWI) or gradient echo (GRE) sequences to assess for hemorrhage. If a stroke is present on a rapid brain MRI study, a full protocol with vascular imaging is often required at some point during the initial hospitalization.
      In general, the presence of diffusion restriction (reduced diffusivity) in the distribution of an arterial territory can confirm stroke, although other entities such as brain tumors, abscesses, white matter diseases, and seizures can exhibit reduced diffusivity or demonstrate hyperintense signal abnormality as well. At the time of initial MR scan, an MRA of the head and neck is most often warranted to evaluate for cervical and/or intracranial arteriopathy or arterial obstruction because of thrombus. In a child who is medically unstable, in whom a contraindication to MRI is present, who presents to a center without MRI capabilities, or in whom sedation will delay MRI, CT with CT angiogram (CTA) of the head and neck may be preferable. Alternatively, some centers administer anesthesia by rapid induction for scanning. In patients with a strong suspicion of arteriopathy in whom MRA or CTA is nondiagnostic, conventional catheter angiogram might help elucidate the stroke etiology. In addition, in children with moyamoya disease or syndrome in whom revascularization is being considered, four vessel angiogram including the extracranial circulation may be required.
      With the advent of newer therapeutic techniques and recent guidelines in adult stroke management, there is an increased need to balance comprehensive, but time intensive stroke imaging studies (for example inclusion of MRA neck) with a rapid protocol that allows for quick diagnosis and potentially acute treatment (e.g., thrombolysis). The development of acute therapies that are time-dependent underscores the need for more uniform, consensus-based practical neuroimaging algorithms focused on rapid and accurate diagnosis. These challenges have been addressed to an extent with shorter abbreviated protocols for acute stroke discussed subsequently.
      A suggested imaging protocol for evaluating children with suspected stroke was devised by the International Paediatric Stroke Study (IPSS) Neuroimaging Subgroup (Fig 5). They are based on analysis of current literature, expert opinion, and formal consensus.
      Figure thumbnail gr5
      Figure 5Please refer to the for detailed parameters on suggested imaging. Sequences may be modified as clinically warranted. (The color version of this figure is available in the online edition.)

      Imaging and therapy timing for AIS

      Hyperacute therapy for AIS—intravenous or intra-arterial tissue plasminogen activator (tPA) and mechanical thrombectomy—has not been prospectively studied in children but is used on a case-by-case basis. Consideration of tPA for treatment of acute childhood AIS requires imaging confirmation of the diagnosis of AIS with occlusion or partial occlusion of an artery in the distribution of the stroke or thrombus in a vessel that corresponds to the territory of the clinical deficit.
      • Rivkin M.J.
      • deVeber G.
      • Ichord R.N.
      • et al.
      Thrombolysis in pediatric stroke study.
       In addition, assessment of stroke size on DWI or early change on CT and confirmation that intracranial hemorrhage is not present are required. Assessing AIS onset time is critical for the determination of whether a patient can be treated with tPA or thrombectomy; mismatch between DWI and fluid attenuated inversion recovery (FLAIR) changes may prove useful in determination of the time window for potential intervention.
      • Wouters A.
      • Lemmens R.
      • Dupont P.
      • Thijs V.
      Wake-up stroke and stroke of unknown onset: a critical review.
      • Aoki J.
      • Kimura K.
      • Iguchi Y.
      • Shibazaki K.
      • Sakai K.
      • Iwanaga T.
      FLAIR can estimate the onset time in acute ischemic stroke patients.
       In adults, DWI appears to reliably predict the core infarct
      • Purushotham A.
      • Campbell B.C.
      • Straka M.
      • et al.
      Apparent diffusion coefficient threshold for delineation of ischemic core.
      and the DWI–perfusion-weighted imaging (PWI) mismatch may then be used to assess tissue at risk (penumbra).
      • Wouters A.
      • Lemmens R.
      • Dupont P.
      • Thijs V.
      Wake-up stroke and stroke of unknown onset: a critical review.
       A number of dynamic susceptibility contrast (DSC) perfusion parameters including mean transit time, time to peak, and time to maximum (Tmax) using various thresholds have been used for determination of the ischemic penumbra in adults. Although these measures can be used in children, further validation is needed. Although arterial spin labeling (ASL) is correlated with DSC perfusion parameters in assessment of mismatch in AIS, the correlation is imperfect and improved methods are needed before routine clinical use in acute stroke.
      • Zaharchuk G.
      • El Mogy I.S.
      • Fischbein N.J.
      • Albers G.W.
      Comparison of arterial spin labeling and bolus perfusion-weighted imaging for detecting mismatch in acute stroke.
       In our experience, prominence of veins on SWI is not yet a suitable method for assessing mismatch as it may be seen in the infarct core, and delineation of the region of venous prominence is difficult. Intravenous tPA administration is predicated on rapid diagnosis of stroke by neuroimaging within 4.5 hours of its documented onset in children, just as in adults. As in adults, it may be reasonable to perform noncontrast CT and CTA if MRI with MRA cannot be obtained in a timely fashion and would create delays that would make the child ineligible for therapy.
      In addition to stroke confirmation, neuroimaging can detect stroke risk factors that require prompt treatment such as cervical artery dissection, other large and medium vessel cerebral arteriopathies with risk of vessel-to-vessel emboli, and aneurysm. Surgical revascularization may decrease risk of further stroke in children with moyamoya syndrome or disease but can be associated with increased risk of stroke during the perioperative period. Children with malignant MCA infarction or cerebellar herniation after stroke may be candidates for decompressive craniectomy.
      • Smith S.E.
      • Kirkham F.J.
      • Deveber G.
      • et al.
      Outcome following decompressive craniectomy for malignant middle cerebral artery infarction in children.

