Growing Heart Failure Burden of Hypertensive Heart Disease: A Call to Action

The contribution of HHD to the pathophysiology of HF in hypertension is the result of a sequential and complex process (reviewed in Messerli et al¹⁴). Initially, LV adaptation to high BP-mediated hemodynamic wall stress (ie, pressure overload) consists of LV wall thickening and LV mass (LVM) increase with development of concentric LVH. At this stage, LV diastolic dysfunction is the first manifestation of cardiac dysfunction. When pressure overload is sustained, diastolic dysfunction progresses, and HF with preserved LV ejection fraction ensues. The end stage of HHD, usually the result of longstanding pressure overload with or without the concurrence of myocardial infarction, consists of dilated cardiomyopathy with reduced LVEF and development of HF with reduced LV ejection fraction.

Accumulating data highlight the contributing role of nonhemodynamic mechanisms in the pathophysiology of HHD. Specifically, interconnected mechanisms related to alterations of immunity, oxidation/antioxidation balance, and endothelial function may be potential contributors to cardiac (and other target-organ) damage in hypertension (reviewed in Rizzoni et al¹⁵ and Griendling et al¹⁶). Additionally, it is important to consider that the timing and manifestations of the process leading to HHD and its clinical evolution may be individually modulated by several factors including demographics. For instance, lifetime BP evolution differs in women compared to men, potentially resulting in an increased HF risk at lower BP thresholds in women.¹⁷

Evidence collected during the last years support that changes in LV morphology and function developing along the clinical course of HHD are accompanied by broad alterations in the three histological components of the myocardium (the cardiomyocyte [ie, hypertrophy and death], the interstitium [ie, inflammation and fibrosis] and the microvasculature [ie, wall thickening with lumen reduction of small arteries and arterioles, and capillary rarefaction]) leading to its structural remodeling (reviewed in González et al¹⁸).

As reviewed elsewhere,¹⁹ it is increasingly recognized that the diffuse excessive accumulation of collagen fibers within the myocardial interstitium or MIF plays an important role in impaired LV systolic and diastolic function and that its severity is determinant in the development of HF and arrhythmias in patients with HHD and other cardiac diseases (Figure 1). Briefly, MIF is the main result of alterations in the extracellular processing of fibrillary collagen type I that lead to the excessive formation and subsequent deposition of highly cross-linked collagen type I collagen fibers. These alterations are due to the presence of activated fibroblasts and differentiated myofibroblasts originated from the resident cardiac fibroblasts as a reactive response to biomechanical stress or as a reparative response to damage-associated molecular pattern–dependent pathways. The composition of the secretome of activated fibroblasts and myofibroblasts favors the synthesis of highly cross-linked fibrillar type I collagen fibers against its degradation. MIF may contribute to LV diastolic dysfunction and HF with preserved LV ejection fraction by directly increasing LV stiffness and impairing LV filling during early diastole and may participate in LV systolic dysfunction and HF with reduced LV ejection fraction by realignment of fibers relative to cardiomyocytes, impairing the transmission of force generated by cardiomyocytes to the ventricular chamber, and/or by encircling and restricting the stretching of cardiomyocytes in diastole, thereby reducing their length-dependent generation of force in systole.

Current hypertension guidelines^4,5 recommend ECG or 2-dimensional standard doppler echocardiography for basic or more detailed screening of the heart, respectively. However, neither method provides accurate information on MIF. Although the histopathologic analysis of endomyocardial biopsy samples is the gold-standard method to detect and characterize myocardial remodeling (ie, to quantify the volume of myocardial tissue occupied by collagen fibers or collagen volume fraction [CVF]), several limitations preclude the routine implementation of endomyocardial biopsy for patients with HHD.^20,21 Therefore, imaging- and biochemical-based approaches have been suggested to assess noninvasively MIF in HHD.

Cardiac magnetic resonance imaging techniques based on native T1 mapping and extracellular volume (ECV) quantification have been developed to detect and characterize MIF given that several studies have shown direct correlations between the ECV and CVF in cardiac patients with and without HF.²² However, none of these studies included patients with HHD. It has been reported that hypertensive patients with LVH exhibited increased ECV values compared to normotensive subjects and hypertensive patients without LVH (Figure 2).^23–30 Whereas ECV was similar in patients with HHD and HF than in patients with HHD without HF in one study,²³ it was higher in HF patients than in non-HF patients in another study.²⁵ ECV has been shown to be directly correlated with LVM and LV wall thickness and inversely correlated with circumferential strain, longitudinal strain, and LVEF in patients with HHD without HF,^{23,26,27,30,31} and inversely correlated with peak oxygen consumption and directly correlated with brain natriuretic peptide in patients with HHD and HF.²⁵

The main limitation of ECV is that pathologies other than MIF can cause an increase in ECV, for example, interstitial inflammation that is part of myocardial remodeling in HHD.¹⁸ In this regard, Pan et al²⁹ reported that in patients with HHD the increase in ECV was significantly associated with circulating biomarkers of inflammation. However, the lack of studies that analyze the relationship between ECV and CVF in patients with HHD makes it difficult to establish an ECV cutoff value to discriminate the severity of MIF in these patients.

