Association of ankle-brachial index level with stroke

There are conflicting data on the relationship between peripheral arterial disease (PAD) as defined by an ankle-brachial index (ABI) of <0.9 and stroke risk, and these prior data have largely focused on a single race or gender. Furthermore, the association between subsequent ABIs above the established 0.9 threshold for PAD vs. stroke has not been studied. This study assessed stroke occurrence across the full spectrum of ABI values among a broad cohort of stroke survivors.

Data from the National Health and Nutrition Examination Survey from 1999-2004, a nationally representative sample of United States adults, were used to assess prevalence of PAD as evidenced by ABI of <0.9, and to simultaneously identify the independent associations of ABI levels (<0.90, 0.90-0.99, 1.00-1.09, >1.3) with stroke prevalence.

Among 6382 adults, 4602 (72%) had full and complete data of which 277 (6%) reported stroke occurrence. Stroke survivors were older and more likely to have diabetes, hypertension, coronary artery disease, elevated blood pressure and glycohemoglobin, than non-stroke respondents. More subjects with ABI <0.9 vs. > or =0.9 had experienced a stroke (12.7% vs. 5.4%, p<0.001). Within the full ABI range, stroke presence rose with decreasing ABI (p=0.0013). After adjustment for potential confounders, only ABI <0.9 (OR 1.9, 95% CI=1.2-3.1) and 0.9-0.99 (OR 2.3, 95% CI=1.4-3.6) compared with the referent group (1.10-1.29) were significantly associated with presence of stroke.

Data from this nationally representative cohort support an independent relationship between established PAD and stroke occurrence, but also indicate that even borderline PAD is strongly linked to stroke.