Redefinition of uraemic cardiomyopathy with cardiac magnetic resonance imaging

Patients with end stage renal disease (ESRD) have a 20-100 fold risk of premature cardiovascular death compared to age matched controls from the general population. These patients have many ‘conventional’ cardiovascular risk factors such as diabetes, ischaemic heart disease, hypertension, cigarette smoking and hyperlipidaemia. However, the relationship between the presence of these risk factors and cardiovascular outcomes is less clear in ESRD than in the general population. In the cases of hyperlipidaemia and hypertension a paradoxical relationship has been demonstrated where lower cholesterol or blood pressure is associated with an increased risk of cardiovascular events. One factor previously demonstrated to be associated with poor prognosis is the presence of uraemic cardiomyopathy, found in approximately 70% of ESRD patients at initiation of dialysis therapy, usually defined echocardiographically as the presence of left ventricular (LV) abnormalities, including left ventricular hypertrophy (LVH), LV dilatation and LV systolic dysfunction. However, echocardiography makes assumptions regarding LV geometry, which is frequently distorted in patients with ESRD. Furthermore any errors in measurements are amplified by the changes in hydration status which occur during the dialysis cycle, leading to changes in LV chamber dimensions. For these reasons, cardiac magnetic resonance imaging (CMR), by providing high fidelity measurements, potentially offers a ‘volume independent’ method of quantifying LV dimensions. Furthermore, by using gadolinium based contrast agents, tissue abnormalities particularly myocardial fibrosis, indicated by late gadolinium enhancement (LGE) may by identified. The work contained in this thesis examines the relationship between cardiac dimensions, as defined by CMR and cardiovascular risk factors (both conventional and specific to uraemia). In a study of 145 patients with ESRD using CMR with gadolinium, two specific pathological processes were demonstrated. First, the presence of subendocardial LGE indicating previous myocardial infarction was associated with the presence of conventional cardiovascular risk factors such as previous ischaemic heart disease and diabetes. Patients with subendocardial LGE frequently had LV systolic dysfunction. Second, diffuse LGE, representing fibrosis throughout the LV wall was identified in patients with LVH. This was an unexpected finding and appears specific to uraemia. Using CMR, isolated LV dilatation was rare. These findings suggest that in uraemia two forms of cardiomyopathy exist- LV systolic dysfunction due to underling myocardial ischaemia and LVH which is a true ‘uraemic cardiomyopathy’ associated with diffuse myocardial fibrosis. Attempts were made to reassess the relationship between CMR and echocardiographic measures of cardiac dimensions. In keeping with a previous study, it was demonstrated that M-mode echocardiography overestimates LV mass compared to CMR in this population. Thus, CMR may be used to optimise echocardiographic formulae to calculate LV mass. Furthermore, it appeared that either by echocardiography or by CMR the chief determinant of LVH in this population was blood pressure, in particular systolic blood pressure. This has implications for treatment as recent studies aimed at correcting anaemia, previously associated with LVH, either to reduce LV mass or to improve survival, have generally demonstrated increased cardiovascular events with higher haemoglobin. Therefore, if LV mass is a goal of treatment, attempts should be made to reduce blood pressure further in this population. The patients studied in these investigations were candidates for renal transplantation, the definitive treatment for chronic renal failure. Cardiovascular disease is the leading cause of death both in patients on the renal transplant list, as well as post successful transplantation. There is a great deal of interest in identifying patients at high cardiovascular risk, to allow strategies to be adopted to minimise this risk, frequently by undertaking invasive investigation such as coronary angiography. In a survival study of 300 potential renal transplant recipients, factors associated with increased risk of mortality were increased age, ischaemic heart disease whilst receipt of a renal transplant was protective. Although the presence of LGE was associated with poorer outcome, this finding was not independent of other variables. One interesting finding was that patients with greater exercise tolerance, measured objectively using the full Bruce exercise test had better outcomes. This observation represents a simple pragmatic method to risk-stratify such patients. A study using the biomarker brain natriuretic peptide (BNP) a peptide released from the LV in response to stretch and hypertrophy, in 114 patients, demonstrated that whilst BNP has potential as a diagnostic tool for the presence of uraemic cardiomyopathy, in particular LVH, this peptide added little prognostic value. As familiarity with CMR techniques developed, it became clear that vascular function could be investigated with this imaging modality. Previous studies using alternative measures of vascular function have suggested that arterial stiffness is an important predictor of long term outcome in patients with ESRD. A study of 147 uraemic patients using aortic distensibilty and aortic volumetric arterial strain as CMR measures of aortic stiffness demonstrated that both these parameters were associated with an increased risk of cardiovascular events and mortality. To date there do not appear to be any similar outcome studies using these measures, although a number of authors have noted an association between aortic distensibilty and cardiovascular risk factors. These factors may represent potential targets for therapy aimed at reduction of cardiovascular risk in patients with ESRD. One unfortunate development during the period during which these studies were undertaken, was the emergence of a link between exposure to gadolinium based contrast agents and nephrogenic systemic fibrosis (NSF), a potentially life threatening skin disorder in patients with advanced renal failure. This finding lead to the cessation of contrast CMR studies. A retrospective investigation of factors present in patients in North Glasgow affected by NSF, confirmed that patients with NSF were more likely to have undergone contrast based imaging than unaffected patients, frequently undergoing multiple scans, with high doses of gadodiamide used. Until this issue is clarified, future scans using these agents in this population should be undertaken with caution. These studies have characterised for the first time the relationship between both uraemic cardiomyopathy and uraemic arterial stiffness and both cardiovascular risk factors and long term outcome. CMR measures of cardiac dimensions and vascular function represent future targets for interventions aimed at reducing cardiovascular risk in patients with advanced renal failure