A New Quantitative Imaging Biomarker for Identifying Critical Coronary Artery Stenosis with Myocardial BOLD MRI

Introduction: Blood-oxygen-level dependent (BOLD) MRI may be used for detecting myocardial oxygenation changes secondary to coronary artery stenosis (CAS) (1-3). Under pharmacological stress, areas of the myocardium supplied by a stenotic coronary artery appear hypointense relative to healthy regions in BOLD images. Manual windowing is often required to visualize these changes and segmentation of the myocardium according to the American Heart Association’s (AHA) recommendation is used to confirm and characterize the BOLD effect. However, the identification of critical (clinically significant) CAS on the basis of BOLD MRI has mostly been met with limited sensitivity. The purpose of this work is to present a fundamentally new direction for visualizing and quantifying regional myocardial BOLD signal changes. The proposed approach, tested on a canine model, identifies the affected territory of the myocardium based on a statistical framework, correlates strongly with reference flow measurements, and most importantly, leads to a significant increase in sensitivity to microvascular flow changes compared to previous approaches.