An elevated ankle-brachial index is not a valid proxy for peripheral medial arterial calcification

Abstract

BACKGROUND AND AIMS: The ankle brachial index (ABI) is often used as a proxy for medial arterial calcification (MAC) in studies investigating MAC as a cardiovascular risk factor, but evidence supporting this hypothesis is sparse. This study aims to investigate the use of an elevated ABI as proxy for MAC, as visualized with computed tomography (CT). METHODS: Cross-sectional data of 718 participants with, or at risk of cardiovascular disease was used. The ABI was calculated using cutoffs >1.4 and > 1.3. The presence of MAC was assessed in the crural and femoral arteries by CT imaging. Modified Poisson regression was used to assess the association between an elevated ABI and the presence of MAC, and test characteristics were calculated. RESULTS: MAC was found in 25.0% of participants. An ABI >1.4 was found in 8.7% of participants, of whom 45.2% had MAC. An elevated ABI was significantly associated with the presence of MAC (RR 1.74, CI: 1.26-2.40). However, poor positive specific agreement (23.3%, CI: 13.9-34.3), sensitivity (15.7%, CI: 10.4-21.1) and positive predictive value (45.2%, CI: 32.8-57.5) were found. Despite good specificity (93.6%, CI: 91.6-95.7) the area under the receiving operator curve remained poor (54.7%, CI: 51.8-57.6). Negative specific agreement (84.5%, CI: 81.4-87.0) and negative predictive value (77.0%, CI: 73.7-80.2) were acceptable. CONCLUSIONS: An elevated ABI is insufficient to serve as a true diagnostic proxy for MAC. Studies that have drawn conclusions on the association between MAC and cardiovascular disease, solely based on the ABI, are likely to underestimate the found effects.