Published online before print July 17, 2008, doi: 10.1161/STROKEAHA.108.189696
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
(Stroke. 2008;39:2644.)
© 2008 American Heart Association, Inc.
AHA Scientific Statement |
Management of Stroke in Infants and Children
A Scientific Statement From a Special Writing Group of the American Heart Association Stroke Council and the Council on Cardiovascular Disease in the Young
 |
Abstract |
|---|
Top
Abstract
IntroductionPurpose— The purpose of this statement is to review the literature on childhood stroke and to provide recommendations for optimal diagnosis and treatment. This statement is intended for physicians who are responsible for diagnosing and treating infants, children, and adolescents with cerebrovascular disease.
Methods— The Writing Group members were appointed by the American Heart Association Stroke Council’s Scientific Statement Oversight Committee. The panel included members with several different areas of expertise. Each of the panel’s recommendations was weighted by applying the American Heart Association Stroke Council’s Levels of Evidence grading algorithm. After being reviewed by panel members, the manuscript was reviewed by 4 expert peer reviewers and by members of the Stroke Council Leadership Committee and was approved by the American Heart Association Science Advisory and Coordinating Committee. We anticipate that this statement will need to be updated in 4 years.
Results— Evidence-based recommendations are provided for the prevention of ischemic stroke caused by sickle cell disease, moyamoya disease, cervicocephalic arterial dissection, and cardiogenic embolism. Recommendations on the evaluation and management of hemorrhagic stroke also are provided. Protocols for dosing of heparin and warfarin in children are suggested. Also included are recommendations on the evaluation and management of perinatal stroke and cerebral sinovenous thrombosis in children.
Key Words: AHA Scientific Statements • children • stroke
|
Introduction |
|---|
Stroke has been increasingly recognized in children in recent years, but diagnosis and management can be difficult because of the diversity of underlying risk factors and the absence of a uniform treatment approach. Children and adolescents with stroke have remarkable differences in presentation compared with older patients. Stroke type also varies according to age. In Western countries, 80% to 85% of strokes among adults are ischemic, and the rest are hemorrhagic. In children,
55% of strokes are ischemic, and the remainder are hemorrhagic. The World Health Organization definition of stroke (a clinical syndrome of rapidly developing focal or global disturbance of brain function lasting >24 hours or leading to death with no obvious nonvascular cause) is far from ideal for children. Children with symptoms compatible with a transient ischemic attack (TIA), for example, commonly have a brain infarction shown by brain imaging despite the transient nature of their symptoms. Children with cerebral venous sinus thrombosis (CVST) commonly present with headache or seizures. "Stroke-like episodes" without an obvious vascular cause may occur in migraine or metabolic disease but may require specific treatment. Prior illness (eg, infection) or events (eg, head trauma) need not preclude a diagnosis of stroke. Although extra-axial hematomas, neonatal intraventricular hemorrhages (IVHs), and periventricular leukomalacia arise from cerebrovascular dysfunction in a broad sense, they are not considered in detail here.
Our purpose is to review the literature on stroke in children and, whenever possible, to make recommendations for the diagnosis and management of these children. Writing Group members were appointed by the American Heart Association Stroke Council’s Scientific Statement Oversight Committee. The panel included members with several different areas of expertise. The panel reviewed relevant articles on stroke in children using a computerized search of the medical literature through 2006. These articles were supplemented by other articles known to the authors. Each recommendation was weighted by applying the American Heart Association Stroke Council’s Levels of Evidence grading algorithm (Table 1 and the Figure). After being reviewed by the panel members, the manuscript was reviewed by 4 expert peer reviewers and members of the Stroke Council Leadership Committee and was subsequently approved by the American Heart Association’s Science Advisory and Coordinating Committee.
|
|
Although some information about the cause and clinical manifestations of childhood stroke is included for the convenience of readers who may be unfamiliar with these topics, the group’s recommendations emphasize issues regarding treatment. The recommendations in this article represent a consensus of the authors. Because some aspects of stroke in children have been studied more thoroughly than others, some topics receive more attention than others. Despite major progress in the study of stroke in children in recent years, much of the literature remains descriptive. Continued research is essential if we are to better understand the diagnosis and treatment of stroke in children.
|
Overview of the Cause of Childhood Stroke |
|---|
About half of the children presenting with an acute focal neurological deficit have a previously identified risk factor, and
1 additional risk factors often are uncovered in the remaining patients.1 For arterial ischemic stroke, the most common underlying conditions are sickle cell disease (SCD) and congenital or acquired heart disease. Heart disease and chronic anemia (including SCD and β-thalassemia) also are risk factors for CVST, but the list of associated conditions ranges from head and neck infections to systemic conditions such as inflammatory bowel disease and autoimmune disorders.2 Head trauma appears to be a trigger for arterial stroke1,3 and dehydration for venous stroke,4–6 whereas infections, including varicella, meningitis, tonsillitis, and otitis media,1 and anemia, leukocytosis, and prothrombotic disorders are probably risk factors for both.1,5–7 It is increasingly evident that many children have multiple risk factors that together determine the risk of stroke or stroke recurrence. Various factors influencing stroke recurrence risk have been documented in individuals with both symptomatic and idiopathic stroke.5 Some arteriopathies are transient,8 and occluded venous sinuses often recanalize.5,6 An estimated 10% of intracranial hemorrhages (ICHs) in the young result from CVST.
