Recent evidence supports a central role for calcium channel blockers (CCBs), and in particular the dihydropyridine (DHP) agents, in the treatment of hypertension and cardiovascular disease (CVD). We review here the findings of recent trials, systematic reviews and meta-analyses that rebut those of the single, yet widely-publicised, earlier unfavourable meta-analysis. In these latest reports, CCBs emerge as effective and safe antihypertensive agents that markedly reduce the risk of cerebrovascular and cardiovascular events and also may have beneficial effects in the kidney. Thus, since most hypertensive patients need more than one agent to achieve recommended blood pressure (BP) targets, there are strong arguments often to include a CCB in the combination.
Numerous comparative studies have demonstrated CCBs to be generally at least as effective as other classes of antihypertensive agents. Nifedipine gastrointestinal therapeutic system (GITS), for example, achieved BP reductions similar to those of co-amilozide in the Intervention as a Goal in Hypertension Treatment (INSIGHT) trial in hypertensive patients with at least one additional cardiovascular risk factor. In the Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial (ALLHAT), in a somewhat similar patient population, amlodipine, lisinopril and chlorthalidone all lowered BP to a similar extent: although the 5-year systolic BP levels were 0.8mmHg higher in the amlodipine than the chlorthalidone group (p=0.03), 5-year diastolic BP was significantly lower (0.8mmHg, p<0.001). In the Anglo-Scandinavian Cardiac Outcomes Trial (ASCOT) (amlodipine, adding perindopril as required versus atenolol, adding bendroflumethiazide as required), BP values were lower throughout the trial in patients receiving the amlodipine-based rather than the atenolol-based regimen.
In the Valsartan Antihypertensive Long-term Use Evaluation (VALUE) trial, in hypertensive patients at high cardiovascular risk, amlodipine-based therapy was shown to be significantly more effective in controlling BP than that based on the angiotensin AII receptor blocker (ARB) valsartan. The amlodipine-based regimens achieved greater BP control within one month of starting therapy and, although a 24-hour ambulatory BP monitoring (ABPM) sub-study showed similar reductions in 24-hour BP after one year's treatment with either treatment regimen, night-time BP during the last hours of the dosing interval tended
Patients with a high salt intake respond well to CCBs, due to their diuretic and natriuretic properties, in contrast to those on ACE inhibitors or ARBs. Nifedipine, at a dose of 90mg (as the sustained-release GITS preparation) was as effective as 50mg of hydrochlorothiazide in a cross-over study in a group of 10 mild hypertensives treated for a period of eight weeks. Probably, most patients in VALUE were not restricting their salt intake, which may explain the efficacy of amlodipine-based versus valsartan-based treatments in that study.
Effects on Stroke
Ever since the Syst-Eur study showed a dramatic 42% reduction in stroke rate versus placebo, CCBs have been identified as efficacious antihypertensive agents in patients with cerebrovascular disease. Compared with placebo, CCBs lower the relative risk of stroke by 38-39%. This reduction is considerably greater than the 19% achieved with beta blockers, which is half that expected from previous hypertension trials. Furthermore, although stroke incidence in the control group varied considerably between the individual trials, a trend towards a reduction with CCB treatment was observed in most of them, albeit reaching significance in only one, the NORDIL (Nordic diltiazem) study, with diltiazem versus a diuretic/beta blocker. However, all DHP agents studied consistently reduced stroke risk. In the randomised, double-blind, placebo-controlled A Coronary disease Trial investigating Outcome with Nifedipine GITS (ACTION) study in which nifedipine GITS was compared with placebo, significant reductions of 28% and 27% were seen the incidence of any stroke or transient ischaemic attack (TIA) in both the total population and the hypertensive subgroup, respectively. Moreover, in the hypertensive patients, nifedipine GITS significantly reduced by 33% versus placebo the incidence of debilitating stroke (see Table 1). In contrast, verapamil, in both the Controlled ONset Verapamil INvestigation of Cardiovascular Endpoints (CONVINCE) study and Verapamil in Hypertension and Atherosclerosis Study (VHAS), was associated with an increased stroke incidence compared with the active comparator (atenolol/hydrochlorothiazide and chlorthalidone, respectively). It should be noted, however, that both CONVINCE and VHAS were inadequately powered and that the results of INternational VErapamil SR-Trandolapril Study (INVEST) (see below) demonstrated a reduction in the risk of stroke with verapamil-based versus beta-blocker- based therapy.
