Present HT guidelines are not concordant on the role of beta-blockers (BBs). JNC 8 no longer recommends BBs for initial antihypertensive therapy (James et al. 2014); this confirms the previous JNC 7, that at the same time listed some compelling indications for the use of BBs in hypertensive patients with other cardiovascular diseases (e.g. HF or previous myocardial infarction) (Chobanian et al. 2003). Similarly, the NICE guidelines state that BBs are not a preferred initial treatment; exceptions are considered for patients intolerant to ACE-Is and angiotensin II receptors (ARBs), women of child-bearing potential and people with evidence of increased sympathetic drive: apart from these conditions, BBs are relegated to fourth- or even fifth-line choices (National Institute for Health and Clinical Excellence, http://​www.​nice.​org.​uk/​guidance/​cg127). On the contrary, ESC guidelines do not exclude BBs as one of the options for fist-line antihypertensive pharmacological strategy (Mancia et al. 2013); similarly to JNC 7, European guidelines reported some suggested indications for the use of BBs and the other antihypertensive agents in specific conditions (see Table 1).

The beta-adrenergic signaling system acts through different receptors (Brodde et al. 2006). In brief, it should be taken into account that beta₁-adrenoreceptors are situated on the cardiac sarcolemma and their stimulation increases heart rate, the rate and force of myocardial contraction, the reuptake of cytosolic calcium (lusitropic effect) and the rate of conduction. Beta₂-receptors, on the other hand, are mainly located on bronchial and vascular smooth muscle cells, where they mediate the dilating effect of epinephrine; there is a significant proportion of beta₂-receptors also in the myocardium that might have additional inhibitory action modulating the level of adrenergic activation. Beta₃-receptors are a more recently recognized type of beta-receptor that have been found in many tissues: even if their specific role remains to be fully elucidated, they appear to be involved, among the others, in the control of thermogenesis and urinary bladder relaxation (Balligand 2013). A main cardiovascular effect of beta₃-receptors seems to be the endothelium-dependent vasorelaxation by production of nitric oxide. Moreover, even if further research is needed, there are some evidences that these receptors may have a cardioprotective role in myocardial hypertrophy and HF, counteracting the known adverse effects of beta₁ and beta₂ overactivation and modulating left ventricular relaxation and inotropism (Balligand 2013).

BBs have several antihypertensive mechanisms: along with their chronotropic and inotropic negative effects that reduce cardiac output, they exert central adrenergic inhibition and decrease the renal release of renin (negatively influencing also the activity of RAAS). At the beginning, these mechanisms do not lead to a significant fall in blood pressure, since there is a reflex increase in peripheral vascular resistance with tendency to normotension: alfa-adrenergic vasoconstriction, in particular, is not properly counteracted by the vasodilator component of epinephrine because of the vascular beta₂ blockade (Vatner and Hintze 1983). A later action of BBs on prejunctional receptors of the sympathetic terminal neurons inhibits the release of norepinephrine, possibly explaining the subsequent reduction of peripheral resistance and late antihypertensive effect. BBs retaining a vasodilatory property (see ensuing paragraph) determine an early decrease in vascular resistance and blood pressure. Additionally, like most other antihypertensive agents, beta-blockers probably also lower blood pressure through interference with a not yet identified vasoconstrictor mechanism (Man in’t Veld et al. 1988). Apparently, no significant effect on circulating blood volume is exerted (Weidmann et al. 1976). In relation to RAAS activation, BBs have been shown to yield either a neutral (Balansard et al. 1977; Fagard et al. 1976) or a mildly inhibitory effect (Ishizaki et al. 1983; Pitkäjärvi et al. 1979), with apparently no relation with blood pressure fall.

According to their pharmacological characteristics, different subclasses of BBs can be distinguished. The first-generation molecules (e.g. propranolol) are non-selective agents, blocking beta₁ and beta₂ receptors. They have been superseded by cardioselective blockers (such as atenolol, metoprolol, bisoprolol): the greater β1 than β2 affinity is dose dependent and is less with higher doses. A third generation class of BBs is represented by agents with vasodilatory activity that depends by two main mechanisms: direct vasodilation mediated by the release of nitric oxide (as for nebivolol and carvedilol) (Chen et al. 2013) and additional alfa-blockade (as for labetalol and carvedilol). Some BBs (pindolol and acebutolol) have an intrinsic sympathomimetic activity, stimulating smooth muscle cells to relax. As described further in the text, vasodilating agents carry a better metabolic profile that might represent an added value in treating hypertensive patients.