|
Epidemiology of Childhood Stroke |
|---|
After excluding neonatal strokes and strokes related to trauma and infection, Schoenberg et al9 found 3 hemorrhagic strokes and 1 ischemic stroke among 15 834 children in Rochester, Minn, between 1965 and 1974. Their estimated annual stroke incidence rate for children <15 years of age was 2.52 per 100 000 per year or 1.89 per 100 000 per year and 0.63 per 100 000 per year for hemorrhagic and ischemic strokes, respectively. In this population, hemorrhagic strokes occurred more often than ischemic strokes, whereas in the Mayo Clinic referral population, ischemic strokes were more common. Another study found a similar overall stroke incidence of 2.7 cases per 100 000 per year in the greater Cincinnati, Ohio, area,10 although their combined incidence rate for intraparenchymal brain hemorrhage and subarachnoid hemorrhage (SAH), 1.5 cases per 100 000 per year, was similar to the incidence of ischemic infarction (1.2 cases in 100 000 per year).10
More recent US studies have found a similar incidence,11 although a few studies report higher rates.12 Analyzing California hospital discharge data for a 10-year interval for children 1 month through 19 years of age, Fullerton and colleagues11 estimated the stroke incidence to be 2.3 per 100 000 children per year (1.2 per 100 000 per year for ischemic lesions and 1.1 per 100 000 per year for hemorrhagic lesions). Boys were at higher risk than girls, and black children were at higher risk than white and Asian children, even after adjustment for trauma and the presence of SCD.11 A study of Chinese children in Hong Kong found a similar overall stroke risk (2.1 per 100 000 per year), but only 28% of those children had hemorrhagic strokes.13 Peak age for both ischemic stroke and intraparenchymal brain hemorrhage is the first year of life, with a third of the cases presenting in this age group, whereas SAH is more common among teenagers.11 There also appears to be an excess of strokes in boys and in those of black ethnicity; this excess is not fully explained by the prevalence of SCD in this population.14
The frequency of ischemic stroke may be greater than previously suggested. Giroud and colleagues15 calculated the stroke risk among individuals <16 years of age to be 13 per 100 000 per year. Their estimated incidence of hemorrhagic lesions in children, however, was only slightly higher than in earlier reports. Data from the National Hospital Discharge Survey from 1980 to 1998 indicate that the risk of ischemic stroke in individuals from birth through 18 years of age is 7.8 per 100 000, with a hemorrhagic stroke risk of 2.9 per 100 000.16
Recent estimates suggest that ischemic stroke occurs in 1 per 4000 live births,17 clearly a much higher rate than in older children. Approximately 80% of these are ischemic, and the rest are due to CVST or hemorrhage (excluding SAH and IVH in premature babies). Schoenberg18 estimated that the incidence of nontraumatic perinatal ICH in neonates was 1.1 per 1000 live births. However, this study was done before the widespread use of computed tomography (CT), cranial ultrasound, and magnetic resonance imaging (MRI); the fact that all 12 identified babies had an autopsy suggests that only the most severe hemorrhages were identified. However, more recent (1980 to 1998) data from the National Hospital Discharge Survey indicate that the rate of hemorrhagic stroke for term infants is 6.7 per 100 000 per year.16
The incidence of childhood CVST is 0.3 per 100 000 children per year for term birth to 18 years of age, and neonates make up 43% of the patients.4
|
Arterial Ischemic Stroke |
|---|
Perinatal Stroke
Definitions and Risk Factors
Neonatal stroke encompasses both ischemic and hemorrhagic events resulting from disruption of either arteries or veins from early gestation through the first month of life. The term perinatal stroke describes cerebrovascular lesions that occur from 28 weeks’ gestation through the first 7 days of life, although some authors broaden this range from 20 weeks’ gestation to 28 days after birth, and lesions occurring even before 20 weeks have been documented.17,19 Approximately 80% of these are ischemic, and the remainder are due to CVST or hemorrhage. Risk factors for neonatal stroke include cardiac disorders, coagulation disorders, infection, trauma, drugs, maternal and placental disorders, and perinatal asphyxia.17 Multiple risk factors have been documented, especially blood disorders with asphyxial stress.20–22 Suspected maternal risk factors for perinatal arterial ischemic stroke include a history of infertility, chorioamnionitis, premature rupture of membranes, and preeclampsia.23 In fact, the rate of perinatal arterial ischemic stroke increased dramatically with the increasing number of risk factors in population-based studies.
Clinical Presentation
Both neonatal arterial and venous strokes often present with seizures, typically focal motor seizures involving only 1 extremity. Stroke accounts for an estimated 10% of the seizures in term neonates. Some children with perinatal stroke do not present in the neonatal period and appear quite normal at this stage. These children usually present with early handedness or developmental delay, suggesting a process that began the first few months of life.
Diagnostic Evaluation
The optimal imaging study is somewhat dependent on the child’s clinical stability. Cranial ultrasound is safe and readily available, but it may miss superficial and ischemic lesions.24 CT is relatively quick, accurately depicts superficial or hemorrhagic lesions, and confirms the lesion location. However, venous thrombosis and early arterial ischemic stroke (AIS) are easily missed with CT.
Like ultrasound, MRI techniques do not expose the neonate to the potentially harmful effects of ionizing radiation. MRI, magnetic resonance angiography (MRA), and magnetic resonance venography (MRV) may more accurately define the site of an arterial or venous occlusion. Additionally, MR studies often demonstrate associated parenchymal abnormalities more clearly, including nonischemic lesions that clinically mimic arterial or venous stroke. Diffusion-weighted imaging can confirm the presence and location of an infarction earlier than other MRI sequences or CT.21 CT angiography (CTA) is an accurate means of identifying primary vascular abnormalities when there is an unexplained hemorrhagic lesion. Catheter angiography (CA) is technically more difficult in babies and tends to be done only when endovascular surgical intervention is anticipated.
Residual Effects of Perinatal Stroke
Children frequently have significant long-term disabilities after a perinatal stroke, including cognitive and sensory impairments, cerebral palsy, and epilepsy. The few studies of perinatal stroke outcome tend to be limited by small sample size, heterogeneity of age of onset (eg, neonates with perinatal stroke frequently are combined with older children with stroke), or confounding by other conditions such as hypoxic-ischemic encephalopathy. Only studies that include neonates with clearly defined AIS, CVST, or ICH and outcome data are discussed here.
