Myocardial remodeling is a process reactive to physiologic or pathologic stimuli. It manifests as gradual increases in left ventricular (LV) end-diastolic and end-systolic volumes, wall thinning and the changing of chamber geometry. The increase in chamber volumes is commonly associated with a continuous decline in ejection fraction. The concept was first described by Pfeffer et al.1 as a progressive myocardial transformation, which occurs following myocardial infarction. Since then, the remodeling process has been recognized as a common pathway for heart failure (HF) progression, as reviewed in ref. 2. The response to stress, pressure overload, chronic myocarditis and various forms of cardiomyopathies is an adaptive transformation of cardiac tissue during an initial phase, necessary to maintain adequate ventricular function. Over time, the process becomes maladaptive leading to functional decline and subsequent failure of cardiac function.3, 4 At cellular level, the hallmark of maladaptive transformation is alteration in myocyte phenotype. Induction of cardiomyocyte hypertrophy parallels the activation of a fetal gene program, which results in altered protein synthesis, defective excitation–contraction coupling, disturbed intracellular Ca2+ handling, loss of cardiac energy reserve and apoptosis. The mechanical and electrophysiological impairment of cardiomyocytes is accompanied by transformation of the extracellular matrix, activation and proliferation of fibroblasts, as well as the rearrangement of the coronary microvascular structure. More recently, the crucial role of non-coding RNAs in the regulation of these and many other processes has been suggested (reviewed in ref. 5). The consequences of cardiac remodeling include progressive worsening of systolic and diastolic function.

A clinically highly relevant part of the full spectrum of heart failure symptomatic is the increased propensity for arrhythmias due to the electrophysiological or arrhythmogenic remodeling.6 Effective contraction at a regular and adequate rate depends on appropriately timed generation and conduction of cardiac electrical activity. Through altered expression and regulation, the remodeling process directly affects the operation of the basic elements of electrophysiological activity, such as ion channels and transporters (potassium, sodium and calcium channels, calcium transporters and hyperpolarization-activated nonselective cation channels) and gap junction proteins like connexins (Figure 1). As a result, disturbances in cardiac rhythm may develop. Indeed, patients with heart failure experience a variety of rhythmic abnormalities.7 Occurrence of ventricular tachyarrhythmias and associated high risk of sudden cardiac death even when heart failure is hemodynamically well compensated is clinically the most relevant. In addition, atrial arrhythmias, particularly atrial fibrillation, are also common, and substantially contribute to morbidity and mortality. Bradycardias resulting from abnormal sinoatrial node function leads to cardiac dysfunction, weakness, syncope or circulatory collapse. In summary, sudden death, generally due to arrhythmic causes, is responsible for up to 50% of deaths among patients with cardiac failure.6

Figure 1
figure 1

Key determinants of the electrophysiological properties of the heart: L-type calcium channel. Red arrows indicate changes during pathological modeling. See reviews for more details (for example, in ref. 6). ATP, sodium-potassium pump; NCX, sodium-calcium exchanger; RyR, ryanodine receptor; SERCA, sarcoplasmic reticulum; TnC, Troponin C.

Both, preventing or reversing maladaptive remodeling represents therapeutics aim for heart failure patients. Preventive approaches such as treatment of causative conditions (hypertension or valve diseases) or early reperfusion therapy after myocardial infarction are the most effective ways to avert pathological remodeling.

Cardiac remodeling can be reversed due to the remarkable structural and functional plasticity of the myocardium. The reversal of remodeling encompasses the process of regression of pathological cardiac hypertrophy, improvement of systolic function and ideally normalization of chamber size and shape (as reviewed in ref. 8). Favorable reverse remodeling may occur in response to therapeutic interventions, hence its therapeutic relevance. Indeed, large bodies of experimental and clinical evidence indicate that pharmacological or surgical treatment or device therapy approaches that effectively attenuate progressive LV enlargements or improve LV ejection fraction are concomitantly reducing all-cause mortality or HF progression (for example, the MADIT-CRT trial9). Though reverse remodeling is an appropriate therapeutic goal for heart failure patients, the precise mechanisms underlying the structural improvements are still not well understood. The questions raised by Hallevell et al.8 in their review 5 years ago are yet to be answered, we still lack details regarding the underlying biology of the cardiac regression capacity and still cannot fully explain its diversity among heart failure patients. Even though many novel biomarkers of the remodeling process have been identified, we likewise do not have reliable markers that can predict at what stage is the heart still recoverable, or markers that can guide pharmacological or device treatment.