As stated above, even if BBs have been commonly used as first-line treatment for HT, at present they have been downgraded in most guidelines since several evidences have questioned their efficacy in preventing death and cardiovascular outcomes. A first meta-analysis by Messerli et al. found that BBs were ineffective in preventing CAD, cardiovascular and all-cause mortality in comparison with diuretics in patients older than 60 years (Messerli et al. 1998). Another systematic review by Calberg et al., including 9 randomized clinical trials, showed that atenolol had a blood pressure-lowering effect similar to that of other antihypertensive drugs, but cardiovascular mortality and stroke were higher with atenolol (Messerli et al. 1998): these results were confirmed also by two subsequent larger meta-analysis evidencing that BBs are no better than any other antihypertensive at preventing heart attacks and are less effective at preventing strokes (Lindholm et al. 2005; Bradley et al. 2006). A more recent Cochrane meta-analysis, updating a previous one, was performed on 13 randomized controlled trials and reported that patients treated with BBs showed higher total mortality and more cardiovascular events than CCBs (not diuretics and RAAS blockers) and higher incidence of stroke than CCBs and RAAS inhibitors; the different classes of agents were found similar for CAD (Wiysonge et al. 2007). It should be underlined that in the majority of the studies included in the meta-analysis atenolol was the most used BB, therefore any extrapolation to other BBs should be done with caution: in particular, as addressed in the ensuing text, some limitations of the old BBs seem not to be shared by third-generation molecules (Ram 2010). Moreover, the Authors acknowledged the low quality of the available evidence. The main arguments supporting the fact that ESC guidelines did not exclude BBs from the first-line pharmacological options are similar to these observations: in general, it has been recognized that evidence against their use appeared to be mixed and globally not strong enough (Mancia et al. 2013). In fact, differently from the analysis presented above, another large meta-analysis on 147 trials showed that all classes of antihypertensive drugs had a similar effect in reducing myocardial infarction and stroke for a given reduction in blood pressure and, in particular, BBs were highly effective in preventing cardiovascular events in patients with recent myocardial infarction and HF (Law et al. 2009). Moreover, incidence of cardiovascular outcomes was found to be similar for BBs and other drugs in the revision of trials realized by the Blood Pressure Lowering Treatment Trialists’ Collaboration, giving the message that reducing blood pressure is what apparently really counts (Czernichow et al. 2011). Nevertheless, as a consequence of all observations, several authorities consider that at present even if it would be incorrect to affirm that BBs have no effect in patients with primary HT, probably their effect should be considered suboptimal (Lindholm et al. 2005). BBs appear to be less effective than RAAS blockers and CCBs also in delaying or reverting hypertensive organ damage (e.g. left ventricular hypertrophy, arterial remodeling..). In any case, the importance of sympathetic nervous system activation in the pathogenesis of HT and the utility of BBs in certain compelling indications for cardiovascular diseases indicate that these drugs still have an important role for many hypertensive patients.

Several reasons have been considered to explain the relative lower efficacy of BBs found in some studies. One possible mechanism has been suggested by the Conduit Artery Function Evaluation (CAFE) study (a substudy of the Anglo-Scandinavian Cardiac Outcomes -ASCOT- Trial): treatment with BBs resulted in reduced brachial blood pressure with a lesser reduction of central (aortic) and pulse pressure in comparison to the other agents (Williams et al. 2006). Indeed, it has been shown that distinct antihypertensive drugs produce different blood pressure effects peripherally vs centrally (Morgan et al. 2004) and that changes in peripheral artery pressure do not accurately reflect changes in central pressure following different drug interventions. In particular, it seems that the effect of BBs on central aortic pressure is overestimated by brachially-measured blood pressure whereas the effects of ACE-i and CCBs are underestimated. Moreover, from a hemodynamic point of view, it was shown that BBs increased blood pressure variability and this correlated with trends in stroke risk (Rothwell et al. 2010).

