Prethrombotic Disorders in Children With Arterial Ischemic Stroke and Sinovenous Thrombosis FREE

ARTERIAL ISCHEMIC stroke (AIS) and sinovenous thrombosis (SVT) are relatively uncommon events in children,^{1 - 2} yet few such epidemiological data have been reported.^{1 - 5} Although the pathogenesis of vascular occlusion is partially understood, inherited and acquired prethrombotic disorders have been identified as a cause of stroke in young people.^{6 - 26}

We performed a prospective study, including a complete hemostatic evaluation, to determine the frequency of prethrombotic disorders in a pediatric population with AIS and SVT who were treated at a single center.

From May 1992 to April 1997, a prospective study was performed of 30 consecutive children (13 girls and 17 boys) with AIS and 10 children (3 girls and 7 boys) with SVT, aged 3 months to 18 years, who were treated at the Hospital de Pediatría "Prof. Dr. Juan P. Garrahan", Buenos Aires, Argentina. Cardiological evaluation and echocardiography were performed in all cases. Children with AIS and congenital heart diseases were excluded from further analysis, and newborns were not studied because the authors were not involved in their treatment. Data concerning clinical presentation, underlying diseases and/or circumstantial risk factors, radiological findings, location, and family history of thrombosis were recorded.

Radiological diagnosis of the thrombotic episode and magnetic resonance imaging were made by computed tomography and magnetic resonance angiography, respectively, in all the patients, and by conventional angiography in 15 of 40 patients. Moyamoya syndrome was considered to be present when angiographic findings showed bilateral stenosis of the supraclinoid portion of the internal carotid arteries, profuse collateral network, and progression to complete obstruction of the main arterial channels by sparing of the distal branches.

On admission, laboratory studies included blood cell count; peripheral blood smear; sickle solubility test; serum complement C3 and C4; antinuclear antibodies; rheumatoid factor; and anti-DNA and fasting serum cholesterol, triglycerides, high-density lipoprotein, and low-density lipoprotein levels.

After parental informed consent was obtained, blood samples were collected into 3.2% sodium citrate at a ratio of 9:1 by clean venopuncture. Blood was centrifuged twice at 2500g for 15 minutes. Platelet poor plasma was immediately processed or stored at –70°C. Evaluation for prethrombotic disorders included the following assays in all patients: prothrombin time; activated partial thromboplastin time; thrombin time; reptilase time; fibrinogen; and factors V, VII, and XII activity by standard procedures. Functional activities of protein C, antithrombin, and plasminogen were measured by amidolytic assays (chromogenic substrates; Chromogenix AB, Mölndal, Sweden) and protein C and protein S by clotting assays using a model ST4 coagulometer (Diagnostica Stago, Asnières, France). Immunologic measurements of protein C, protein S (total and free), antithrombin, and plasminogen were made by the Laurell technique²⁷ using polyclonal rabbit antibodies against the respective antigens (protein C, protein S, and antithrombin; Dakopatts, Denmark; Assera plasminogen; Diagnostica Stago). Free protein S was assayed by precipitating bound protein with polyethylene glycol 8000 (3.75%) for 30 minutes at 4°C, followed by electroimmunoassay according to the method by Comp et al.²⁸ Presence of activated protein C resistance (APC-R) was tested using a commercial kit with factor V–deficient plasma as a prediluent for plasma samples (Chromogenix AB). DNA analysis for the factor V Leiden mutation was performed by standard polymerase chain reaction.²⁹

Children having a thrombotic episode before 1994 were called again at a later date for the last 2 tests. The presence of a lupus anticoagulant was determined, based on abnormal screening assay results, mixing studies, and confirmatory assay results, based on the method giving abnormal screening test results.³⁰ Anticardiolipin antibodies IgG and IgM isotypes were performed (Coaliza anticardiolipin IgG and IgM, Chromogenix AB) and calibrated against Harris standards from the University of Louisville, Ky. Serum samples with IgG and IgM values less than 10 µg/mL were defined as negative, 10 to 20 µg/mL as positive at low levels, 20 to 80 µg/mL as positive at moderate levels, and above 80 µg/mL as positive at high levels. Age-appropriate references were used to interpret hemostatic variables.³¹ Any abnormal result was confirmed with repeat assay 3 to 6 months after the acute thrombosis and 10 days after anticoagulation therapy was ceased.

According to proposed criteria for the antiphospholipid antibody syndrome, only those patients with positive lupus anticoagulant test results and/or positive IgG and/or IgM anticardiolipin antibodies at moderate or high levels in 2 determinations performed more than 8 weeks apart were considered positive for antiphospholipid antibody syndrome.³²

Parents were included in the study if any relevant hemostatic abnormality was detected in the child. A positive family history of thrombosis required a first- and/or second-degree relative with thrombosis at a relatively young age.

