Pathophysiological role of aldosterone in cardiac remodelling after myocardial infarction

Acute myocardial infarction (AMI) remains a common and serious manifestation of coronary artery disease. The development of heart failure (HF) and/or evidence of left ventricular systolic dysfunction (LVSD) following AMI increases both in-hospital and longer-term mortality. A series of structural and functional changes occur within the heart in general and within the left ventricle (LV) in particular following AMI, initially providing a stabilising mechanism to maintain the cardiac output but over time becoming maladaptive and leading to progressive ventricular dilatation, dysfunction, HF and premature death. This process is termed remodelling. It is now understood that a complex series of mechanical, genetic and neurohormonal factors, including the mineralocorticoid hormone aldosterone, are implicated in its pathogenesis. Aldosterone antagonism reduces cardiovascular morbidity and mortality in patients with advanced chronic HF and survivors of large AMI who develop HF and/or are diabetic. These benefits could, at least in part, be due to an anti-remodelling action, although this was uncertain at the time I began my research project. The work contained in this thesis examines cardiac remodelling, and the effects of the mineralocorticoid receptor antagonist eplerenone, in a cohort of 100 patients admitted with AMI, with depressed LV ejection fraction (LVEF) but without HF or diabetes mellitus, using a gold standard modality for LV functional assessment and infarct imaging: late gadolinium-enhanced cardiac magnetic resonance (LGE-CMR). Patients were treated with (double-blinded) eplerenone or placebo for 24 weeks, and underwent serial LGE-CMR scanning and measurement of haematological, urinary and genetic markers thought to be of pathophysiological importance in post-infarction remodelling. There was, by chance, a significant imbalance in LV function at baseline between the randomised groups. After pre-specified covariate-adjustment, however, I found a significant effect of eplerenone on LV remodelling. Moreover I found that the use of eplerenone in addition to an angiotensin-converting enzyme inhibitor or angiotensin receptor blocker was well-tolerated. I also found an effect of eplerenone on two matrix metalloproteinases (MMPs) considered key enzymes in extracellular matrix turnover. Specifically, eplerenone decreased MMP-2 and attenuated the drop in MMP-9 seen in the placebo group, changes that are protective against remodelling. These findings suggest a potential anti-remodelling effect of eplerenone in ‘asymptomatic’ LVSD after AMI, i.e. patients in whom eplerenone is not currently indicated. Patients with limited functional recovery early after AMI resulting in a persistently reduced LVEF require stringent monitoring and may qualify for an implantable cardioverter defibrillator. The use of predictive biomarkers to identify such patients is gaining popularity. I found that two biomarkers, tissue plasminogen activator (tPA) antigen and tissue inhibitor of metalloproteinase-4 (TIMP-4), measured in a blood sample taken a mean of 3 days after AMI, were independent predictors of adverse remodelling. These findings are novel, provide further pathophysiological insights into the inter-related biological systems that underlie remodelling, and may inform future trials aimed at modulating these pathways in order to attenuate remodelling. LGE-CMR affords detailed characterisation of myocardium. In keeping with previous studies, infarct volume, endocardial extent and transmurality predicted LV remodelling. In addition I also found that the presence of microcirculatory dysfunction, defined as persistent microvascular obstruction (MVO) within the infarcted region on baseline LGE-CMR, divided my study population into two distinct groups with opposite remodelling outcomes. Patients with late MVO progressively remodelled, while those without reverse remodelled over 24 weeks. From these results, I propose that late MVO be used as an indicator of adverse ventricular remodelling. This may enhance the risk-stratification of survivors of AMI. Aldosterone has a number of detrimental effects on the cardiovascular system and is strongly implicated in the pathogenesis of remodelling. I observed direct correlations between aldosterone sampled at baseline and CMR parameters of remodelling. Analysis by treatment group revealed an association between change in aldosterone over time and parameters of remodelling in placebo- but not eplerenone-treated patients, despite higher circulating aldosterone concentrations in the latter group. We propose that the cardiac effects of aldosterone display a temporal variation after AMI, specifically that circulating aldosterone in the first few days after infarction is key in selecting a remodelling pathway but that over the following weeks and months circulating aldosterone is less influential in potentiating the remodelling process. I also found a novel relationship between aldosterone and infarct volume, which merits further investigation as the role of aldosterone in the pathophysiology of remodelling is further described. My trial design afforded the opportunity to examine the relationships of certain novel biomarkers with LV function and other established biomarkers after AMI. I demonstrated that circulating concentrations of the peptide apelin were reduced over 24 weeks after AMI compared to healthy controls but bore no relationship to LV function. Separately, I showed that concentrations of the soluble interleukin-1 receptor family member ST2 fell significantly over time after AMI and correlated with early and medium-term LV function and infarct volume. I detected a novel relationship between ST2 and aldosterone, which may suggest a pathophysiological role for ST2 in post-infarction remodelling and merits further investigation. These studies provide further insights into the roles of aldosterone and of selective mineralocorticoid receptor antagonism in cardiac remodelling in a relatively under-studied population: survivors of AMI with ‘asymptomatic’ LVSD. The primary results may inform clinical trials powered to detect a mortality benefit in this patient group. The data provided on the use of established and recently-discovered biomarkers in the prediction of medium-term LV function after AMI represents a highly topical area for further studies. Finally, pre-discharge CMR was safe in AMI patients, and facilitated the detection of additional findings that positively influenced the management of almost one-quarter of the trial cohort. These findings may lead to greater uptake of this increasingly-available modality, to enhance the early management and risk stratification of survivors of large AMI