2. Pathophysiology of Hypertension
The pathophysiological mechanisms responsible for HTN are complex, and act on a genetic background. Further, the probability of developing HTN increases with aging, due to the progressive stiffening of the arterial wall caused by, among other factors, slowly developing changes in vascular collagen and increases in atherosclerosis
[24][23].
The mosaic theory of hypertension, which has prevailed to the present day, proposes that HTN pathophysiology is multifaceted. Accordingly, HTN is caused by multiple factors, including genetics, environment, adaptive, neural, mechanical, and hormonal perturbations, which intertwine to elevate BP
[25,26][24][25]. Over the years, the Mosaic paradigm has been modified, and new concepts such as oxidative stress, inflammation, sodium homeostasis, and the microbiota have arisen, giving rise to further refinements of the ,osaic theory
[27][26].
One of the recently proposed pathways for BP elevation involves the gut microbiota. The microbiota and the brain (i.e., the microbiota–gut–brain axis) communicate via various routes including the immune system, the vagus, and the enteric nervous system. Many factors can influence microbiota composition in early life, including infection, type of birth delivery, antibiotics, nutritional provision, environmental stressors, and host genetic factors. On the other hand, microbial diversity decreases with aging
[28][27].
There is compelling evidence that shifts in gut microbiota play a key role in BP regulation. In this regard, intestinal bacteria synthesize metabolites, the most important being short-chain fatty acids (SCFAs), vasoactive hormones, trimethylamine (TMA) and trimethylamine N-oxide (TMAO) and uremic toxins, such as indoxyl sulfate (IS) and p-cresyl sulfate (PCS)
[29][28]. Microbiota derangements are causally associated with HTN
[30][29]. Hypertensive stimuli (stress, diet, salt, maternal factors, environmental toxins, etc.) significantly impact the microbiota-gut-brain axis and influence ANS brain regions to affect sympathetic, endocrine, and immune pathways
[31][30]. Chronic overactivation of immune cells either directly or through the gut microbiota ultimately produces chronic inflammation and HTN. T cells are central to the immune responses underlying HTN, as activated T cells infiltrate tissues and produce cytokines including interleukin 17A, which promote renal and vascular dysfunction as well as end-organ damage, thereby leading to HTN
[32][31].
In the gut itself, hypertensive stimuli cause gut microbial dysbiosis, gut barrier weakening, and inflammation
[31][30]. Weakened barrier function allows previously excluded gut contents (e.g., bacteria and their metabolites) to come in contact with the immune system and generate inflammation, which in turn exacerbates leakiness and inflammation and recruits pro-inflammatory bone marrow progenitor cells to the gut generating a vicious cycle. The factors crossing the weakened gut barrier and inflammatory mediators generated in the gut reach the brain via the circulation and contribute to the development of local (neuroinflammation) and systemic inflammation
[33][32].
Finally, there is bidirectional interaction between the microbiota and the renin angiotensin system (RAS), a major determinant of BP levels. Gut bacteria and their metabolites modulate gastrointestinal and systemic RAS, and at the same time, changes in the intestinal habitat caused by alterations in RAS may shape microbiota metabolic activity and composition
[34][33]. Thus, an inflammatory milieu developed by hypertensive stimuli precedes, and is causally related to HTN development regardless of the presence or absence of other morbidities, rendering the theory of multimorbidity-induced inflammation leading to HFpEF obsolete
[35][34]. It comes, therefore, as no surprise that HFpEF in the absence of HTN is rare
[36][35].
The effect of the aforementioned perturbations on BP levels is also affected by genetic factors. Several genes have been identified to play a role in the pathophysiology of HTN, including those involved in the renin–angiotensin–aldosterone system (RAAS), catecholamine/adrenergic system, renal kallikrein–kinin system, epithelial sodium channel, adducin, and those involving lipoprotein metabolism, hormone receptors, and growth factors
[37][36]. Genome-wide association studies (GWAS) have exposed more than 100 variants associated with blood pressure in the general population
[38][37], and some studies have reported genes associated with hypertensive heart disease’s complications such as cardiomyopathy and HF
[39][38]. Lastly, post-genomic biomarkers, from the emerging fields of transcriptomics, proteomics, glycomics, and lipidomics, have provided new insights into the molecular underpinnings of hypertension
[37][36].
3. From Hypertension to Hypertensive Heart Failure
HTN is characterized by chronic LV pressure overload and increased intravascular volume, both affecting the LV structure and function
[40][39]. Nevertheless, the conventional concept, described more than 120 years ago by William Osler, has been that HTN leads to concentric LV hypertrophy (LVH), which in turn is followed by eccentric LVH and HF
[41][40]. In this regard, Messerli et al. proposed that HHF develops in four stages:
[42][41] (a) stage I: isolated LV diastolic dysfunction with no LV hypertrophy; (b) stage II: LV diastolic dysfunction with concentric LVH; (c) stage III: clinical HF (dyspnea and pulmonary edema) with concentric LVH; and (d) stage IV: eccentric LVH with HF and reduced ejection fraction.
