Secondary Logo

Journal Logo

A single-site retrospective study of pediatric arterial ischemic stroke etiology, clinical presentation, and radiologic features

SUN, Dan; WU, Xiao-man; WANG, Zeng-wu; JIN, Run-ming; LIU, Zhi-sheng; LIU, Fan; HUANG, Sui; WANG, Hai-qin; HU, Jia-sheng

doi: 10.3760/cma.j.issn.0366-6999.20123633
Original article

Background Stroke occurs upon obstruction of cerebral blood circulation and is clinically characterized by sudden onset symptoms. Advanced age is the main risk factor of stroke, but cases of pediatric stroke have been rarely reported. This study aimed to determine the etiology, clinical presentation, and radiologic features of neurological deficit for pediatric arterial ischemic stroke (PAIS).

Methods The medical records of 42 PAIS patients (age range: 9 months to 13 years) treated at Wuhan Children's Hospital between July 2007 and January 2011 were retrospectively reviewed. Infarction location was first determined by craniocerebral computed tomography and magnetic resonance (MR) imaging. The stenotic or occluded main cerebral arteries and/or branches were determined by MR angiography and digital subtraction angiography.

Results The majority of the 42 PAIS cases (66.7%, n=28) were ≤3 years old (vs. >3 years old: 33.3%, n=14; P<0.05), but the male: female ratio was similar in both groups (P>0.05). The most frequently reported signs and symptoms for both age groups were limited physical activity followed by convulsions and delirium, but convulsions were more prevalent in children ≤3 years-old. Children>3 years-old mainly experienced the limited physical activity symptoms, including hemiparalysis, aphasia, and ataxia. For all 42 cases, the most frequent etiologies were infections (38.1%, n=16), iron deficiency anemia (16.7%, n=7), and moyamoya syndrome (11.9%, n=5). The predominant infarcts among all cases were middle cerebral artery (63.6%, n=21) and basal ganglia (64.3%, n=27).

Conclusions PAIS occurs more frequently in younger children and this group most frequently presents with convulsion as the initial symptom. The overall etiologies of PAIS may be different from those of adult stroke and the involved regions may be distinguishing features of PAIS or its different forms, but more research is required.

Department of Pediatrics, Union Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, Hubei 430022, China (Sun D and Jin RM)

Department of Neurology, Wuhan Children's Hospital, Wuhan, Hubei 430016, China (Sun D, Liu ZS, Liu F, Huang S, Wang HQ and Hu JS)

Department of Pediatrics, Fourth Hospital, Wuhan, Hubei 430016, China (Wu XM)

Department of Neurosurgery, People's Hospital, Weifang, Shandong 261500, China (Wang ZW)

Correspondence to: JIN Run-ming, Department of Pediatrics, Union Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, Hubei 430022, China (Email:; LIU Zhi-sheng, Department of Neurology, Wuhan Children's Hospital, Wuhan, Hubei 430016, China ( SUN Dan, WU Xiao-man and WANG Zeng-wu contributed equally to this work.

This study was supported by a grant from the Clinical Key Special Subject of Wuhan City (No. 2007–43).

(Received February 26, 2013)

Edited by CHEN Li-min

Pediatric arterial ischemic stroke (PAIS) is a relatively rare disorder, with the estimated incidence ranging from 3–8 individuals per 100 000.1 However, the prognosis of PAIS is very poor and many of the sufferers experience lifelong effects on physical growth (e.g. claudication, muscle atrophy), intelligence, behavior, psychosocial function, and other parameters affecting quality of life.

PAIS is distinctive from adult stroke in many regards. Due to the comparatively high incidence of adult stroke and concomitant amount of focused research efforts, the risk factors are well-defined and include advanced age, atherosclerosis, hypertension, hyperlipidemia, diabetes mellitus, intemperance and smoking. Such factors are rare in the PAIS case population. Instead, the most commonly reported precipitating factors are associated with blood and blood dynamics. For example, PAIS cases in western countries are often related to sickle cell diseases and other clotting disorders involving protein C and S, anticoagulant enzyme III, and Leiden V.2 A subsequent report of Han Chinese PAIS patients did not find a significant association with these risk factors,3 suggesting the possibility of distinct etiologies and risk factors for the different ethnicities.

