Optimal%20management%20 hta%202011%20ehj

REVIEW

Novel therapeutic concepts

Hypertension management 2011: optimalcombination therapyPeter S. Sever1* and Franz H. Messerli 2

1International Centre for Circulatory Health, Imperial College London, 59 North Wharf Road, London W2 1LA, UK; and 2Division of Cardiology, St Luke’s and Roosevelt Hospitals,Columbia University College of Physicians and Surgeons, New York, NY, USA

Received 11 August 2010; revised 16 March 2011; accepted 13 May 2011; online publish-ahead-of-print 22 June 2011

Raised levels of blood pressure result from the complex interplay of environmental and genetic factors. The complexity of blood pressurecontrol mechanisms has major implications for individual responsiveness to antihypertensive drugs. The underlying haemodynamic disorderin the majority of cases is a rise in peripheral vascular resistance. This observation led to the discovery and development of increasinglysophisticated and targeted vasodilators, although many of the earlier antihypertensive drugs, by virtue of their actions blocking the sympath-etic nervous system, had a vasodilator component to their mode of action. A recent meta-analysis of placebo controlled trials of monother-apy in unselected hypertensives, reports average (placebo-corrected) blood pressure responses to single agents of 9.1 mmHg systolic and5.5 mmHg diastolic pressure. These average values disguise the extremely wide ranging responses in individuals across a fall of 20–30 mmHgsystolic at one extreme, to no effect at all, or even a small rise in blood pressure at the other. The second factor determining individualresponses to monotherapy is the extent to which initial falls in pressure are opposed by reflex responses in counter regulatory mechanismsthat are activated following the blood pressure reduction. Thus, a satisfactory blood pressure response is rarely reached with monotherapyalone. What then is the next step if blood pressure is not a goal after the patient has been treated with monotherapy for a few weeks? Shouldyou uptitrate, substitute or combine?- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -Keywords Hypertension † Combination therapy

IntroductionRaised levels of blood pressure result from the complex interplayof environmental and genetic factors leading to the activation orsuppression of one or more of a host of physiological systemsinvolved in blood pressure regulation (Figure 1). The complexityof blood pressure control mechanisms, first hypothesized byIrvine Page,1 has major implications for individual responsivenessto antihypertensive drugs (Figure 2), because of the inevitablevariety of hypertensive phenotypes, the identification of which,with some notable exceptions, remains elusive to the practicingphysician involved in making treatment decisions for individualpatients.2

Hypertension is, by definition, a haemodynamic disorder. Themajor haemodynamic finding associated with higher levels ofblood pressure is a rise in peripheral vascular resistance. Thisobservation led to the discovery and development of increasinglysophisticated and targeted vasodilators, although many of the

earlier antihypertensive drugs, by virtue of their actions blockingthe sympathetic nervous system, had a vasodilator component totheir mode of action. The first non-specific vasodilator, hydrala-zine, was followed by vasodilatation which involved blockadeof calcium channels on vascular smooth muscle cells [thecalcium channel blockers (CCBs)], blockade of post-synapticalpha-adrenoceptors on peripheral sympathetic neurones (thealpha blockers) and, finally, vasodilatation achieved byblockade of the renin–angiotensin–aldosterone system (RAAS)[angiotensin-converting enzyme (ACE) inhibitors, angiotensinreceptor blockers (ARBs), direct renin inhibitors (DRIs)] (Figure 3).

The nature of these molecules, and in most cases their single siteof action, dictates that when administered to a heterogeneouspopulation, encompassing many hypertensive phenotypes, bloodpressure responses will be largely unpredictable and wideranging (Figure 4). If, in a particular case, blood pressure levelsare largely determined by activation of the RAAS, for example inrenal artery stenosis, marked falls in blood pressure with

* Corresponding author. Tel: +44 207 594 1100, Fax: +44 207 594 1145, Email: [email protected]

Published on behalf of the European Society of Cardiology. All rights reserved. & The Author 2011. For permissions please email: [email protected]

European Heart Journal (2011) 32, 2499–2506doi:10.1093/eurheartj/ehr177

by guest on March 7, 2012

http://eurheartj.oxfordjournals.org/D

ownloaded from

mailto:[email protected]




http://eurheartj.oxfordjournals.org/

impairment of renal function may follow the administration of anACE-Inhibitor.3 On the other hand, in the elderly and in thoseof African origins, where the activity of the RAAS is generally sup-pressed, blood pressure reductions4,5 with an ACE-Inhibitor maybe small. In general, however, the phenotype is not known.

A recent meta-analysis of placebo-controlled trials of monother-apy, in unselected hypertensives,6 reports average (placebo cor-rected) blood pressure responses to single agents of 9.1 mmHg

systolic and 5.5 mmHg diastolic pressure. These average values dis-guise the extremely wide ranging responses in individuals across afall of 20–30 mmHg systolic at one extreme, to no effect at all, oreven a small rise in blood pressure at the other7 (Figure 4).

The second factor determining individual responses to mono-therapy is the extent to which initial falls in pressure areopposed by reflex responses in counter regulatory mechanismsthat are activated following the blood pressure reduction. In

Figure 2 A modified and updated mosaic theory of blood pressure regulation derived from the original Paige mosaic.1

Figure 1 A schematic to demonstrate the interaction of environmental factors with underlying genetic predisposing factors to increase bloodpressure through the activation of a variety of pathogenetic mechanisms.

P.S. Sever and F.H. Messerli2500



ownloaded from


extreme cases, these reflex responses can nullify any fall inpressure (Figure 5).

Thus, a satisfactory blood pressure response is rarely reachedwith monotherapy alone. What then is the next step if bloodpressure is not at goal after the patient has been treated withmonotherapy for a few weeks? Should you uptitrate, substitute,or combine?

