Deep venous thrombosis (DVT) remains a major health problem. Although thrombolytic therapies are effective in recanalising the veins, restoring blood flow, preventing pulmonary embolism and post-thrombotic complications, there is still no consensus on the selection criteria for this invasive treatment. Experimental data suggest that thrombus rich in fibrin has better response to thrombolysis than red cell rich (acute phase) or collagen rich (chronic) thrombus. Thus, there is a need for a diagnostic technique that provides better information on the stage of thrombus organization in-vivo and allows identification of thrombus suitable for thrombolysis. Current imaging modalities for the diagnosis of DVT do not provide information on the biological stage of thrombus organization. Contrast venography is still considered the gold standard for the diagnosis of DVT, even though thrombi are not clearly visualized and they are indirectly detected due to alterations in blood flow. The aim of this thesis was to develop and validate new imaging methodology for the better evaluation of DVT using Magnetic Resonance Imaging (MRI). The first aim was to investigate the potential of non-contrast enhanced MRI sequences including T1 mapping, T2* mapping, Magnetization Transfer Contrast (MTC), and Apparent Diffusion Coefficients (ADC) maps for the detection of thrombus in a murine model of deep venous thrombosis. The second aim was to investigate the merit of a fibrin and macrophage specific MR contrast agent for the detection of DVT. Both fibrin and macrophages play a major role in thrombus organization. The third aim was to investigate the merits of the fibrin binding contrast agent for the guidance of thrombolysis. We also developed two new non-contrast enhanced venous spin labelling approaches in order to obtain venograms without the need of a contrast agent and thereby to improve non-invasive DVT diagnosis using MRI. In conclusion, this thesis proposes a new imaging methodology for the accurate staging of thrombus organization and the successfully detection of thrombus amenable for thrombolysis in a murine model of DVT. The translation of this technique into the clinic should have great potential to change clinical evaluation and treatment of patients with DVT.

Background: The response of the human heart to ischaemic stress is not uniform and adaptive mechanisms play a role in attenuating myocyte damage and improving performance. The mechanisms of such adaptations are poorly understood and likely multifactorial. The main aim of this thesis was to examine these mechanisms using the models of exercise induced myocardial ischaemia and acute myocardial infarction. Methods: Using a specially adapted supine ergometer, we used invasive physiological measurements and high-resolution cardiac magnetic resonance imaging to assess changes in coronary blood flow, central haemodynamics and transmural myocardial perfusion during exercise. In a separate group, we sought to examine the role of post-conditioning as a potential therapeutic tool in a randomised controlled trial involving patients undergoing primary percutaneous revascularisation for acute myocardial infarction. Results: We were able to demonstrate that the reduction of ischaemia seen on second exercise in patients with stable coronary artery disease is associated with synergistic changes in central and coronary haemodynamics, with a fall in myocardial microvascular resistance and enhanced vascular-ventricular coupling. High-speed CMR perfusion imaging using k-t acceleration is a feasible tool to investigate these differences, with sufficient spatial resolution to detect transmural flow heterogeneity. The data from the postconditioning study did not show a difference in infarct size between the groups but numbers were small. Conclusion: The mechanisms of adaptation of the heart to ischaemic stress are complex and likely multifactorial. These results suggest that synergistic changes in systemic and coronary circulations as part of a generalised reactive hyperaemic vasodilatory response to exercise results in improved myocardial perfusion and overall performance. Transmural flow redistribution to the subendocardium is likely to play an important role in attenuating myocardial ischaemia on repeat exercise although we await the results of ongoing work. Innate myocardial protection, such as that afforded by postconditioning remains a possibility, although the results from this study are inconclusive.

