
* Neuroepidemiology Branch
Family Studies Unit, National Institute of Neurological Disorders and Stroke, Bethesda, Maryland
| ABSTRACT |
|---|
|
|
|---|
Methods.A series of 59 children (age 018 years) with arterial ischemic stroke or porencephaly were referred to the National Institutes of Health. A blood sample, buccal smear sample, questionnaire, and pedigree were requested for each child. Blood samples were analyzed for protein C (PC); protein S; antithrombin (AT); activated PC resistance (APCR); lipoprotein (a) [Lp(a)]; lupus anticoagulant; anticardiolipin antibodies; and the methylenetetrahydrofolate reductase C677T (MTHFR), factor V G1619A, factor II G20210A (PT), plasminogen activator inhibitor-1 4G6755G, and tissue factor pathway inhibitor C536T mutations. The frequency of each coagulation abnormality was compared with published international pediatric stroke case and control rates.
Results.At least 1 prothrombotic abnormality was identified in 63% (36 of 57) of children studied, including plasminogen activator inhibitor-1 4G6755G (15 of 56), MTHFR (12 of 56), elevated Lp(a) (12 of 59), APCR (11 of 58), factor V G1619A (5 of 57), PT (3 of 57), PC deficiency (1 of 59), and AT deficiency (1 of 59). The MTHFR mutation, elevated Lp(a), the PT mutation, and AT deficiency rates were similar to rates in cases and more common than control subjects in previously published studies. The rate of children with APCR or multiple abnormalities was higher than in previous pediatric stroke studies. A family history of early thrombosis was identified in one third of the children with a prothrombotic abnormality.
Conclusions.Two thirds of children in this study had at least 1 of the prothrombotic risk factors tested, and several children had multiple risk factors. These results provide additional evidence that prothrombotic abnormalities are common among children with AIS or porencephaly.
Key Words: cerebrovascular diseases children coagulation genetic polymorphisms
Abbreviations: PC, protein C PS, protein S AT, antithrombin FVL, factor V G1619A PT, factor II G20210A PAI-1, plasminogen activator inhibitor -1 4G6755G MTHFR, methylenetetrahydrofolate reductase C677T AIS, arterial ischemic stroke APCR, activated protein C resistance Lp(a), lipoprotein (a) LA, lupus anticoagulant ACLA, anticardiolipin antibodies TFPI, tissue factor pathway inhibitor C536T Ig, immunoglobulin
During the past decade, interest has grown in the contribution of inherited coagulation and metabolic factors to the risk for stroke in adults and children. Genetic factors probably play a larger role in individuals who present with stroke at an early age as compared with onset in later years. Studies of risk factors for stroke in children, who lack the accumulated comorbidities that contribute heavily to stroke in adults, may yet provide important insights into the pathogenesis of stroke in adults. For example, clinical observations of children with homocystinuria and vascular disease led to epidemiologic studies and clinical trials that resulted in the identification of hyperhomocysteinemia as an important contributor to vascular disease in adults.1
Stroke in children is a heterogenous disorder associated with significant morbidity and mortality. Incidence has been estimated at
1 per 4000 for neonates and has ranged from 1 per 7000 to 1 per 70000 for older children (1 month to 18 years).2,3 The most frequently reported risk factors for stroke in children include cardiac disorders, prothrombotic disorders, metabolic disorders, vascular disorders, and infection.4,5 In up to two thirds of children, no specific cause has been determined.6
Many genetic and acquired prothrombotic abnormalities have been evaluated in children with cerebrovascular disease, but there has been a wide variation in the prevalence of these abnormalities among different age groups, stroke subtypes, and international populations. The most common prothrombotic factors studied include (1) deficiencies in anticoagulant proteins (protein C [PC], protein S [PS], and antithrombin [AT]), (2) antibodies to phospholipids associated with thrombosis, (3) genetic polymorphisms encoding proteins associated with thrombosis (factor V G1619A [FVL], factor II G20210A [PT], and plasminogen activator inhibitor-1 4G6755G [PAI-1]) and homocysteine metabolism (methylenetetrahydrofolate reductase C677T [MTHFR]), and (4) lipoproteins that predispose to thrombosis. Several case-control studies, using hospital-based adult and population-based child control subjects, have shown an association between many of these abnormalities and stroke,79 whereas other studies have been negative or too small to show a difference.1013 The most consistently reported associations are for PC deficiency, elevated lipoprotein (a), and the FVL mutation. Recent evidence also suggests that children with specific prothrombotic abnormalities are at a greater risk for recurrent stroke.4
Currently, it is unclear to what degree these abnormalities contribute to the development of childhood stroke, either individually or in the presence of other genetic or environmental risk factors. Although a thorough evaluation for thrombophilia seems reasonable for children with stroke, the diagnostic yield is low in adults.14 There are no evidence-based guidelines for screening or treatment of children with stroke and a prothrombotic abnormality. The purpose of this pilot study was to compare the frequencies of genetic and functional coagulation abnormalities in children with arterial ischemic stroke or porencephaly with previously published population levels to plan for future genetic association studies.
