Aorto-Coronary Haemodynamics: Studies in Cardiovascular Magnetic Resonance

Abstract

Large artery stiffness has been shown to independently predict cardiovascular events and all-cause mortality in a broad range of cohorts. Despite its description almost half a century ago, the cardiovascular research community continues to grapple with a robust model for ventriculo-vascular interactions, let alone a sound basis for its predictive strength. A mechanistic understanding of the key sequelae arising from increased arterial stiffness is likely to provide instructive pathophysiological insight, but also pave the way for novel risk stratification and possible therapeutic targets. One powerful, putative effect of increased arterial stiffness is a reduction in coronary blood flow, however a reproducible, simultaneous and non-invasive methodology to evaluate both has been elusive. This thesis examines the relationship between myocardial perfusion and arterial stiffness. In Chapter 2, invasive assessment of both coronary blood flow and coronary flow (velocity) reserve was undertaken with Doppler FloWires and then compared with cardiovascular magnetic resonance (CMR) derived assessment of arterial stiffness, aortic distensibility. In a cohort of subjects with a favourable traditional risk factor profile and no significant angiographic disease, aortic distensibility was linearly associated with measures of both resting and hyperaemic coronary blood flow as well as coronary flow velocity reserve. Moreover, aortic stiffness was associated with a reduced response to adenosine. The validation of this relationship was completed in non-invasive methodology by using CMR perfusion imaging (CMR-PI) to evaluate indices of myocardial blood flow and myocardial perfusion reserve in Chapter 3. In subjects with normal perfusion studies, aortic distensibility was again strongly correlated with resting and hyperaemic myocardial blood flow as well as myocardial perfusion reserve. After evaluating ‘normal’ individuals in earlier studies, the later cohorts focused on disease states. In Chapter 4, the behaviour of this relationship in ischaemic and nonischaemic myocardium was evaluated in a highly selected group with a single perfusion defect on CMR-PI confirmed on subsequent coronary angiography. In this group, arterial stiffness (once again, evaluated by CMR derived aortic distensibility) remained associated with resting and hyperaemic blood flow despite critical epicardial coronary disease. In Chapter 5, subjects who had received primary percutaneous coronary intervention for acute ST elevation myocardial infarction (STEMI) were evaluated with CMR perfusion and delayed enhancement imaging on day three and again at three months. The myocardium was dichotomised to infarct and non-infarct territories with blood flow, perfusion reserve and indices of myocardial injury evaluated on a per segment basis. In this cohort, arterial stiffness was associated with reduced early and late hyperaemic response to adenosine in both infarct and non-infarct myocardium. On multiple linear regression aortic distensibility remained an independent predictor of improvement in myocardial perfusion reserve suggesting arterial stiffness may be a key feature for worse outcomes post STEMI. This thesis not only confirms the putative relationship between large artery function and coronary blood flow but also establishes CMR-PI as a reproducible and robust modality in this space. Non-invasive evaluation of the aorto-coronary haemodynamic relationship may provide novel insight, not only as a potential clinical surrogate endpoint, but also a manner to evaluate therapeutic efficacy.