The predictive value of the borderline ankle-brachial index for long-term clinical outcomes: An observational cohort study

Background and aims: Low ankle-brachial index (ABI) is associated with increased mortality and an increased incidence of cardiovascular events. The purpose of this study was to investigate the value of borderline ABI in predicting clinical outcomes. Methods and results: The data were derived from the Shinken Database 2004-2012, from a single hospital-based cohort study (N = 19,994). ABI was measured in 5205 subjects; 4756 subjects whose ABI was 0.91-1.39 and having no history of peripheral artery disease were enrolled. The subjects were classified into two groups as follows: borderline ABI (0.91-1.00; n = 324) and normal ABI (1.01-1.39; n = 4432). Subjects in the borderline ABI group had more comorbidities, including diabetes mellitus, aortic disease, and stroke. Moreover, the borderline ABI group was associated with higher levels of hemoglobin A1c and brain natriuretic peptide, larger diameters of left atrium and left ventricle, and lower levels of estimated glomerular filtration rate and left ventricular ejection fraction. All-cause death and cardiovascular death occurred in 9.3% and 4.6% of subjects in the borderline ABI group, and in 2.0% and 0.8% of subjects in the normal ABI group, respectively. An adjusted Cox regression model showed that borderline ABI was associated with a higher incidence of all-cause death (hazard ratio [HR] 2.27, p = 0.005) and cardiovascular death (HR 3.47, p = 0.003). Conclusion: A borderline ABI was independently associated with worse clinical outcomes in relatively high risk population. Our data should be confirmed in larger populations including those with low risk profiles. (C) 2016 Elsevier Ireland Ltd. All rights reserved.