Many of the molecular details originate from recent studies with left ventricular assisted device (LVAD) - treated hearts. LVADs are used as ‘bridge therapy’ to sustain medically refractory patients awaiting cardiac transplantation, although the limited amount of possible donor hearts increase the ‘bridge to destination’ indications of LVAD therapies. A minority of LVADs (~5%) also can be removed because of interim cardiac improvements. LVAD-support induces cardiac unloading by sustained reductions in both preload and afterload and decreases neurohormonal activation. In addition, access to cardiac tissues, removed at the time of device implantation and at the time of transplantation, has been particularly valuable for studying the biology of reverse remodeling. The outcome of those studies has demonstrated that reductions in chamber mass and dilation are closely associated with decreases in average cardiomyocyte size and that improved cardiomyocyte calcium handling and β-adrenergic responsiveness account for improved cardiac pump function, as reviewed, for example, by Burkhoff et al.10

Nevertheless, LVAD studies may provide details mainly for late stage failing hearts. Medical therapy, including emerging novel approaches, aims to reverse pathological remodeling at an earlier and less severe stage of heart failure. This period is much more challenging to study due to the inability to access the heart directly. Suitable cardiac tissue samples from patients are largely unavailable and adequate surrogate markers are yet to be identified.

Well-characterized experimental models are therefore prerequisite to study the process of reverse myocardial remodeling. Only a few experimental models have been reported so far, among them the most commonly utilized model is pressure overload induced remodeling by aortic banding followed by de-banding, leading to mechanical unloading of the heart in rodents. The aortic banding and de-banding model in rodents provides a unique opportunity to imitate those alterations that occur in heart failure patients after a successful pressure unloading therapy such as aortic valve repair, replacement or pharmacotherapy. This model reliably mimics the fundamental cellular features of cardiomyocyte hypertrophy, interstitial fibrosis and their reversal.11

Using the same model in rats, Ruppert et al.12 in this issue of Hypertension Research provides novel evidence that an almost complete reversal of maladaptive electrophysiological alterations can be achieved during reverse remodeling. In their study, early left ventricular hypertrophy and associated ECG changes (prolonged QT and PQ interval, widened QRS complex) developed in response to aortic banding induced pressure overload. The changes gradually regressed following the unloading. The model allows the simultaneously investigation of electrophysiological changes parallel to morphological and functional alterations in the left ventricle. The timeline of molecular and histopathological alterations has been established during the process of electrical remodeling and reverse electrical remodeling, which helped to establish connections between the observed ECG changes and the cellular and functional alterations.

In summary, studying the various aspects of reverse remodeling as a measure of efficacy and as a therapeutic target is a popular and emerging topic in heart failure research. Changes in LV volume have become a valuable outcome measure that merits consideration, at least in phase II heart failure clinical trials. Moreover, as highlighted in the ‘2016 ESC Guidelines for the diagnosis and treatment of acute and chronic heart failure’,13 there are innovative and new therapeutics on the horizon that target reverse remodeling at the cellular level. These approaches aim to improve cardiomyocyte function and fibrosis acting on novel targets (for example, acto-myosin cross-bridge activation, sarcoplasmic reticulum Ca2+-ATPase activation, ryanodine receptor stabilization, energetic modulation or anti-fibrosis, and anti-matrix remodeling).

Finally, novel next generation concepts of a more mechanistically direct therapeutics such as normalization of protein expression by modulation of derailed regulatory non-coding RNAs may become clinical reality in the foreseeable future.5

Further studies are eagerly needed to both uncover further mechanistic details of the process and to demonstrate the efficacy of approaches targeting reverse remodeling.