Cognitive impairment after perinatal AIS ranges from 0% to 55%.20,25–29 One study of 40 children with perinatal stroke found language delay in 25%.29 A study of 29 children with unilateral perinatal stroke reported that all children >2 years of age had "normal" IQ scores.26 Estimates of generalized "developmental delay" after neonatal CVST range from 28%30 to 58%.31 A study of 19 children with neonatal CVST reported a learning disability in only 1 child (5.3%).31 One report of 27 children with neonatal sinovenous thrombosis found cognitive impairment in 16 (59%), and 9 of the 27 had moderate to severe dysfunction.32 Another study of 11 children with spontaneous ICH reported cognitive delay or speech impairment in 4 (36%).33
Sensory function in young children is difficult to assess. Children with neonatal middle cerebral artery (MCA) territory infarction may develop thalamic atrophy, but whether this has long-term implications for sensory perception is unclear.34 In a study of 12 children and young adults with unilateral perinatal stroke, 4 had abnormal somatosensory-evoked potentials, but these children were unable to cooperate for further evaluation of sensory function.35 In a study of 16 school-aged children with perinatal stroke, 6 (28%) had impaired visual function.36 Several studies of children with unilateral prenatal or perinatal strokes have described increased difficulty with facial recognition37 and other visuospatial tasks in small cohorts.38–40
Estimates of the incidence of cerebral palsy after perinatal stroke vary widely: 9% to 88% after perinatal AIS26,29,41–47 and 6% to 67% after perinatal CVST.4,30,32,44 Outcome of children with perinatal stroke ranges from normal to subtle hemiplegia to severe quadriplegia. However, most children with neonatal AIS and CVST learn to walk independently, usually before 2 years of age.48 Although many neonates with AIS, CVST, or ICH present with seizures,4,30 most do not have epilepsy after the neonatal period. Estimates of the incidence of epilepsy past the neonatal period range from 0% to 46% for neonatal AIS20,25,27,29,43 and 6% to 41% for neonatal CVST.4,32,49 A small portion of perinatal stroke patients undergo surgery because of severe epilepsy syndromes such as infantile spasms50 or intractable epilepsy.51
Outcome
There are few studies of stroke recurrence risk after neonatal stroke. One study followed 215 children with neonatal AIS for a median of 3.5 years. Recurrent symptomatic thromboembolism occurred in 7, AIS in 4, CVST in 2, and deep venous thrombosis in 1 individual. Factors associated with an increased recurrence rate included thrombophilic states and the presence of additional comorbidities such as complex congenital heart disease or dehydration.52 Few data exist on the risk of recurrent thromboembolism after neonatal CVST. In the Canadian Registry, 5 of 61 neonates (8%) had recurrence.4 Prothrombotic risk factors may modify presentation and severity in children with bleeding disorders. Children who have both hemophilia A and a thrombophilic state may have a lower risk of hemorrhage as a neonate and less likelihood of recurrent ICH.53
One study of 46 neonates with AIS noted that vascular territory is not closely correlated with outcome.27 Another study of 24 children with perinatal stroke found that lesions involving the cortex, basal ganglia, and internal capsule on MRI were more likely to cause hemiplegia than strokes involving only one of these regions.42 A study of 40 children with perinatal stroke found that large stroke size and injury to Broca’s area, the internal capsule, Wernicke’s area, or basal ganglia were associated with cerebral palsy.29 In another cohort of 23 preterm and term neonates with perinatal AIS, infarction of an area supplied by a main branch of the MCA predicted subsequent hemiplegia.46 A report of 62 children with neonatal AIS and 25 children with neonatal CVST noted that bilateral infarctions decreased the likelihood of walking.48 Neonates with CVST without infarction have a better outcome than those with both CVST and infarction.4,32,48,54
Although most neonates with stroke do not subsequently develop epilepsy, those who present with seizures may be at a higher risk of an abnormal neurodevelopmental outcome. In 1 cohort of 46 neonates with AIS, the presence of seizures in the neonatal period predicted the development of 1 disabilities in the first years of life.27 Another study of 24 children with perinatal AIS found that an abnormal electroencephalography background predicted childhood hemiplegia.42,55
Neonatal encephalopathy may predict poor outcome after AIS.56 However, children who appear normal in the neonatal period but develop a hand preference or seizures after 2 months of age resulting from perinatal AIS may have a worse prognosis than children who have neurological signs as neonates. The presenting findings (hemiparesis or seizures) may be more likely to persist.20,29 Prothrombotic disorders may lead to poorer outcome after neonatal AIS.55 Not surprisingly, term neonates who develop idiopathic ICH or IVH appear to have a better prognosis than those with ICH associated with hypoxic-ischemic insults or trauma.57
Management
Acute
Supportive care is important for all types of perinatal stroke. Anecdotal reports suggest that surgical evacuation of a hematoma may reduce extremely high intracranial pressure, but it is not clear whether surgery improves the eventual outcome. Ventricular drainage and, if indicated, later shunting for progressive hydrocephalus caused by IVH is appropriate.33
The optimal therapy for neonates with AIS or CVST has not been determined. Thrombolytic therapy has been used for peripheral thrombi in neonates but rarely for neonatal CVST, so neither its safety nor its effectiveness has been ascertained in these patients. Neither unfractionated heparin (UFH) nor low-molecular-weight heparin (LMWH) is used widely in children with perinatal AIS, although children with severe prothrombotic disorders or with cardiac or multiple systemic thrombi may benefit. No major complications occurred in preliminary studies of LMWH in neonates with CVST,58 but it is not clear whether anticoagulation is beneficial in these neonates, except in the setting of multiple thrombosed sinuses and radiological evidence of propagating thrombosis despite supportive therapy. Case reports also have described the use of antithrombin concentrate59,60 and protein C concentrate61 to prevent venous thrombosis in neonates with congenital or iatrogenic coagulation factor deficiencies.
Markedly low platelet counts and factor deficiencies should be corrected. Vitamin K deficiency may be an issue in areas of the world where vitamin K is not routinely administered to newborns,62 in infants with biliary atresia, or in babies whose mothers ingested warfarin, phenytoin, or barbiturates during pregnancy. Large doses of vitamin K may be needed to correct factor deficiencies induced by maternal medications.63,64
Chronic
Many studies have documented a positive effect of rehabilitation on outcomes in children with cerebral palsy,65–70 but these studies have not specifically described the subset of children with perinatal stroke. Children who need rehabilitation may be at higher risk of poor outcome.27,44 A study of 18 children with hemiplegic cerebral palsy from several causes suggested that constraint of the normal arm led to increased use of the weak arm.71
There is little information about the long-term use of prophylactic therapies such as LMWH in neonates. Although recurrent stroke is uncommon in these patients, individuals with prothrombotic conditions plus other risk factors (eg, complex congenital heart disease, dehydration, or prolonged bed rest) may have a higher likelihood of recurrent venous and arterial thrombosis, and prophylaxis may be considered in these indviduals.52 It is reasonable to supplement folate and B vitamins for children with a methylenetetrahydrofolate reductase (MTHFR) mutation in an effort to normalize homocysteine levels.
Similar uncertainties arise in the long-term treatment of children with neonatal CVST. It is reasonable to try to prevent dehydration and anemia, 2 known precipitants of sinovenous thrombosis.4 Individuals with ICH caused by bleeding disorders may require prophylactic replacement of coagulation factors.72
Recommendations for Perinatal Stroke
Class I Recommendations
- Markedly low platelet counts should be corrected in individuals with ICH (Class I, Level of Evidence B).
- Neonates with ICH resulting from coagulation factor deficiency require replacement of the deficient coagulation factors (Class I, Level of Evidence B).
- Vitamin K should be administered to individuals with vitamin K–dependent coagulation disorders (Class I, Level of Evidence B). Higher doses of vitamin K may be required in neonates with factor deficiencies resulting from maternal medications.