UptitrationUptitration of the initial drug is reasonable only if definitive,enhanced antihypertensive efficacy of the higher dose has beendocumented and the cost is not prohibitive. Regrettably, most

antihypertensive drugs have a rather shallow dose–responsecurve. In particular, with RAAS inhibitors doubling the dose hasminimal incremental effect on blood pressure. In contrast, withCCBs, additional antihypertensive efficacy can be gained when, forexample, the starting dose of amlodipine is doubled from 5 to10 mg. However, the incidence of pedal oedema also is dose depen-dent and increases with a higher dose of amlodipine. Importantly, theadditional blood pressure fall from combining drugs from two differ-ent classes is �5 times greater than the one from doubling the doseof a single drug.8 Thus, the odds of getting blood pressure to goal areseveral times greater with combining drugs than with up titration ofmonotherapy. From a sheer efficacy point of view, combinationtherefore takes precedence over uptitration.

Figure 3 The history of development of antihypertensive drugs reproduced with kind permission of Thomas Unger.

Figure 4 The frequency distribution of changes in diastolicblood pressure produced by three different antihypertensivedrugs. Negative values represent placebo-corrected reductionsin diastolic pressure. Modified from reference.7

Figure 5 Response to the vasodilator hydralazine followed bythe co-administration of the beta-blocker, atenolol.

To uptitrate, to substitute, or to combine drugs 2501



ownloaded from


SubstitutionSubstituting an antihypertensive drug from a different class should beconsidered only if there is no antihypertensive effect with a reason-able dose, as is occasionally observed with beta-blockers or RAASblockers in black patients, or if there are any intolerable adverseeffects such as angioedema. Fortunately, most modern antihyperten-sive drugs are generally well tolerated and serious adverse effects arefew. However, before resorting to drug substitution one may con-sider that the addition of another drug may unmask the antihyperten-sive efficacy of the initial agent. For instance, the addition of a thiazidediuretic in a patient previously unresponsive to RAAS blockade isprone to stimulate the renin–angiotensin system to the extentthat now both drugs, the RAAS inhibitor as well as the diuretic,have an additive antihypertensive effect.9

Rationale for combination therapyThe rationale for combination therapy in hypertension is thereforestraightforward. First, it is to combine drugs acting on differentphysiological systems in a situation where the phenotype is notknown and where a pharmacological ‘attack’ on two (or more)systems will have a greater impact on blood pressure reductionthan blind monotherapy. Second, it is an attempt to block counter-regulatory responses that are activated by the perturbation of theblood pressure regulatory mechanisms when a physiologicalsystem is blocked with single-drug therapy (Figure 6).

Third, the hypertensive population includes many with levels ofblood pressure categorized as moderate or severe (stage 2 hyperten-sion).10 There is general consensus that those with systolic bloodpressures .160 mmHg and/or diastolic pressures .100 mmHg fallinto this category. They constitute 10–15% of hypertensive popu-lations and are at substantially greater risk of a future cardiovascularevent. For every 20 mmHg increase in systolic blood pressure,there is an approximate doubling of cardiovascular risk.11

Obviously the proportion of the population with hypertensionincreases with age and this also applies to those with stage 2 hyperten-sion. As age advances systolic hypertension predominates and islargely accounted for by loss of elasticity and increasing rigidity oflarge arteries.

Although there are some differences between guidelines, severalnow recommend the initiation of combination therapy as first linein particular circumstances, in view of the associated risks of moresevere hypertension, the recognition that dual (or triple therapy) isinvariably needed to achieve target blood pressures of ,140/90 mmHg, and that there is a degree of urgency in reducingblood pressure to more acceptable levels to combat this risk.

JNC-7 recommends initiating therapy with two drugs whenblood pressure is .20 mmHg above systolic goal or 10 mmHgabove diastolic goal.10 The European Guidelines,12 including theirmost recent update,13 confirm such a recommendation and alsoproposes the initiation of combination therapy in those withmilder degrees of blood pressure elevation in the presence of mul-tiple risk factors, subclinical organ damage, diabetes, renal, orassociated cardiovascular disease. Although combination therapyis not specifically advocated as initial therapy in the 2004 BritishHypertension Society Guidelines14 (largely based on the fact thatthere is a lack of randomized controlled trial evidence tosupport such practice), it is probable that the results of ongoingtrials will provide new evidence in favour of their early introduc-tion into treatment strategies.

Inevitably, there are concerns that initiating therapy with morethan one drug could induce significant hypotension and increasecoronary risk. An analysis of intervention trials in hypertension15,16

provides some evidence for a ‘J-curve’ relationship between themagnitude of blood pressure lowering and coronary heartdisease outcome, but this seems to be confined to high-risk indi-viduals including those with established coronary artery disease,in whom excessive blood pressure lowering compromises coron-ary perfusion. In uncomplicated hypertension, lower pressures arewell tolerated, for example, as seen in the Systolic Hypertension inthe Elderly Study, in which diastolic pressures as low as 60 mmHgwere achieved in the active treatment group.17 Ongoing trials com-paring initiation of dual therapy vs. sequential monotherapy inhypertension will aim to clarify the safety of the former.

Fourth, blood pressure variability has been shown to decreasewith combination therapy18 when compared with monotherapy.In an extensive analysis of several randomized trials, visit-to-visitvariability of systolic blood pressure was documented to be astrong predictor of both stroke and myocardial infarction andthis was independent of mean in-trial blood pressure.18 Interest-ingly enough, CCBs and diuretics were most efficacious in reducingvisit-to-visit blood pressure variability and also were associatedwith the most efficacious stroke prevention.19 In contrast, beta-blockers were shown to increase variability of systolic pressurein a dose-dependent way and also were the least efficacious instroke prevention. The addition of a CCB or to a lesser extentof a diuretic to a RAAS inhibitor diminishes variability of systolicpressure, which makes another strong argument for combinationtherapy.