      Emerging MRI sequences in childhood AIS

      PWI has been used in adult stroke evaluation to determine tissue at risk (penumbra).
      • Fisher M.
      • Albers G.W.
      Advanced imaging to extend the therapeutic time window of acute ischemic stroke.
       Bolus CT perfusion and MRI perfusion techniques are available, but application has been limited in the setting of childhood AIS. PWI with MR can be performed with contrast, most commonly exploiting DSC technique. ASL, a family of perfusion imaging techniques not requiring the administration of an exogenous contrast but using the patient's blood as an endogenous contrast, are becoming more available and are likely to be applied to pediatrics to a greater degree in the future. However, such studies must be interpreted with caution and by experienced practitioners. ASL, for example, will be altered by a longer blood transit course from the neck arteries to the brain parenchyma; hence, ASL may appear abnormal in a child with moyamoya and collateral circulation even if perfusion is adequate. The utility of ASL in determining the penumbra remains unknown, as it may overestimate the penumbra because of arterial transit time artifacts.
      Vessel wall imaging is a developing technique that may prove useful in the differentiation and monitoring of vasculopathies, particularly the different causes of focal cerebral arteriopathy (Fig 3).
      Black-blood T1-weighted imaging after gadolinium contrast can show abnormal vessel wall enhancement in the setting of an active and potentially inflammatory process, differentiating it from a now quiescent or non-inflammatory process. Imaging before gadolinium contrast can be used to identify an arterial wall hematoma, which may indicate an intracranial dissection.

      Follow-up imaging—subacute and chronic childhood AIS

      Subacute imaging may be performed to assess hemorrhagic transformation, infarct extension, edema formation, mass effect, herniation, and stroke recurrence. MRI is optimal, but CT may be indicated for unstable patients. If a child undergoes thrombolysis, a CT or MRI scan 24 hours later is required for surveillance of intracranial hemorrhage.
      Follow-up imaging is often performed at six weeks to three months after the incident AIS. Follow-up imaging is used to screen for silent infarction, evaluate for progression or improvement of existing arteriopathies, and diagnose arteriopathies that were not obvious at the time of initial stroke diagnosis. MRI with MRA is often the modality of choice in the chronic follow-up setting. Long-term complications after dissection can include pseudoaneurysm development; this can be monitored by MRI/MRA, CT/CTA, and if indicated, cervicocranial catheter angiogram.

      Transient ischemic attack

      TIAs occur in children, although the true incidence remains unclear as TIAs often cannot be distinguished from stroke mimics. The significance of TIA in children has not been as thoroughly evaluated as it has in adults. In adults, stroke occurs within 3 months of a TIA in 10% to 15% of patients with TIAs.
      • Kleindorfer D.
      • Panagos P.
      • Pancioli A.
      • et al.
      Incidence and short-term prognosis of transient ischemic attack in a population-based study.
      • Johnston S.C.
      • Gress D.R.
      • Browner W.S.
      • Sidney S.
      Short-term prognosis after emergency department diagnosis of TIA.
       In a recent retrospective cohort of pediatric patients with childhood AIS, 13% had a stroke after TIA with a mean follow-up period of 4.5 years. Female sex, autoimmune disorders, and presence of arteriopathy were significantly associated with stroke following TIA presentation.
      • Lehman L.L.
      • Watson C.G.
      • Kapur K.
      • Danehy A.R.
      • Rivkin M.J.
      Predictors of stroke after transient ischemic attack in children.
       In particular, children with moyamoya often present with TIAs. Adults who experience a TIA have diffusion abnormalities on MRI at the time of their symptoms in 27% to 40% of cases.
      • Giles M.F.
      • Rothwell P.M.
      Systematic review and pooled analysis of published and unpublished validations of the ABCD and ABCD2 transient ischemic attack risk scores.
      • Giles M.F.
      • Albers G.W.
      • Amarenco P.
      • et al.
      Addition of brain infarction to the ABCD2 score (ABCD2I): a collaborative analysis of unpublished data on 4574 patients.
      • Ay H.
      • Arsava E.M.
      • Johnston S.C.
      • et al.
      Clinical- and imaging-based prediction of stroke risk after transient ischemic attack: the CIP model.
      • Calvet D.
      • Touze E.
      • Oppenheim C.
      • Turc G.
      • Meder J.F.
      • Mas J.L.
      DWI lesions and TIA etiology improve the prediction of stroke after TIA.
       Those adults who have diffusion abnormalities at the time of TIA have an increased risk of stroke in the following 7 days.
      • Giles M.F.
      • Rothwell P.M.
      Systematic review and pooled analysis of published and unpublished validations of the ABCD and ABCD2 transient ischemic attack risk scores.
      • Giles M.F.
      • Albers G.W.
      • Amarenco P.
      • et al.
      Addition of brain infarction to the ABCD2 score (ABCD2I): a collaborative analysis of unpublished data on 4574 patients.
      • Calvet D.
      • Touze E.
      • Oppenheim C.
      • Turc G.
      • Meder J.F.
      • Mas J.L.
      DWI lesions and TIA etiology improve the prediction of stroke after TIA.
       Similar observations in pediatric patients have yet to be reported. Imaging evaluation for children with suspected TIA is the same as for children with AIS.