Introduction: Coronary revascularisation is a key component of the management of patients with coronary artery disease (CAD). The importance of combining functional with anatomic information to select appropriate patients is increasingly recognised. Established Cardiovascular Magnetic Resonance (CMR) techniques already provide much of the relevant functional information and the absence of ionising radiation makes CMR ideal for serial examinations. However, new CMR techniques continue to emerge, and require appropriate clinical evaluation. Methods: This thesis comprises a series of clinical studies developing and evaluating techniques for guiding revascularisation: i. Comparison of a high-resolution <em>k-t</em> accelerated perfusion sequence with a standard sequence. ii. Validation of CMR quantification of absolute myocardial perfusion using the high-resolution sequence against PET. iii. Investigation of the relationship between an angiographic score (BCIS-1 Jeopardy score) and CMR estimations of CAD burden. iv. Determination of the inter-study reproducibility of perfusion imaging and strain analysis with CMR feature tracking (CMR-FT). v. Evaluation of a novel scar imaging technique using a dual-inversion recovery (dual-IR) pre-pulse for the first-time in patients. vi. The feasibility of combined CMR and coronary intervention in a hybrid laboratory. Results: The main findings for each compnent were: i. Perfusion imaging with the <em>k-t</em> accelerated sequence resulted in significantly improved image quality, signal and contrast to noise ratios and a reduction in dark rim artefacts compared to the standard sequence. ii. There was good correlation between quantitative myocardial perfusion reserve (MPR) derived from CPR and PET. CMR and PET-derived MPR were both comparable and accurate for the detection of CAD. However, absolute perfusion values from both modalities were only weakly correlated. iii. The correlation between the BCIS-1 Jeopardy score and CMR ischaemic burden was good and the score predicted a prognostically important ischaemic threshold of 12% with high specificity. iv. The inter-study reproducibility of quantitative myocardial perfusion and CMR-FT was reasonable and better for global rather than regional measures. There was no detectable variation in perfusion or strain during the day. v. The dual-IR sequence improved scar imaging compared to the IR technique. vi. Combined CMR and interventional coronary procedures were successful and well tolerated. Conclusions: Quantitative analysis of perfusion is an exciting prospect with considerable potential and has demonstrated clinically utility, however, its application remains challenging. Novel scar and strain imaging techniques have also shown promising results. Further method refinement and appropriate clinical studies should allow the full potential of these tools for guiding  revascularisation to be realised.

Background: The thesis investigated the role of self-compassion, defined as “being open to and moved by one‟s own suffering, experiencing feelings of caring and kindness toward one-self, taking an understanding, non-judgmental attitude toward one‟s own inadequacies and failures and recognising that one‟s experience is part of the common human experience” (Neff, 2003b, p. 244) in adjustment following MI. Adjustment was defined in terms of emotional, cognitive and behavioural response to MI. Previous empirical findings indicate that self-compassion confers resilience dealing with adversity (Leary, Tate, Adams, Batts Allen, & Hancock, 2007; Neff, Hsieh, & Dejitterat, 2005; Neff, Kirkpatrick, & Rude, 2007), and it has recently been hypothesised that self-compassion may be an important mechanism influencing response to health events (Terry & Leary, 2011). Method: The study sample comprised adults referred to an urban Cardiac Rehabilitation Service following MI (N= 87). Participants completed a battery of self-report questionnaires which assessed self-compassion (Self-compassion scale, SCS, Neff, 2003a), anxiety and depression (Hospital Anxiety and Depression Scale, HADS, Zigmond & Snaith, 1983) , illness perceptions (Brief Illness Perception Questionnaire, Brief-IPQ, Broadbent, Petrie, Main, & Weinman, 2006) and cardiac symptoms (Ketterer et al., 2008). Cardiac function post MI was assessed using a measure of left ventricular ejection fraction (LVEF). Attendance at cardiac rehabilitation (decision to opt in or out) was used as an index of behavioural adjustment. Results: As predicted, higher levels of self-compassion were associated with lower levels of anxiety (p =.001), depression (p =.01), and a reduced tendency to perceive that MI is having a negative emotional impact (p<.01), after statistically controlling for cardiac symptoms. Cardiac function was not associated with emotional adjustment following MI. Self-compassion was not associated with beliefs related to the cognitive representation of cardiac illness, after controlling for other factors. Self- compassion was not statistically significantly associated with an increased tendency to opt in to cardiac rehabilitation. Discussion: The results provide preliminary evidence which indicates that the “buffering effect” of self-compassion (observed previously in response to academic and interpersonal stressors) is also observed responding to MI (i.e. a life threatening health event). More specifically, self-compassion is associated with a reduced negative emotional impact of MI but was not associated with the cognitive representation of cardiac illness. The potential effect of self compassion upon the moment to moment processing of cardiac stimuli and adjustment to MI over time are discussed.