Additional observations suggest that most traditional BBs produce several deleterious metabolic and vascular effects, which could negatively affect the prognosis of hypertensive patients (Ram 2010; Fragasso et al. 2009). As stated above, BBs may induce peripheral vasoconstriction that not only can be particularly detrimental in the context of atherosclerosis but is also one of the determinants of insulin resistance, a pathologic condition directly correlated with the severity of HT (Taddei et al. 2001; Ferrannini et al. 1987; Schiffrin and Deng 1996). The loss of insulin sensitivity is also associated with endothelial dysfunction, a pathophysiological factor present in many patients with HT that has a relevant prognostic weight. BBs have been shown to impair vasodilation, to induce endothelial dysfunction (that can contribute to vasoconstriction) and to deteriorate glucose homeostasis in hypertensive patients (Lithell 1991): atenolol has been found to increase the risk of new-onset diabetes in predisposed patients (Gupta et al. 2008), particularly when used in combination with diuretics. Additionally, old generations BBs have also been consistently shown to favor the increase of body weight, because of the inhibition of lipolysis in adipose tissue, and to adversely affect lipid metabolism, with tendency to an increase in plasma triglyceride and lowering of HDL cholesterol (Eliasson et al. 1981). In general, it could be stated that old generation BBs yield a metabolic profile not dissimilar from diuretics. Attention should be paid to diabetic patients not only for the possible metabolic side effects but also for the risk of masking the symptoms of hypoglycemia. Along with all these issues, BBs may affect QOL also by causing insomnia, depression (due to low arousal of central nervous system), erectile dysfunction and easy fatigability (which can limit significantly the ability to perform exercise).

As previously outlined, BBs may yield differential metabolic and vascular effects among different molecules. In particular, in more recent years, the development of the third-generation vasodilatory BBs has determined a reassessment of the ancillary metabolic effects of this class of drugs and some of the past reservations may apply less or not at all to the new agents (Kalinowski et al. 2003). ESC guidelines have also underlined the potential ameliorating effects of new generations BBs and this fact was another valid reason to keep BBs among first-line treatment options (Mancia et al. 2013).

Carvedilol, compared to metoprolol, was shown not to negatively affect glycemic control and to improve some components of the metabolic syndrome in hypertensive patients with diabetes (Bakris et al. 2004a). Similar results were obtained with nebivolol, that was superior in terms of insulin sensitivity and fibrynolitic balance (reducing plasminogen activator inhibitor-1 antigen concentrations) (Ayers et al. 2012); a reduction of oxidative stress was also observed (Kaiser et al. 2006; Celik et al. 2006; Rizos et al. 2003). The glucose-neutral effect of nebivolol has been observed even when added to diuretics. Accordingly, new BBs have been seen to reduce vasoconstriction and endothelial dysfunction (Mason et al. 2005; Pasini et al. 2008). It has also been proven that nebivolol reduces central aortic pressure and induces regression of left ventricular hypertrophy to a greater extent than metoprolol (Kampus et al. 2011). Vasodilation may be important not only for blood pressure reduction, but also for better tolerability; nebivolol has been shown to improve QOL (Van Bortel et al. 2008; Kendall 1989). In any case, apart from these pathophysiological observations, it should be noted that at present there are no strong outcome data to support the use of third-generation BBs as antihypertensives, even if both carvedilol and nebivolol have shown good results in HF (Mancia et al. 2013). General main contraindications to BBs are bronchospasm (in particular BBs are not safe in patients with asthma), sinus/atrio-ventricular node dysfunction, active peripheral vascular disease (also considering the Raynaud phenomenon) and coronary vasospasm. Figure 2 synthesizes the effects of traditional and new generation BBs on principal factors involved in the pathogenesis, maintenance and induction of complications in essential HT.