      CSVT and cortical vein thrombosis

      Childhood CSVT is uncommon, with an incidence of 0.4 to 0.7 per 100,000 children per year.
      • Dlamini N.
      • Billinghurst L.
      • Kirkham F.J.
      Cerebral venous sinus (sinovenous) thrombosis in children.
       However, CSVT is increasingly diagnosed with heightened awareness and increased use of modern neuroimaging. As in adults, etiology of CSVT in children is often multifactorial
      • Heller C.
      • Heinecke A.
      • Junker R.
      • et al.
      Cerebral venous thrombosis in children: a multifactorial origin.
      ; associated risk factors include acute conditions such as infection and trauma, and chronic conditions such as anemia, polycythemia, and prothrombotic disorders. Increased intracranial pressure from obstructed venous outflow can lead to nonspecific symptoms such as headache, encephalopathy, papilledema, or abducens palsies, whereas accompanying hemorrhage or venous infarction can cause seizures or hemiparesis.
      Initial imaging often consists of CT or MRI. Focal brain lesions are found in approximately 40% of children with CSVT
      • Teksam M.
      • Moharir M.
      • Deveber G.
      • Shroff M.
      Frequency and topographic distribution of brain lesions in pediatric cerebral venous thrombosis.
      and include hemorrhage from diapedesis of blood through a congested venous system and ischemia from local compression of arteries and/or reduction of cerebral blood flow from retrograde venous pressure. Vasogenic (increased ADC values) and cytotoxic edema (decreased ADC values) may coexist; thalamic edema is classic for thrombosis of the deep venous system
      • Crombe D.
      • Haven F.
      • Gille M.
      Isolated deep cerebral venous thrombosis diagnosed on CT and MR imaging. A case study and literature review.
      (Fig 6). An atypical or non-arterial pattern of ischemia should prompt further investigation as should ischemia or hemorrhage in the biparietal lobes or bilateral thalami.
      Figure thumbnail gr6
      Figure 6Deep venous thrombosis. (A) Non-contrast sagittal T1-weighted image illustrates abnormal hyperintense signal (thrombus) of the straight sinus, vein of Galen, and internal cerebral veins (white arrows). (B) Axial T2-weighted and (C) axial gradient echo images demonstrate hemorrhagic edema of the thalami bilaterally (black arrows) as well as focal hemorrhage and/or thrombus of the right choroid plexus.
      CT may demonstrate the venous thrombus itself as a hyperdensity within the intracranial dural sinus (“dense clot sign”), and contrast-enhanced CT may reveal a triangular intraluminal filling defect (“empty delta sign”), particularly in the torcular (Fig 7). Multiplanar reformatting can also be helpful.
      Figure thumbnail gr7
      Figure 7Cerebral sinovenous thrombosis. (A) Noncontrast sagittal T1-weighted image reveals abnormal hyperintense signal in the superior sagittal sinus (white arrows). (B) Contrast-enhanced axial T1-weighted image demonstrates the empty delta sign of the superior sagittal sinus with contrast outlining a triangular thrombus (white arrowheads).
      Case series of ICVT report a “cord sign” or “dot sign” on CT in 13% to 51%
      • Singh R.
      • Cope W.P.
      • Zhou Z.
      • De Witt M.E.
      • Boockvar J.A.
      • Tsiouris A.J.
      Isolated cortical vein thrombosis: case series.
      (Fig 8). On CT, high hemoglobin concentration in the setting of dehydration or polycythemia can also be confused with clot, but in these individuals the entire vascular system is dense.
      Figure thumbnail gr8
      Figure 8Cortical vein thrombosis. (A) Axial noncontrast computed tomography image reveals a hematoma with surrounding edema within the left frontal lobe (white arrows) and focal hyperdensity of the superior sagittal sinus (black arrowhead). (B) Contrast-enhanced coronal T1-weighted image confirms the presence of thrombus within the superior sagittal sinus extending into a draining cortical vein (black arrowheads).
      T1- and T2-weighted image findings are variable, depending on the age of the clot. MR SWI is particularly adept at visualizing venous blood and is more sensitive than GRE or T1 spin echo (T1SE) for detecting thrombosis. In one case series, SWI was 90% sensitive for detecting CSVT and 97% sensitive for ICVT within the first week of clinical onset compared with T1SE (71 and 78% sensitive, respectively).
      • Singh R.
      • Cope W.P.
      • Zhou Z.
      • De Witt M.E.
      • Boockvar J.A.
      • Tsiouris A.J.
      Isolated cortical vein thrombosis: case series.
       Three-dimensional volumetric GRE T1-weighted sequences do not suffer from the artifacts that may plague T1SE and are quite adept at demonstrating clot, particularly after the administration of contrast (Fig 8). DWI hyperintensity within the thrombosed sinus has also been described but has poor sensitivity.
      • Favrole P.
      • Guichard J.P.
      • Crassard I.
      • Bousser M.G.
      • Chabriat H.
      Diffusion-weighted imaging of intravascular clots in cerebral venous thrombosis.
       Newer techniques such as black-blood imaging (MRI technique suppressing the signal from normal flowing blood) have demonstrated high sensitivity for venous thrombosis.
      • Yang Q.
      • Duan J.
      • Fan Z.
      • et al.
      Early detection and quantification of cerebral venous thrombosis by magnetic resonance black-blood thrombus imaging.
      Two-dimensional time-of-flight (TOF) MR venography is a sensitive modality for visualizing venous slow flow, particularly flow perpendicular to the plane of acquisition—thus axial, coronal (or any other two planes), and source images should be evaluated to reduce diagnostic error. Three-dimensional phase-contrast magnetic resonance venography (MRV) improves visualization of small veins and of the dural sinuses but has a longer acquisition time, thus rendering it more susceptible to motion artifact. However, newer versions using incomplete k-space acquisition and parallel imaging have made the acquisition times almost comparable with 2D TOF MRV. Follow-up imaging may demonstrate irregular filling of the sinus, indicating incomplete recanalization, or formation of dural collaterals from cortical veins proximal to the site of occlusion. Because MRV indirectly demonstrates clot via impaired flow dynamics, motion artifact is not uncommon (reported in up to 31% of TOF MRV).
      • Yang Q.
      • Duan J.
      • Fan Z.
      • et al.
      Early detection and quantification of cerebral venous thrombosis by magnetic resonance black-blood thrombus imaging.
       Contrast-enhanced MRV can be useful because it can decrease various flow-related artifacts. Vessel hypoplasia, atresia, and arachnoid granulations protruding into the sinus lumen may all be mistaken for thrombosis.
      Although MR may be preferred for its lack of radiation, CT venogram is at least equivalent in sensitivity for CSVT diagnosis
      • Khandelwal N.
      • Agarwal A.
      • Kochhar R.
      • et al.
      Comparison of CT venography with MR venography in cerebral sinovenous thrombosis.
      and is perhaps more reliable for small vessels and the deep venous system.
      • Ozsvath R.R.
      • Casey S.O.
      • Lustrin E.S.
      • Alberico R.A.
      • Hassankhani A.
      • Patel M.
      Cerebral venography: comparison of CT and MR projection venography.
       Directed and timely neuroimaging is essential because failure to recognize CSVT can lead to delayed treatment and poor outcomes.

