Validation of Myocardial Oxygen Extraction Fraction Measurement with MRI

Imbalance of myocardial oxygen supply and consumption precipitates a cascade of physiological changes resulting in ischemic pathology. While assessment of myocardial perfusion alone may allow accurate assessment of myocardial oxygen consumption in some pathophysiological conditions, the perfusion-oxygen consumption relationship is derailed in several conditions, including: myocardial ischemia and infarction, hypertrophic and dilated cardiomyopathies, heart failure, valvular heart disease, and septic cardiomyopathy. Importantly, this oxygen supply/perfusion mismatch occurs early – before mechanical dysfunction. Therefore, evaluation of myocardial consumption independent of perfusion is of importance for early diagnosis and monitoring of these pathophysiological conditions. Myocardial oxygen extraction fraction (mOEF), which relates the biologic coupling of myocardial blood flow (oxygen supply) to oxygen consumption, may provide a more accurate assessment of this balance. For example, in ischemic cardiomyopathy, adequate myocardial perfusion is commonly reduced by high grade epicardial coronary artery stenoses. To avoid ischemia-caused injury, mOEF is likely to be increased to compensate for decrease in myocardial perfusion and oxygen delivery to myocytes. Consequently, the affected myocardial regions drop into a so called “hibernating” state that is effective in the short term. In this respect, an accurate mOEF assessment is a unique tool, with the potential to determine the likelihood of cardiac functional recovery after reperfusion. To date, the reference method for non-invasive quantification of mOEF in vivo is Positron Emission Tomography (PET). We recently developed a novel contrast- free cardiovascular magnetic resonance (CMR) acquisition method to quantify mOEF in vivo that has several advantages over PET: our CMR method has better spatial resolution, shorter acquisition time, does not expose the patient to ionizing radiation, and could be more widely available than PET. The overall objective of this study is to leverage our expertise in CMR imaging to refine and rigorously validate this new mOEF method. In Aim 1. This technique will be developed with assistance of a novel deep learning approach for artifact-free images and then validated using large animal models with and without induced coronary artery disease. Invasive catheter-measured and non-invasive PET-MRI-measured mOEF will be used as reference. In Aim 2, the CMR mOEF method will be validated in a small cohort of patients with hibernating myocardium in vivo, with the availability of 18F-FDG-PET as reference. Although we will study this mOEF technique in hibernating myocardium, this imaging method can be applied to the diagnosis and evaluation of treatments in a wide range of cardiomyopathies. Given the capability of CMR for the comprehensive assessment of myocardial function, tissue characterization, and viability, successful completion of CMR mOEF validation will provide a ‘one-stop shop’ evaluation of metabolic, functional, and structural abnormalities in patients with cardiomyopathy. Furthermore, the advantage of fast imaging without using an intravenous contrast agent is well-suited for repeat quantitative mOEF measurements needed to guide the effects of novel therapeutic interventional strategies.