To generate the first Chinese-related profile of PAIS, this study retrospectively investigated the etiology, clinical presentation, and radiologic features of PAIS cases treated in the Neurology Department of Wuhan Children's Hospital.

Back to Top | Article Outline


Identification of PAIS patients and data collection

The medical records database of the Neurology Department of Wuhan Children's Hospital was searched from 2007 to 2011 for patients presenting with acute neurologic symptoms and receiving subsequent diagnosis of PAIS. Neonatal PAIS patients (<4 weeks-old) were excluded, as were any children with clinical diagnosis of temporary ischemic attack (TIA).

A total of 42 children (26 males and 16 females) aged between 9 months and 13 years were identified for study inclusion. All PAIS diagnoses were made according to a standard adult evaluation protocol using the following criteria: presence of a lesion within a cerebral artery that persists for >24 hours. Lesions were detected by computed tomography (CT), magnetic resonance imaging (MRI) and/ or magnetic resonance angiography (MRA).

Back to Top | Article Outline

Etiology assessment

The following data were collected from each case's medical record: age at onset, sex, geographic location and season at time of onset, and clinical symptoms and signs (including but not limited to limb paralysis, fever, convulsion, headache, disturbance of consciousness (delirium)). The PAIS cases were divided into five groups according to etiology:4,5 cardiac disease: congenital heart disease, acquired heart disease, and arrhythmia; vascular disease: moyamoya syndrome, vasculitis, vascular disease, and vascular malformation; hematologic disease: hemal tumor, anemia, and other blood coagulation disorders; other pre-existing diseases with clear etiology: infection, trauma, and metabolic disease; idiopathic stroke (IS). Designation of pre-existing infection as an etiology required specific pathogen identification by routine laboratory practices or increased levels of some pathogenic antibodies (IgM and IgG) against common virus such as respiratory syncytial virus and adenovirus etc. Trauma was defined as an etiology only if the stroke occurred within two weeks after the body injury and no other co-morbidities were present.

If a PAIS case showed two or more of the above etiologies, then only one factor was selected as the primary etiology based upon its importance in stroke onset; for example, cardiac disease (infection endocarditis)>anemia>infection. Infection endocarditis, as the primary etiology factor, was presumed to explain the existence of the secondary causes and was used to perform the statistical analyses.

Back to Top | Article Outline

Neuroradiological measurements

For all patients, CT or MRI imaging showed infarctions with localization consistent with the neurological symptoms and signs. Cranial MRI examinations were performed at admission using a 1.5 Tesla MR instrument with T1- and T2-weighted spin-echo. The images were retrospectively evaluated by three experienced neuroradiologists who worked independently, were blinded to the clinical data, and used a standardized evaluation form. The locations of infarctions were classified as: basal ganglia, internal capsule, cortex, periventricular white matter, multiple (≥2) sites, or other unusual sites (e.g. cerebellum, thalamus, or cerebral peduncle).

Some PAIS cases underwent additional imaging by magnetic resonance angiography (MRA; n=30) and digital subtraction angiography (DSA; n=18) to locate the precise stenotic or occluded main artery and/or branches. The following regions were identified: middle cerebral artery (MCA), internal carotid artery (ICA), anterior cerebral arteries (ACA), posterior cerebral artery (PCA), and vertebral basal arteries.

Back to Top | Article Outline

Statistical analysis

All data were expressed as mean±standard deviation (SD). The SPSS 13.0 software (SPSS Inc., USA) was used for statistical analysis. The χ2 test was used to exam relationships between incidence of PAIS and age and sex. A P<0.05 was considered statistically significant.

Back to Top | Article Outline


Clinical findings

The clinical findings of the 42 PAIS cases are summarized in Table 1. There were several obvious trends among the cases. The majority of cases were ≤3 years-old (66.7%, n=28), resided in rural areas (88.1%, n=37), and occurred in spring or summer (83.3%, n=35).