Trial evidence for and againstspecific combinationsAn extensive review of first-line drug choices has been publishedby the Blood Pressure Lowering Treatment Trialists’

Figure 6 Renin–angiotensin–aldosterone system and sym-pathetic nervous system activation and suppression by differentclasses of antihypertensive drugs.




ownloaded from


Collaboration20,21 and is based upon prospective meta-analyses oftrials comparing different drug regimens. Similar analyses have beenundertaken by the National Institute for Clinical Excellence (NICE)in the UK.22 The difficulty in extending these analyses to evaluatethe comparative effects of different combinations of drugs is that inmany trials it is not possible to establish which add-on drugs wereused and in what doses. The evidence base for making claims aboutthe comparable or superior efficacy of one regimen vs. anothercomes from trials where the treatment algorithm was clearlydefined and one could conclude with reasonable assurednessthat a particular regimen was similar to, better than or worsethan another. The best evidence, from which claims can bemade of outcomes in favour of a particular regimen, comes fromfour trials, the Losartan Intervention For Event Reduction Trial(the LIFE Trial,23 the Anglo-Scandinavian Cardiac Outcomes Trial(ASCOT).24 The Avoiding Cardiovascular Events through Combi-nation Therapy in Patients Living with Systolic Hypertension Trial(ACCOMPLISH)25 and the Valsartan Antihypertensive Long-termUse Evaluation Trial (VALUE).26

In the LIFE Trial,23 9193 hypertensive patients were random-ized to initial treatment with either an ARB (losartan) or a beta-blocker (atenolol). Hydrochlorothiazide was added in themajority of patients to achieve blood pressure control, alongwith the further addition of common third-line agents in a min-ority of patients. After an average follow-up of 5 years duringwhich there was no discernable difference in blood pressurebetween the two regimens, the composite primary cardiovascu-lar endpoint was reduced by 13% in the losartan-based groupcompared with the atenolol-based group. The major benefitwas seen in the secondary stroke endpoint (a component ofthe primary) which was reduced by 25% in the losartan-basedgroup.

In the second trial, ASCOT,24 over 19 000 hypertensive patientswith no prior history of coronary heart disease were randomizedto either a CCB, amlodipine, or a beta-blocker, atenolol. TheACE-Inhibitor perindopril or the diuretic bendroflumethiazidewas added to each arm, respectively, in an attempt to achieveblood pressure targets. Again, common third-line drugs could beadded to each arm in a minority of patients. After an averagefollow-up of 5.5 years, the trial was stopped prematurely on theadvice of the Data Safety Monitoring Committee, because ofhighly significant outcome benefits in favour of the amlodipine-based regimen. All cardiovascular events were reduced by 26%,stroke by 23%, and all-cause mortality by 11% by the amlodipine-based regimen compared with the atenolol-based regimen. Theprimary endpoint of non-fatal myocardial infarction and fatal cor-onary disease was reduced non-significantly by 10% in favour ofthe amlodipine-based regimen, best explained by the early termin-ation of the trial before the required number of primary endpointshad been reached. In the event, a more comprehensive coronaryendpoint which included coronary revascularizations wasreduced significantly by 13%.

In several subsequent analyses, the small blood pressure differ-ences observed early in the trial did not explain the outcomebenefits in favour of amlodipine-based treatment.27 In recentreports, however, it has been shown that additional haemo-dynamic measurements may be better determinants of

cardiovascular outcome, and differentially affected by differenttreatment strategies. For example, in the CAFE substudy ofASCOT, the amlodipine/perindopril regimen lowered centralaortic blood pressure to a greater extent than the atenolol/thia-zide regimen (by 4 mmHg systolic), and the level of centralpressure was related to cardiovascular and renal outcomes28. Inanother substudy, various measures of blood pressure variabilityduring the trial were strongly associated with both stroke andcoronary outcomes, in that the amlodipine-based treatmentregimen reduced blood pressure variability compared with theatenolol-based regimen. These differences largely accounted forthe observed differences in cardiovascular outcomes betweenthe two-drug regimens.29

In the third trial, ACCOMPLISH,25 11 506 hypertensive patientswere randomized to a combination of the ACE-Inhibitor, benaze-pril, with either hydrochlorothiazide, or the CCB, amlodipine.Patients were followed for 3 years. Blood pressure levels werereduced similarly in the two arms of the trial. Cardiovascularevents were significantly reduced by 20% in benazepril/amlodipinearm compared with the benazepril/hydrochlorothiazide arm.Myocardial infarction was reduced significantly (22%) and strokenon-significantly (16%) by benazepril/amlodipine compared withbenazepril/hydrochlorothiazide. The benefits of the benazepril/amlodipine combination over benazepril/hydrochlorthiazide wereseen in both diabetic and non-diabetic patients.30

In the fourth study, VALUE,26 15 245 hypertensive patientswere randomized to either the ARB, valsartan, or the CCB,amlodipine. Hydrochlorothiazide was added to each limb inattempting to achieve goal blood pressures. Other add-ondrugs were similar in the two treatment arms. Mean follow-upwas 4.2 years. Blood pressures were more effectively and morerapidly reduced in the amlodipine-based treatment arm. Althoughthe primary composite endpoint of cardiac morbidity and mor-tality was similar in the two arms of the trial, myocardial infarc-tion occurred significantly less frequently (risk reduction 19%)and strokes non-significantly less often (risk reduction 15%) inthe amlodipine-based treatment arm compared with thevalsartan-based arm. The authors of the trial attributed earlydifferences in blood pressure as an explanation for the differentialeffects of the two treatments on myocardial infarction andstroke.

The cumulative evidence from these trials strongly supports theview that, in hypertensive patients, combination therapy with CCB/ACE-I or CCB/ARB is likely to be associated with better cardiovas-cular outcomes, including myocardial infarction and stroke, thanregimens containing beta-blockers and thiazide diuretics and thatCCB/ACE-I combinations are preferable to diuretic/ACE-I combi-nations on major cardiovascular endpoints. Added to this shouldbe the cost-effectiveness analysis from the NICE Guidelineswhich clearly demonstrates that CCBs and ACE-Is or ARBs aremore cost-effective treatment choices than beta-blockers orthiazide diuretics.22

The above recommendations apply, in general, to those subjectswith uncomplicated hypertension. In hypertensives with associatedcardiovascular disease such as heart failure or coronary heartdisease, the guidelines are consistent in recommending specificdrugs with compelling indications, based on randomized controlled




ownloaded from


trial evidence, that should be incorporated into treatmentstrategies.

Specific drug combinationsGiven that there are seven major classes of antihypertensive drugsand numerous members of each class, the number of possiblecombinations is extensive. In the following, we subdivide combi-nations as preferred, acceptable or unacceptable/ineffective combi-nations, based on outcome, antihypertensive efficacy, safety, and/ortolerability.