Table 1 shows baseline demographics, underlying diseases and/or circumstantial risk factors, presenting features, and radiological findings in our cohort. Underlying diseases and/or circumstantial risk factors were detected in 15 children (50%) with AIS and in all children with SVT, 5 of whom had 2 risk factors (Table 1). Seven children had multiple arterial infarcts, and 4 (3 girls and 1 boy) had moyamoya syndrome (1 also had common variable immunodeficiency and 1 presented with Down syndrome). Two of the other 3 patients had acute lymphoblastic leukemia and received Escherichia coli L-asparaginase therapy, and the remaining 1 was a girl with systemic lupus erythematosus (Table 1).

Radiologically, multiple occlusions of the anterior and middle cerebral arteries were detected in the 4 children with moyamoya syndrome and the girl with systemic lupus erythematosus. Multiple occlusions of the middle and posterior cerebral arteries were found in the 2 patients with acute lymphoblastic leukemia and E. coli L-asparaginase therapy (Table 1). Family history of thrombosis was positive for 5 children (Table 1). The laboratory findings of these patients are summarized in Table 2 and Table 3, and their clinical and radiological features are summarized in Table 4.

The incidence of stroke in children is much lower than in adults.^{1 - 2} However, the interest in studying this subject in children has increased in recent years.^{1 - 26 ,33} In our study, some epidemiological data concerning sex predominance, frequency according to age, and frequency of risk factors were similar to other series.^{1 - 4} Prethrombotic disorders may cause or contribute to the pathogenesis of thrombosis.³⁴ Inherited or acquired prethrombotic disorders have been reported in children with AIS or SVT.^{6 - 26} Only one study²⁶ prospectively evaluated a large series of pediatric patients to define the contribution of prethrombotic conditions to cerebral thromboembolism. Therefore, the prevalence of different hemostatic abnormalities in the pediatric population is not well determined.

In our series, 9 (30%) of the 30 children with AIS had prethrombotic disorders. No APC-R or factor V Leiden mutation was found in them. An association between APC-R and factor V Leiden mutation and cerebrovascular disease remains to be determined.^{35 - 39} Only a few reports have found a relationship in young children between AIS and APC-R and factor V Leiden mutation.^{23 - 25} We found acquired antithrombin deficiency in children with acute lymphoblastic leukemia or lymphoma who had received E coli L-asparaginase. This drug interferes with protein synthesis, resulting in acquired deficiency of this coagulation protein.⁴⁰ Plasminogen deficiency is frequently associated with thrombophilia.⁴¹ Available studies³⁴ still do not permit definitive statements to be made on the association with a thrombotic risk. In the child of our series with AIS and plasminogen deficiency, the development of thrombosis is thought to be caused by an interaction of this defect with an acquired factor such as antiphospholipid antibodies.³⁴

Compared with previous reports^{2 ,4} showing that SVT occurs less often in the absence of predisposing factors, our study showed that all the children had some additional risk factor, and in 5 (50%) of the 10 children, a prethrombotic disorder was detected. The APC-R test and factor V Leiden mutation have been associated with increased risk for venous thrombosis.⁴² One report⁴³ described these defects as a frequent coagulation abnormality in adults with SVT. Considering the allelic frequency of 1.25% in healthy subjects in Argentina,⁴⁴ no final conclusion could be obtained in our series because few children were studied.

Lower frequency (13%) of lupus anticoagulant and/or anticardiolipin antibodies was found in our patients with AIS and SVT compared with other pediatric series,^{13 - 14 ,18 - 21} most of which evaluated children with systemic lupus erythematosus.

We did not examine the prothrombin gene that could potentially increase the frequency of abnormalities in young patients with arterial or venous thrombosis.⁴⁵ Such prethrombotic disorders were regarded as inherited if the family had another member with the same abnormality, whereas a persistent antiphospholipid antibodies positivity pointed to an acquired disorder.

Although no family history of thrombosis was found in 88% of our patients, inherited thrombophilia should not be excluded from consideration because the defect may be caused by a fresh mutation (which is rare) or because affected relatives are still asymptomatic.³⁴

In this prospective study, most children evaluated had underlying risk factors and prethrombotic disorders. Therefore, we recommend performing a hemostatic evaluation of all children with AIS and SVT, despite the presence of obvious clinical risk factors or lack of family history of thrombosis.

Accepted for publication September 12, 1998.

We thank Maureen Andrew, MD, for critical review of the manuscript and Paul Monagle, MD, for helpful comments.

Reprints: Mariana Bonduel, MD, Hospital de Pediatría "Prof. Dr. Juan P. Garrahan," Combate de los Pozos 1881, Buenos Aires, Argentina 1245 (e-mail: garahan@giga.com.ar).