Among the many circulating molecules proposed as biomarkers of MIF, several peptides derived from collagen metabolism have been described that can provide valuable information on MIF (reviewed in López et al³²). For instance, the carboxy-terminal propeptide of procollagen type I (PICP) which is formed during the extracellular conversion of procollagen type I into mature fibril-forming collagen type I by the enzyme procollagen type I carboxy-terminal proteinase or PCP.³³ The catalytic activity of this enzyme is increased up to 15 times by the glycoprotein PCP enhancer-1³⁴ which, in turn, is stimulated by angiotensin II under conditions of cardiac pressure overload.³⁵ PICP plays a key role in the assembly of collagen type I molecules into collagen type I fibrils and can, therefore, be considered as a fibrogenic factor.³⁶ Serum concentrations of PICP are abnormally increased in patients with HHD, especially those with HF, and exhibit a strong correlation with CVF in patients with HHD (Figure 3).^37–39 In addition, the performance of PICP in predicting the severity of MIF in patients with HHD makes it possible to establish a cutoff value of 110.8 ng/mL (as measured by enzyme-linked immunoabsorbent assay) above which endomyocardial biopsy shows CVF values >6% (ie, severe MIF).³⁹ Interestingly, after a median follow-up of 5.31 years (range, 0.24–7.21 years), a composite end point of first HF hospitalization after enrollment or death from cardiovascular cause occurred in 27.0% and 48.1% of patients with PICP values < or ≥110.8 ng/mL, respectively.⁴⁰

The main limitation of circulating biomarkers of MIF is their lack of specificity for the myocardium, which can confound fibrosis in noncardiac organs/tissues throughout the body. However, serum PICP detected in peripheral blood of patients with HHD seems to be mainly of cardiac origin because a positive gradient exists for its concentration from the coronary sinus towards antecubital vein in these patients but not in normotensive subjects, and the concentrations of peripheral and coronary PICP are highly correlated in patients with HHD (Figure 3).³⁹ In addition, it has been shown that the pathophysiological and clinical value of this biomarker may be enriched (eg, in prognostic terms) by combining its assessment with the assessment of other collagen-derived biomarkers reflecting not only the quantity but also the quality of MIF.⁴⁰

Despite the accumulating evidence supporting the growing HF burden of HHD, current guidelines for treating hypertension recommend no specific pharmacological strategies in patients with HHD beyond BP lowering.^4,5 However, in a position paper⁴¹ of the Heart Failure Association in collaboration with the European Association of Preventive Cardiology, the authors recommended the use of diuretics, angiotensin-converting enzyme inhibitors, and angiotensin receptor blockers to prevent HF in hypertensive patients. This recommendation is mainly supported by the results from a network meta-analysis based on 26 trials showing that the 3 aforementioned classes of antihypertensive drugs were the most effective to reduce the incidence of HF compared with placebo, whereas calcium channel blockers and β-blockers were found to be the less effective.⁴²

However, since HF risk is higher in patients with LVH compared with those with persistently normal LVM,⁴³ regressing LVH has been proposed as an important therapeutic goal in hypertensive patients. However, the results of a meta-analysis, based on 80 studies, indicate that the ability to reduce LVH by antihypertensive treatment is highly variable between different classes of BP-lowering drugs and independent of changes in BP.⁴⁴ Furthermore, data from 2 ancillary studies of SPRINT (Systolic Blood Pressure Intervention Trial) show that, although a more intense reduction in BP results in the prevention of HF events,⁴⁵ it is not associated with a greater reduction in LVM.⁴⁶

