Abstract TMP55: Abdominal Obesity Predicts Stroke Risk in the Framingham Study

Abstract

Background: Previous studies have suggested that measurements of abdominal fat, such as waist circumference or waist-to-hip ratio, more accurately predict risk of stroke than body mass index (BMI). We assessed the risk of ischemic stroke associated with BMI, waist circumference, and waist-to-hip ratio in a community sample. Methods: We pooled Framingham Heart Study (FHS) participant data from FHS clinic examinations for the Original cohort (Exam 21, date range 1988-1992) and Offspring cohort (Exam 4, 1987-1991, and Exam 7, 1998-2001). Offspring participants who did not have stroke at the end of the first 10-year observation period could contribute data to the subsequent observation period. We included stroke-free participants 45 years of age or older who obtained measurements of BMI (kg/m 2 ) and both waist and hip circumference. Multivariable Cox proportional hazards regression models were used to separately relate BMI, waist circumference, and waist-to-hip ratio to risk of incident ischemic stroke over 10 years. Results: We analyzed data from 6,533 observation periods (mean age of participants 62±10 years, 54% female) with mean follow up time of 9.2 ±2 years. There were 240 (3.7%) incident ischemic stroke events. Each of the three body weight measurements was associated with increased stroke risk in models that adjusted for age and sex (Table). In full models, additionally adjusting for vascular risk factors that are more frequent in obese persons, only waist-to-hip ratio predicted risk of ischemic stroke. The hazard ratio (HR) per standard deviation increase in waist-to-hip ratio was 1.18, 95% CI 1.02-1.37, p=0.030. Conclusions: Higher waist-to-hip ratio predicted 10-year risk of ischemic stroke. Waist-to-hip ratio is a simple measure of abdominal obesity that should be used in future studies investigating weight loss interventions for stroke prevention.