Preferred combinationsRenin–angiotensin–aldosterone system inhibitorsand calcium channel blockersAdditive blood pressure reduction has been documented with thecombination of an ACE-Inhibitor, ARB, or DRI with a CCB. Thecommon dose-dependent adverse effect of CCB monotherapy isperipheral oedema. The addition of a RAAS blocker has beenshown to mitigate this adverse effect. A recent meta-analysis hasshown that ACE-Inhibitors are somewhat more efficacious thanARBs in decreasing peripheral oedema associated with CCBtherapy.31 As stated above, the ACCOMPLISH trial showed thatfixed combination of an ACE-Inhibitor (benazapril) with a CCB(amlodipine) was more beneficial with regard to morbidity andmortality reduction than the fixed combination of the sameACE-Inhibitor with hydrochlorthiazide.25 Generally, similar end-point reductions have been demonstrated with ACE-Inhibitorsand ARBs, although there is a suggestion that ACE-Inhibitorsmay be slightly more cardioprotective and that ARBs may confersome advantages in stroke prevention.32

The International Verapamil-Trandolapril Study (INVEST)33 wasa comparison of ‘new’ vs. ‘old’ drugs in that a regimen of the non-dihydropyridine, verapamil, to which trandolapril was added ifnecessary, was compared with atenolol to which hydrochlorthia-zide was added if necessary to achieve blood pressure goals. Atotal of 22 576 hypertensives with established coronary arterydisease were enrolled and followed up for a mean of 2.7 years.The combined cardiovascular outcome was similar in the twogroups. Perhaps the most logical explanation for these findings isthat the disadvantage of the beta-blocker regimen observed inhypertension trials in uncomplicated patients was offset by theknown advantages of beta blockade in the context of establishedcoronary artery disease.

Renin–angiotensin–aldosterone system inhibitorsand diureticsNumerous factorial design studies have shown that the combi-nation of a thiazide diuretic with an ACE-Inhibitor, an ARB, or aDRI result in fully additive blood pressure reduction. Diuretics,by depleting intravascular volume, activate the RAAS whichcauses salt and water retention as well as vasoconstriction. Theaddition of a RAAS blocker attenuates this counter regulatoryresponse. Moreover, diuretic induced hypokalaemia as well asglucose intolerance is mitigated by the addition of a RAASblocker. Chlorthalidone has been shown to be more effectivethan hydrochlorthiazide in reducing blood pressure and should

therefore be the preferred agent to be combined with a RAASblocker. Unfortunately most RAAS inhibitors are available only ina fixed-dose combination (FDC) with hydrochlorthiazide.

In a recently reported study in a very elderly (.80 years) hyper-tensive population, the Hypertension in the Very Elderly Study(HYVET),34 a thiazide-like diuretic, indapamide, to which anACE-Inhibitor, perindopril, was added, was found to reducestroke incidence (30%) and the incidence of heart failure (64%),compared with placebo.

Acceptable combinationsBeta-blockers and diureticsThe addition of diuretics has been shown to improve the antihy-pertensive efficacy of beta-blockers in African-American patientsand other populations with low-renin hypertension. However,both of these drug classes have been shown to have similaradverse effects in that they increase the risk of glucose intolerance,the development of new-onset diabetes,22 fatigue, and sexual dys-function. Outcome studies have shown a morbidity and mortalityreduction with diuretics and beta-blockers in combination.22

Calcium channel blockers and diureticsMost physicians are somewhat reluctant to combine a CCB with adiuretic. However, in the VALUE trial,26 hydrochlorthiazide wasadded as a second step in patients randomized to amlodipineand the diuretic/CCB combination was well tolerated, althoughthere was a higher risk of new onset diabetes and hyperkalaemiawhen compared with the valsartan arm. Nevertheless, morbidityand mortality reductions were at least as good in the amlodipineas in the valsartan arm of the VALUE study.

Calcium channel blockers and beta-blockersThe combination of a beta-blocker with a dihydropyridine CCBhas additive blood pressure reduction and, in general, is well toler-ated. In contrast, beta-blockers should not be combined with non-dihydropyridine calcium blockers such as verapamil or diltiazem.The negative chronotropic effect of both of these drugs mayresult in heart block or bradycardia.

Dual calcium channel blockadeThe combination of a dihydropyridine CCB with either verapamilor diltiazem has been shown in a recent meta-analysis 35 to havean additive effect on blood pressure lowering without significantlyincreasing adverse events. Dual CCB blockade may be useful inpatients with documented angioedema on RAAS inhibitors or inpatients with advanced renal failure at risk for hyperkalaemia.However, no outcome data are available with dual CCB therapyand long-term safety remains undocumented.

Unacceptable/ineffective combinationsDual renin–angiotensin–aldosterone system blockadeFor the treatment of hypertension per se, dual RAAS blockade, ingeneral, is not recommended.36 In the ONTARGET study,37 therewere more adverse events with a combination of telmisartan andramipril than with individual agents and cardiovascular endpoints,despite a small additional blood pressure reduction, were notimproved compared with monotherapy. Thus, there is little if any




ownloaded from


reason to combine an ARB with an ACE-Inhibitor for the treatmentof hypertension. However, as blockade of the renin–angiotensincascade by either an ACE-Inhibitor or an ARB increases plasmarenin activity, the argument has been put forward that the additionof a DRI could have additional benefits. Indeed, the combination ofaliskiren with an ARB has been shown to have a small, significantadditional effect on blood pressure in a double-blind study of 1797patients.38 However, this fall in blood pressure with dual RAASblockade was less than one would have expected by the additionof either a thiazide diuretic or a CCB. Of note, in an open label pro-spective crossover study in patients with resistant hypertension, thealdosterone antagonist spironolactone was shown to lower bloodpressure more effectively than conventional dual RAAS blockade.39

At the present time, no outcome data are available to supportbenefits of the combination of a DRI with either an ACE-Inhibitoror an ARB. Nevertheless, a randomized double-blind trial (ALTI-TUDE) has been designed to look into this question and is currentlyin progress.

Renin–angiotensin–aldosterone system blockersand beta-blockersIn patients having suffered a myocardial infarction or in those inheart failure, these two drug classes are commonly combinedbecause both have been shown to reduce reinfarction rates andto improve survival. However, their combination produces littleadditional blood pressure reduction compared with either mono-therapy. Thus, for the treatment of blood pressure per se, there isno reason to combine these two drug classes.