- Patients who develop hydrocephalus after an ICH should undergo ventricular drainage and later shunting if significant hydrocephalus persists (Class I, Level of Evidence B).
Class II Recommendations
- It is reasonable to treat dehydration and anemia in neonates with stroke (Class IIa, Level of Evidence C).
- It is reasonable to use rehabilitation and ongoing physical therapy in an effort to reduce neurological dysfunction in individuals with perinatal stroke (Class IIa, Level of Evidence B).
- It is reasonable to give folate and B vitamins to individuals with an MTHFR mutation in an effort to normalize homocysteine levels (Class IIa, Level of Evidence C).
- It is reasonable to evacuate an intraparenchymal brain hematoma to reduce very high intracranial pressure, although it is not clear whether this approach always improves the outcome (Class IIa, Level of Evidence C).
- Anticoagulation with LMWH or UFH may be considered in selected neonates with severe thrombophilic disorders, multiple cerebral or systemic emboli, or clinical or radiological evidence of propagating CVST despite supportive therapy (Class IIb, Level of Evidence C). Until additional information on its safety and efficacy is available, a recommendation on the use of anticoagulation in other neonates with CVST is not possible.
Class III Recommendations
- Thrombolytic agents are not recommended in neonates until more information about the safety and effectiveness of these agents is known (Class III, Level of Evidence C).
AIS in Infants and Older Children
Risk Factors for First Stroke
Almost half of children with AIS are known to have a stroke risk factor at the time of infarction, and 1 vascular risk factors can be identified in at least two thirds of children after a thorough evaluation.1,7 A detailed family history will document cerebrovascular disease among first-degree relatives, and a thorough physical examination will help to identify systemic diseases that increase stroke risk. Even after extensive investigations, however, no cause can be discovered in up to 30% of children with AIS. A discussion of all potential stroke risk factors is beyond the scope of this article, but pediatric stroke risk factors have been reviewed extensively.73
Some patients have >1 risk factor, but the presence of multiple risk factors may compound the stroke risk for some children. It may be impractical to investigate every conceivable stroke risk factor in each child, and some physicians try first assessing for the common causes of stroke and then eliminating less common stroke risk factors (Tables 2 through 5
) on the basis of the patient’s clinical findings. The evaluation of stroke in children is presented in more detail in Diagnostic Evaluation of Children With Stroke.
|
|
|
|
Risk of Recurrent Stroke
Clinical and radiological recurrence of AIS is seen in 6% to 14% of children with a new infarction,74,75 but many more have TIAs or silent reinfarctions. There are data suggesting that a vascular lesion plus prothrombotic risk factors predict recurrence risk. A large German population-based cohort74 found that a vascular origin was a risk factor for recurrence, as were elevated lipoprotein(a) and protein C deficiencies. In another series of 212 children from the United Kingdom, the most important predictors of clinical recurrence were moyamoya syndrome and low birth weight.75 The presence of at least 1 genetic thrombophilic state was associated with a higher risk of recurrence, as were previous TIAs and bilateral infarctions at initial presentation.
Vascular and Nonvascular Risk Factors for Arterial Ischemic Stroke
There are many risk factors for ischemic stroke in infants and children, including preexisting illnesses such as congenital cardiac disease, SCD, infections (eg, meningitis or varicella), and various prothrombotic states (Table 3). Other patients have cervicocephalic arterial dissection (CCAD), fibromuscular dysplasia (FMD), vasculitis (Table 5), moyamoya disease, or other vasculopathies (Table 4).76 Yet, the cause of stroke in up to one third of children with AIS goes undetermined. What follows is a summary of some of the more important risk factors for childhood ischemic stroke.
Sickle Cell Disease
Stroke is one of the major complications of SCD.77 Rates of stroke in SCD are much higher than rates of stroke in children in general. The Baltimore-Washington Cooperative Young Stroke Study was a retrospective study that identified all cases of ischemic and hemorrhagic stroke among children and young adults within a catchment area that totaled 46 hospitals over a 3-year period between 1988 and 1991.14 The total population of the region was based on 1990 census data. The population of children with SCD was estimated from the National Newborn Screening Report. A total of 35 strokes, 18 ischemic infarcts and 17 ICHs were identified. The overall incidence of childhood stroke was calculated at 1.29 per 100 000 per year in the Baltimore-Washington area, but for those with SCD, the incidence rate was 285 per 100 000 per year (0.28% per year). The risk of both ischemic infarction and hemorrhage was increased. The overall incidence of AIS was 0.58 per 100 000 and ICH was 0.71 per 100 000. Conversely, for children with SCD, the incidence rates per 100 000 were 238 and 47.5 for AIS and ICH, respectively.
Similar incidence rates in SCD were obtained in the Cooperative Study of Sickle Cell Disease (CSSCD).78 The CSSCD collected data on 4082 patients with SCD at 23 clinical centers within the United States over a 10-year period from 1978 to 1988. In this study, TIA was included with infarction and ICH. The annual incidence of first stroke was 0.46% per year. Children homozygous for the sickle cell gene mutation (SCD-SS) had an even higher rate, 0.61% per year. The highest rate of first stroke was in children between 2 and 5 years of age, followed by those from 6 to 9 years of age, with incidences of 1.02% and 0.68% per year, respectively. In contrast, the incidence of ICH was highest in adults between 20 and 29 years of age.
Acute Stroke in SCD.
SCD promotes different forms of cerebrovascular disease. The clinical presentation depends primarily on the size and location of the lesion, although many asymptomatic individuals have small infarctions on MRI. Individuals with carotid vasculopathy often present with an acute deficit resulting from a large ischemic infarction in the MCA territory. Some individuals develop progressive vasculopathy of the intracranial internal carotid artery (ICA) with distal collateral vessels or moyamoya syndrome (see below). Large infarctions within the anterior cerebral artery (ACA) or posterior cerebral artery territories occur less often. Small infarctions are common and typically involve the basal ganglia and deep white matter within the anterior circulation. Border-zone infarctions are not as common as large infarctions, but both are probably traceable to large-artery disease. Occasionally, individuals with SCD may develop sinovenous thrombosis or anterior spinal artery syndrome. SAH and intracerebral hemorrhage also occur in the context of sinovenous thrombosis and after rupture of aneurysms usually located at the bifurcations of major vessels, particularly in the vertebrobasilar circulation, or of fragile moyamoya vessels. Reversible posterior leukoencephalopathy has been described after acute chest syndrome79 but may result in infarction, typically occipital.
Risk Factors for First Stroke in SCD.