| METHODS |
|---|
|
|
|---|
Diagnosis of Stroke and Porencephaly
Arterial ischemic stroke (AIS) was defined as a new focal neurologic deficit that lasted 24 hours or longer and was presumably due to a vascular process; porencephaly was defined as a fluid-filled cavity within the cerebral hemispheres that may communicate with cerebrospinal fluid spaces. The diagnosis of AIS or porencephaly was confirmed by computed tomography or MRI for each child.
Study children were classified into 3 categories: (1) perinatal stroke, defined as systemic or neurologic symptoms with radiographic evidence of AIS or porencephaly that occurred before birth or in the first 30 days of life; (2) delayed perinatal stroke, defined as no diagnosis of stroke in the perinatal or neonatal period, focal motor dysfunction or seizures diagnosed after 2 months of age, and neuroimaging evidence of remote focal infarction, encephalomalacia, and or porencephaly; and (3) childhood stroke, defined as an acute onset of systemic or neurologic symptoms with radiographic evidence of AIS that occurred between 30 days and 18 years of age. Children with a history of cancer, heart disease, trauma, chromosomal or metabolic disorders, sickle cell disease, or central nervous system infection were excluded.
Data Abstraction
Study investigators reviewed available medical records for each child. Parents completed a questionnaire regarding maternal and birth history, stroke history, and functional status of their child with a diagnosis of stroke. Family histories were obtained by a certified genetic counselor (L.N.) through telephone interviews for a majority of the children in the study. Parents of probands were instructed to contact family members to determine a family history of a major vascular event. A positive family history was defined as a major thrombotic event (myocardial infarction, deep venous thrombosis, pulmonary embolus, or ischemic stroke) before 60 years of age in first- or second-degree relatives of the probands.
Laboratory Methods
Blood samples were collected at least 3 months after the incident stroke for each child. Blood was collected by peripheral venipuncture into plastic tubes that contained 0.105 M buffered sodium citrate (3.2%). Citrated blood was placed immediately on ice and was processed within 1 hour. Citrated blood was centrifuged at 1700g for 15 minutes at room temperature. Plasma supernatant was removed, transferred to plastic tubes, recentrifuged for 5 minutes to ensure platelet-free plasma, and then stored at 70°C. The buffy coat layer was subsequently removed and stored at 70°C before DNA extraction. Blood samples were frozen at 70°C and stored in batches until they were shipped on dry ice to the Hemostasis Reference Laboratory at the Henderson Research Centre (Hamilton, ON, Canada) for analysis.
Blood samples were analyzed for PC, PS, AT, activated PC resistance (APCR), lipoprotein (a) [Lp(a)], lupus anticoagulant (LA), anticardiolipin antibodies (ACLA), and the MTHFR, FVL, PT, PAI-1, and tissue factor pathway inhibitor C536T (TFPI) mutations. Age-related standardized ranges were used to determine normal and abnormal levels for PC, PS, AT, APCR, Lp(a), LA, and ACLA. In some cases, coagulation testing had been performed before enrollment in the study, but this information was not available to investigators at the time of screening for inclusion.
PC was measured on a BCT or BCS analyzer using a functional assay (Protein C Reagent Kit; Dade Behring, Marburg, Germany), AT was measured by chromogenic assay (Coamatic Antithrombin Assay; Diapharma, West Chester, OH), and PS and Lp(a) were measured by enzyme-linked immunosorbent assay (Asserchrom Free Protein S, Diagnostica Stago, Asnieers, France; and Tintelize Kit, Biopool, Sweden, now Trinity Biotech, County Wicklow, Ireland). APCR was assayed by Dade Actin FS aPTT reagent (Dade Behring) and activated PC (American Diagnostics, Greenwich, CT).15 LA was tested using PTT-LA reagent (Diagnostica Stago) and dilute Russell Viper Venom (American Diagnostics, Greenwich, NJ, or BioMerieux, Durham, NC). LA samples with abnormal results were repeated using a confirmatory test, either Hexogonal Phospholipid or dRVV confirm. ACLA, immunoglobulin (Ig) G, IgM, and IgA, were measured by enzyme-linked immunosorbent assay (Corgenix, Westminster, CO). The MTHFR, FVL, PT, PAI-1, and TFPI mutations were identified by amplification and polymerase chain reactions followed by digestion of amplified fragments with restriction enzymes, as previously reported.1620
Published Population Cases and Controls
For comparisons, we identified reports, using a search of the MEDLINE database, published in English between 1995 and May 2004, that evaluated the presence of genetic coagulation abnormalities in populations of children with cerebrovascular disease. We abstracted information on type and prevalence of genetic coagulation abnormality, study design, patient and control populations, age range, and methods of analysis.
