Ankle-brachial index predicts stroke in the general population in addition to classical risk factors

Abstract

Background: Predictors of future stroke events gain importance in vascular medicine. Herein, we investigated the value of the ankle-brachial index (ABI), a simple non-invasive marker of atherosclerosis, as stroke predictor in addition to established risk factors that are part of the Framingham risk score (FRS).

Methods: 4299 subjects from the population-based Heinz Nixdorf Recall study (45-75 years; 47.3% men) without previous stroke, coronary heart disease or myocardial infarcts were followed up for ischemic and hemorrhagic stroke events over 109.0±23.3 months. Cox proportional hazard regressions were used to evaluate ABI as stroke predictor in addition to established vascular risk factors (age, sex, systolic blood pressure, LDL, HDL, diabetes, smoking).

Results: 104 incident strokes (93 ischemic) occurred (incidence rate: 2.69/1000 person-years). Subjects suffering stroke had significantly lower ABI values at baseline than the remaining subjects (1.03±0.22 vs. 1.13±0.14, p<0.001). In a multivariable Cox regression, ABI predicted stroke in addition to classical risk factors (hazard ratio=0.77 per 0.1, 95% confidence interval=0.69-0.86). ABI predicted stroke events in subjects above and below 65 years, both in men and women. ABI specifically influenced stroke risk in subjects belonging to the highest (>13%) and intermediate (8-13%) FRS tercile. In these subjects, stroke incidence was 28.13 and 8.13/1000 person-years, respectively, for ABI<0.9, compared with 3.97 and 2.07/1000 person-years for 0.9≤ABI≤1.3.

Conclusions: ABI predicts stroke in the general population, specifically in subjects with classical risk factors, where ABI identifies subjects at particularly high stroke risk.

Keywords: Peripheral artery disease; Stroke prediction; Subclinical atherosclerosis.