CCBs have gained a great role among the treatments for HT and are recommended as a first-line option by all the guidelines, in particular for older and black patients (Mancia et al. 2013; National Institute for Health and Clinical Excellence, http://​www.​nice.​org.​uk/​guidance/​cg127). This class of agents comprises two main subclasses, dihydropyridines (DHPs, e.g. nifedipine and amlodipine) and nondihydropyridines (Non-DHPs, verapamil and diltiazem), extensively studied in hypertensive patients (Nathan et al. 2005). The pharmacological property shared by all the CCBs is the selective inhibition of L-type calcium channel, whose main role is to allow the entrance of calcium ions required for initiation of contraction via calcium-induced calcium release from the sarcoplasmic reticulum. This channel is present in vascular smooth muscle cells and in the myocardium: distinction between DHPs and Non-DHPs depends on the different molecular binding site (Opie 1996). DHPs show a greater vascular selectivity (with modest direct cardiac effects), while Non-DHPs act also on the atrioventricular and sinoatrial node (with negative chronotropic effect and conduction delay) and have significant negative inotropic effect, more evident in patients with already altered myocardial function. The main anti-hypertensive mechanism of CCBs is arteriolar dilation due to blockade of vascular calcium channels (Braunwald 1982). A certain diuretic effect is also observed (particularly with short-acting DHPs as nifedipine): DHPs may increase glomerular filtration rate and renal plasma flow by antagonizing the intrarenal effects of angiotensin II and/or norepinephrine with vasodilatation of renal afferent arterioles, thereby resulting in increased natriuresis and diuresis (Loutzenhiser and Epstein 1985; Chellingsworth et al. 1990; Madeddu et al. 1987). The initial diuretic and natriuretic effects of most calcium channel blockers probably persist with long-term usage (Kaplan 1989; Epstein and De Micheli 1992) but the natriuresis resulting from nifedipine may be transient (Ene et al. 1985). In any case, CCBs acts independently of sodium intake.

The reduction in blood pressure achieved by CCBs stimulates a reflex activation of counter-regulatory mechanisms, with increased sympathetic activity and stimulation of RAAS: these systems determine tachycardia, increased myocardial contractility and tendency to increased systemic vascular resistance. The reflex release of norepinephrine is particularly stimulated by DHPs (Hamada et al. 1998), whereas verapamil tends to decrease plasma catecholamines levels (Bonadue et al. 1997; Kailasam et al. 1995; Vaage-Nilsen and Rasmussen 1998), yielding a sympatholytic effect. Moreover, verapamil and diltiazem control reflex tachycardia through their action on sinoatrial node and by reducing the influence of sinoaortic baroreceptors on heart rate, facilitating the reflex vagal control on the cardiac pacemaker by the afferent cardiopulmonary vagal receptors (Staszewska-Woolley 1987) and attenuating the sympathetic and parasympathetic components of the baroreceptor reflex (Giudicelli et al. 1984).

CCBs of both subclasses appear to be effective antihypertensive agents and most clinical trial data showed that they compare well with other drugs (Nathan et al. 2005); moreover, some meta-analysis have suggested that CCBs determine a better protection against stroke (Law et al. 2009; Blood Pressure Lowering Treatment Trialists’ Collaboration 2005; Verdecchia et al. 2005), although the reason is not fully understood. Potential harmful effects of CCBs have been previously reported (Psaty et al. 1997; Furberg et al. 1995). Observational studies and systematic overviews led to the suspect of an increased risk for cardiovascular events with some agents of this therapeutic class. In particular, short-acting nifedipine may precipitate myocardial ischemia as a consequence of the acute adrenergic reflex stimulation and sudden blood pressure fall: this pathophysiological mechanism might be enhanced in presence of CAD, with special hazard in high-risk patients in the immediate post-myocardial infarction period or in those with unstable angina (Marwick 1996; Turnbull 2003; Opie et al. 1995). The possibility of a “coronary steal” phenomenon has been considered too. Initial data suggested also pro-arrhythmic and pro-hemorrhagic effects.