      Hemorrhagic stroke

      Hemorrhagic stroke refers to nontraumatic intracerebral hemorrhage (with or without intraventricular extension), IVH, and subarachnoid hemorrhage (https://commondataelements.ninds.nih.gov/stroke.aspx#tab=Data_Standards). Childhood hemorrhagic stroke is typically defined as occurring after 28 days of life to age 18 years, and perinatal hemorrhagic stroke (discussed in a separate article) is defined as those occurring in neonates greater than 36 weeks' gestation at birth to 28 days of life to differentiate these from IVH of prematurity. A study from a California-wide discharge database estimated the incidence of hemorrhagic stroke among patients aged one month to 19 years as 1.1 per 100,000 per year.
      • Fullerton H.J.
      • Wu Y.W.
      • Zhao S.
      • Johnston S.C.
      Risk of stroke in children: ethnic and gender disparities.
       The most frequent presenting symptom in children is headache, present in 60% to 80%.
      • Beslow L.A.
      • Licht D.J.
      • Smith S.E.
      • et al.
      Predictors of outcome in childhood intracerebral hemorrhage: a prospective consecutive cohort study.
      • Meyer-Heim A.D.
      • Boltshauser E.
      Spontaneous intracranial haemorrhage in children: aetiology, presentation and outcome.
      • Al-Jarallah A.
      • Al-Rifai M.T.
      • Riela A.R.
      • Roach E.S.
      Nontraumatic brain hemorrhage in children: etiology and presentation.
      • Lo W.D.
      • Lee J.
      • Rusin J.
      • Perkins E.
      • Roach E.S.
      Intracranial hemorrhage in children: an evolving spectrum.
       Other frequently occurring symptoms include altered mental status, nausea and emesis, neck pain, seizures, and focal neurological deficits such as hemiparesis, aphasia, and ataxia. Hydrocephalus can develop rapidly or slowly because of IVH or direct compression by the hemorrhage on the ventricular system. In one study, more than half of children presented acutely but nearly half presented more insidiously,
      • Meyer-Heim A.D.
      • Boltshauser E.
      Spontaneous intracranial haemorrhage in children: aetiology, presentation and outcome.
      which can lead to delays in diagnosis and treatment.
      Vascular malformations, most frequently arteriovenous malformations (AVMs), cavernous malformations, and aneurysms, are the most common causes of pediatric hemorrhagic stroke reported in tertiary care settings
      • Beslow L.A.
      • Licht D.J.
      • Smith S.E.
      • et al.
      Predictors of outcome in childhood intracerebral hemorrhage: a prospective consecutive cohort study.
      • Meyer-Heim A.D.
      • Boltshauser E.
      Spontaneous intracranial haemorrhage in children: aetiology, presentation and outcome.
      (Figure 9, Figure 10, Figure 11); however, other causes of hemorrhage include brain tumors and coagulopathy (acquired or congenital).
      Figure thumbnail gr9
      Figure 9Arteriovenous malformation. (A) Axial noncontrast computed tomography image shows a hematoma within the posterior left periventricular white matter (black asterisk) with extension of blood products into the left lateral ventricle. (B) Anteroposterior digital subtraction angiography of a left vertebral artery injection illustrates a vascular nidus (white asterisk) primarily supplied by the left posterior cerebral artery (black arrow) with early venous drainage (black arrowhead).
      Figure thumbnail gr10
      Figure 10Cavernous malformation. (A) Axial T2-weighted image reveals a mixed signal intensity, popcorn-like cavernous malformation of the pons (white arrows) with surrounding edema because of recent intralesional bleeding. (B) Axial susceptibility-weighted image illustrates “blooming” of the hemosiderin rim surrounding the lesion (black arrows).
      Figure thumbnail gr11
      Figure 11Ruptured aneurysm. (A) Axial noncontrast computed tomography (CT) image demonstrates a right frontal intraparenchymal hematoma (black asterisk) with extension of blood products into the right lateral ventricle and the presence of subarachnoid hemorrhage. (B) Axial contrast-enhanced CT image reveals an aneurysm (black arrow) just posterior to the right frontal hematoma. Note the performance of a right decompressive craniotomy in the interval between the two scans.
      Just as in ischemic stroke, in a child with a presentation concerning for hemorrhagic stroke, rapid neuroimaging is critical for identifying hemorrhage and for differentiating hemorrhage from other stroke subtypes and from stroke mimics. CT is often the first neuroimaging modality performed because of its sensitivity for detecting hemorrhage, its short scan time (almost never needing moderate sedation or anesthesia), and its availability in the emergency setting. Rapid acquisition of neuroimaging is especially important for children with altered mental status or coma or in whom the airway is not stable. However, in a stable and cooperative child, MRI brain with DWI, SWI or GRE, FLAIR, MRA, and MRV can diagnose hemorrhage, differentiate hemorrhagic transformation of arterial or venous infarction from primary hemorrhage, and evaluate the brain parenchyma. Cavernous malformations can sometimes be identified on CT as hyperdense round lesions or are conspicuous on T2*-weighted MRI sequences as a round hypointense area with blooming.
      • Fortuna A.
      • Ferrante L.
      • Mastronardi L.
      • Acqui M.
      • d'Addetta R.
      Cerebral cavernous angioma in children.
      • Lee J.W.
      • Kim D.S.
      • Shim K.W.
      • et al.
      Management of intracranial cavernous malformation in pediatric patients.
       ASL has been useful preoperatively and postoperatively in the setting of pediatric AVMs for detecting focal increased cerebral blood flow through the AVM nidus and within the draining veins.
      • Nabavizadeh S.A.
      • Edgar J.C.
      • Vossough A.
      Utility of susceptibility-weighted imaging and arterial spin perfusion imaging in pediatric brain arteriovenous shunting.
      With suspected or diagnosed hemorrhagic stroke, it is critical to evaluate the integrity of the cerebral vasculature. CTA or MRA can be performed at the time of initial head CT or brain MRI. In many cases, these modalities can diagnose underlying AVMs or aneurysms; however, if no vascular malformation is detected and no hematologic cause or brain tumor is identified, a conventional catheter angiogram should be considered in most cases because CTA and MRA can miss small AVMs and aneurysms, particularly those less than 2 mm in size. During AVM resection, intraoperative catheter angiography can help to identify residual AVM, allowing the surgeon to complete the resection at the time of the initial procedure.
      • Lang S.S.
      • Beslow L.A.
      • Bailey R.L.
      • et al.
      Follow-up imaging to detect recurrence of surgically treated pediatric arteriovenous malformations.
      If no vascular malformation is noted on catheter angiography, repeat neuroimaging should be obtained after the hematoma has resolved because small vascular lesions can be compressed and concealed by the hematoma. In one report of 28 childhood AVMs that were surgically resected, some AVMs were identified up to two years after the incident hemorrhage, and recurrent AVMs were identified in some children even after complete surgical resections.
      • Lang S.S.
      • Beslow L.A.
      • Bailey R.L.
      • et al.
      Follow-up imaging to detect recurrence of surgically treated pediatric arteriovenous malformations.
       Therefore follow-up vascular imaging is recommended in most children in whom no cause for hemorrhage was identified or in whom a vascular lesion like an AVM was treated.
      The timing, modality, and frequency of follow-up imaging are often center-dependent. Some centers obtain noninvasive imaging like MRI with MRA at three months and one year after the incident hemorrhage. Some pediatric centers, like many adult centers, advocate a conventional angiogram in the follow-up period. If this imaging does not show an underlying or recurrent vascular malformation, performance of additional imaging at age five years or 18 years is sometimes performed. More frequent neuroimaging follow-up may be indicated in some children, for example, those with cerebral cavernous malformation gene mutations or multiple cavernous malformations.