Beta-blockers and antiadrenergic drugsLittle if any antihypertensive efficacy can be gained when beta-blockers are combined with antiadrenergic drugs such as clonidine.In fact, an exaggerated rebound in BP has been observed with thiscombination.40

Other drug classes in combination therapy: alpha-blockersand spironolactoneAlpha-adrenoceptor antagonists have been widely used as add-ondrugs in combination regimens to achieve target blood pressures.The availability of extended release formulations has improvedtheir tolerability profile. Data from an observational analysis ofthe ASCOT trial showed that doxazosin gastrointestinal thera-peutic system (GITS) used as third-line therapy lowered bloodpressure and caused a modest reduction in serum lipids.41 In con-trast to earlier findings in ALLHAT,42 doxazosin use in ASCOT wasnot associated with an increased incidence of heart failure.

For subjects with resistant hypertension, defined as failureto achieve target blood pressure (,140/90 mmHg) despitemaximum doses or maximum tolerated doses of three antihyper-tensive drugs including a RAAS blocker, a CCB, and a thiazidediuretic, quadruple therapy is frequently required. Recent reportsdemonstrate that spironolactone added to triple therapy is associ-ated with substantial further reductions in blood pressure of,on average, 22/9.5 mmHg.43 Spironolactone is therefore rec-ommended as a component of combination therapy in patientswith resistant hypertension.

Adverse effectsThe adverse reactions associated with combination treatments arelargely predicted from the known side effects of the individualcomponents. However, in older combinations of vasodilators(hydralazine) with beta-blockers and diuretics, the side effects ofvasodilatation (tachycardia and fluid retention) were mitigated bythe additional drugs. There is some evidence that the oedemacommonly associated with dihydropyridine CCBs is partiallyrelieved by co-administration of RAAS blockers44,45 and RAASblockers may reduce the incidence of hypokalaemia induced bythiazides.46 On the other hand, it seems likely that the increasein incidence of new-onset diabetes commonly associated withbeta-blockers is exacerbated when these drugs are given in con-junction with thiazide diuretics. A meta-analysis of the increasedincidence of new-onset diabetes with beta-blocker and thiazidetreatment, compared with ‘newer’ drugs, is provided by theNICE Guidelines.22

These conclusions assume that there are no differencesbetween individual drugs within a particular drug class in relationto their effects on long-term morbidity and mortality. Amongthe CCBs, the best evidence is for amlodipine. Among theACE-Is and ARBs, several different drugs have been used bothwithin and without combination trials in hypertensive patientsand in other cardiovascular patient groups, and no clear benefitsof one drug over another are evident. For thiazide and thiazide-likediuretics, there persists an opinion that the evidence base for long-term benefits is best for moderate doses of chlorthalidone,17,42,47

compared with other thiazides in lower doses. Regrettably, thereare unlikely to be future trials comparing drugs within this class.

For the beta-blockers, atenolol has been the drug most oftenused and claims have been made that had other drugs in thisclass been used in the trials then perhaps different results wouldhave occurred.48 This is unlikely since the adverse effects of ateno-lol, observed in ASCOT, on blood pressure variability,29 and anincrease in central aortic pressures compared with amlodipine 28

(both of which were associated with an increase in cardiovascularrisk), would be likely to occur with most other beta-blockers.

Outcome trials in hypertension with beta-blockers possessingadditional pharmacological properties have not been conducted.

Fixed-dose combinations andoutcome benefitsIn a recent review of the potential advantages of FDC formulationsover their corresponding free drug components given separately, itwas shown that the FDCs were associated with significantly bettercompliance and a non-significant improvement in persistence withtreatment.49 Similarly, in a meta-analysis of nine studies comparingthe administration of FDCs with their separate components, theadherence rate was improved by 26% in patients receiving FDCs.50

In trials in which blood pressure data were reported, use ofFDCs was associated with a non-significant lowering of systolicand diastolic pressure (4.1 and 3.1 mmHg, respectively) comparedwith the corresponding drugs administered separately.49 Such




ownloaded from


differences in blood pressure if sustained long term wouldundoubtedly confer advantages on cardiovascular outcomes.

Blood pressure control in practiceWorldwide surveys of blood pressure control to targets rec-ommended by national and international guidelines have consist-ently revealed that in clinical practice the conventional goal of ablood pressure ,140/90 mmHg is reached by only a minority ofpatients.51 Data from several countries are shown in Figure 5.While there are several explanations for physicians failing toachieve target blood pressures, including poor compliance or con-cordance with drug taking by patients, white coat hypertension,undiagnosed secondary causes of hypertension, and true resistanthypertension, in the majority of cases therapeutic inertia on thepart of the physician plays a major role. There is good evidencethat when physicians are faced with patients on treatment forhypertension, but who have not reached goal blood pressures,they are reluctant to increase drug doses or initiate second- andthird-line combination therapy.52

The issues surrounding these observations are complex. Clearlylack of education and failure to appreciate the importance oflowering blood pressure to targets to prevent cardiovascularoutcomes associated with uncontrolled blood pressure are impor-tant issues. The historical focus on diastolic pressure as the basisfor initiation of therapy and as a treatment target is another. Inpractice, diastolic targets of ,90 mmHg are far more commonlyattained than systolic targets of ,160 mmHg.53

Lastly, and importantly, true therapeutic inertia—the reluctanceto change medications when faced with a patient whose bloodpressures remain above goals. Excuses such as the followingexample—‘It’s a little bit higher today (cold weather, rush toclinic, stress at work, domestic problems etc) but we will seewhat it’s like in a few weeks/months time’ are all too frequent.This major problem can be overcome (as we observe in trials)when physicians or nurses are obliged to follow goal directedtreatment algorithms dictated by a trial protocol, and when‘excuses’ cannot be made to avoid changes in medications whenblood pressures are not at target.

