A Unified View of Ventricular Remodelling

This editorial refers to ‘Cardiac remodelling and myocardial recovery: lost in translation?’ by Douglas L. Mann et al., published in this issue on pages 789–796. The association between cardiac enlargement and heart failure (HF) has long been recognized. In 1975, Grossman et al.1 hypothesized a fundamental relationship between myocardial wall stress and ventricular hypertrophy and chamber enlargement in which pressure overload would be associated with sarcomere replication in parallel and increased wall thickness-to-chamber radius ratio (concentric hypertrophy), allowing for normalization of left ventricular (LV) peak systolic meridional wall stress. Conversely, volume overload has been associated with sarcomere replication in series and preserved wall thickness-to-chamber radius ratio (eccentric hypertrophy). The term ‘ventricular remodelling’ was first coined to describe the progressive cavity dilatation following chronic myocardial infarction (MI),2 but has come to be used more generally to refer to changes in LV structure in response to physiological and pathological haemodynamic stress. Numerous studies have demonstrated not only the prognostic importance of increased LV volume3 in response to acute myocyte loss, but also the prognostic importance of architectural changes that occur in response to more chronic myocardial stressors, such as hypertension and diabetes. Increased LV mass4 is a powerful predictor of cardiovascular morbidity and mortality, and the relationship between LV mass and chamber size—LV geometry—provides incremental prognostic information beyond either parameter alone.5 Importantly, these characteristic patterns of response, eccentric and concentric remodelling, often coexist and interact with regard to patient risk. Among patients who sustained an MI complicated by LV dysfunction or HF, specific patterns of LV geometry have been associated with death or HF hospitalization, even after adjusting for ejection fraction (EF) and LV end-diastolic volume.6 Additionally, concentric hypertrophy at baseline predisposes to LV enlargement post-MI.7 Thus, a more comprehensive understanding of ventricular remodelling, incorporating not only chamber enlargement but also the pattern of ventricular hypertrophy, is essential to understanding the myocardial response to insult and its prognostic implications (Figure Figure 1). Patients with diabetes and hypertension, for example, have worse outcomes following MI for the same degree of ventricular function or enlargement compared with patients without these disorders.8,9 Indeed, the relationship between ventricular size and outcome in diabetics is different from that relationship in non-diabetics,10,11 suggesting that we must incorporate more sophisticated measures of structure to fully understand the remodelling process and to identify the patients at highest risk. In this month's issue of the European Journal of Heart Failure, Mann et al.12 present a timely review highlighting the importance of LV structure in the management of patients with HF. Specifically, they question the emphasis generally placed on EF at the expense of parameters of ventricular structure and the resulting missed opportunities to potentially better select HF patients for medical and device therapies. One example of where this emphasis on EF, to the exclusion of LV structure, is particularly problematic is the population of patients with HF and a normal or near-normal EF. Surprisingly, little is known about the spectrum of abnormalities in cardiac structure that characterize the heterogeneous syndrome of HF with preserved EF (HFpEF). As reflected in recent guidelines, it is believed to be primarily a disorder of diastolic function, with preserved to decreased LV cavity size.13 However, HFpEF patients encountered in clinical practice represent a much more heterogeneous group. No single EF cut-off defines a distinct HFpEF syndrome. Rather, EF has a unimodal distribution in patients with HF14 and falls below the clearly normal range in a sizeable minority of subjects enrolled in large clinical trials. A relatively high prevalence of concomitant ischaemic disease is also noted in these studies. Just as structural changes typical of patients with HFpEF (e.g. hypertrophy) are also present in HF with reduced EF patients, these findings suggest that many HFpEF patients share, to varying degrees, pathophysiological features of both concentric and eccentric remodelling. A better understanding of the spectrum of perturbations in cardiac structure underlying HFpEF, and particularly of the interaction between concentric and eccentric patterns of remodelling, may improve our ability to identify relevant patient subgroups and to develop effective disease-specific therapies to improve prognosis. Left ventricular structural parameters are not static—the myocardium can display tremendous plasticity in terms of both structure and function. In the acute MI setting, interventions to limit infarct size are associated with recovery of function and reverse remodelling.15 Recent studies raise the possibility that novel interventions like stem cell therapy may also attenuate early post-MI remodelling.16 Medical therapies, such as angiotensin-converting enzyme inhibitors17 and β-blockers18 that lower systolic and diastolic load and block adverse neurohumoral signalling, are associated with modest degrees of ventricular reverse remodelling and more profound degrees of reverse remodelling have been observed with cardiac resynchronization therapy (CRT).19 Similarly, regression of LV hypertrophy and concentric remodelling occurs with the treatment of chronic hypertension.20 What processes underlie reverse remodelling? Changes to both cellular and interstitial components of the myocardium characterize the remodelling process. Changes in myocyte volume occur, along with alterations in function characterized by decreased β-adrenergic receptor density and signalling, abnormal excitation–contraction coupling, and myosin heavy chain isoform switching.21 In addition, complex changes in the interstitial space, potentially mediated by matrix metalloproteinases and tissue inhibitors of metalloproteinases, are characterized by interstitial and perivascular fibrosis, increased chamber stiffness, and altered chamber geometry.22 It may therefore be expected that both cellular and interstitial myocardial components are involved in the reverse remodelling process. Indeed, reverse remodelling is associated with a characteristic myocardial gene expression profile which has been observed in the context of medical therapy,23 ventricular assist devices,24 and CRT.25 Reverse remodelling—both with respect to chamber size reduction and wall thickness reduction—is associated with improved cardiovascular outcomes. Multiple large clinical trials have demonstrated concordant improvements in ventricular size and all-cause mortality or HF progression associated with effective medical therapy26 or device therapy.27 Studies of direct modification of LV structure by surgical ventricular reconstruction demonstrate a similar relationship between the degree of reduction in LV volume and improvement in clinical outcomes.28 Moreover, reduction in the risk of cardiovascular events has been associated with LV mass regression with antihypertensive therapies.20 An important area for further investigation is the identification of factors contributing to the heterogeneity in the degree of reverse remodelling experienced with any given therapy. Given what is already known, gender, race/ethnicity, underlying aetiology of HF, and medical co-morbidities will likely play significant roles. Importantly, at least in the context of CRT, measures of myocardial function (strain) are emerging as important predictors of structural improvement. Cardiac structure and function are intimately related. The heart compensates to acute and chronic insult so as to maintain stroke volume and cardiac output. Falling EF therefore reflects compensatory chamber dilatation. However, ventricular volume and EF are distinct parameters29 and ventricular volumes alone do not capture the variety of morphological changes that the ventricle can undergo in response to stress or insult. Across the spectrum of EF, traditionally dichotomized patterns of concentric and eccentric remodelling coexist and interact with regard to patient risk. As Mann et al. outline, non-invasive methods to assess cardiac structure continue to improve and should allow for a more mature appreciation of the full spectrum of ventricular geometry in patients with HF, regardless of EF. Ultimately, greater understanding of the interaction between the pattern of remodelling, myocardial function, and patient response will be essential to translate current and emerging HF therapies into optimal clinical outcomes.