Background: Sickle cell disease (SCD) is a hereditary hemoglobinopathy that causes chronic complications due to repetitive vaso-occlusive events and hemolysis, and can lead to multiorgan failure and shortened life expectancy. Among a spectrum of cardiovascular manifestations in these patients, pulmonary hypertension (PHT) has been stated to pose the highest concern. The exact prevalence of PHT in SCD is controversial; clinical studies using echocardiography have suggested PHT to be highly prevalent and a major determinant of outcome. The contribution of cardiac and endothelial dysfunction to SCD also remains poorly understood. Aims: (1) To characterise cardiac function in SCD patients using new 3-dimensional echocardiographic imaging techniques. (2) To determine the prevalence of PHT in SCD patients by using advanced echocardiographic methods and establish mechanisms involved in the development of PHT. (3) To assess whether SCD patients have endothelial dysfunction independent of other vascular risk factors. Methods: Detailed 2D, 3D and Doppler echocardiography studies were performed in 122 consecutive unselected SCD patients and 30 healthy age- gender- and ethnicity-matched controls to assess cardiac function and look for PHT. A sub19 group of patients underwent invasive assessment of pulmonary vascular resistance (PVR) by right heart catheterisation. CT lung scan and lung function tests were also performed in patients suspected to have PHT. Vascular endothelial function was assessed by flow-mediated forearm vasodilatation in twenty stable adolescent SCD patients without other risk factors and fifteen healthy age-matched controls. Results: Results indicate that patients with SCD have significantly enlarged cardiac dimensions associated with elevated cardiac index (CI) that correlated with the degree of anaemia. Cardiomegaly in SCD patients was not associated with significant contractile dysfunction as assessed by regional myocardial deformation. A high proportion (&gt;30%) of SCD patients had a tricuspid regurgitation jet velocity ! 2.5m/s but non-invasive estimation of PVR revealed that only a minor proportion (&lt;5%) had elevated values. These findings were confirmed by right heart catheterisation. The raised tricuspid regurgitant velocities in SCD patients may be driven more by elevated CI than by elevated PVR. Finally, it was found that young patients with SCD had evidence of subclinical vascular endothelial dysfunction. Taken together, these results provide new information about the prevalence and underlying mechanisms of cardiovascular dysfunction in patients with SCD.

Background: Coronary Allograft Vasculopathy (CAV) is the leading cause of late death for children after heart transplantation. It often presents late due to lack of ischaemic symptoms from the denervated allograft and it is difficult to detect using conventional angiography. A non-invasive diagnostic tool would be invaluable for the early detection and therefore, prevention of disease. MRI has the potential to achieve this by imaging coronary lumen, vessel wall and central vascular stiffness in a child-friendly free-breathing protocol lasting thirty minutes. Methods: A three-dimensional steady-state-free-precession (3D SSFP) imaging sequence for coronary artery lumen imaging was optimised for its use in children. Patient based studies were performed to show the clinical potential for this approach. Furthermore, contrast-enhanced inversion-prepared coronary vessel wall imaging was investigated for its ability to detect intimal disease in CAV, using intravascular ultrasound to validate the findings. Finally, MRI was used to study the potential of aortic screening for rapid assessment of CAV. Results: MRI coronary artery lumen imaging is feasible in children and reference values are provided. Contrast-enhanced vessel wall imaging has 91% accuracy in the detection of significant CAV. Further improvements in the technique can be achieved using high-field MRI (3.0 Tesla) or sublingual nitroglycerin. Central aortic stiffness is increased in children after heart transplantation and a relationship between this and CAV is demonstrated. Conclusions: Coronary artery lumen and vessel wall MRI can be used to follow up children after heart transplantation. Non-invasive follow-up can be more frequent and has the potential to allow early prevention. From this research, there are wider implications for congenital heart disease, Kawasaki disease and cardiovascular atherosclerosis imaging.

Hieutrophil cytoplasmic antibody (ANCA) associated vasculitis is a systemic disease i affects the kidneys, lungs, and other tissues. ANCA were first described in patients i focal necrotising glomerulonephritis in 1982, with myeloperoxidase )) and proteinase 3 (PR3) subsequently shown to be the antigenic targets responsible rthe perinuclear and cytoplasmic staining patterns, respectively. Infection is thought to erbate disease partially through the production of the proinflammatory cytokine TNFot i primes neutrophils for respiratory burst. In this thesis, the role of another cytokine, Milocyte-colony stimulation factor (GCSF) is examined both in vitro and in vivo. Previous i have implicated the complement system in ANCA vasculitis. Furthermore, TNFot •d neutrophils which have been activated with ANCA in vitro are known to release a >r into the supernatant which causes complement activation in normal serum. This w has yet to be identified, although many candidates such as the alternative pathway aonent properdin have been suggested. In this work it is shown that GCSF antrations are elevated in patients with acute ANCA vasculitis and that GCSF can prime ted neutrophils for anti-MPO induced respiratory burst. A passive antibody transfer i model of anti-MPO vasculitis was established and GCSF administration was shown to te disease. Experiments also explored other models of anti-MPO vasculitis based •PO-deficient mice. The mouse model was also used to investigate the effect of icy of either properdin or MASP2 in disease. Using the passive anti-MPO passive Sr model, properdin deficiency was shown to have no effect on the extent of disease ! MASP2-deficiency exacerbated disease by a mechanism which has yet to be identified. iwork has established GCSF as a key cytokine and possible therapeutic target, and I novel observations on complement in ANCA vasculitis.