      Stroke mimics

      In adults, presentation with sudden onset hemiparesis with or without facial weakness and speech problems are hallmark presenting features, which clinch the diagnosis of stroke without unnecessary delay in most cases. In children, the stroke diagnosis is not as straightforward. Despite a growth in awareness about childhood stroke, when children present with acute neurological deficits, stroke is often not the first diagnosis considered by health personnel. The diagnosis is even more challenging in children who present with nonlocalizing and nonspecific signs, such as headache and vomiting. Also, localizing signs of stroke such as lateralized weakness after seizure or ataxia are often overlooked. In children with AIS, studies report total delay from symptom onset to AIS diagnosis of 16 to 24.8 hours, in-hospital delay of 9.6 to 12.7 hours, and neuroimaging delay of over eight hours.
      • Rafay M.F.
      • Pontigon A.M.
      • Chiang J.
      • et al.
      Delay to diagnosis in acute pediatric arterial ischemic stroke.
      • Gabis L.V.
      • Yangala R.
      • Lenn N.J.
      Time lag to diagnosis of stroke in children.
      • Braun K.P.
      • Rafay M.F.
      • Uiterwaal C.S.
      • Pontigon A.M.
      • DeVeber G.
      Mode of onset predicts etiological diagnosis of arterial ischemic stroke in children.
      • McGlennan C.
      • Ganesan V.
      Delays in investigation and management of acute arterial ischaemic stroke in children.
       In a study of 209 children with acute AIS, 70% of children reached a hospital within six hours of stroke symptom onset, but only 20% were diagnosed with stroke within six hours. Stroke was not suspected in more than 62% of children at initial presentation.
      • Rafay M.F.
      • Pontigon A.M.
      • Chiang J.
      • et al.
      Delay to diagnosis in acute pediatric arterial ischemic stroke.
       Failure to consider stroke in the differential diagnosis of children who present with signs suggesting it continues to delay its diagnosis until well beyond the time that acute interventional therapy can be effectively administered. The relative frequency of other diagnoses that can present similarly to stroke contribute to stroke diagnosis delay in children (Fig 12).
      Figure thumbnail gr12
      Figure 12Hemiplegic migraine. (A) Three-dimensional time-of-flight magnetic resonance angiography maximum intensity projection image illustrates diffuse attenuation of the left middle cerebral artery and branch vessels (white asterisks). (B) Axial susceptibility-weighted image shows increased conspicuity of the left cerebral sulcal veins (white arrowheads) because of increased deoxyhemoglobin concentration in a hypoperfused vascular territory. (C) Multiple axial arterial spin labeled images reveal decreased perfusion to the left cerebral hemisphere (white arrows). (The color version of this figure is available in the online edition.)
      Shellhaas et al.
      • Shellhaas R.A.
      • Smith S.E.
      • O'Tool E.
      • Licht D.J.
      • Ichord R.N.
      Mimics of childhood stroke: characteristics of a prospective cohort.
      reported that among 30 children with stroke mimics, presentations included focal weakness in 14 (47%), seizure in 11 (36%), headache in nine (30%), focal sensory abnormality in seven (23%), and mental status change in six (20%). Most children (63%) had other serious neurological diagnoses (posterior reversible leukoencephalopathy (Fig 13), epilepsy, intracranial infection, inflammation, focal lesions, drug toxicity) (Fig 14); few had benign diagnoses (migraine, psychogenic, and musculoskeletal disorders). In another recent prospective observational study of 124 children who presented to a tertiary pediatric emergency department and in whom a stroke alert was activated, 40% had a stroke or other neurological emergency. Thirty children (24%) had confirmed ischemic strokes, two (1.6%) had a TIA, 20 (17%) had complicated migraine syndromes, 19 (15%) had seizures, five (4%) had meningitis or encephalitis, and four (3%) had intracranial neoplasms.
      • Ladner T.R.
      • Mahdi J.
      • Gindville M.C.
      • et al.
      Pediatric acute stroke protocol activation in a children's hospital emergency department.
      Figure thumbnail gr13
      Figure 13Posterior reversible encephalopathy syndrome (PRES). (A) Axial FLAIR image demonstrates patchy parieto-occipital cortical and subcortical edema (white arrows). (B) Axial diffusion-weighted image shows no restricted diffusion.
      Figure thumbnail gr14
      Figure 14Methotrexate toxicity. (A) Axial diffusion-weighted image and (B) Apparent diffusion coefficient map demonstrate patchy areas of restricted diffusion within the centrum semiovale (white arrows on diffusion-weighted image).