However, doubt has been raised that the previously mentioned sequence of events is typical in hypertensives, based on the following arguments: (a) concentric LVH may not be the most frequent geometric pattern, and is less commonly seen than eccentric LVH in studies enrolling hypertensive subjects
[43][42]; (b) the transition from concentric LVH to eccentric LVH is uncommon in the absence of CAD
[44][43]; (c) the risk varies by LV geometric pattern, with eccentric and concentric LVH predisposing individuals to HFrEF and HFpEF, respectively
[45][44].
However, recent studies have provided compelling evidence that cardiac remodeling is a dynamic process, with phenotypic transitions occurring frequently regardless of the presence or absence of CAD. For example, in the Swedish Heart Failure Registry (SwedeHF), the proportion of patients with HFpEF (approximately 70% were hypertensive) that worsened during follow-up (from HFpEF to HF with mid-range LVEF [HFmrEF] or HF with reduced LVEF [HFrEF]) was 31.2% in the absence of baseline CAD, 33.1% in the presence of baseline CAD, and 57.2% in the case of interim CAD
[46][45]. Similar were the findings of another study including 1082 patients (approximately 70% were hypertensive) admitted to hospital due to decompensated HFpEF (LVEF > 50% at the first LVEF assessment at discharge). At LVEF reassessment within 6 months in the outpatient setting, 758 patients (70%) had an LVEF > 50%, 138 patients (13%) had an LVEF of 40–49% (HFmrEF), and 186 patients (17%) had an LVEF of <40% (HFrEF)
[47][46]. In addition, antihypertensive treatment attenuates or even reverses cardiac remodeling
[48,49][47][48].
The port of entry in the HF spectrum (HHF entry phenotype) depends on (a) HTN severity, duration and antihypertensive treatment effectiveness; (b) the balance between LV pressure and LV volume overload; (c) the coexistence of morbidities such as obesity, DM, and CAD that preexist and/or modify LVH development; and (d) disease modifiers (age, sex, genes, other). The eventual HHF phenotype results from transitions across the HF spectrum, whose direction depends on disease severity and antihypertensive treatment, which shift towards the lower end of upper end of the HF spectrum.
4. Cardiac Autonomic Imbalance
HTN evolves over the lifespan, from predominant sympathetic nervous system (SNS)-driven HTN with elevated mean BP in early and mid-life to a late-life phenotype of increasing systolic and falling diastolic BP, associated with increased arterial stiffness and aortic pulsatility
[53][49]. However, growing evidence indicates that the SNS is also capable of modulating arterial stiffness independently of prevailing hemodynamics and vasomotor tone
[54,55][50][51]. The contribution of arterial stiffness to HTN and HF development is high in the elderly and patients with CKD
[53,56][49][52].
The autonomic dysregulation observed in HTN escalates in HHF. SNS overactivity has long been appreciated as a compensatory mechanism initially supporting the failing heart, which, however, in the long term, triggers a sequence of unfavorable remodeling processes, causing HF progression and the occurrence of major cardiovascular events
[57,58][53][54].
The adverse cardiac effects of SNS overactivity in HF have been predominantly studied in dilated LV with eccentric LVH, in which it manifests as an increase in norepinephrine spillover from the cardiac sympathetic endings, leading to chronic β-adrenergic receptor (AR) hyperstimulation and maladaptive GRK2 upregulation (GPCR kinase 2)
[59][55], thereby promoting β-AR down-regulation, cardiac hypertrophy, and myocyte apoptosis. Further, GRK2 recruits β-arrestin, which then competes with G-proteins for interaction with the β-AR and limits their activation
[60][56]. The chronic SNS overactivity In HF with eccentric LVH has been attributed to several neurogenic disturbances acting in concert
[61][57].
SNS overactivity is also present in HF with nondilated hearts and lack of eccentric LVH. In this regard, earlier studies suggested that the SNS overdrive which is present in essential HTN is augmented in HF without eccentric LVH
[62,63][58][59]. These findings have also been confirmed in recent HFpEF studies, in which most patients were hypertensive. Kaye et al. evaluated 14 healthy volunteers and 20 HFpEF patients (65% hypertensive), and found systemic sympathoexcitation in HFpEF patients, as indicated by the increased plasma arterial norepinephrine concentration and plasma levels of dihydroxyphenylglycol (a major intraneuronal metabolite of recaptured norepinephrine) compared with normal controls
[64][60]. Seo et al., using iodine-123-labeled metaiodobenzylguanidine (123I-MIBG) single-photon emission computed tomography (SPECT) imaging in 148 patients admitted for acute decompensated nonischemic HFpEF (91% hypertensive)
[65][61], observed that during a mean follow-up period of 2.4 ± 1.6 years, those with a high total defect score (TDS) levels had a significantly greater risk of cardiac events than those with middle or low TDS levels.