Single ablative therapy for PAF has moderate success and many patients present with recurrent arrhythmia. We propose that the structure of the RF lesion applied during ablation is important in determining recurrences. The nature of the RF lesion was studied using MRI with gadolinium delayed enhanced (DE) imaging and high signal T2 weighted imaging. Levels of DE and T2 were low in pre-procedural scans but rose dramatically immediately following the procedure. Acute DE was greater in patients without recurrences compared to those with recurrences. Conversely T2 levels were lower in patients without recurrences and higher in those with recurrences. On the late scans, T2 reduced to baseline. DE however remained and was greater in patients without recurrences. We therefore propose that acute RF ablation injury is composed of two types of tissue damage. DE infers largely necrotic tissue injury which lasts longer and causes persistent conduction block. T2 is a transitory phenomenon co-existing with DE, causing acute conduction block. We propose that resolution of oedema is associated with recurrences of PV connection and therefore arrhythmia recurrences. Modifications in our ablative techniques to achieve more DE at the acute ablation would potentially be important in conferring better ablation outcomes. The role of DE imaging was utilised to compare left atrial catheter ablation with robotic assisted navigation and standard navigation. A greater circumferential lesion extent by DE was observed in the robotic group . This suggests that catheter stability improves tissue contact permitting the creation of more contiguous durable scar around the PV antrum. We also sought to improve DE imaging sequences to optimise scar visualisation and tested the feasibility of an automatic scar quantification tool to improve reproducibility whilst maintaining accuracy.

Cardiovascular magnetic resonance (CMR) perfusion imaging has been shown to accurately detect significant coronary artery stenoses and is of potential use to detect patients for early treatment and to improve prognosis. New techniques yield a superb spatial resolution and may allow for full quantification of perfusion. Novel CMR techniques and imaging biomarkers are often validated in small animal models or empirically in patients. The direct translation of small animal magnetic resonance (MR) imaging protocols to humans is rarely possible, while validation in humans is often difficult, slow and occasionally not possible due to ethical considerations. -- The aim of the thesis was to develop an MR-compatible isolated blood-perfused pig heart model, which closely resembles human physiology, anatomy and size and to utilize it for controlled validation of quantitative perfusion at the segmental and voxel level using standard clinical sequences and MR scanners. To enable accurate quantification a universal dual-bolus method was developed. The design of the heart allowed exquisite control regarding overall and regional blood-flow and imaging by identical equipment used for humans. Quantitative perfusion imaging showed a good correlation with microspheres, which was most apparent with Fermi function constrained deconvolution regardless of sequence or field strength. Fermi deconvolution based voxel-wise quantitative perfusion values also correlated well with microspheres throughout the myocardial wall. The validated sequences proved useful for the detection of significant coronary artery disease in a small feasibility study in patients analysing perfusion at the segmental level. In conclusion this work has resulted in an accurate validation of quantitative perfusion CMR at a segmental and voxel level at common clinical field strengths.

Aims: To determine whether new pacing lead technology can be used to overcome problems with left ventricular (LV) lead placement during cardiac resynchronisation therapy (CRT). To establish whether cardiac magnetic resonance (CMR) data can be used to guide LV lead placement in real-time at CRT implant. To establish the best method of multi-site pacing. Methods: We investigated the incidence of problems with phrenic nerve stimulation (PNS) and high capture thresholds at implant and at 4 and 6 month follow-up periods in 40 patients who underwent CRT with a new quadripolar lead. In 20 patients we used a pressure wire to assess the acute haemodynamic response (AHR) to pacing within different regions of the coronary sinus (CS) to determine whether problems with poor AHR can be overcome with electronic repositioning. In 23 patients we used CMR acquisition, processing, overlay and registration tools to guide LV lead placement in real-time during CRT. In 12 patients we turned on the multi-site function of a quadripolar lead, implanted temporary endocardial and epicardial pacing leads and measured the AHR whilst pacing in multiple different ways. Results: Quadripolar lead technology successfully overcame problems with PNS and high capture thresholds at implant and at follow-up but not poor AHR. A CMR dyssynchrony-guided approach to LV lead placement gave an AHR comparable to the best that can be achieved anywhere and was associated with improved reverse remodelling at 6 months. Endocardial pacing gave the best overall AHR but in different patients different methods of multi-site pacing were best. Conclusions: New lead technology can be used to overcome some LV lead problems but not poor AHR. Real-time CMR dyssynchrony guided CRT may be better than conventional empirical LV lead placement. Endocardial pacing gives excellent overall AHR but different methods of pacing (including multi-site) may be better in individual patients.