Coronary artery disease is responsible for an estimated 1.5 million myocardial infarctions and 520 thousand deaths per year. Despite the widespread clinical use of coronary angiography for the diagnosis of coronary artery disease, assessment of the physiologic significance of a coronary lesion identified with angiography remains very subjective. Measurement of the coronary flow reserve (CFR), defined as the ratio of hyperremic to resting coronary blood flow, has been advocated as a method capable of providing information regarding the physiologic significance of a stenosis. However, existing techniques for measuring CFR all have significant drawbacks that limit the usefulness of CFR measurements in the clinical environment. Flow measurement techniques using magnetic resonance (MR) imaging have the potential to provide information regarding coronary blood flow and CFR. MR techniques based on the phase-difference (PD) velocity method have been used in several other regions of the body to assess blood flow. However, in small vessels the PD technique is affected by partial volume and other effects which introduce errors in the flow determination. We have developed an alternative strategy for measuring coronary blood flow using a MR technique called complex difference (CD) flow measurement. Simulations and phantom experiments show that the CD method is less sensitive to artifacts in small vessels than phase difference (PD) flow measurement techniques. Preliminary data obtained from phantoms, animals, and human volunteers suggest that the CD and PD techniques are capable of providing CFR measurements in vivo. We propose to investigate the PD and our recently developed CD techniques as potential non-invasive methods for assessing coronary flow and coronary flow reserve. First we will develop two new MR image acquisition strategies that are capable of obtaining data for coronary flow measurements in vivo. The two new acquisition strategies are based on modifications to a segmented, view shared gradient echo acquisition as well as a segmented echo-planar acquisition. Then we will test the accuracy and precision of absolute flow and flow ratio measurements obtained with these sequences using a computer controlled flow phantom on a computer-controlled motion state. Third, we will evaluate these techniques using an open-chest coronary artery animal model that will provide velocities, flow rates, arterial sizes, arrhythmias, and motion comparable to those expected in humans. Finally, we will conduct a study in 30 patients with single vessel coronary artery disease to show that the techniques validated in the animal model are reproducible in humans and can demonstrate abnormal human coronary flow reserve measurements in patients.