At present, only long-acting CCBs are recommended for HT and several long-term studies have shown their safety and efficacy. Data from the ASCOT-Blood Pressure Lowering Arm (ASCOT-BPLA) trial showed that amlodipine (with possible addition of ACE-I as required) reduced major cardiovascular events and diabetes development compared to atenolol-based strategy (Dalhöf et al. 2005). Moreover, the more recent Avoiding Cardiovascular Events in Combination Therapy in Patients Living with Systolic Hypertension (ACCOMPLISH) trial, a large morbidity and mortality study comparing the effects of two different antihypertensive combination strategies on major fatal and nonfatal cardiovascular events, was stopped earlier because treatment with the ACE-I benazepril plus amlodipine was more effective than treatment with the ACE-I and diuretic (Jamerson et al. 2008). The trend to increased HF in patients treated with CCBs compared to ACE-I and BBs remains an important debated issue. This observation has been found in several trials and confirmed in a recent systematic review (Shibata et al. 2010): certainly, these findings should be considered in accordance with the important role of BBs and ACE-i in established HF treatment and with the superior results of CCBs-ACE-i combination in hypertensive patients. It should be also considered that, as reported as a criticism for the ALLHAT trial, the results against CCBs might be due to the design of the trials, with withdrawal of previous therapies in compensated patients (Mancia et al. 2013; Zanchetti 2012). Nevertheless, the reflex sympathetic activation stimulated by DHPs might indeed yield a negative role in HF patients (Muiesan et al. 1986).

As already recognized when interpreting the results of the ASCOT-BPLA trial, the effects of CCBs might not be entirely explained only by the better control of blood pressure. A first point to underline is that CCBs have also been shown to interact with endothelial function. The lack of functional calcium channels on endothelial cells suggests that CCBs might modify the function of mediators specifically released by the endothelium: it has been observed that some CCBs promote the release of nitric oxide (Salomone et al. 1995; 1996). Furthermore, nitric oxide and endothelin, initially identified as specific products of endothelium, are also produced by other cells: this production may be altered by some CCBs and not necessarily related to calcium channel blockade (Godfraind and Salomone 1996). Some experimental data suggest data CCBs might modify the expression of the endothelin gene (Huang et al. 1993). Moreover, it has been shown that amlodipine, nifedipine and lacidipine (two DHPs) determined a lower progression of carotid atherosclerosis than BBs. (Zanchetti et al. 2002) In the end, CCBs have no effects on potassium, glucose, uric acid and lipid metabolism: this metabolic-neutral profile and the beneficial effects on endothelial function could explain, at least in part, the lower incidence of new metabolic syndrome and diabetes observed in patients treated with CCBs in comparison to other agents (particularly BBs and diuretics). CCBs have shown also a greater effectiveness in reducing left ventricular hypertrophy (Fagard et al. 2009).

A final important point to consider is the role of CCBs in relation to progression of renal dysfunction and their use in patients with chronic kidney disease. CCBs determine no impairment of renal function. Moreover, the ALLHAT trial showed that renal function was better preserved in patients treated with amlodipine (ALLHAT Officers and Coordinators for the ALLHAT Collaborative Research Group 2002). On the other hand, the value of CCBs in patients with proteinuria and elevated creatinine is uncertain (Kloke et al. 1998), with a differential renoprotective effect between DHPs and Non-DHPs. In people with proteinuric renal disease, progressive increases in proteinuria and a more rapid decline in kidney function were noted in patients treated with DHPs in comparison to those treated with ACE-I or ARB (Lewis et al. 2001; Wright et al. 2002). In contrast, several small long-term clinical studies using Non-DHPs have demonstrated reductions in proteinuria and slower declines in glomerular filtration rate but not reduced progression towards end-stage renal disease (Bakris et al. 1996; 1998; 2004b; Smith et al. 1998). In agreement with these observations, the Kidney Disease Outcomes Quality Initiative Blood Pressure guidelines recognized that Non-DHPs alone or in combination with an ACE-I or an ARB are preferable in hypertensive patients with proteinuria of >300 mg/dl, as well as in those with impaired kidney function (Kidney Disease Outcomes Quality Initiative (K/DOQI) 2004). In general, if DHPs are used in hypertensive patients with chronic kidney disease, it is considered prudent to add CCBs after ACE-I or ARB, to achieve a balanced dilation of the afferent and efferent renal arterioles avoiding excessive glomerular pressure and flow.