An alternative scheme, practised in the UK since 2004, has beento remunerate doctors based on the extent to which they achievea number of clinical targets, one of which is dictated by the pro-portion of their hypertensive patients whose blood pressures arelowered to an audit standard of ,150/90 mmHg. This has contrib-uted to improvements in the levels of blood pressure control inthe population and has been accompanied by the increasing useof combination therapies.54

ConclusionsThe use of combinations of drugs in therapeutic practice iscommon place in contemporary medicine in a wide variety ofdisease categories, for example, in infectious disease, to cover mul-tiple organisms and to overcome drug resistance; in respiratoryillness such as chronic bronchitis or asthma to target multiplepathophysiological mechanisms of disease and in neurological con-ditions to interfere with different abnormalities of neurotransmit-ter function. In fact throughout medicine, combination therapy is

often the norm rather than the exception. In hypertension, theunderlying rationale for combination therapy is somewhat differ-ent. Since we do not know the cause of the blood pressureelevation, therapy is essentially blind and a shotgun approachmay be more efficacious than targeted therapy. This is particularlytrue because monotherapy invariably triggers a variety of counterregulatory mechanisms which are mitigated by combinationtherapy. Thus, a strong case can be made for the early introductionof combination therapy and conceivably, the time will come whencombination therapy in low doses will be the preferred option forfirst-line treatment in patients with hypertension.

Take home message andrecommendations(1) Many, if not most patients, need two or more drugs from

different classes to achieve blood pressure control.(2) Combination therapy should be initiated if the patient’s blood

pressure is .20/10 mmHg above target level unless cardiovas-cular status is brittle.

(3) Preferred or acceptable two drug combinations should beused (Table 1).

(4) Whenever convenience and cost outweigh other consider-ations fixed-dose combinations rather than individual drugsshould be used.

Conflict of interest: P.S.S. has received grant income and honor-aria from Pfizer and Servier. F.H.M. is an ad hoc consultant for thefollowing organizations: Novartis, Daiichi Sankyo, Pfizer, Takeda,Abbott. F.H.M. received grant support from Forest, DaiichiSankyo and Boehringer Ingelheim.

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Table 1 Drug combinations in hypertension:recommendations

Preferred

ACEInhibitor/diuretic

ARB/diuretic

ACE-Inhibitor/CCB

ARB/CCB

Acceptable

Beta-blocker/diuretic

CCB (dihydropyridine)/beta-blocker

CCB/diuretic

Renin inhibitor/diuretic

Renin inhibitor/CCB

Dihydopyridine CCB/non-dihydropyridine CCB

Unacceptable

ACE-Inhibitor/ARB

Renin inhibitor/ARB

Renin inhibitor/ACE-Inhibitor

RAS inhibitor/beta-blocker

CCB (non-dihydropyridine)/beta-blocker

Centrally acting agent/beta-blocker




ownloaded from


References1. Page IH. The MOSAIC theory. In Page IH, ed. Hypertension Mechanisms.

New York: Grune and Stratton, 1987 p910–923.2. Sever PS. The heterogeneity of hypertension: why doesn’t every patient respond

to every antihypertensive drug? J Hum Hypertens 1995;9:S33–S36.3. Mimran A, Ribstein J, Du Cailar G. Converting enzyme inhibitors and renal func-

tion in essential and renovascular hypertension. Am J Hypertens 1991;4(Suppl. 1):7S–14S.

4. Dickerson JE, Hingorani AD, Ashby MJ, Palmer CR, Brown MJ. Optimisation ofantihypertensive treatment by cross-over rotation of four major classes. Lancet1999;353:2008–2013.

5. Matterson BJ, Reda DJ, Cushman WC. Department of Veterans Affairs.Single-therapy of hypertension study. Revised figures and new data. Departmentof Veterans’ Affairs Cooperative Study Group on Antihypertensive Agents. Am JHypertens 1995;8:189–192.

6. Law MR, Wald NJ, Morris JK, Jordan RE. Value of low dose combination treat-ment with blood pressure lowering drugs: analysis of 354 randomised trials.BMJ 2003;326:1427–1435.

7. Attwood S, Bird R, Burch K, Casadei B, Coats A, Conway J, Dawes M, Ebbs D,Farmer A, Robinson J, Sherlock C. Within-patient correlation between the anti-hypertensive effects of atenolol, lisinopril and nifedepin. J Hypertens 1994;12:1053–1060.

8. Wald DS, Law M, Morris JK, Bestwick JP, Wald NJ. Combination therapy versusmonotherapy in reducing blood pressure: meta-analysis on 11,000 participantsfrom 42 trials. Am J Med 2009;122:290–300.

9. Jamerson KA. Rationale for angiotensin II receptor blockers in patients with low-renin hypertension. Am J Kidney Dis 2000;36(Suppl. 1):S24–S30.

10. Chobanian AV, Bakris GL, Black HR, Cushman WC, Green LA, Izzo JL Jr,Jones DW, Materson BJ, Oparil S, Wright JT Jr,, Roccella EJ; Joint National Com-mittee on Prevention, Detection, Evaluation, and Treatment of High BloodPressure. National Heart, Lung, and Blood Institute; National High Blood PressureEducation Program Coordinating Committee. Seventh Report of the JointNational Committee on Prevention, Detection, Evaluation, and Treatment ofHigh Blood Pressure. Hypertension 2003;42:1206–1252.

11. MacMahon S. Blood pressure and the risks of cardiovascular disease. In Swales JD,ed. Textbook of Hypertension. Blackwell Scientific Publication, 1994, p46–57.

12. Mancia G, De Backer G, Dominiczak A, Cifkova R, Fagard R, Germano G,Grassi G, Heagerty AM, Kjeldsen SE, Laurent S, Narkiewicz K, Ruilope L,Rynkiewicz A, Schmieder RE, Struijker Boudier HA, Zanchetti A; EuropeanSociety of Hypertension; European Society of Cardiology. 2007 ESH-ESC Guide-lines for the management of arterial hypertension: the task force for the manage-ment of arterial hypertension of the European Society of Hypertension (ESH) andof the European Society of Cardiology (ESC). Blood Press 2007;16:135–232.