Risk factors for stroke in individuals with SCD include high blood flow velocity on transcranial Doppler (TCD), low hemoglobin value, high white cell count, hypertension, silent brain infarction, and history of chest crisis.78
Identification by TCD of those at highest risk provided an opportunity to prevent first stroke.80 Children with TCD ultrasound evidence of high cerebral blood flow velocity rates (time-averaged mean velocity 200 cm/s) have a stroke rate of at least 10% per year.81,82 Although few strokes occur in the ACA distribution, there is some evidence that elevated blood flow velocity in the ACA predicts a higher stroke risk in individuals with SCD than would similar velocities in the ICA or MCA.83
There appears to be a familial predisposition to stroke in individuals with SCD,84 and some genetic risk factors have been identified,85,86 although siblings may also share adverse environmental conditions, including poverty, air pollution, and poor nutrition. Nocturnal hypoxemia also is a predictor of central nervous system (CNS) events in SCD87 and might be a modifiable risk factor.
Predictors of Stroke Recurrence in SCD.
Recurrence of stroke is high in SCD, but no population or clinical trial data are available. Pegelow et al88 pooled retrospective data from 6 clinics and compared rates of recurrent stroke with and without regular transfusion and reported that at 50 months the recurrent stroke-free survival was only 30% without transfusion. In another multicenter cohort, Scothorn and colleagues89 reported on data from 137 children who had had a first stroke between 1.4 and 14 years of age. Twenty-six individuals had an antecedent or concurrent event (hypertension, fever, chest syndrome, acute anemia, or exchange transfusion). Thirty-one of these patients (23%) had at least 1 recurrent stroke with a mean time to recurrence of 4 years and a recurrence rate of 2.2 per 100 patient-years. After 2 years, recurrence continued only in those with no antecedent events. Dobson et al90 described the findings from a single center managing children with homozygous SCD with stroke who were <18 years of age from 1980 to 1999. Forty-one percent of the patients had recurrent stroke or TIAs. Recurrent stroke was more common in patients with moyamoya syndrome.
Subclinical Brain Disease.
The CSSCD confirmed that 20% of children with SCD have "silent" brain lesions on MRI predominantly in frontal and parietal cortical, subcortical, and border-zone locations.91 These so-called "silent infarcts" are important because they are associated with deterioration in cognitive function with effects on learning and behavior.92 Further evidence from the CSSCD confirms that the risk of clinical stroke is increased if MRI is abnormal from the background rate of 0.5% to 1% to 2% per year.93 Silent lesions are evidence of brain injury, and it is reasonable to reassess these patients’ histories for symptoms that were not previously recognized and to re-examine the patients’ clinical and laboratory risks for stroke. The rate of stroke in children with positive MRI but TCD findings that do not reach current treatment guidelines is not clear, and regular blood transfusions are not recommended on the basis of MRI alone. The risks and benefits of prophylactic transfusion based on silent MRI lesions are being tested in a randomized clinical trial (SITT: http://www.clinicaltrials.gov/ct/show/NCT00072761?order=1).
ICH Resulting From SCD.
Children with SCD develop all types of ICH.77,78 Intraparenchymal bleeding may be associated with large-vessel vasculopathy, especially if moyamoya formation is present. ICH also can occur as a result of CVST. IVH is unusual but may occur if moyamoya vessels are present near the ventricular wall. One retrospective study suggested that systemic hypertension, corticosteroid use, and recent transfusion increased the likelihood of an ICH in individuals with SCD.94
There are reports of epidural hematomas in the absence of significant head trauma in SCD.95 The cause is not known, but epidural hematomas may be related to hypervascular areas of bone.
SAH is relatively common in individuals with SCD. Although the cause of many of these hemorrhages is unknown, an aneurysm often is present in adolescents and adults with SCD and an SAH.96 Because of the potential for rebleeding in individuals with an aneurysm, it is reasonable to fully evaluate these individuals with CA. However, there is some concern that CA might facilitate sickling in individuals with SCD, and the typical approach is to defer the CA until after the percentage of sickle hemoglobin has been reduced with transfusions.
Treatment of Acute Stroke Caused by SCD.
The usual acute treatment of acute ischemic infarction resulting from SCD is hydration and exchange transfusion, although there are no controlled studies to prove the value of this approach.97 Exchange transfusion avoids the theoretical risk of increasing blood viscosity that could accompany a rapid increase in hematocrit. Hypoxemia and hypotension should be treated and normoglycemia maintained. Even though intracranial arterial vasculopathy characteristic of SCD is the most likely cause of stroke in this setting,98 it may be appropriate to consider other disorders such as infection, cardiac embolism, and CVST.
Primary and Secondary Prevention of Stroke Resulting From SCD.
Although no clinical trial has addressed this specific issue, regular blood transfusions have been used for several years in individuals with SCD for secondary stroke prevention.88 The targeted reduction of the sickle hemoglobin to <30% of the total hemoglobin is based on in vitro viscosity experiments, has not been critically evaluated, but has become the standard therapeutic target.97 Pegelow et al88 estimated from retrospective clinical data that stroke-free survival with chronic transfusion was 80% at 50 months compared with 30% when no transfusion is given. Scothorn et al89 estimated an annual rate of recurrent stroke of 2% despite ongoing transfusion. Several transfusion regimens are in use, including simple transfusions of 10 to 15 mL/kg of packed red blood cells every 3 to 4 weeks and the use of apheresis machines to remove blood while adding donor red cells. Chronic transfusion brings with it the requirement to manage iron overload, and the reader is referred to other reviews for more information and for a discussion of transfusion method options.97 In the absence of more robust data, it has been recommended that transfusion should be continued for at least 5 years or at least until the child reaches 18 years of age.97
A randomized trial (Stroke Prevention Trial in Sickle Cell Anemia [STOP]) compared periodic blood transfusion with standard care in 130 children with SCD ranging from 2 to 16 years of age (mean age, 8 years) who were selected for high stroke risk with TCD. Blood transfusions were given an average of 14 times per year for >2 years in the treatment group, with a target reduction of sickle hemoglobin from a baseline of >90% to <30%. The trial was halted 16 months early after 11 strokes occurred in the standard-care arm compared with 1 stroke in the transfusion-treated group. The risk of stroke was reduced from 10% to <1% per year. A Clinical Alert issued by the National Heart, Lung, and Blood Institute of the National Institutes of Health recommended screening all children with SCD who have no history of stroke, with consideration of transfusion for those with 2 abnormal TCD ultrasound studies. The rates of first stroke among children with SCD have declined in California since the STOP study and the Clinical Alert called for screening and treatment, suggesting that this approach may be having an impact.99