In terms of main side effects, DHPs are more problematic, since peripheral edema (especially ankle edema) and flushing may be a problem for some patients (Papavassiliou et al. 2001; Weir 2003). Conversely, Non-DHPs appear better tolerated, at least when compared to old generation BBs (Fletcher et al. 1989), even though constipation may represent a relevant problem in a significant proportion of patients treated with verapamil, particularly in the elderly. Major contraindications to DHPs are represented by severe aortic stenosis or obstructive hypertrophic cardiomyopathy (to avoid excess pressure gradient). Left ventricular systolic dysfunction is a contraindication to CCBs. Non-DHPs should not be used in known sinoatrial or atrioventricular nodal disease, in almost all cases of ventricular tachycardia and in patients with pre-excitation. The use of Non-DHPs with concomitant BBs therapy is discouraged, since cardiac effects might be harmfully enhanced. On the contrary, CCBs may be useful in patients with stable and vasospastic angina, Raynaud phenomenon (to obtain peripheral vasodilation) or supraventricular tachy-arrhythmias (to control heart rate with Non-DHPs). Figure 3 synthesizes the effects of DHPs and Non-DHPs on principal factors involved in the pathogenesis, maintenance and induction of complications in essential hypertension.

RAAS plays a central role in the control of blood pressure as well as global cardiovascular and renal function and certainly its deregulation is a major pathophysiological factor involved in HT and in the pathogenesis of many other cardiovascular and renal diseases: it is clear that RAAS should be a main pharmacological target (Castro-Chaves et al. 2010). Renin controls the first and rate-limiting step of the RAAS, determining the conversion of angiotensinogen to angiotensin-I and representing an interesting point to reduce the activity of the whole system. A direct renin blocker (aliskiren) has been recently introduced (Azizi et al. 2006): despite several theoretical attractive features and initial enthusiasm, its development for clinical use has been limited and some clinical trials have been stopped for adverse events (see ensuing paragraph), underlining the need for further research. On the contrary, over the years, important results have been obtained with ACE-Is and ARBs. The angiotensin-converting enzyme (ACE) realizes the central conversion of angiotensin-I to angiotensin-II, the metabolite that activates angiotensin receptors. ACE-Is, therefore, reduce the complex and diffuse effects of angiotensin-II and RAAS, lowering arteriolar resistance and increasing venous capacity, finally increasing cardiac output. Moreover, renal vascular resistance is lowered (with specific relaxation of the efferent arterioles) and aldosterone release decreased, enhancing diuresis and natriuresis with concomitant reduction of blood volume (Sánchez et al. 1985; Atlas et al. 1979). ACE is found mainly in the vascular endothelium of the lungs, even if present in all vascular beds, and is a relatively nonspecific enzyme. Bradykinin and other tachykinins are further substrates of ACE and thus the inhibition of conversion results in accumulation of these molecules: even if they cause some of the most frequent side effects of this class of drugs, it has been suggested that decreased degradation of bradykinin might also play a role in ACE-Is induced vasodilation (Remme 1997; Su et al. 2000). In the long term competitive inhibition of ACE could result in a reactive increase in renin and angiotensin I levels, possibly reducing the effect of ACE-Is. Moreover, it should be noted that angiotensin-II is partially generated through non-ACE pathways, that are unaffected by ACE-Is (Dzau 1989). For these reasons, to overcome some problems of ACE-Is and to achieve a more intense RAAS blockade, ARBs have been developed: these molecules act selectively blocking the major angiotensin-II receptor subtype (AT-1) (Burnier and Brunner 2000). The main antihypertensive mechanisms are similar to those of ACE-Is.