In contrast to SNS overactivity, there is attenuation of the parasympathetic nervous system (PNS) activity and its physiological effects in HF
[66][62], including the PNS-mediated anti-inflammatory reflex
[67][63]. The vagus nerve also plays a crucial role in this reflex, providing the afferent and efferent pathways which underly the communication between the brain and peripheral organs, including the heart
[67,68][63][64]. Vagal sensory afferents are activated by proinflammatory cytokines in peripheral tissues and convey the signal to the brain. Subsequently this signal causes the release of acetylcholine from vagal efferents into the reticuloendothelial system, which inhibits proinflammatory cytokine synthesis and release. Thus, vagal withdrawal contributes to the creation of the inflammatory milieu in HF
[69][65].
In patients with HF, treatment of HTN has been unanimously recommended, but BP targets have been inappropriately defined
[70][66]. However, as RAAS and SNS overactivity play key roles in HTN and HF as well in the development of major complications such as kidney dysfunction, their inhibition is of outmost importance for the management of HHF
[71][67]. Further, another class of agents, the sodium glucose cotransporter 2 inhibitors (SGLT-2i), which have proved tremendously effective in HF management, have a complex pleiotropic mechanism of action, including a reduction of neurohormonal overactivity
[72][68].
5.1. Blood Pressure Targets
In the major outcome trials of hypertension, comparison of BP reduction with antihypertensive medications against placebo or no treatment, incident HF was the outcome that showed the largest intergroup risk reductions
[73,74,75][69][70][71]. Effective HTN treatment in HF should target not only average BP, but the time in therapeutic range pressure range (TTR) as well.
5.1.1. Target Average Blood Pressure
It has been suggested that there is a J-shaped curve describing an inverse relationship between BP and cardiovascular complications, and that this association is more pronounced in patients with preexisting CAD, HTN, or LVH
[76][72]. However, several lines of evidence dispute the presence of such a curve.
The target BP which should be pursued with medical treatment was evaluated in the pivotal SPRINT (Systolic Blood Pressure Intervention Trial), which demonstrated that among patients at high risk for cardiovascular events in the absence of DM, targeting a systolic BP < 120 mmHg (intensive treatment group; mean systolic BP at one year 121.4 mmHg), as compared with < 140 mmHg (standard treatment group; mean systolic BP at one year 136.2 mmHg), decreases the incidence of the primary outcome (myocardial infarction, other acute coronary syndromes, stroke, HF, or death from cardiovascular causes)
[77][73]. Nevertheless, analyses comparing the effects of intensive and standard BP treatment in the ACCORD (Action to Control Cardiovascular Risk in Type 2 Diabetes) trial showed that the diabetic patients who received standard glycemic therapy and intensive BP control had benefits similar to those seen in SPRINT
[78,79][74][75].
The results of SPRINT have been confirmed in several subsequent investigations. which demonstrated a significant and direct dose–response between systolic BP levels and ischemic heart disease risk across all systolic BP exposure values (100–200 mmHg), without evidence for a J-shaped curve
[80,81][76][77].
According to the 2018 ESC/ESH guidelines, in hypertensive patients with HF, BP-lowering treatment should be considered if an individual’s BP is ≥140/90 mmHg, and systolic BP should be lowered to a range of 120–130 mmHg
[7].
5.1.2. Time in Therapeutic Range
Although BP is a continuous and dynamic variable, a single or average BP value has been used for BP monitoring in clinical practice and HTN studies. To overcome this limitation, it has been recommended that doctors should pursue every point of BP monitoring instead of measuring a single value of BP
[82][78]. In this regard, the term TTR was introduced, which expresses the percentage of BP measurements recorded within a certain window (e.g., TTR for BP window 120–140 mmHg) and reflects, therefore, the prevailing BP during the follow-up period and the magnitude of BP variability
[83][79] (
Table 1). The importance of TTR in HTN management was documented in a study including 371,996 hypertensive patients, in which the mortality rate increased from the most consistently controlled quartile (>75% in TTR) towards the less consistently controlled quartiles
[83][79]. Likewise, a secondary analysis of the TOPCAT (Treatment of Preserved Cardiac Function Heart Failure With an Aldosterone Antagonist) trial, in which the TTR was calculated with the target range of systolic BP defined as 110 to 130 mmHg
[84][80], a greater time in the systolic BP target range was associated with a decreased risk of cardiovascular outcomes and mortality events beyond BP level, especially among younger patients. Finally, in a recent post hoc analysis of HHF patients both from TOPCAT and BEST (Beta-Blocker Evaluation of Survival Trial) showed a linear relationship between TTR and the primary outcome (cardiovascular death or HF hospitalization), and similar patterns were observed in the individual trials
[85][81]. Moreover, sensitivity analyses redefined target range as 110 to 130 mmHg for systolic BP or 70 to 80 mmHg for diastolic BP.
Table 1. Example illustrating the importance of time in therapeutic range (TTR) in hypertension management. Although the average systolic blood pressure (BP) achieved with treatment is similar among the three patients, the TTR (target range for systolic BP: 110–130 mmHg) significantly differs.
These BP targets may also be applied in elderly HHF patients, as well as in those with coexistent DM
[86][82] or kidney disease
[87][83].
5.2. Medications
The drugs used in the management of HF, except for diuretics, have both cardioprotective and blood pressure-lowering properties (
Table 2).
Table 2.
Therapeutic targets and medications in the management of hypertensive heart failure.