Table 1

Table 1

The most commonly reported signs and symptoms among all 42 PAIS cases involved limited physical activity. Specifically, 95.2% (n=40) of cases presented with paralysis, which manifested as hemiplegia (n=32), alternating hemiplegia (n=5), and monoplegia (n=3). The second most common sign overall was convulsion (26.2%, n=11). Intriguingly, the majority of cases (n=11) experiencing convulsions were 3 years-old or younger (72.7%, n=8). Eight children experienced delirium, 75.0% (n=6) of whom were under 3 years-old. The less frequent symptoms reported were fever (n=12), headache (n=5), and ataxia (n=1). Hemiplegia was the most common sign in children over 3 years-old, while seizures (partial or generalized) was the most common for children 3 years-old and younger.

Back to Top | Article Outline

Etiology analysis

The etiologies for all 42 children are summarized in Table 2. Only five cases had unknown etiology (IS: 11.9%, n=5). The etiologies in the remaining 88.1% (n=37) of cases were other diseases with clear etiology in 47.6% (n=20), hematologic diseases in 19.5% (n=8), vascular diseases in 16.7% (n=7), and cardiac diseases in 4.8% (n=2). Among the cases with clear etiology of PAIS, infectious disease (38.1%, 16/42) was the most common, followed by iron deficiency anemia (16.7%, 7/42), moyamoya (11.9%, 5/42), and head trauma (aneurysm) and vascular malformation (arteriovenous malformation) (one case each; 4.8%, 2/42).

Table 2

Table 2

Back to Top | Article Outline

Radiological findings

As shown in Table 3, of the 42 children who underwent MRI, infarcts were detected in basal ganglia for 27 cases and in cortex for seven cases. Fourteen cases showed multifocal infarcts, involving the cortex of the frontal, temporal and parietal regions, the internal capsule, and other regions such as deep white matter. In the four children who underwent both CT (three children examined within one day of onset and one child examined three days later) and MRI, the CT images were normal in two (both of whom received imaging within 24 hours) and showed small infarction in the location of the MCA under MRI and MRA image.

Table 3

Table 3

Thirty cases underwent MRA or DSA examination or both, and it was found that all cases involved the carotid arteries and branches (total of 33 regions identified), but no case involved the vertebrobasilar arteries as well (Table 4). Unilateral involvement was the most common (n=27; Figure 1), whereas only three cases showed bilateral involvement. In the 27 cases with unilateral involvement, right-sided infarction was present in 19, and left-sided in 14 with the main artery involved being the middle cerebral artery (MCV) (39.4%, 13/33) and its branch lenticulostriate artery (27.3%, 9/33). Infarction of the ACA was never seen in isolation and often accompanied an MCV lesion.

Table 4

Table 4

Figure 1.

Figure 1.

Back to Top | Article Outline


PAIS is clinically defined as cerebral infarction focal disorders of nervous functions lasting for more than 24 hours, and is diagnosed by neural imaging evidence of ischemic lesions. TIAs are also caused by nerve dysfunction but persist less than 24 hours, and this feature is used for differential diagnosis. Besides these basic diagnostic features very little is known about the pathogeneses of pediatric strokes, especially PAIS. The current retrospective study of PAIS cases from a single institute revealed a trend in PAIS incidence towards the younger pediatric population, with incidence being remarkably lower after 3 years-old. This finding agrees with the results from a previous study of PAIS neonate and adolescent (up to 16 years-old) patients by Steinlin et al,6 which showed that 50% of cases had occurred in infancy. In our overall case population, the most common signs and symptoms were limited physical activity, followed by convulsions and delirium. However, the initial clinical manifestation was different among the different age groups, which may indicate distinctive pathogeneses related to development or yet unrecognized forms of PAIS. This hypothesis is based on our finding that children ≤3 years old mainly present with convulsions, but the overall incidence of convulsions in our case population was remarkably lower than that reported in another previous study (26.2% vs. 83.0%7). It is possible that our results are reflective of the Chinese ethnicity or due to an unknown factor that will be proven or disproven by future studies with larger sample sizes.

