Non-Invasive Assessment of Right Ventricular Function in Health and in Acute and Chronic Pulmonary Arterial Hypertension

Abstract

This thesis investigates the utility of cardiac magnetic resonance imaging (CMR) to assess right ventricular function and dynamic right ventricular (RV) reserve, in health and in acute and chronic pulmonary hypertension. Right ventricular function is a strong predictor of prognosis, in patients with pulmonary arterial hypertension (PAH), but there are currently no available clinical tests that can predict which right ventricle is destined to fail, despite medical therapy. RV reserve, or the ability of the RV to augment during exercise, has recently gained more attention as a potential marker vascular-ventricular uncoupling, and CMR potentially offers a novel, accurate and reproducible method to assess this. Exercise CMR is in its developmental stages, and while several groups have used it in healthy volunteers and patients to study dynamic biventricular function, approaches have varied. We set out to develop a novel and robust exercise CMR methodology and demonstrate its accuracy and reproducibility. Furthermore, we aimed to show how timing of acquisition and respiratory variation, are important considerations, potentially affect key pathophysiological changes. In Chapter 3, we outline the developmental considerations and demonstrate intra-observer, inter-observer, inter-study and inter-test reproducibility. In Chapter 4, we show how the RV remodels after pulmonary thromboendarterectomy in patients with chronic thromboembolic pulmonary hypertension, in a retrospective analysis of CMRs acquired before and after surgery. Importantly, we demonstrate how biventricular interactions, expressed as a composite measure of RV end diastolic volume to left ventricular (LV) end diastolic volume ratio, correlates with change in 6MWD, but parameters of each ventricular alone do not. Furthermore, we demonstrate a correlation between left atrial size and left ventricular end diastolic index, showing that under-filling is likely to be an important pathophysiological explanation of a reduction in LV size, in these patients. LV under-filling from increased RV pressure, and biventricular interactions, continue to remain key themes throughout the following experiments. While exercise CMR is a new technique, so is assessment of RV reserve and there are no normal values available. Furthermore, there is conflicting evidence in the literature, as to normal cardiovascular changes that occur during exercise in normal hearts. We demonstrate in Chapter 5 that an increase in RV forward flow is due to a decrease in RV end systolic volume and an increase in RV ejection fraction, leading to an increase in LV end diastolic volume. LV ejection fraction increases as a result of an increase in left ventricular end diastolic volume and a decrease in left ventricular end systolic volume. Understanding normal exercise physiology, during continuous steady state exercise, is key to interpreting pathophysiological changes. Acute normobaric hypoxia causes hypoxic pulmonary vasoconstriction and an increase in pulmonary vascular resistance and pulmonary pressures. It offers a model in which RV reserve in healthy controls can be studied. It has been hypothesised that the reduction in VO2 max in acute hypoxic exercise is a consequence of a reduction in RV forward flow, however this has never been definitively documented in an imaging study. We show, for the first time using exercise CMR in Chapter 6, that exercise during acute hypoxia leads to a reduction in RV forward flow, a blunting of the rise in RV ejection fraction and LV under-filling with a reduction in LV forward flow. We also show that there is a blunting of the rise in MPA average blood velocity on exercise during acute hypoxia, and hypothesise that this could be a novel CMR parameter to assess pulmonary vascular distensibility. Identifying patients whose RV will continue to fail despite medical therapy has remained elusive. We go on to demonstrate in Chapter 7 that our approach to exercise CMR is not only feasible in patients with pulmonary arterial hypertensoin, but that exercise in these patients, considered stable on medical therapy with normal resting RV ejection fraction, unmasks impaired right ventricular reserve. Furthermore, we demonstrate that there is heterogeneity of response that cannot be predicted by routine clinical tests and resting biventricular function. This information could potentially be clinically valuable, and we outline where this research will take us to next in Chapter 8, where we hope to demonstrate change in right ventricular reserve before and after medical therapy, in patients with inoperable chronic thromboembolic pulmonary hypertension, predicts long term clinical outcomes on follow up. Together, these studies demonstrate how CMR can accurately and non-invasively assess RV remodeling and RV reserve, its impact on the LV, and unmask right ventricular – vascular uncoupling which is otherwise not present at rest, in patients with acute and/or chronic pulmonary arterial hypertension.