Assessment of myocardial viability with magnetic resonance imaging

Accurate assessment of myocardial viability is a central goal in clinical diagnosis of ischemic heart disease, as well as in basic research into the underlying mechanisms of cardiac pathology. Clinical assessment of myocardial viability would have a significant impact on the management of patients with ischemic heart disease, by avoiding unnecessary revascularization procedures, as well as assessing the outcome of such interventions. In this work, two different approaches towards the assessment of myocardial viability using magnetic resonance imaging (MRI), will be investigated: (1) Development of near real time MR cardiac stress testing in humans. (2) Development of new methodologies towards estimates of tissue metabolic viability by determination of regional oxygen consumption.; Towards the development of an MR cardiac stress test, an ultra-fast spoiled gradient echo (SPGR) imaging sequence was written for the new 5.6 GE scanner. Through the use of improved gradient hardware and the hardware optimized trapezoid (HOT) algorithm, this sequence reduced imaging times by half, from previous fast SPGR sequences.; A detailed SNR analysis of SPGR imaging, was conducted and it was shown that SNR depends on relatively few parameters, most notably sequence efficiency. Most notably it was shown that image SNR is independent of TR when scan time is held constant.; Designed for stress-testing with parallel tagging capabilities, the readout acquisition of this sequence was modified to form a multi-echo hybrid sequence. Significant reductions in total scan time are possible with this sequence which is approximately 30–40% faster than single echo sequences. A theoretical analysis of echo train length and the measurement of $T*2$ and field inhomogeneities in vivo showed that the optimal echo train length for multi-echo cardiac sequences is approximately 2–3.; The effects of amplitude and phase discontinuities, as well as time delays on ghosting artifacts was investigated in the context of echo planar imaging (EPI). A novel reference technique was formulated and its ability to measure time delays and phase discontinuities was investigated. It was shown that excellent artifact reduction is possible using reference scans.; Time delays observed in obliquely oriented echo planar imaging were mathematically characterized and experimentally validated. It was shown that time delays measured in oblique coordinates are easily predicted knowing the image orientation and the delay contributions from individual physical gradients. “Compensation blips” were shown to be an effective means of removing the effects of anisotropic gradient delays for EPI.; Estimation of oxygen consumption was pursued by developing and validating a novel spin-labelling methodology that non-invasively measured myocardial arterial coronary perfusion in isolated, perfused, arrested rabbit hearts at 4.7T. Through the utilization of fast spin echo (FSE) imaging, 32 fold reductions in scan time permitted perfusion measurements over the entire ventricle (6–8 slices) in less than 15 minutes. By T₂ weighting the FSE images, indirect estimation of oxygen concentration was made from the same images used for perfusion measurements. This work demonstrates the feasibility of non-invasive myocardial oxygen consumption measurement using MRI.