Abstract

Abstract 3993

Poster Board III-929

Background

The incidence of stroke in children is estimated at about 2.6 per 100,000 per year. Risk factors include congenital heart malformations, trauma, hemolytic anemias, collagen tissue diseases, inborn metabolic disorders, and infectious diseases. Apart from acquired thrombophilic risk factors, such as the presence of antiphospholipid antibodies, inherited thrombophilias (IT) have been found to be associated with stroke in infants and children. However, results of single studies on the risk of stroke onset associated with IT have been contradictory or inconclusive, mainly due to lack of statistical power. The aim of this study was to estimate the impact of thrombophilia (IT) on risk of childhood stroke via meta-analysis of published observational studies.

Methods and Results

A systematic search of electronic databases (Medline, EMBASE, OVID, Web of Science, The Cochrane Library) for studies published from 1970 to 2009 was conducted using key words in combination both as MeSH terms and text words. Citations were independently screened by two authors and those meeting the a priori defined inclusion criteria were retained. Data on year of publication, study design, country of origin, number of patients/controls, ethnicity, stroke type (arterial ischemic stroke [AIS]; cerebral venous sinus thrombosis [CSVT]) were abstracted. Publication bias indicator and heterogeneity across studies were evaluated, and summary odds ratios (ORs) and 95% confidence intervals (CIs) were calculated using both fixed-effects and random-effects models. Twenty-one of 185 references found met inclusion criteria. 1698 patients (AIS: 1291; CSVT: 407) and 2913 controls aged neonate to 18 years were enrolled. No significant heterogeneity was discerned across studies, and no publication bias was detected. A statistically significant association with stroke onset was demonstrated for each IT trait evaluated, with no difference found between AIS (table) and CSVT. Summary ORs/CIs (random-effects model) for AIS & CSVT cohorts were as follows: Protein C-deficiency (8.76/4.53-16.96), FV G1691A (3.34/2.66-4.26), FII G20210A (2.50/1.67-3.74), MTHFR T677T (1.61/1.21-2.14), antiphospholipid antibodies (5.84/3.06-11.18), elevated lipoprotein (a) (6.24/4.51-8.64), and combined ITs (8.85/3.32-23.57). Carrier rates reported for antithrombin- or protein S deficiency among patients were 1.5% and 1.6% as compared with 0.06% (p<0.001) and 0.4% (p=0.003) in healthy controls.

Conclusions

The present meta-analysis indicates that IT serve as risk factors for incident stroke. However, the impact of IT upon outcome and recurrence risk needs to be further investigated.

Table 1

Summary odds ratios (ORs)/95%-confidence intervals (CIs: meta-analysis) including testing for heterogeneity (I2), non-combinability, and publication bias for thrombophilic risk factors associated with a first AIS onset in children

IT [N of studies] patients / controls (N) OR/ CI; p-value: fixed model OR/ CI; p-value: random model I2 [%], p-value bias indicator (Harbord): 
acquired risk factors      
protein C-def. [n=8] 844/1207 11.0/5.13-23.59 9.74/4.61-20.57 0, p=0.96 p=0.60 
APS/LA* [n=8] 930/1194 6.95/3.67-13.14 5.84/3.05-11.17 0, p=0.82 p=0.86 
genetic risk factors      
lipoprotein (a) [n=5] 722/909 6.27/4.52-8.69 6.24/4.51-8.64 0.0, p=0.91 p=0.64 
FV G1691A [n=15] 981/2190 3.70/2.82-4.85 3.75/2.85-4.92 0, p=0.67 p=0.22 
FII G20210A [n=11] 1000/1858 2.60/1.66-4.08 2.73/1.70-4.39 0, p=0.46 p=0.78 
MTHFR TT** [n=10] 718/1287 1.58/1.20-2.08 1.61/1.21-2.14 3.5, 0.40 p=0.42 
> 2 genetic traits [n=9] 701/1265 18.75/6.49-54.14; 12.58/3.89-40.61; 0, p=0.59 p=0.61 
IT [N of studies] patients / controls (N) OR/ CI; p-value: fixed model OR/ CI; p-value: random model I2 [%], p-value bias indicator (Harbord): 
acquired risk factors      
protein C-def. [n=8] 844/1207 11.0/5.13-23.59 9.74/4.61-20.57 0, p=0.96 p=0.60 
APS/LA* [n=8] 930/1194 6.95/3.67-13.14 5.84/3.05-11.17 0, p=0.82 p=0.86 
genetic risk factors      
lipoprotein (a) [n=5] 722/909 6.27/4.52-8.69 6.24/4.51-8.64 0.0, p=0.91 p=0.64 
FV G1691A [n=15] 981/2190 3.70/2.82-4.85 3.75/2.85-4.92 0, p=0.67 p=0.22 
FII G20210A [n=11] 1000/1858 2.60/1.66-4.08 2.73/1.70-4.39 0, p=0.46 p=0.78 
MTHFR TT** [n=10] 718/1287 1.58/1.20-2.08 1.61/1.21-2.14 3.5, 0.40 p=0.42 
> 2 genetic traits [n=9] 701/1265 18.75/6.49-54.14; 12.58/3.89-40.61; 0, p=0.59 p=0.61 
Disclosures:

Manco-Johnson:Baxter BioScience: Honoraria; Bayer HealthCare: Honoraria; CSL Behring: Honoraria; NovoNordisk: Honoraria; Octapharma: Honoraria. Off Label Use: Enoxaparin (LMWH) is used off-label in children to prevent symptomatic thromboembolism.

Author notes

*

Asterisk with author names denotes non-ASH members.