Abstract 39: Timing Of Infection And Risk Of Arterial Ischemic Stroke In Children

Abstract

Background. Minor acute infections have been associated with arterial ischemic stroke (AIS) in both adults and children, and exposure to multiple infections over time is thought to contribute to atherosclerosis in adults. Our goal was to examine the association between infection and stroke, and assess whether timing or burden of infection is a stronger predictor of childhood AIS. Methods. We performed a case-control study nested in a cohort of 2.5 million children (<20y) enrolled in a large integrated health care plan (1993-2007). Non-neonatal AIS cases were identified from searches of diagnostic and radiology databases, and confirmed through independent chart review; 3 controls per case were randomly selected from the cohort, matched by age and primary care facility. A medical records analyst abstracted data on all out-patient visits for infection during the 2 years prior to the stroke, or index date in controls (stroke date for the paired case). The proportion of subjects who had an infectious visit in each period: from 0-2 days, 3-7 days, 8-28 days, 1-3 months, 3-6 months, 6-12 months, and 12-24 months prior to stroke was determined. Cases were compared to controls using conditional logistic regression and the Kaplan-Meier method. Results. We identified 126 cases (median age 10.5 years) and 378 controls. Infectious visits within the prior 2 years were documented in 79% of cases and 62% of controls. A striking disparity in number of cases and controls with an infectious visit was observed in the first 2 days prior to stroke: 29% of cases vs. 1% of controls (p<0.0001). From 3-7 days prior, 13% of cases and 2% of controls had a visit for infection (p<0.0001). Analysis with Kaplan-Meier methods, defining time-to-event as days prior to stroke/index date of most proximate infection, yielded a significant difference between survival functions for cases vs. controls (p<0.0001). In conditional logistic regression adjusted for gender, and hematologic, autoimmune, and cardiac disease, a visit with infection in the 2 days prior to stroke was associated with a relative hazard of 182 (p<0.0001), which dropped to 10 (p<0.0001) and 2 (p=0.046) for periods of 3-7 and 8-28 days prior. Conclusions. The risk of stroke is substantially elevated among children in the week after a visit for infection. The risk was greatest in the first 2 days after infection but remained elevated for the first month, even after adjustment for the presence of chronic disease. After the first month, infectious burden was not significantly different between cases and controls, and does not appear to be associated with an increased risk of childhood AIS.