Neuroimaging is an important diagnostic method for PAIS. Our finding that the most commonly involved infarcted artery is the MCA (39.4%, 13/33) agreed with previous reports.8,9 The ACA lesion was never seen in isolation and often accompanied an MCV infarction. Further studies in a larger sample should assess whether this is a consistent feature of PAIS. Most of the cases involving cerebral artery infarction in our cohort were due to the lenticulostriate, which presented as an isolated basal ganglia lesion. Arteriopathies of small vessels induced by infections and Moyamoya, such as lenticulostriate arteries, are more common in childhood AIS than in adult AIS related to thrombolysis.10,11

The fact that two of the children in our cohort had normal CT findings confirmed that CT in the early phase after stroke is not sufficient for diagnosis of PAIS.12 Thus, this fact should be considered in clinical practice, particularly in children with small infarctions. In addition, these conclusions indicate that MRI or/and MRA should be considered as necessary standard methods to detect the involved locations of PAIS.

The etiologies of our PAIS cases were different from the known etiologies of adult stroke, as expected. We were, however, surprised to find that 38.1% of our PAIS cases were related to infectious disease, indicating that pathogen infection and immune defense may be key factors of PAIS. The systemic inflammatory response causes a hypercoagulable state and may also cause direct damage to vascular endothelial cells,13,14 which may promote onset of PAIS. Additionally, in our case population, iron-deficiency anemia was the second most frequent etiology. However, a previous study of Africans found a higher percentage of pediatric stroke cases caused by sickle cell anemia.15 It is believed that anemia disorders may promote stroke via one of the following pathways:16,17 (1) low iron storagerelated disease primary thrombocythemia and accumulation of deformed red blood cells may increase blood viscosity, leading to PAIS; (2) anemia-related reduction of cerebral perfusion, known as anemic anoxia or anoxemia, may disrupt neurological pathways. Iron-deficiency anemia is relatively easy to treat, by administration of oral iron supplements or high-iron food sources, thereby reducing any associated risk of PAIS.

Moyamoya syndrome is another putative risk factor of PAIS that was identified in our Chinese case population as well as case studies from non-Chinese countries.18,19 Moyamoya syndrome is a chronic, non-inflammatory, intracranial vascular occlusion, characterized by stenotic or occluded internal carotid artery, hyperplasia of the carotid artery siphon and deep perforators, and new formation and expansion of abnormal vascular networks. Worldwide, this syndrome has the highest incidence in Japan, but incidence in other Asian countries, such as China, are steadily increasing. In addition, the increased diagnosis of moyamoya syndrome has been accompanied by an increase in its identification as an etiology of PAIS. The underlying etiology of moyamoya syndrome itself remains unclear. Genetic studies have found an association with other genetic disorders related to chromosome 3 3p24.2–26, chromosome 6 D6s441, and chromosome 17.20 It is possible that moyamoya syndrome may play a role in genetic risk of PAIS, but further studies need to be conducted.

In conclusion, PAIS is a rare but severe childhood disease that remains to be fully understood the etiology of PAIS appears to be distinctive for Chinese and Caucasian populations, possibly due to genetic differences, or even differences in lifestyle or environment of Western and Asian countries. Moreover, the Chinese PAIS patients showed distinctive features according to younger age. Multicenter collaborative studies are needed to better understand the ethnic- and age-specific clinical presentations, radiologic characteristics, and etiologies of PAIS.