A meta-analysis of a number of studies including those in which CCBs were compared with conventional therapy (diuretics or beta blockers) also showed CCBs to provide better protection against fatal and non-fatal stroke than the older agents. The earlier report considered nine CCB trials for which, when combined, the odds ratio (OR) for stroke was 0.92 (95% confidence interval (CI): 0.84-1.01, p=0.07). Indeed, after excluding the only trial (CONVINCE) out of the nine to use a non-DHP CCB, the OR for stroke was 0.9 and reached significance (95% CI: 0.82-0.98, p=0.02). The later publication added Systolic Hypertension in the Elderly Long-term Lacipidine (SHELL) trial and INVEST. SHELL showed lacidipine and chlorthalidone to be similarly effective while, in INVEST, the reduction in stroke favoured the CCB-based over the beta-blocker-based strategy. Not surprisingly, therefore, Staessen and colleagues in 2005 found an even greater protective effect for CCBs, increasing the benefit over older agents from 7.6% to 8% (p=0.03). However, even this recent analysis did not include ASCOT in which the CCB-based treatment significantly reduced by 25% the proportion of patients with a first fatal or non-fatal stroke.
The Blood Pressure Lowering Treatment Trialists' Collaboration overview has revealed CCBs as not only superior to diuretics/beta blockers in preventing stroke but also at least as effective as ACE inhibitors. Moreover, the Trialists' 2003 and 2005 overviews included neither ASCOT nor INVEST nor VALUE. VALUE (amlodipine versus valsartan) showed a trend towards a reduction with the CCB-based treatment of 15% (hazard ratio (HR) 1.15 (0.98-1.35), p=0.08) in the proportion of patients suffering a first fatal or non-fatal stroke. Thus the 2005 publications from both Staessen and the Blood Pressure Lowering Treatment Trialists' Collaboration may have underestimated the superior protective effect against strokes of CCBs in comparison not only with beta blockers/diuretics but also ARBs.
Effects on Cardiovascular Events
The 2003 meta-analysis by Staessen and colleagues found no significant differences in pooled ORs between CCB and conventional treatment groups for mortality, cardiovascular death, all cardiovascular events and myocardial infarction. However, this meta-analysis included neither VALUE and ASCOT, as mentioned previously, nor ACTION and CAMELOT. The 2005 update acknowledged that, in these more recent trials, DHP CCB use was associated with a reduction in the risk of myocardial infarction (MI) versus either placebo (as in CAMELOT) or an ARB (as in VALUE).
In the VALUE trial, the amlodipine-based regimen was shown to be significantly more effective than that based on valsartan in reducing the incidence of fatal and non-fatal myocardial infarction (see Figure 1). This benefit was probably related to the better antihypertensive effect of the amlodipine-based strategy, particularly during the first few months of therapy. Furthermore, for the same level of BP control, there was no significant difference between treatments in the primary composite end-point of cardiac morbidity and mortality, or for all-cause mortality, although the incidence both of hospital admission for heart failure and of new-onset diabetes were both significantly lower in the valsartan-based group. The results therefore failed to support VALUE's primary hypothesis, i.e. that, for the same level of BP control, valsartan would be more effective than amlodipine in reducing cardiac morbidity and mortality. Both treatment strategies were well tolerated; oedema and hypokalaemia were significantly more common in the amlodipine group and dizziness, headache and diarrhoea were reported more frequently in the valsartan group. Angina was also more frequent with the ARB-based strategy.
The ACTION trial assigned patients with treated stable angina to nifedipine GITS (60mg once daily) or placebo. Follow up was 97.3% complete. There were no significant differences between groups in the frequency of the primary efficacy end-point (the combination of death, acute MI, refractory angina, new overt heart failure, debilitating stroke, and peripheral revascularisation. Furthermore, there was no difference between groups in the primary safety end-point (deaths, MI and stroke), underlining the safety of nifedipine GITS. There was also, in ACTION, an 11% reduction in the combined rate of death, major cardiovascular events, revascularisation and coronary angiography (a predefined secondary end-point; p=0.001) (see Figure 3), and a significant 29% reduction in the frequency of new overt heart failure (p=0.02). ACTION therefore demonstrates that, in patients with symptomatic angina and stable CAD, nifedipine GITS is safe and reduces the need for coronary interventions and the incidence of heart failure.
The CAMELOT study compared the effects of amlodipine or enalapril versus placebo on cardiovascular events in patients with angiographically documented CAD and normal BP. Both active treatments lowered BP to the same extent. Nevertheless, time to a major cardiovascular event was significantly longer with amlodipine than with either enalapril or placebo and, compared with placebo, amlodipine reduced adverse cardiovascular events by 31%, coronary revascularisation by 27% and hospitalisation for angina by 42%.