13. Mancia G, Laurent S, Agabiti-Rosei E, Ambrosioni E, Burnier M, Caulfied MJ,Cifkova R, Clement D, Coca A, Dominiczak A, Erdine S, Fagard R, Farsang C,Grassi G, Haller H, Heagerty A, Kjeldsen SE, Kiowski W, Mallion JM,Manolis A, Narkiewicz K, Nilsson P, Olsen MH, Rahn KH, Redon J, Rodicio J,Ruilope L, Schmieder RE, Struijker-Boudier HA, Van Zwieten PA, Viigimaa M,Zanchetti A. Reappraisal of European guidelines on hypertension managements:a European Society of Hypertension Task Force document. J Hum Hypertens2009;27:2121–2158.

14. Williams B, Poulter NR, Brown MJ, Davis M, McInnes GT, Potter JF, Sever PS,Thom SM; BHS guidelines working party for the British Hypertension Society.Guidelines for the management of hypertension: report of the fourth workingparty of the British Hypertension Society, 2004—BHS IV. J Hum Hypertens2004;18:139–185.

15. Messerli FH, Mancia G, Conti CR, Hewkin AC, Kupfer S, Champion A, Kolloch R,Benetos A, Pepine CJ. Dogma disputed: can aggressively lowering blood pressurein hypertensive patients with coronary artery disease be dangerous? Ann InternMed 2006;144:884–893.

16. Bangalore S, Messerli FH, Wun C, Zuckerman AL, DeMicco D, Kostis JB, LaRosa JC. Treating to New Targets Steering Committee and Investigators.J-Curve revisited: an analysis of the Treating to New Targets (TNT) Trial. Am JColl Cardiol 2009;53:A219.

17. SHEP Cooperative Research Group. Prevention of stroke by antihypertensivedrug treatment in older persons with isolated systolic hypertension. Finalresults of the Systolic Hypertension in the Elderly Programme (SHEP). JAMA1991;265:3255–3264.

18. Rothwell PM, Howard SC, Dolan E, O’Brien E, Dobson JE, Dahlof B, Sever PS,Poulter NR. Prognostic significance of visit-to-visit variability, maximum systolicblood pressure, and episodic hypertension. Lancet 2010;375:895–905.

19. Webb AJ, Fischer U, Mehta Z, Rothwell PM. Effects of antihypertensive-drug classon interindividual variation in blood pressure and risk of stroke: a systematicreview and meta-analysis. Lancet 2010;375:906–915.

20. Blood Pressure Lowering Treatment Trialists’ Collaboration. Effects of ACEinhibitors, calcium antagonists, and other blood-pressure-lowering drugs; resultsof prospectively designed overviews of randomised trials. Lancet 2000;356:1955–1964.

21. Blood Pressure Lowering Treatment Trialists’ Collaboration. Effects of differentblood-pressure-lowering regimens on major cardiovascular events: results ofprospectively-designed overviews of randomised trials. Lancet 2003;362:1527–1535.

22. Nice guidelines. Management of hypertension in adults in primary care. 2004.www.nice.org.uk.

23. Dahlof B, Devereux RB, Kjeldsen SE, Julius S, Beevers G, Faire U, Fyhrquist F,Ibsen H, Kristiansson K, Lederballe-Pedersen O, Lindholm LH, Nieminen MS,Omvik P, Oparil S, Wedel H; LIFE Study Group. Cardiovascular morbidity andmortality in the Losartan Intervention For Endpoint reduction in hypertensionstudy (LIFE): a randomised trial against atenolol. Lancet 2002;359:995–1003.

24. Dahlof B, Sever PS, Poulter NR, Wedel H, Beevers DG, Caulfield M, Collins R,Kjeldsen SE, Kristinsson A, McInnes GT, Mehlsen J, Nieminen M, O’Brien E,Ostergren J; for the ASCOT Investigators. Prevention of cardiovascular eventswith an antihypertensive regimen of amlodipine adding perindopril as requiredversus atenolol adding bendroflumethiazide as required, in theAnglo-Scandinavian Cardiac Outcomes Trial-Blood Pressure Lowering Arm(ASCOT-BPLA): a multicentre randomised controlled trial. Lancet 2005;366:895–906.

25. Jamerson K, Weber MA, Bakris GL, Dahlof B, Pitt B, Shi V, Hester A, Gupte J,Gatlin M, Velazquez EJ; ACCOMPLISH Trial Investigators. Benazepril plus amlo-dipine or hydrochlorothiazide for hypertension in high-risk patients. NEJM2008;359:2417–2428.

26. Julius S, Kjeldsen SE, Brunner H, Hansson L, Platt F, Ekman S, Laragh JH,McInnes G, Schork AM, Smith B, Weber M, Zanchetti A; VALUE Trial. VALUEtrial: Long-term blood pressure trends in 13,449 patients with hypertensionand high cardiovascular risk. Am J Hypertens 2003;7:544–548.

27. Poulter NR, Wedel H, Dahlof B, Sever PS, Beevers DG, Caulfield M, Kjeldsen SE,Kristinsson A, McInnes GT, Mehlsen J, Nieminen M, O’Brien E, Ostergren J,Pocock S; for the ASCOT Investigators. Role of blood pressure and othervariables in the differential cardiovascular event rates noted in theAnglo-Scandinavian Cardiac Outcomes Trial-Blood Pressure Lowering Arm(ASCOT-BPLA). Lancet 2005;366:907–913.

28. Williams B, Lacy PS, Thom SM, Cruickshank K, Stanton A, Collier D, Hughes AD,Thurston H, O’Rourke M; CAFE Investigators; Anglo-Scandinavian Cardiac Out-comes Trial Investigators; CAFE Steering Committee and Writing Committee.Differential impact of blood pressure-lowering drugs on central aortic pressureand clinical outcomes: principal results of the Conduit Artery Function Evaluation(CAFE) study. Circulation 2006;113:1213–1225.

29. Rothwell PM, Howard SC, Dolan E, O’Brien E, Dobson JE, Poulter NR, Sever PS;on behalf of the ASCOT-BPLA and MRC Trial Investigators. Effects of b-blockersand calcium-channel blockers on within-individual variability in blood pressureand risk of stroke. Lancet Neurol 2010;9:469–480.

30. Weber MA, Bakris GL, Jamerson K, Weir M, Kjeldsen SE, Devereux RB,Velazquez EJ, Dahlof B, Kelly RY, Hua TA, Hester A, Pitt B; for the ACCOMPLISHInvestigators. Cardiovascular events during differing hypertension therapies inpatients with diabetes. JACC 2010;56:77–85.