In STOP, there was no evidence of transfusion-related infection, but iron overload and alloimmunization remained important transfusion-related risks. A randomized controlled trial (STOP II) of withdrawal of transfusion was initiated in 2000 to test whether chronic transfusions for primary stroke prevention could be safely discontinued after at least 30 months in children who had not had an overt stroke and who had reversion to low-risk TCD on transfusion. In both treatment arms, children received close clinical and TCD surveillance for the first occurrence of reversion of TCD to abnormal, confirmed with 2 TCDs of 200 cm/s, indicating a return of high risk for overt stroke. Clinical stroke also was part of the composite end point. It was hoped that most end points would be TCD reversions rather than clinical stroke and that close monitoring with TCD, at least quarterly, would provide acceptable safety monitoring by signaling the return of high risk, allowing reinstitution of transfusion. The patients in the transfusion arm received periodic simple or exchange blood transfusions every 3 to 4 weeks in an effort to maintain the sickle hemoglobin level <30%. After 79 of a planned sample size of 100 children had been randomized, the Data and Safety Monitoring Board recommended closure of the study for safety concerns after an interim analysis showed a highly significant difference between the transfusion and nontransfusion treatment arms with respect to the composite end point of TCD reversion to high risk and overt stroke. Among the 41 individuals randomized to halt transfusion, there were 16 end points, 14 of which were TCD reversions to high-risk TCD (without stroke), and 2 were ischemic strokes that occurred shortly after the first TCD reverted to abnormal but before a confirmatory TCD could be obtained and transfusion therapy resumed. Six other subjects were returned to periodic transfusion therapy for other clinical reasons before end points took place. No strokes or reversions to high-risk TCD were observed in those subjects who were in the chronic transfusion treatment arm. The reversions to abnormal TCD velocity were seen only 4 to 9 months after randomization. Eight children (20%) tolerated removal from chronic transfusion therapy without apparent problems, but the National Heart, Lung, and Blood Institute does not recommend discontinuation of transfusion after 30 months based on STOP II because of the high TCD reversion rate and the small risk of overt stroke despite frequent TCD surveillance.100
There are no data to support the use of chronic transfusion to prevent recurrent ICH in SCD.
Hydroxyurea is used to reduce painful episodes in adults with SCD, but whether it reduces the risk of stroke in children with SCD is unknown.101 Data from nonrandomized clinical series suggest that hydroxyurea might be an alternative to transfusion for primary stroke prevention, but a definitive study has not been done.102
A single study has looked at stroke rates in children and young adults who were first transfused after stroke and then later switched to hydroxyurea in an open-label study with historical controls,103 with encouraging results. Hydroxyurea with phlebotomy is being tested in a randomized clinical trial (the SWITCH study; see http://www.clinicaltrials.gov/ct/show/NCT00122980?order=1) on the basis of preliminary evidence that it may offer a comparable stroke risk reduction after transfusion has been given for several years.103
Bone marrow transplantation reportedly stabilizes the cerebrovascular disease caused by SCD,104 but data are limited. Because of the unavailability of suitable donors and other issues, bone marrow transplantation is not always feasible.
A few patients with moyamoya syndrome resulting from SCD have undergone surgical bypass procedures in an effort to prevent stroke.105,106 In the largest series, encephaloduroarteriosynangiosis was performed in 6 children, 4 of whom had had a stroke (2 occurred while on a chronic transfusion program). One patient who was initially free of cerebrovascular symptoms had an infarction 2 weeks after surgery.106 The natural history of children with SCD and moyamoya has not been well studied, and the risk and benefit of this procedure should be evaluated in a randomized trial. In the meantime, revascularization surgery may be considered as a last-resort option for SCD patients who cannot be treated otherwise or who continue to have brain infarctions despite medical therapy.
Screening Patients With SCD
It is reasonable to monitor younger children (2 to 10 years of age) and those with relatively high velocities with TCD more frequently.107 On the basis of an analysis of the STOP TCD data, Wang108 proposed that patients with a normal TCD (
170 cm/s) be restudied annually and that patients with an abnormal TCD (200 cm/s) be restudied in 1 month. He suggested a repeat TCD in 3 months for individuals with a blood flow velocity of 185 to 199 cm/s and every 6 months for those with flow velocities in the range of 170 to 184 cm/s.108
Recommendations for Children With SCD
Class I Recommendations
- Acute management of ischemic stroke resulting from SCD should include optimal hydration, correction of hypoxemia, and correction of systemic hypotension (Class I, Level of Evidence C).
- Periodic transfusions to reduce the percentage of sickle hemoglobin are effective for reducing the risk of stroke in children 2 to 16 years of age with an abnormal TCD resulting from SCD and are recommended (Class I, Level of Evidence A).
- Children with SCD and a confirmed cerebral infarction should be placed on a regular program of red cell transfusion in conjunction with measures to prevent iron overload (Class I, Level of Evidence B).
- Reducing the percentage of sickle hemoglobin with transfusions before performing CA is indicated in an individual with SCD (Class I, Level of Evidence C).
Class II Recommendations
- For acute cerebral infarction, exchange transfusion designed to reduce sickle hemoglobin to <30% total hemoglobin is reasonable (Class IIa, Level of Evidence C).
- In children with SCD and an ICH, it is reasonable to evaluate for a structural vascular lesion (Class IIa, Level of Evidence B).
- In children with SCD, it is reasonable to repeat a normal TCD annually and to repeat an abnormal study in 1 month (Class IIa, Level of Evidence B). Borderline and mildly abnormal TCD studies may be repeated in 3 to 6 months.
- Hydroxyurea may be considered in children and young adults with SCD and stroke who cannot continue on long-term transfusion (Class IIb, Level of Evidence B).
- Bone marrow transplantation may be considered for children with SCD (Class IIb, Level of Evidence C).
- Surgical revascularization procedures may be considered as a last resort in children with SCD who continue to have cerebrovascular dysfunction despite optimal medical management (Class IIb, Level of Evidence C).
Moyamoya Disease and Moyamoya Syndrome
Moyamoya syndrome is characterized by chronic progressive stenosis of the distal intracranial ICA and, less often, stenosis of the proximal ACA and MCA, the basilar artery, and the posterior cerebral arteries. The term moyamoya is a Japanese word meaning "hazy, like a cloud of smoke drifting through the air," referring to the often hazy angiographic appearance of the distal collateral network on angiography. Traditionally, individuals with a well-recognized associated condition are categorized as having moyamoya syndrome, and those with no known risk factors are said to have moyamoya disease.
