Comparison of different ankle-brachial indices in the prediction of overall and cardiovascular mortality

Abstract

Background and aims: Low ankle-brachial index (ABI) calculated using systolic blood pressure (SBP) is associated with poor prognosis. However, there is no study assessing ABI calculated using mean artery pressure (MAP) and diastolic blood pressure (DBP) in predicting mortality.

Methods: Two cohort populations were enrolled. The first population comprised 379 patients (106 patients with angiography-proved peripheral artery disease (PAD) and 273 relative normal patients) to evaluate the best cutoff values of ABImbp and ABIdbp for prediction of PAD. The second population included 941 patients undergoing echocardiographic examinations to assess the ability of different ABIs in predicting mortality. ABIs were measured using an ABI-form device.

Results: The best cutoff values of ABImbp and ABIdbp for prediction of PAD were 0.92 and 0.88. In our second population, median follow-up to mortality was 93 months. There were 87 cardiovascular and 228 overall deaths. Multivariable analysis showed ABIsbp, ABImap, ABIdbp, ABIsbp <0.9, and ABImap <0.92 could predict overall and cardiovascular mortality (all p < 0.001). ABIdbp <0.88 could only predict CV mortality (p = 0.033). In a direct comparison of 6 multivariable models, the basic model consisting of significant variables in the univariable analysis plus ABImap <0.92 had the highest predictive value for overall and cardiovascular mortality (all p < 0.001).

Conclusions: In a direct comparison of 6 multivariable models, the basic model + ABImap < 0.92 was the best model in predicting overall and cardiovascular mortality. Hence, calculation of ABI using MAP except SBP might provide extra benefit in survival prediction.

Keywords: Ankle-brachial index; Diastolic blood pressure; Mean arterial pressure; Mortality; Systolic blood pressure.