      Future areas of investigation

      Pediatric stroke is a growing field with many avenues to explore. An initial focus should be directed toward aid to centers for streamlining childhood stroke imaging protocols that minimize delays and diagnose stroke syndromes rapidly while at the same time evaluating the vasculature for stroke risk factors. In addition, given that there are many other neurological diseases that can present similarly to stroke, sequences that diagnose stroke mimics should also be included when possible.
      Future areas of investigation in the setting of pediatric stroke include the role of PWI to aid in defining stroke onset and ischemic penumbra in order to help identify children who might be candidates for thrombolysis. Vessel wall imaging may improve diagnosis and characterization of arteriopathies. Understanding the pathophysiology of the various arteriopathies may lead to the rational design of specific treatment plans.
      Ultimately, imaging protocols must be designed and standardized across pediatric centers and must address the challenges of imaging the pediatric brain and of differentiation of stroke syndromes from other entities to inform treatment, clinical trials, and evidence-based guidelines. A benefit to consensus-based neuroimaging is that it will facilitate multicenter treatment trials and allow for research collaborations that address clinical outcomes.
      We would like to thank Gabrielle's Ride for funding this one-day workshop (http://www.gabriellesride.com/).
      We wish to acknowledge the following International Paediatric Stroke Study Members who contributed to the manuscript: Drs. Andrew Demchuk, Christopher Filippi, Heather Fullerton, Manu Goyal, Kristin Guilliams, Manoelle Kossorotoff, Guillaume Sebire, Mukta Sharma, Nicholas Stence, and Arastoo Vossough.
      The Recommended Imaging Protocols for evaluating neonates and children with suspected stroke were devised by the International Paediatric Stroke Study (IPSS) Neuroimaging subgroup and Pediatric Stroke Neuroimaging Consortium, led by Drs. Max Wintermark, Stanford and Michael Rivkin, Boston Children's Hospital.
      Significant contributors to this effort included Aashim Bhatia, Monroe Carell Jr. Children's Hospital at Vanderbilt; Adam Kirton, Alberta Children's Hospital; Allen Newton, Monroe Carell Jr. Children's Hospital at Vanderbilt; Arastoo Vossough, Children's Hospital of Philadelphia; Bruce Bjornson, British Columbia Children's Hospital; Chris Hess, University of California San Francisco; Christine Fox, University of California San Francisco; David Mirsky, Children's Hospital Colorado; Elka Miller, Children's Hospital of Eastern Ontario; Gabrielle deVeber, The Hospital for Sick Children; Helen Carlson, Alberta Children's Hospital; Jerome Rusin, Nationwide Children's Hospital; Jessica Carpenter, Children's National Medical Center; Jonathan Murnick, Children's National Medical Center; Kate Lefond, Seattle Children's Hospital; Khaled Mohamed, Alberta Children's Hospital; Kristin Guilliams, St. Louis Children's Hospital; Lauren Beslow, Yale University School of Medicine; Manoelle Kossorotoff, Hôpital Necker-Enfants Malades; Manu Goyal, St. Louis Children's Hospital; Manu Shroff, The Hospital for Sick Children; Mark Halverson, Nationwide Children's Hospital; Max Wintermark, The Stanford University Medical Center; Michael Rivkin, Boston Children's Hospital; Mubeen Rafay, Winnipeg Children's Hospital; Mukta Sharma, Children's Mercy Hospitals and Clinics; Nany Rollins, University of Texas Southwestern Medical Center; Nicholas Stence, Children's Hospital Colorado; Noma Dlamini, The Hospital for Sick Children; Paola Pergami, Children's National Medical Center; Pradeep Krishnan, The Hospital for Sick Children; Risto Filippi, Northwell Health; Ryan Felling, Johns Hopkins Hospital; Sahar Hassanein, Ains Shams University; Sarah Lee, The Stanford University Medical Center; Suzanne Laughlin, The Hospital for Sick Children; Trish Domi, The Hospital for Sick Children; Warren Lo, Nationwide Children's Hospital; and Wayne Lee, The Hospital for Sick Children.
      Other contributors included Chris Watson, Boston Children's Hospital; Andrew Demchuk, Foothills Medical Center; Sumit Pruthi. Monroe Carell Jr. Children's Hospital at Vanderbilt; Gamil Fteeh, Loma Linda University School of Medicine; Dennis Shaw, Seattle Children's Hospital; and Tim Zinkus, Children's Mercy Hospitals and Clinics.