The relative rarity of moyamoya disease (an estimated 0.086 per 100 000 children in the United States) has limited the studies on its diagnosis, treatment, and outcome in this country.109 Although there is a vast literature on moyamoya, randomized clinical trials are not available to guide therapy. A recent meta-analysis of the surgical treatment of pediatric moyamoya syndrome offers the most complete review of the literature to date.110
Clinical Features and Diagnosis.
A Japanese research committee issued the following guidelines for the diagnosis of moyamoya: (1) stenosis involving the region of the distal ICA bifurcation (C1) and proximal portions of the ACA (A1) and MCA (M1), (2) appearance of dilated basal collateral arteries, and (3) bilateral abnormalities. If any of the conditions listed above is present and the angiographic pattern is found on 1 side only, the diagnosis is probable.
Moyamoya disease accounts for 6% of childhood strokes in Western countries. Half of the patients present before 10 years of age.111,112 Some patients have rare, intermittent ischemic events or even extended periods of clinical stability; other individuals have a more fulminant rapid neurological decline.113,114 Children with moyamoya typically present with ischemic stroke or TIAs. Ischemic episodes also occur in older individuals, but adults are more likely than children to develop an ICH. Ischemic strokes often are multiple and recurrent, involving predominantly the carotid circulation. Infarctions may be superficial or deep and often are found in watershed territories. Ischemic symptoms may follow hyperventilation, crying, coughing, straining, or fever. A characteristic electroencephalographic finding is slowing of the background rhythm after cessation of hyperventilation ("rebuildup" phenomenon).2
Epidemiology of Moyamoya Disease.
First described in Japan, moyamoya syndrome has now been observed throughout the world and affects individuals of many ethnic backgrounds, with increasing detection of this disease in American and European populations.115,116 In Japan, it is the most common pediatric cerebrovascular disease, affecting girls almost twice as often as boys with a prevalence of 3 of 100 000.111,117 In Europe, a recent study cited an incidence of 0.3 patients per center per year, which is 1/10 the incidence in Japan.118 Results from a 2005 US study suggest an incidence of 0.086 of 100 000 persons. The ethnicity-specific incidence rate ratios compared with whites were 4.6 (95% CI, 3.4 to 6.3) for Asian Americans, 2.2 (95% CI, 1.3 to 2.4) for blacks, and 0.5 (95% CI, 0.3 to 0.8) for Hispanics.109
In the United States and Korea, reports have corroborated historical claims of a bimodal age distribution of moyamoya, 1 group in the pediatric age range (around the first decade of life) and a second group of adults in the 30- to 40-year-old range. Children appear to be more likely to present with ischemic events (either strokes or TIAs) and adults with hemorrhage, leading to more rapid diagnosis.119,120 The significance of ischemic events in children usually proves more difficult to diagnose because of the patient’s age and limited communication skills, leading to delayed recognition of the underlying moyamoya condition.3
Associated Conditions.
Although the cause and pathogenesis of moyamoya disease are poorly understood, genetic factors play a major role. The familial incidence of affected first-degree relatives in Japan is 7% to 12%, and a similar rate of 6% was found in the Children’s Hospital, Boston, series.113,121–123 Moyamoya has been linked to several genetic loci.124–126 Moyamoya has been associated with specific human leukocyte antigen (HLA) haplotypes, including the HLA-B40 antigen in patients <10 years of age and the HLA B52 antigen in those >10 years of age. Moyamoya also has been associated with the AW24, BW46, B51-DR4, and BW54 antigens. Elevated levels of fibroblast growth factor may play a role in its pathogenesis. Increased levels of fibroblast growth factor have been found in the cerebrospinal fluid, and a strong fibroblast growth factor receptor immunoreactivity has been demonstrated in superficial temporal vessels.
Several clinical conditions have been reported in conjunction with moyamoya syndrome, although for conditions with only 1 or 2 reported cases, the link is at best tenuous.114 Frequently reported risk factors include cranial radiotherapy, Down syndrome, neurofibromatosis type 1, and SCD. Table 6 summarizes the clinical associations noted in several published series.127–131
|
Natural History and Prognosis of Moyamoya.
The prognosis of moyamoya disease is difficult to predict because its natural history is not firmly established. By some estimates, 50% to 66% of untreated moyamoya patients have progressive neurological dysfunction and a poor outcome132–134 compared with a 2.6% deterioration rate for children summarized in a more recent meta-analysis of 1156 surgically treated patients.110
The overall prognosis of patients with moyamoya syndrome depends on the rapidity and extent of vascular occlusion, the patient’s ability to develop effective collateral circulation, age at onset of symptoms, severity of presenting neurological deficits and degree of disability, and the extent of infarction on CT or MRI at presentation.135 Some authors suggest that neurological status at the time of treatment, more than age of the patient, predicts long-term outcome.113,136
Diagnostic Studies.
Diagnosis is based on a distinct arteriographic appearance characterized by bilateral stenosis of the distal ICA extending to the proximal ACA, MCA, and posterior cerebral artery with frequent involvement of the circle of Willis and development of an extensive collateral network at the base of the brain. The typical abnormal vessels of moyamoya disease are noted with 3-dimensional CTA, postcontrast MRI, and MRA. MRA may be useful for screening high-risk subjects. TCD has been used to assist with the diagnosis and postoperative follow-up of these patients. Small areas of hypodensity suggestive of stroke are commonly observed in cortical watershed zones, basal ganglia, deep white matter, or periventricular regions.137,138
There are no data to support routine vascular screening for moyamoya syndrome, but screening may be considered in individuals with relatively common and high-risk disorders such as neurofibromatosis type 1, Down syndrome, and SCD.73,139–141 There is little evidence to justify screening first-degree relatives of patients with moyamoya when a single individual in a family is affected.
