Abstract P107: Thirty-Year Risk of Incident and Recurrent Ischemic Stroke in Individuals With Sickle Cell Trait and Modification by Midlife Risk Factors: The Atherosclerosis Risk in Communities Study

Abstract

Introduction: Sickle cell trait (SCT) has been associated with ischemic stroke in some, but not all, epidemiologic cohorts. Heterogeneity of the association may be related to differences in demographic subgroups or midlife risk factors. Methods: The ARIC study is an observational cohort based on 4 US communities. Participation has included 6 clinical visits and ongoing surveillance of acute cardiovascular events. Black participants were genotyped for SCT, excluding any with sickle cell anemia or hemoglobin SC disease. Global genetic ancestry was determined from whole genome variation. Midlife factors were ascertained at ARIC visit 1 (1987-1989), or if missing imputed from visit 2 (1990-1992). Acute ischemic stroke was classified by physician review of the medical record. Associations between SCT and ischemic stroke were analyzed using repeat-events Cox regression with robust estimators, counting all occurrences of ischemic stroke per participant. Models were adjusted for sex, age, global genetic ancestry, and midlife factors (smoking, hypertension, hyperlipidemia, diabetes, chronic kidney disease, and previous myocardial infarction). Effect modification was tested by the multiplicative interaction of SCT with each midlife factor. Results: Our population was composed of 3,827 black adults with complete midlife data (32 [0.8%] excluded). SCT was identified in 255 (7%). The mean age at visit 1 was 53 years; the median follow-up time was 26 years. Most (63%) were women. At midlife, individuals with SCT had a greater prevalence of hyperlipidemia (32% vs. 26%; P =0.04) and chronic kidney disease (20% vs. 13%; P =0.003). From 1987-2016, 579 ischemic strokes occurred in 456 individuals. SCT was more strongly associated with ischemic stroke in participants with chronic kidney disease [HR=2.01 (1.13 - 3.60)] than those without [HR=1.10 (0.77 - 1.57)], P for interaction = 0.05, Figure 1 . Conclusions: The association between SCT and ischemic stroke is heterogeneous and appears to be stronger in those with chronic kidney disease.