      Supplementary data

      References

        • Giroud M.
        • Lemesle M.
        • Gouyon J.B.
        • Nivelon J.L.
        • Milan C.
        • Dumas R.
        Cerebrovascular disease in children under 16 years of age in the city of Dijon, France: a study of incidence and clinical features from 1985 to 1993.
        J Clin Epidemiol. 1995; 48: 1343-1348
        • Lynch J.K.
        • Hirtz D.G.
        • DeVeber G.
        • Nelson K.B.
        Report of the National Institute of Neurological Disorders and Stroke workshop on perinatal and childhood stroke.
        Pediatrics. 2002; 109: 116-123
        • Fullerton H.J.
        • Wu Y.W.
        • Zhao S.
        • Johnston S.C.
        Risk of stroke in children: ethnic and gender disparities.
        Neurology. 2003; 61: 189-194
        • Singh R.
        • Cope W.P.
        • Zhou Z.
        • De Witt M.E.
        • Boockvar J.A.
        • Tsiouris A.J.
        Isolated cortical vein thrombosis: case series.
        J Neurosurg. 2015; 123: 427-433
        • De Schryver E.L.
        • Kappelle L.J.
        • Jennekens-Schinkel A.
        • Boudewyn Peters A.C.
        Prognosis of ischemic stroke in childhood: a long-term follow-up study.
        Dev Med Child Neurol. 2000; 42: 313-318
        • deVeber G.
        • Roach E.S.
        • Riela A.R.
        • Wiznitzer M.
        Stroke in children: recognition, treatment, and future directions.
        Semin Pediatr Neurol. 2000; 7: 309-317
        • Ganesan V.
        • Hogan A.
        • Shack N.
        • Gordon A.
        • Isaacs E.
        • Kirkham F.J.
        Outcome after ischaemic stroke in childhood.
        Dev Med Child Neurol. 2000; 42: 455-461
        • Fullerton H.J.
        • Chetkovich D.M.
        • Wu Y.W.
        • Smith W.S.
        • Johnston S.C.
        Deaths from stroke in US children, 1979 to 1998.
        Neurology. 2002; 59: 34-39
        • Fullerton H.J.
        • Wu Y.W.
        • Sidney S.
        • Johnston S.C.
        Risk of recurrent childhood arterial ischemic stroke in a population-based cohort: the importance of cerebrovascular imaging.
        Pediatrics. 2007; 119: 495-501
        • Fullerton H.J.
        • Wintermark M.
        • Hills N.K.
        • et al.
        Risk of recurrent arterial ischemic stroke in childhood: a Prospective International Study.
        Stroke. 2016; 47: 53-59
        • Mackay M.T.
        • Wiznitzer M.
        • Benedict S.L.
        • Lee K.J.
        • Deveber G.A.
        • Ganesan V.
        Arterial ischemic stroke risk factors: the International Pediatric Stroke Study.
        Ann Neurol. 2011; 69: 130-140
        • Wintermark M.
        • Hills N.K.
        • deVeber G.A.
        • et al.
        Arteriopathy diagnosis in childhood arterial ischemic stroke: results of the vascular effects of infection in pediatric stroke study.
        Stroke. 2014; 45: 3597-3605
        • Lehman L.L.
        • Watson C.G.
        • Kapur K.
        • Danehy A.R.
        • Rivkin M.J.
        Predictors of stroke after transient ischemic attack in children.
        Stroke. 2016; 47: 88-93
        • Rafay M.F.
        • Armstrong D.
        • Deveber G.
        • Domi T.
        • Chan A.
        • MacGregor D.L.
        Craniocervical arterial dissection in children: clinical and radiographic presentation and outcome.
        J Child Neurol. 2006; 21: 8-16
        • Chabrier S.
        • Husson B.
        • Lasjaunias P.
        • Landrieu P.
        • Tardieu M.
        Stroke in childhood: outcome and recurrence risk by mechanism in 59 patients.
        J Child Neurol. 2000; 15: 290-294
        • Amlie-Lefond C.
        • Bernard T.J.
        • Sebire G.
        • et al.
        Predictors of cerebral arteriopathy in children with arterial ischemic stroke: results of the International Pediatric Stroke Study.
        Circulation. 2009; 119: 1417-1423
        • Fullerton H.J.
        • Johnston S.C.
        • Smith W.S.
        Arterial dissection and stroke in children.
        Neurology. 2001; 57: 1155-1160
        • McCrea N.
        • Saunders D.
        • Bagkeris E.
        • Chitre M.
        • Ganesan V.
        Diagnosis of vertebral artery dissection in childhood posterior circulation arterial ischaemic stroke.
        Dev Med Child Neurol. 2016; 58: 63-69
        • Chabrier S.
        • Rodesch G.
        • Lasjaunias P.
        • Tardieu M.
        • Landrieu P.
        • Sebire G.
        Transient cerebral arteriopathy: a disorder recognized by serial angiograms in children with stroke.
        J Child Neurol. 1998; 13: 27-32
        • Earley C.J.
        • Kittner S.J.
        • Feeser B.R.
        • et al.
        Stroke in children and sickle-cell disease: Baltimore-Washington Cooperative Young Stroke Study.
        Neurology. 1998; 51: 169-176
        • Ohene-Frempong K.
        • Weiner S.J.
        • Sleeper L.A.
        • et al.
        Cerebrovascular accidents in sickle cell disease: rates and risk factors.
        Blood. 1998; 91: 288-294
        • Dobson S.R.
        • Holden K.R.
        • Nietert P.J.
        • et al.
        Moyamoya syndrome in childhood sickle cell disease: a predictive factor for recurrent cerebrovascular events.
        Blood. 2002; 99: 3144-3150
        • Hulbert M.L.
        • McKinstry R.C.
        • Lacey J.L.
        • et al.
        Silent cerebral infarcts occur despite regular blood transfusion therapy after first strokes in children with sickle cell disease.
        Blood. 2011; 117: 772-779
        • Kennedy B.C.
        • McDowell M.M.
        • Yang P.H.
        • et al.
        Pial synangiosis for moyamoya syndrome in children with sickle cell anemia: a comprehensive review of reported cases.
        Neurosurg Focus. 2014; 36: E12
        • Hall E.M.
        • Leonard J.
        • Smith J.L.
        • et al.
        Reduction in overt and silent stroke recurrence rate following cerebral revascularization surgery in children with sickle cell disease and severe cerebral vasculopathy.
        Pediatr Blood Cancer. 2016; 63: 1431-1437
        • Kassim A.A.
        • Galadanci N.A.
        • Pruthi S.
        • DeBaun M.R.
        How I treat and manage strokes in sickle cell disease.
        Blood. 2015; 125: 3401-3410
        • Aviv R.I.
        • Benseler S.M.
        • Silverman E.D.
        • et al.
        MR imaging and angiography of primary CNS vasculitis of childhood.
        AJNR Am J Neuroradiol. 2006; 27: 192-199
        • Twilt M.
        • Benseler S.M.
        The spectrum of CNS vasculitis in children and adults.
        Nat Rev Rheumatol. 2011; 8: 97-107
        • Kirton A.
        • Crone M.
        • Benseler S.
        • et al.
        Fibromuscular dysplasia and childhood stroke.
        Brain. 2013; 136: 1846-1856
        • Rafay M.F.
        • Pontigon A.M.
        • Chiang J.
        • et al.
        Delay to diagnosis in acute pediatric arterial ischemic stroke.
        Stroke. 2009; 40: 58-64
        • Shellhaas R.A.
        • Smith S.E.
        • O'Tool E.
        • Licht D.J.
        • Ichord R.N.
        Mimics of childhood stroke: characteristics of a prospective cohort.
