Ischemic stroke during infancy and childhood has the potential for life-long morbidity. Information on the neurologic outcome of children who survive ischemic stroke is lacking. Children surviving ischemic stroke between January 1, 1995 and July 1, 1999 were prospectively followed. Neurologic deficit severity was based on the Pediatric Stroke Outcome Measure (PSOM) developed in this study and parental response to two recovery questions. Predictor variables for poor outcome were tested. One-hundred twenty-three children with arterial ischemic stroke and 38 with sinovenous thrombosis were followed for a mean of 2.1 years (range, 0.8 to 6.6 years). The primary outcome based on PSOM assessment was: normal, 37%; mild deficit, 20%; moderate deficit, 26%; and severe deficit, 16%. The secondary outcome was full recovery in 45% of patients, based on parental response. The primary and secondary outcome measures were moderately correlated (P < .001; K = 0.5). In bivariate analysis, arterial stroke type, male gender, age of at least 28 days, presence of associated neurologic disorders, and need for rehabilitation therapy after stroke were predictors of poor outcome (P < .05). Multivariate analysis showed that only arterial ischemic stroke, associated neurologic disorders, and presence of rehabilitation therapy were independent predictors of poor outcome (P < .02). Poor outcome in children after ischemic stroke is therefore frequent and more likely in the presence of arterial stroke, rehabilitation therapy, and associated neurologic disorders, which justifies clinical trials of treatment strategies in childhood ischemic stroke.

A subgroup of children with arterial ischemic stroke in the pre- or perinatal period present with delayed diagnosis. We identified 22 children who met the following criteria: (1) normal neonatal neurological history, (2) hemiparesis and/or seizures first recognized after two months of age, and (3) computed tomography or magnetic resonance imaging showing remote cerebral infarct. Laboratory evaluations included protein C, protein S, antithrombin, activated protein C resistance screen (APCR), Factor V Leiden (FVL), prothrombin gene defect, methylene tetrahydrofolate reductase variant (MTHFR), anticardiolipin antibody (ACLA), and lupus anticoagulant. Not all children received all tests. Age at last visit ranged from 8 months to 16.5 years (median 4 years). Twelve were boys. Fourteen had left hemisphere infarcts. Median age at presentation was 6 months. Eighteen had gestational complications. Fourteen children had at least transient coagulation abnormalities (ACLA = 11, ACLA + APCR = 1, APCR = 2 with FVL + MTHFR = 1); six of these children had family histories suggestive of thrombosis. Cardiac echocardiogram was unremarkable in the 15 tested. Outcomes included persistent hemiparesis in 22; speech, behavior, or learning problems in 12; and persistent seizures in five, with no evidence of further stroke in any patient. The persistence and importance of coagulation abnormalities in this group need further study.

Background and Purpose-The purpose of this study was to determine whether infection with varicella is causal for arterial ischemic stroke (AIS) in children. Methods-First, a prospective cohort study was conducted in young children (aged 6 months to 10 years) with AIS at 2 institutions (cohort study). The presence of varicella infection Ͻ12 months before AIS was determined and compared with the published frequency of varicella infection in the healthy pediatric population. The clinical and radiographic features of AIS were compared between the varicella and nonvaricella study cohorts. Second, a literature search of varicella-associated AIS was conducted, and the clinical and radiographic features were compared with the study nonvaricella cohort. Results-In the cohort study, 22 (31%) of 70 consecutive children with AIS had a varicella infection in the preceding year compared with 9% in the healthy population. Children in the varicella cohort were more likely to have basal ganglia infarcts (PϽ0.001), abnormal cerebral vascular imaging (PϽ0.05), and recurrent AIS or transient ischemic attacks (PϽ0.05) than those in the nonvaricella cohort. The pooled literature analysis of 51 cases of varicella-associated AIS showed similar findings to the varicella cohort. Conclusion-In young children with AIS, there is a 3-fold increase in preceding varicella infection compared with published population rates, and varicella-associated AIS accounts for nearly one third of childhood AIS. Varicellaassociated AIS has characteristic features, including a 2-fold increase in recurrent AIS and transient ischemic attacks. Varicella is an important risk factor for childhood AIS.