Abstract TMP8: Varicella Zoster Virus Infection in Children With Arterial Ischemic Stroke: The Chicken or the Egg?

Abstract

Introduction: Serologic evidence of varicella zoster virus (VZV) infection has been associated with childhood arterial ischemic stroke (AIS). Questions of causality persist: does VZV infection cause AIS, or does AIS reactivate VZV? The Vascular effects of Infection in Pediatric Stroke (VIPS II) study aimed to examine this relationship. Methods: This North American 22-center prospective cohort study enrolled 205 children (28 days-18 years) with AIS (2017-2022). Blood samples were hyperacute (≤72hrs; n=194), acute (4-7 days; n=180) and convalescent (1-6 weeks; n=73). A virology research lab used ELISA to measure VZV IgM and IgG titers. A pediatric virologist (CG) interpreted the results and classified the evidence of infection. Low-level IgG seropositivity is consistent with prior vaccination/infection. Highly elevated IgM in hyperacute/acute samples indicated reactivation coincident with stroke . Rising IgG titers between hyperacute and convalescent samples, occasionally with a high IgM in convalescent sample, indicated reactivation after the stroke . Results: Median age (IQR) was 11.6 years (5.3, 15.6). All 205 cases were low-level IgG positive; 160 (78%) reported prior VZV vaccination and 3 reported remote chicken pox. Serologies for 15 (7.3%; 95% CI 4.1, 11.8%) indicated VZV reactivation coincident with stroke; prior VZV vaccination reported in 14 and unknown in one. Stroke etiology amongst these 15 was definite arteriopathy in 4 (27%), possible arteriopathy in 4 (27%), cardioembolic in 5 (33%), and idiopathic in 2 (13%) (compared to 37%, 37%, 10%, 14%, respectively, for the other 190). Among 73 with convalescent samples, 17 (23%) had VZV reactivation after stroke. All cases of VZV reactivation were asymptomatic zoster sine herpete (herpes zoster without rash). Conclusions: While stroke triggered VZV reactivation in 23% of cases, 7.3% had evidence of zoster sine herpete at the time of their stroke, implying a causal role. We could not distinguish antibodies to vaccine virus versus wild-type; these zoster cases could represent reactivation of vaccine virus or a break-through wild-type varicella infection. Either scenario would be unexpected in this current era of routine childhood VZV vaccination and low rates of pediatric chicken pox or zoster.