In a recent experimental study, it was shown that MIF, evaluated histologically and by cardiac magnetic resonance imaging, appears to be the main determinant of LV dysfunction in HHD both before and after reversal of systemic hypertension,⁴⁷ making the regression of this injury an important goal in the prevention of HF. Since the pioneering works of Brilla et al⁴⁸ and Schwartzkopff et al⁴⁹ in patients with HHD, it is known that MIF present in this disease is reversible with pharmacological treatment. Therefore, it can be of interest to consider the effects of antihypertensive treatment on the 2 surrogate biomarkers of MIF here considered: ECV and PICP. In patients with HHD, 6 months of effective BP normalization with either an angiotensin-converting enzyme or an angiotensin receptor blocker was associated with a significant decrease in most ECV values despite no change in LVM (Table).²³ However, in one study performed in patients with resistant hypertension who underwent renal denervation, there was a significant reduction in ECV associated with marked BP reduction but no change in LVM.⁵⁰ On the contrary, in another study in patients with resistant hypertension submitted to renal denervation in whom BP and LVM decreased significantly, ECV remained unchanged.⁵¹

As regards to serum PICP, it has been reported that its concentration and myocardial CVF decreased in parallel in response to the angiotensin receptor blocker losartan but did not change in response to the calcium channel blocker amlodipine, in patients with HHD without HF treated for 12 months (Figure 4).⁵² Significant parallel reductions in serum PICP and CVF were found in response to the loop diuretic torasemide, but not to the loop diuretic furosemide, in patients with HHD and HF after 8 months of treatment (Figure 4).⁵³ The effects of losartan and torasemide were not influenced by either BP or LVM changes and were accompanied by reduced LV chamber stiffness and improved LV function.^{38,45,52–55} The reduction of serum PICP by losartan and torasemide may be related to their ability to inhibit the enzyme PCP as suggested by experimental³⁵ and clinical⁵⁵ data, respectively.

Although mineralocorticoid receptor antagonists exert biopsy-proven antifibrotic effects in HF patients with idiopathic dilated cardiomyopathy,⁵⁶ no data are yet available on their effects on CVF in patients with HHD. As regards to the surrogate biomarkers of MIF, spironolactone treatment was associated with reduction of serum PICP in one study performed in patients with resistant hypertension,⁵⁷ and no change of this parameter in another study of patients with HHD.⁵⁸ The effects of mineralocorticoid receptor antagonists on ECV in patients with HHD remain to be investigated.

Finally, although there are clinical data showing that sodium-glucose cotransporter-2 inhibitors reduce BP⁵⁹ and LVM,^60,61 and experimental data showing their ability to attenuate MIF,⁶² no evidence is still available on their effects on either PICP or ECV and on their effectiveness to reduce the risk of HF in HHD.

As the HF burden of HHD is growing, reducing its clinical, economic, and societal impact is a key issue in modern cardiovascular medicine that needs more targeted prevention and therapy efforts than those recommended by the current guidelines. This, in turn, requires modifying the handling of HHD (Figure 5). The development of this disease can no longer be considered only from the perspective of the LV chamber response to hemodynamic overload, but must also incorporate the view of the detrimental structural impact of hemodynamic and nonhemodynamic factors on myocardial tissue compartments, in particular on the interstitium leading to MIF. However, the diagnosis of HHD must go beyond changes in LV morphology and function assessed by classical imaging methods and include also the assessment of MIF-related parameters provided by novel imaging methods (eg, ECV assessed by cardiac magnetic resonance imaging) and blood assays (eg, serum PICP assessed by immunoassay methods). This integrated approach of imaging and chemical biomarkers may enable clinicians and investigators to more accurately prevent HF in patients with HHD. Finally, the time has come to incorporate precision medicine in the management of patients with HHD through the selection of personalized therapies aimed not only at BP control and LVH regression but also at the prevention/reduction of MIF (eg, the combination of torasemide and losartan) and, by doing so, reduce the risk of HF.

However, it is important to recognize that this new myocardial-focused paradigm is based on information provided by not many small clinical trials. For instance, despite the clinical evidence on the diagnostic potential of serum PICP presented here, we acknowledge that for this circulating peptide to be established as a definitive biomarker in HHD, there is a need for large prospective randomized clinical trials to assess whether PICP-targeted therapy leads to decreased development of HF in patients with HHD. Therefore, we make a strong call to action for large trials evaluating the aforementioned approach to assess incremental results. Furthermore, therapies targeting remodeled myocardium and/or identification of selective targets of myocardial remodeling will require fundamental research to better understand the molecular basis of hypertensive myocardium.

In summary, academia, industry, and regulators are urged to develop a broad research agenda that spans the spectrum of basic, translational, and clinical studies repurposing known diagnostic tools and drugs for new indications and investigating novel therapies and their effects among HHD patient subgroups defined according to biomarkers of myocardial remodeling (eg, MIF) in the application of personalized/precision medicine to prevent the transition to HF.