        Pediatrics. 2006; 118: 704-709
        • Rivkin M.J.
        • deVeber G.
        • Ichord R.N.
        • et al.
        Thrombolysis in pediatric stroke study.
        Stroke. 2015; 46: 880-885
        • Ladner T.R.
        • Mahdi J.
        • Gindville M.C.
        • et al.
        Pediatric acute stroke protocol activation in a children's hospital emergency department.
        Stroke. 2015; 46: 2328-2331
        • Wouters A.
        • Lemmens R.
        • Dupont P.
        • Thijs V.
        Wake-up stroke and stroke of unknown onset: a critical review.
        Front Neurol. 2014; 5: 153
        • Aoki J.
        • Kimura K.
        • Iguchi Y.
        • Shibazaki K.
        • Sakai K.
        • Iwanaga T.
        FLAIR can estimate the onset time in acute ischemic stroke patients.
        J Neurol Sci. 2010; 293: 39-44
        • Purushotham A.
        • Campbell B.C.
        • Straka M.
        • et al.
        Apparent diffusion coefficient threshold for delineation of ischemic core.
        Int J Stroke. 2015; 10: 348-353
        • Zaharchuk G.
        • El Mogy I.S.
        • Fischbein N.J.
        • Albers G.W.
        Comparison of arterial spin labeling and bolus perfusion-weighted imaging for detecting mismatch in acute stroke.
        Stroke. 2012; 43: 1843-1848
        • Smith S.E.
        • Kirkham F.J.
        • Deveber G.
        • et al.
        Outcome following decompressive craniectomy for malignant middle cerebral artery infarction in children.
        Dev Med Child Neurol. 2011; 53: 29-33
        • Fisher M.
        • Albers G.W.
        Advanced imaging to extend the therapeutic time window of acute ischemic stroke.
        Ann Neurol. 2013; 73: 4-9
        • Kleindorfer D.
        • Panagos P.
        • Pancioli A.
        • et al.
        Incidence and short-term prognosis of transient ischemic attack in a population-based study.
        Stroke. 2005; 36: 720-723
        • Johnston S.C.
        • Gress D.R.
        • Browner W.S.
        • Sidney S.
        Short-term prognosis after emergency department diagnosis of TIA.
        JAMA. 2000; 284: 2901-2906
        • Giles M.F.
        • Rothwell P.M.
        Systematic review and pooled analysis of published and unpublished validations of the ABCD and ABCD2 transient ischemic attack risk scores.
        Stroke. 2010; 41: 667-673
        • Giles M.F.
        • Albers G.W.
        • Amarenco P.
        • et al.
        Addition of brain infarction to the ABCD2 score (ABCD2I): a collaborative analysis of unpublished data on 4574 patients.
        Stroke. 2010; 41: 1907-1913
        • Ay H.
        • Arsava E.M.
        • Johnston S.C.
        • et al.
        Clinical- and imaging-based prediction of stroke risk after transient ischemic attack: the CIP model.
        Stroke. 2009; 40: 181-186
        • Calvet D.
        • Touze E.
        • Oppenheim C.
        • Turc G.
        • Meder J.F.
        • Mas J.L.
        DWI lesions and TIA etiology improve the prediction of stroke after TIA.
        Stroke. 2009; 40: 187-192
        • Dlamini N.
        • Billinghurst L.
        • Kirkham F.J.
        Cerebral venous sinus (sinovenous) thrombosis in children.
        Neurosurg Clin N Am. 2010; 21: 511-527
        • Heller C.
        • Heinecke A.
        • Junker R.
        • et al.
        Cerebral venous thrombosis in children: a multifactorial origin.
        Circulation. 2003; 108: 1362-1367
        • Teksam M.
        • Moharir M.
        • Deveber G.
        • Shroff M.
        Frequency and topographic distribution of brain lesions in pediatric cerebral venous thrombosis.
        AJNR Am J Neuroradiol. 2008; 29: 1961-1965
        • Crombe D.
        • Haven F.
        • Gille M.
        Isolated deep cerebral venous thrombosis diagnosed on CT and MR imaging. A case study and literature review.
        JBR-BTR. 2003; 86: 257-261
        • Favrole P.
        • Guichard J.P.
        • Crassard I.
        • Bousser M.G.
        • Chabriat H.
        Diffusion-weighted imaging of intravascular clots in cerebral venous thrombosis.
        Stroke. 2004; 35: 99-103
        • Yang Q.
        • Duan J.
        • Fan Z.
        • et al.
        Early detection and quantification of cerebral venous thrombosis by magnetic resonance black-blood thrombus imaging.
        Stroke. 2016; 47: 404-409
        • Khandelwal N.
        • Agarwal A.
        • Kochhar R.
        • et al.
        Comparison of CT venography with MR venography in cerebral sinovenous thrombosis.
        AJR Am J Roentgenol. 2006; 187: 1637-1643
        • Ozsvath R.R.
        • Casey S.O.
        • Lustrin E.S.
        • Alberico R.A.
        • Hassankhani A.
        • Patel M.
        Cerebral venography: comparison of CT and MR projection venography.
        AJR Am J Roentgenol. 1997; 169: 1699-1707
        • Beslow L.A.
        • Licht D.J.
        • Smith S.E.
        • et al.
        Predictors of outcome in childhood intracerebral hemorrhage: a prospective consecutive cohort study.
        Stroke. 2010; 41: 313-318
        • Meyer-Heim A.D.
        • Boltshauser E.
        Spontaneous intracranial haemorrhage in children: aetiology, presentation and outcome.
        Brain Dev. 2003; 25: 416-421
        • Al-Jarallah A.
        • Al-Rifai M.T.
        • Riela A.R.
        • Roach E.S.
        Nontraumatic brain hemorrhage in children: etiology and presentation.
        J Child Neurol. 2000; 15: 284-289
        • Lo W.D.
        • Lee J.
        • Rusin J.
        • Perkins E.
        • Roach E.S.
        Intracranial hemorrhage in children: an evolving spectrum.
        Arch Neurol. 2008; 65: 1629-1633
        • Fortuna A.
        • Ferrante L.
        • Mastronardi L.
        • Acqui M.
        • d'Addetta R.
        Cerebral cavernous angioma in children.
        Childs Nerv Syst. 1989; 5: 201-207
        • Lee J.W.
        • Kim D.S.
        • Shim K.W.
        • et al.
        Management of intracranial cavernous malformation in pediatric patients.
        Childs Nerv Syst. 2008; 24: 321-327
        • Nabavizadeh S.A.
        • Edgar J.C.
        • Vossough A.
        Utility of susceptibility-weighted imaging and arterial spin perfusion imaging in pediatric brain arteriovenous shunting.
        Neuroradiology. 2014; 56: 877-884
        • Lang S.S.
        • Beslow L.A.
        • Bailey R.L.
        • et al.
        Follow-up imaging to detect recurrence of surgically treated pediatric arteriovenous malformations.
        J Neurosurg Pediatr. 2012; 9: 497-504
        • Gabis L.V.
        • Yangala R.
        • Lenn N.J.
        Time lag to diagnosis of stroke in children.
        Pediatrics. 2002; 110: 924-928
        • Braun K.P.
        • Rafay M.F.
        • Uiterwaal C.S.
        • Pontigon A.M.
        • DeVeber G.
        Mode of onset predicts etiological diagnosis of arterial ischemic stroke in children.
        Stroke. 2007; 38: 298-302
        • McGlennan C.
        • Ganesan V.
        Delays in investigation and management of acute arterial ischaemic stroke in children.
        Dev Med Child Neurol. 2008; 50: 537-540