Statistical Methods
The percentages of each coagulation abnormality were compared with published pediatric stroke case and control rates. Fisher exact tests were used for internal comparisons between children with and without a family history of thrombosis and between perinatal or delayed perinatal and childhood stroke.
| RESULTS |
|---|
|
|
|---|
|
At least 1 prothrombotic abnormality was identified in 63% (36 of 57) of children studied, and 28% (16 of 57) had multiple abnormalities. Several prothrombotic abnormalities were identified, including PC deficiency (1 of 59), AT deficiency (1 of 59), APCR (11 of 58), elevated Lp(a) (12 of 59), MTHFR (12 of 56), PAI-1 (15 of 56), FVL (5 of 57), and PT (3 of 57). No child was identified with PS deficiency, ACLA, LA, or the TFPI mutation. Tests for prothrombotic abnormalities were incomplete for 4 children as a result of inadequate DNA samples. Of the 4 children with incomplete testing, 2 did not have an identified abnormality (Table 2).
|
A pedigree was completed for 37 of the 59 children. Thrombosis that occurred before 60 years of age was identified in 4% of family members, none (0 of 118) of first-degree relatives of these children, and 5.7% (16 of 281) of second-degree relatives. There was a family history of thrombosis before 60 years of age in 31% (11 of 35) of children overall and among 33% (6 of 18) of children with a prothrombotic abnormality (P = 1.00).
| DISCUSSION |
|---|
|
|
|---|
The present study is the largest series of prothrombotic risk factors in American children with stroke. Two thirds of children in this study had at least 1 of the prothrombotic risk factors tested, and several children had multiple risk factors.
Prothrombotic Abnormalities and Children With Stroke
Several international case-control studies have revealed an association between prothrombotic abnormalities and stroke in children,79 whereas other studies were negative or too small to show a difference1013 (Table 3). The most consistent associations have been reported for PC deficiency, elevated Lp(a), and the FVL mutation. In this study, we found that rates of AT deficiency, elevated Lp(a), the PT mutation, and the MTHFR mutation were higher than those in published rates in pediatric stroke control subjects.
|
Lp(a) is structurally homologous to plasminogen and has a similar effect on coagulation, fibrinolysis, and endothelial function.23 An increased level of Lp(a) has been reported in 3 German studies9,24,25 and 1 Portuguese26 study to be a risk factor for stroke in neonates and children. Most studies in adults have yielded similar findings.27 We found elevated Lp(a) in 20% of the children in this study, a rate much higher than in published pediatric stroke control subjects (5.5%).25
The PT mutation is a genetic variant of the 3-prime untranslated region of the PT gene that leads to increased plasma PT levels. The PT mutation is an autosomal dominant mutation that varies widely in prevalence and alone is associated with a modest increase in risk for ischemic stroke, myocardial infarction, and peripheral vascular disease in adults.28 Studies of children with stroke have yielded mixed results. Case-control studies from Turkey7 and Germany9 were positive, whereas other international studies were negative or underpowered.8,13,26 We identified the PT mutation in 5% of children in this study, a rate slightly higher than the 0.9% to 2.8% reported in published pediatric stroke control subjects.
The association between elevated homocysteine and stroke is widely recognized. The MTHFR mutation has been shown to increase the risk for stroke in children and adults. A recent meta-analysis of patients who were younger than 55 years and had AIS revealed a modest increased risk for stroke among homozygous carriers of the MTHFR mutation.28 Despite these findings in younger adults and high rates in pediatric stroke patients, only 1 case-control study in children was sufficiently powered to show an association.9 We identified homozygosity for the MTHFR mutation in 21% of the children in this study, a rate higher than that among published pediatric stroke control subjects (5.615.2).7,8
Almost 30% of the children in this study had >1 of the prothrombotic abnormalities evaluated. Five children had 3 or more prothrombotic abnormalities. Other studies using an extensive prothrombotic panel have reported a combination of prothrombotic abnormalities in up to 23% of children with stroke.21 A combination of prothrombotic factors substantially increases the risk for stroke in children.9,24,29 Prothrombotic combinations that have been reported with an increased risk for stroke in children include the FVL mutation and elevated Lp(a) or the MTHFR mutation.9
Recurrent Stroke and Prothrombotic Abnormalities
Recurrent stroke is rare among infants with perinatal stroke30 but in children ranges from 6% to 30%.21,31,32 Combined data from Germany, Canada, and the United Kingdom of 565 patients with AIS revealed a recurrence rate of 10% after several years of follow-up.33 The MTHFR mutation, elevated homocysteine, elevated Lp(a), and PC deficiency have been associated with recurrent stroke in infants and children.4 In this study, 19% (3 of 16) of children with childhood stroke had a recurrence, all of which had a coagulation abnormality and 2 a family history of thrombosis.