Acute infarcts are best seen with diffusion-weighted imaging. Chronic infarcts are better demonstrated with T1 and T2 imaging. Cortical oligemia may be inferred from fluid-attenuated inversion recovery sequences, which demonstrate linear high signal following a sulcal pattern, thought to represent slow flow in poorly perfused cortical circulation.142 Most suggestive of moyamoya on MRI is the absence of flow voids in the ICA, MCA, and ACA coupled with abnormally prominent flow voids from basal ganglia and thalamic collateral vessels. These imaging findings are virtually diagnostic of moyamoya syndrome.138,143–147
Because of its excellent diagnostic yield and noninvasiveness, MRA has largely supplanted CA as the primary diagnostic imaging modality for moyamoya syndrome.143,148–153 However, although MRA affords the ability to detect stenosis of major intracranial vessels and to visualize the basal collateral vessels, MRA is less reliable when applied to smaller vessel occlusions. In a study of 190 angiograms from pediatric patients, the complication rate from angiography in children with moyamoya syndrome was no higher than the risk of angiography in nonmoyamoya populations with other forms of cerebrovascular disease.154
Techniques such as TCD, perfusion CT, xenon-enhanced CT, positron emission tomography (PET), MR perfusion imaging,155–157 and single-photon emission CT (SPECT) with acetazolamide challenge have been used to evaluate moyamoya syndrome. TCD provides a noninvasive way to follow changes in blood flow velocities in larger cerebral vessels over time, whereas xenon-enhanced CT, PET, and SPECT can detect inadequate resting perfusion and poor blood flow reserve before treatment and can help determine the extent of the improvement in functional perfusion after therapy.158–166 The role of SPECT and PET scans in the evaluation and management of moyamoya syndrome has increased over the past decade.164,166
Several reports suggest that periodic clinical and radiographic reexaminations of patients with moyamoya disease may be helpful in some clinical settings.110,167 One study found that 27% (17 of 64) of the moyamoya patients with unilateral disease eventually developed bilateral involvement.168 There is also evidence that disease progression is more likely to occur in younger patients.168 Of those patients who had surgery (for either bilateral or unilateral disease), the need for reoperation as a result of refractory disease ranged from 1.8% to 18%.169
Treatment of Moyamoya Disease and Moyamoya Syndrome.
Surgical revascularization procedures are widely used for moyamoya syndrome, particularly for patients with cognitive decline or recurrent or progressive symptoms.110
Direct anastomosis procedures, most commonly a superficial temporal artery to MCA anastomosis (bypass), often are technically difficult to perform in children because of the small size of scalp donor vessels or MCA recipient vessels. Indirect revascularization procedures have included encephaloduroarteriosynangiosis and encephalomyoarteriosynangiosis.170–172 There are multiple variations of these procedures, including simply drilling burr holes without vessel synangiosis173,174 and craniotomy with inversion of the dura in hopes of enhancing new dural revascularization of the brain.175 Some patients stabilize without intervention, but unfortunately this can occur after they have already sustained a debilitating neurological disability. Potential complications of surgery for moyamoya include postoperative ischemic stroke, spontaneous or traumatic subdural hematoma, and ICH.
A number of groups have reported improved results in the use of combined direct and indirect anastamoses.169,170,176,177 A modification of the encephaloduroarteriosynangiosis procedure called pial synangiosis has been used with encouraging results.113 A review of 143 children with moyamoya syndrome treated with pial synangiosis demonstrated marked reductions in their stroke frequency after surgery, especially after 1 month postoperatively. Sixty-seven percent had strokes preoperatively, 7.7% had strokes in the perioperative period, and only 3.2% had strokes after at least 1 year of follow-up. The long-term stroke rate was 4.3% (2 of 46 patients) in individuals with a minimum of 5 years of follow-up.113 Despite extensive anecdotal evidence supporting the role for surgical treatment of moyamoya syndrome, there is a need for further research to validate these data.
Two large studies with long-term follow-up demonstrated a good safety profile for surgical treatment of moyamoya (4% risk of stroke within 30 days of surgery per hemisphere) with a 96% probability of remaining stroke free over a 5-year follow-up period.113,132 A recent meta-analysis of 1156 pediatric moyamoya patients treated with surgery concluded that 87% (1003 patients) derived symptomatic benefit from surgical revascularization but that there was no difference between the indirect and direct/combined groups.110 Some physicians believe that revascularization surgery is less useful in patients presenting with ICH,119 although additional studies of patients presenting with hemorrhage are ongoing.
Moyamoya patients are at additional risk of ischemic events during the perioperative period. Crying and hyperventilation, common occurrences in children during hospitalization, can lower PaCO2 and induce ischemia secondary to cerebral vasoconstriction. Any techniques to reduce pain—including the use of perioperative sedation, painless wound dressing techniques, and absorbable wound suture closures—may reduce the likelihood of stroke and shorten the hospitalization.178 Similarly, it is good to avoid hypotension, hypovolemia, hyperthermia, and hypocarbia both intraoperatively and perioperatively.113 Slight elevation of the systemic blood pressure may be beneficial, and postoperative patients may be given intravenous fluids at 1.5 times the normal maintenance rate based on weight for 48 to 72 hours.178
Few studies have compared medical and surgical therapies for moyamoya. A large survey from Japan in 1994 found no significant differences in outcome between medically and surgically treated moyamoya patients.179 Another study indicated that 38.4% of 651 moyamoya patients who were not initially treated with surgery eventually came to surgery as a result of progressive symptoms.169
Platelet antiaggregants are sometimes given for moyamoya syndrome, particularly when the patient is considered a poor operative risk or has relatively mild disease, but there are few data demonstrating either short-term or long-term efficacy. Antiplatelet agents have been used in individuals whose ischemic symptoms seem to arise from emboli from microthrombus formation at sites of arterial stenosis and routinely for all patients in many operative series.113,114,127,180 Anticoagulants such as warfarin are rarely used because of the difficulty of maintaining therapeutic levels in children and the risk of hemorrhage after inadvertent trauma, but low-dose LMWH has been used.181 Calcium channel blockers180 have been used in a few patients in an attempt to improve intractable headaches and to reduce the frequency and severity of refractory TIAs.
Recommendations for Treatment of Moyamoya in Children
Class I Recommendations
- Different revascularization techniques are useful to effectively reduce the risk of stroke resulting from moyamoya disease (Class I, Level of Evidence B). However, despite a vast literature on moyamoya, there are no controlled clinical trials to guide the selection of therapy.
- Indirect revascularization techniques are generally preferable and should be used in younger children whose small-caliber vessels make direct anastomosis difficult, whereas direct bypass techniques are preferable in older individuals (Class I, Level of Evidence C).
- Revascularization surgery is useful for moyamoya (Class I, Level of Evidence B). Indications for revascularization surgery include progressive ischemic symptoms or evidence of inadequate blood flow or cerebral perfusion reserve in an individual without a contraindication to surgery (Class I, Level of Evidence B).
Class II Recommendations
- TCD may be useful in the evaluation and follow-up of individuals with moyamoya (Class IIb, Level of Evidence C).
- Techniques to minimize anxiety and pain during hospitalizations may reduce the likelihood of stroke caused by hyperventilation-induced vasoconstriction in individuals with moyamoya (Class IIb, Level of Evidence C).
- Management of systemic hypotension, hypovolemia, hyperthermia, and hypocarbia during the intraoperative and perioperative periods may reduce the risk of perioperative stroke in individuals </