Back to Top | Article Outline


1. Lynch J, Hirtz D, DeVeber G, Nelson KB. Report of the national institution of neurological disorders and stroke workshop on perinatal and childhood stroke. Pediatrics 2002; 109: 16-123.
2. Fullerton H, Chetkovick DM, Wu YW, Smith WS, Johnston SC. Deaths of stroke in US children. Neurology 2002; 59: 34-39.
3. Matta AP, Galvao KR, Oliveira BS. Cerebrovascular disorders in childhood: etiology, clinical presentation, and neuroimaging findings in a case series study. Arq Neuropsiquiatr 2006; 64: 181-185.
4. Goeggel Simonetti B, Steinlin M, Arnold M, Fischer U. Arterialischaemic Stroke in childhood. Ther Umsch 2012; 69: 536-542. Ganesan V, Hogan A, Shack N, Gordon A, Isaacs E, Kirkham FJ. Outcome after ischaemic stroke in childhood. Dev Med Child Neurol 2000; 42: 455-461.
5. Steinlin M, Pfister I, Pavlovic J, Everts R, Boltshauser E, CaponeMori A, et al. The first three years of the Swiss Neuropediatric Stroke Registry (SNPSR): a population-based study of incidence, symptoms, and risk factors. Neuropediatrics 2005; 36: 90-97.
6. Prengler M, Sturt N, Krywawych S, Surtees R, Liesner R, Kirkham F. Homozygous thermolabile variant of the methylenetetrahydrofolate reductase gene: a potential risk factor for hyperhomocysteinaemia, CVD, and stroke in childhood. Dev Med Child Neurol 2001; 43: 220-225.
7. Delsing BJ, Catsman-Berrevoets CE, Appel IM. Early prognostic indicators of outcome in ischemic childhood stroke. Pediatr Neurol 2001; 24: 283-289.
8. Matta AP, Galvao KR, Oliveira BS. Cerebrovascular disorders in childhood: etiology, clinical presentation, and neuroimaging findings in a case series study. Arq Neuropsiquiatr 2006; 64: 181-185.
9. Amlie-Lefond C, Chan AK, Kirton A. Thrombolysis in acute childhood stroke: design and challenges of the thrombolysis in pediatric stroke clinical trial. Neuroepidemiology 2009; 32: 279- 286.
10. Ganesan V. Thrombolysis in paediatric arterial ischaemic stroke. Dev Med Child Neurol 2009; 51: 90-91.
11. Buerki S, Roellin K, Remonda L, Mercati DG, Jeannet PY, Keller E, et al. Neuroimaging in childhood arterial ischaemic stroke: evaluation of imaging modalities and aetiologies. Dev Med Child Neurol 2010; 52: 1033-1037.
12. Lindsberg PJ, Grau AJ. Inflammation and infections as risk factors for ischemic stroke. Stroke 2003; 34: 2518-2532.
13. Eleftheriou D, Ganesan V, Hong Y, Klein NJ, Brogan PA. Endothelial injury in childhood stroke with cerebral arteriopathy: a cross-sectional study. Neurology 2012; 79: 2089-2096.
14. Ganesan V, Prengler M, McShane MA, Wade AM, Kirkham FJ. Investigation of risk factors in children with arterial ischemic stroke. Ann Neurol 2003; 53: 167-173.
15. Jomova K, Valko M. Importance of iron chelation in free radicalinduced oxidative stress and human disease. Curr Pharm Des 2011; 17: 3460-3473.
16. Mehta PJ, Chapman S, Jayam-Trouth A, Kurukumbi M. Acute ischemic stroke secondary to iron deficiency anemia: a case report. Case Rep Neurol Med 2012; 2012: 487080.
17. Salih MA, Murshid WR, Al-Salman MM, Abdel-Gader AG, Al- Jarallah AA, Alorainy IA, et al. Moyamoya syndrome as a risk factor for stroke in Saudi children. Novel and usual associations. Saudi Med J 2006; 27: s69-s80.
18. Rana KS, Behera Mk, Adhikari KM. ischemic stroke following mild head injury is the cause. India Pediatr 2006; 43: 994-997.
19. Sakurai K, Horiuchi Y, Ikeda H, Ikezaki K, Yoshimoto T, Fukui M, et al. A novel susceptibility locus for moyamoya disease no chromosome 8q23. J Hum Genet 2004; 49: 278-281.

pediatric; arterial ischemic stroke; etiology; neuroimaging; diagnostic features

© 2013 Chinese Medical Association