Although the more recent meta-analysis by Staessen et al. would appear to suggest that nifedipine GITS does not share the protective effect of amlodipine against MI, it should be emphasised that in VALUE, patients were treated to a target BP whereas those in CAMELOT were normotensive; ACTION was designed to resolve whether nifedipine GITS favourably affects the long-term clinical outcome of patients with stable angina, and a target BP level was not defined. This would account for the apparent superiority of amlodipine over nifedipine GITS if, as noted by Staessen et al., BP reduction alone is sufficient to account for all the observed cardiovascular benefits of antihypertensive agents.
The Staessen group's 2005 update did not include ASCOT, presumably because it was published too late to be considered. The primary objectives of the blood-pressure-lowering arm (BPLA) of this multicentre, prospective trial were to compare the effect on non-fatal MI and fatal coronary artery disease (CAD) of the standard antihypertensive regimen (beta blocker (atenolol) adding a diuretic (bendroflumethiazide) as required) with a more contemporary regimen (a CCB (amlodipine), adding an ACE inhibitor (perindopril) as required). Eligible patients had hypertension and at least three other cardiovascular risk factors. The study was stopped prematurely after 5.5 years' median follow-up because, compared with the atenolol-based regimen, the CCB/ACE inhibitor resulted in fewer non-fatal MI and fatal CAD (10%, not significant); lower all-cause mortality (15%, p≤0.005) (see Figure 2); fewer coronary events (15%, p≤0.005), all cardiovascular events and procedures (15%, p≤0.001), cardiovascular mortality (24%, p≤0.005), and new-onset diabetes (30%, p≤0.001); and the previously mentioned reduction in fatal and non-fatal stroke (25%, p≤0.001). These findings demonstrate that the amlodipine-based therapy conferred an advantage over that based on atenolol on all major cardiovascular end-points, all-cause mortality and new-onset diabetes. Indeed, ASCOT is the only hypertension intervention trial in which a significant difference in total mortality has been shown between active treatments, in this case favouring the CCB-based regimen.
The cardiovascular benefits of DHP CCBs may be at least partly attributable to improvements in endothelial function, as demonstrated by the Evaluation of Nifedipine and Cerivastatin On Recovery of coronary Endothelial function (ENCORE) trial. In this study, in 343 patients undergoing percutaneous coronary interventions, six months' nifedipine treatment improved coronary endothelial function in the most constricted segment. The effect was particularly marked in patients taking concurrent ACE inhibitors.
Effects on the Kidney
In ALLHAT, the rate of decline in kidney function (as measured by the slopes of the reciprocal of serum creatinine over time) was significantly less in the amlodipine-based group than in the group receiving chlorthalidone-based treatment (p<0.001), although the incidence of end-stage renal disease was similar. ASCOT demonstrated a significant reduction with the amlodipine-based regimen in new-onset renal impairment that was evident after two years; the difference between treatment groups continued to increase over the rest of the study. This continuing and increasing benefit over time suggests a role for protective mechanisms that are not directly related to the BP reduction. There is evidence, moreover, that dihydropyridine CCBs have beneficial effects on endothelial function and fibrosis in the kidney, possibly extending to the tubules, microcirculation and mesangial podocytes.
CCBs in Combination Therapy
CCBs are highly effective antihypertensive agents. However, used as monotherapy and as demonstrated by both the VALUE and INSIGHT studies, CCBs are insufficient to attain target BP in 50-70% of patients, hence the need for combination therapy.
The addition of a thiazide diuretic to on-going DHP CCB-based therapy proved effective and safe in both VALUE and ALLHAT. A large minority of patients in VALUE also were successfully treated with amlodipine combined with beta blockade. In addition, the findings of the Nifedipine and Candesartan Combination (NICE Combi) study support the use of a CCB/ARB combination. In NICE Combi, low-dose combination therapy with controlled-release nifedipine and candesartan was demonstrated to provide better control of BP and renal protection than up-titrated candesartan monotherapy. CCBs were also used as add-on therapy in most of the landmark studies of ARBs in diabetic nephropathy. Combinations of CCBs and ACE inhibitors capitalise on the strengths of both agents while reducing the incidence of class-specific side effects, particularly oedema, and are supported by data from INVEST, ASCOT and ENCORE I. Thus, although many clinicians currently choose an ACE inhibitor or an ARB first and then add a diuretic or a CCB (or vice versa), there is an equally good argument for starting therapy with a CCB/ACE inhibitor or CCB/ARB combination.
A number of recent studies and meta-analyses have consistently demonstrated CCBs to be effective and safe antihypertensive agents that reduce cardiovascular morbidity and mortality. DHP CCBs significantly reduce the risk of stroke and, in patients with angina, reduced the risk of heart failure and the need for coronary angiography. These findings therefore allay earlier concerns about the use of these agents as first-line therapy for hypertension. Indeed, the efficacy and safety of DHP CCBs argue for their inclusion in any combination therapy used to treat hypertensive patients.