Age-Related Differences in Frequency of Prothrombotic Abnormalities
The prevalence of prothrombotic abnormalities among individuals with stroke has varied according to age and suggests that genetic risk factors are more important in individuals who present with stroke at an early age. In studies that used an extensive screen for thrombophilia, prothrombotic abnormalities were more common in neonates25,34 as compared with children12,35 or adults with stroke.36 This study found no differences in the frequency of prothrombotic factors among neonates compared with older children, but this likely reflects the small number for comparison.
Family History of Thrombosis and Stroke in Children
A family history of disease provides information on disease susceptibility reflected through common genes, behaviors, and shared environment.37 A family history of vascular disease is an independent risk factor for stroke in young adults, and the risk that it confers varies by stroke subtype.38,39 There is limited information on the risk for stroke in children and neonates with a family history of stroke or thrombosis. Case series and controlled studies of children with stroke have found a family history of thrombosis among 5% to 42% of children studied.21,26,40 A family history of thrombosis was identified in one third of the children in this study and, similar to other studies, was not predictive of the presence of a coagulation abnormality.8 None of the children in this study had a sibling with a history of AIS or porencephaly.
Treatment of Children With Stroke and a Prothrombotic Abnormality
Prothrombotic abnormalities are associated with an increased risk for stroke and recurrent stroke in children,4,31 but there are as yet no evidence-based strategies or consensus guidelines for the treatment of children with stroke and a prothrombotic abnormality. Recommendations for antithrombotic therapy in children with thrombophilia have been based on small, nonrandomized studies and expert opinion.41 Antithrombotic therapy is recommended in adults for acute events and as prophylaxis in high-risk situations and may be warranted in children with specific prothrombotic abnormalities.42 The addition of other genetic risk factors for thrombosis in symptomatic individuals probably warrants long-term therapy, but this must be weighed against the risk for anticoagulation in young children. The response and long-term complications of antithrombotic therapy are not the same in adults and children.43 Clinical trials are needed to determine the indications and best treatment for children with a prothrombotic abnormality. The screening of unaffected relatives of probands with a prothrombotic mutation has been advocated because the risk for thrombosis can be reduced by prophylaxis in high-risk situations.44
| CONCLUSIONS |
|---|
|
|
|---|
| ACKNOWLEDGMENTS |
|---|
| FOOTNOTES |
|---|
Reprint requests to (J.K.L.) National Institute of Neurological Disorders and Stroke, Building 10, Room 5S220, 10 Center Dr, MSC 1447, Bethesda, MD 20892-1447. E-mail: lynchj{at}ninds.nih.gov
No conflict of interest declared.
PEDIATRICS (ISSN 0031 4005). Published in the public domain by the American Academy of Pediatrics.
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
R. Laugesaar, A. Kolk, T. Tomberg, T. Metsvaht, M. Lintrop, H. Varendi, and T. Talvik Acutely and Retrospectively Diagnosed Perinatal Stroke: A Population-Based Study Stroke, August 1, 2007; 38(8): 2234 - 2240. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. J.N.L. Benders, F. Groenendaal, C. S.P.M. Uiterwaal, P. G.J. Nikkels, H. W. Bruinse, R. A.J. Nievelstein, and L. S. de Vries Maternal and Infant Characteristics Associated With Perinatal Arterial Stroke in the Preterm Infant Stroke, June 1, 2007; 38(6): 1759 - 1765. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. Skardoutsou, K. A. Voudris, S. Mastroyianni, E. Vagiakou, G. Magoufis, and P. Koukoutsakis Moya Moya Syndrome in a Child With Pyruvate Kinase Deficiency and Combined Prothrombotic Factors J Child Neurol, April 1, 2007; 22(4): 474 - 478. [Abstract] [PDF] |
||||
![]() |
K. B. Nelson Perinatal Ischemic Stroke Stroke, February 1, 2007; 38(2): 742 - 745. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. P. Miller, Y. W. Wu, J. Lee, E. J. Lammer, D. M. Iovannisci, D. V. Glidden, S. L. Bonifacio, A. Collins, G. M. Shaw, A. J. Barkovich, et al. Candidate Gene Polymorphisms Do Not Differ Between Newborns With Stroke and Normal Controls Stroke, November 1, 2006; 37(11): 2678 - 2683. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||