Measuring in vivo Regional Myocardial Function Using High-Field MRI

Heart failure (HF) is one of the major causes of mortality in the Western world, recognized by a compromised ability of the heart to supply the body with blood. The poor understanding of the disease mechanisms and lack of adequate therapy strategies are reflected in the grim prognosis of HF. Great research efforts over the last decades have been aimed at revealing the factors responsible for the reduction in function the left ventricle (LV) of the heart. In this, small-animal models of cardiac disease plays an irreplaceable role, enabling isolation and identification of structural and functional alterations on cellular and/or subcellular level. However, to be able to relate findings on the microscopic scale to alterations in cardiac function, there is a great need for methodology, preferably noninvasive, that allows detailed assessment of in vivo regional myocardial function. The hearts of mice and rats are more than two orders of magnitudes smaller than human hearts by weight, and beats up to ten times faster. Measurement of cardiac function in mice and rats thus understandably requires considerably higher resolution than in humans to offer comparable data yield. Phase contrast magnetic resonance imaging (PC-MRI) is a well-established and versatile noninvasive imaging technique allowing measurement of time-resolved 3D motion. We have developed an improved PC-MRI technique able to measure myocardial motion in small animals with improved accuracy and resolution compared to earlier approaches (Paper I). A major consequence of pushing the limits of achievable spatiotemporal resolution in PC-MRI is increased generation of eddy currents in the systems, which introduces severe baseline shifts in the measured motion that may render the data unusable. This required further development of eddy current correction techniques (Paper II). We developed this imaging technique further and introduced and validated a protocol for calculating regional myocardial strain from PCMRI velocity data. We applied this protocol, as a proof-of-concept, in normal and regionally dysfunctional rat hearts (Paper III). Finally, we incorporated a mathematical model allowing calculation of regional myocardial work from PC-MRI data, in combination with identification of the mitral and aortic valve events and a simple measurement of peak blood pressure. This protocol was also demonstrated in normal and dysfunctional rat hearts (Paper IV). The work in this thesis demonstrates that PC-MRI allows noninvasive measurement of regional myocardial motion, strain and work in small-animal models of cardiac disease with high resolution. The results are readily extendable to human applications, ultimately allowing higher sensitivity and/or resolution and extended data yield in functional cardiac MRI