Consistently with the central pathogenic role of RAAS, ACE-Is have been found to be beneficial for treating HF (The SOLVD Investigators 1991), achieving effective secondary prevention in CAD and slowing the progression toward end-stage renal disease (Hannedouche et al. 1994). ACE-Is are a main first-line option for HT treatment, particularly for younger patients, those with high-renin HT (more than a third of all hypertensive patients) and chronic kidney disease patients. ACE-I should also be preferentially used in hypertensive patients with diabetes, HF or CAD (particularly post-infarction) (Mancia et al. 2013). Meta-analysis of outcome trials in hypertensive patients have shown that ACE-Is are effective in reducing events (particularly HF and heart attacks) and mortality (Blood Pressure Lowering Treatment Trialists’ Collaboration 2007), even if apparently less protective against stroke in comparison with diuretics with or without BBs and with CCBs. The HOPE (Heart Outcomes Prevention Evaluation) trial found that the use of ramipril provided substantial additional cardiovascular protection in high-risk patients (HOPE (Heart Outcomes Prevention Evaluation) Study Investigators 2000). This result was reproduced by the EUROPA (Efficacy of perindopril in reduction of cardiovascular events among patients with stable coronary artery disease) trial in patients with known CAD, where a significant reduction in myocardial infarction was observed compared to placebo (Fox and EURopean trial On reduction of cardiac events with Perindopril in stable coronary Artery disease Investigators 2003). On the other hand, the perindopril protection against recurrent stroke study (PROGRESS), enrolling patients with a previous stroke or transient ischemic attack, found that only the subgroup receiving both perindopril and indapamide had reduced stroke recurrence: it has therefore been suggested that in that context the association of ACE-i and diuretic may be more effective in reducing blood pressure and, therefore, events (PROGRESS Collaborative Group 2001). However, the most recent ACCOMPLISH trial, as state above, showed that ACE-I plus amlodipine was more effective than treatment with the ACE-I and diuretic in reducing morbidity and mortality (Jamerson et al. 2008). It has been suggested by the results of several studies that ACE-Is have beneficial effects that are at least partially independent from the blood pressure-lowering action. They yield no effects on total cholesterol, LDL cholesterol and triglycerides, while some evidences suggest that HDL fraction is increased (Sasaki and Arakawa 1989). Moreover, improved insulin sensitivity (Paolisso et al. 1992) and reduced diabetes incidence have also been observed after ACE-I therapy (Andraws and Brown 2007). This better metabolic profile might be associated also to endothelial function improvement (De Gennaro et al. 2005) and reduced sympathetic activity induced by ACE-Is (Imai et al. 1982). Last but not least, they have been found to inhibit cardiac and vascular remodeling associated with chronic cardiovascular diseases (Abdulla et al. 2007). Figure 4 summarizes the effects of ACE-Is on the principal mechanisms and correlates of HT.

ACE-Is are generally well tolerated and improve QOL (Croog et al. 1986), provided that their administration does not cause cough, the most frequent side effect (due to bradykinin metabolites accumulation). Angioedema is the most severe, but rare, side effect (Israeli and Hall 1992). Additionally they may induce hyperkalemia and contribute to renal function deterioration, especially in patients with an already impaired renal function. Nevertheless, since early mild worsening of renal function in the setting of ACE-Is initiation appears to represent a benign event that is not associated with a loss of benefit from continued therapy, ACE-Is administration should not be discontinued and is advised also in these conditions (Testani et al. 2011). ACE-Is are contraindicated in patients with bilateral renal artery stenosis and during pregnancy.

ARBs have been tested in large outcome trials, proving their role as antihypertensive agents. A major systematic review comparing ACE-Is and ARBs reported similar blood pressure control and outcomes (Matchar et al. 2008a): even if there is evidence for better tolerability, at present no superiority of ARBs have been proven concerning hard endpoints. ACE-Is remain a preferred choice for many physicians because of the lower cost and long-term experience with their use, reserving ARBs for ACE-Is intolerant patients. ARBs are also approved for chronic kidney disease (particularly diabetic nephropathy) (Kidney Disease Outcomes Quality Initiative (K/DOQI) 2004) and as an alternative to ACE-Is in HF (McMurray et al. 2012). Moreover, the two classes of drugs share similar contraindications. Similarly to ACE-Is, ARBs have a beneficial effect on endothelial function, probably because of their antioxidant effect (Huang et al. 2007; Flammer et al. 2007), and a good metabolic profile (Kintscher et al. 2007; Vitale et al. 2005; Aksnes et al. 2007). In general, ARBs are well tolerated (Weber et al. 2003): in particular, the occurrence of cough is significantly lower (Matchar et al. 2008b). It should be noted that some data have suggested a possible association between ARBs and cancer (Sipahi et al. 2010): a subsequent large meta-analysis found no evidence of increased cancer incidence (Pfeffer 2013). Another controversy regarded the possible increase of myocardial infarction (Strauss et al. 2006), but also in this case a more comprehensive analysis of available data dismissed initial doubts (Bangalore et al. 2011). Figure 5 summarizes the effects of ARBs on the principal mechanisms and correlates of HT.