31. Makani H, Bangalore S, Romero J, Wever-Pinzon O, Messerli FH. Effect ofrenin-angiotensin-system blockade on calcium channel blockers associatedperipheral edema. Am J Med 2011;124:128–135.

32. Sever PS, Poulter NR. Management of hypertension: is it the pressure or the drug?Blood pressure reduction is not the only determinant of outcome. Circulation2006;113:2754–2774.

33. Pepine CJ, Handberg EM, Cooper-DeHoff RM, Marks RG, Kowey P, Messerli F,Mancia G, Cangiano JL, Garcia-Barreto D, Keltai M, Erdine S, Bristol HA,Kolb HR, Bakris GL, Cohen JD, Parmley WW; INVEST Investigators. A calciumantagonist vs. a non-calcium antagonist hypertension treatment strategy forpatients with coronary artery disease. The International Verapamil-TrandolaprilStudy (INVEST): a randomised controlled trial. JAMA 2003;290:2805–2819.

34. Beckett NS, Peters R, Fletcher AE, Staessen JA, Liu L, Dumitrascu D,Stoyanovsky V, Antikainen RL, Nikitin Y, Anderson C, Belhani A, Forette F,Rajkumar C, Thijs L, Banya W, Bulpitt CJ, for the HYVET Study Group. Treatmentof hypertension in patients 80 years of age and older. NEJM 2008;358:1887–1898.

35. Alviar CL, Devarapally S, Romero J, Benjo AM, Nadkarni G, Javed F,Suryadevara R, Kang H, Messerli FH. Efficacy and Safety of Dual Calcium ChannelBlocker Therapy for the Treatment of Hypertension: A Meta-analysis. ASH, 2010.

36. Messerli FH, Staessen JA, Zannad F. Of fads, fashion, surrogate endpoints and dualRAS blockade. Eur Heart J 2010; Aug 3. [Epub ahead of print].

To uptitrate, to substitute, or to combine drugs 2506a



ownloaded from

www.nice.org.uk

www.nice.org.uk

www.nice.org.uk

www.nice.org.uk


37. Yusuf S, Teo KK, Pogue J, Dyal L, Copland I, Schumacher H, Dagenais G, Sleight P,Anderson C; the ON-TARGET Investigators. Telmisartan, ramipril, or both inpatients at high risk for vascular events. NEJM 2008;358:1547–1559.

38. Oparil S, Yarrows SA, Patel S, Zhang J, Satlin A. Dual inhibition of the renin systemby aliskiren and valsartan. Lancet 2007;370:1126–1127.

39. Alvarez-Alvarez B. Management of resistant arterial hypertension: role of spiro-nolactone versus double blockade of the renin-angiotensin-aldosterone system.J Hypertens 2010; Jul 21. [Epub ahead of print].

40. Bailey RR, Neale TJ. Rapid clonidine withdrawal with blood pressure overshootexaggerated by beta-blockade. BMJ 1976;6015:942–943.

41. Chapman N, Chang CL, Dahlof B, Sever PS, Wedel H, Poulter NR, for theASCOT Investigators. Effect of doxazosin gastrointestinal therapeutic system asthird-line antihypertensive therapy on blood pressure and lipids in theAnglo-Scandinavian Cardiac Outcomes Trial. Circulation 2008;118:42–48.

42. ALLHAT Officers and Coordinators for the ALLHATT Collaborative ResearchGroup. Diuretic versus a-blocker as first-step antihypertensive therapy. Finalresults from the Antihypertensive and Lipid-Lowering Treatment to PreventHeart Attack Trial (ALLHAT). Hypertension 2003;42:239–246.

43. Chapman N, Dobson J, Wilson S, Dahlof B, Sever PS, Wedel H, Poulter NR, onbehalf of the ASCOT Trial Investigators. Effect of spironolactone on bloodpressure in subjects with resistant hypertension. Hypertension 2007;49:839–845.

44. Lv Y, Zou Z, Chen GM, Jia HX, Zhong J, Fang WW. Amlodipine and angiotensin-converting enzyme inhibitor combination versus amlodipine monotherapy inhypertension: a meta-analysis of randomised control trials. Blood Press Monit2010;15:195–204.

45. Messerli FH, Grossman E. Pedal edema-not all dihydropyridine calcium antagon-ists are created equal. Am J Hypertens 2002;15:1019–1020.

46. Kaplan N. Clinical Hypertension. In Kaplan NM, ed. 8th ed. Lippincott Williamsand Wilkins, 2002, p247.

47. Hypertension Detection and Follow-up Programme Cooperative Group (HDFP).Five-year findings of the Hypertension Detection and Follow-Up programme:reduction in mortality in persons with high blood pressure, including mild hyper-tension. JAMA 1979;242:2562–2571.

48. Lindholm LH, Carlberg B, Samuelssson O. Should b blockers remain first choicein the treatment of primary hypertension? A meta-analysis. Lancet 2005;366:1545–1553.

49. Gupta AK, Arshad S, Poulter NR. Compliance, safety and effectiveness of fixed-dose combinations of antihypertensive agents: a meta-analysis. Hypertension 2010;55:399–407.

50. Bangalore S, Kamalakkannan G, Parkar S, Messerli FH. Fixed-dose combinationsimprove medication compliance: a meta-analysis. Am J Med 2007;120:713–719.

51. Wolf-Maier K, Cooper RS, Kramer H, Banegas JR, Giampaoli S, Joffres MR,Poulter NR, Primatesta P, Stegmayr B, Thamm M. Hypertension treatment andcontrol in five European countries, Canada, and the United States. Hypertension2004;43:10–17.

52. Redon J, Coca A, Lazaro P, Dolores Aguila M, Cabanas M, Gil N,Sanchez-Zamarano MA, Arand P. Factors associated with therapeutic inertia inhypertension: validation of a predictive model. J Hypertens 2010;28:1770–1777.

53. Sever PS. Is systolic blood pressure all that matters? BMJ (Head to Head Debate)2009;339:138–139.

54. Falaschetti E, Chaudhury M, Mindell J, Poulter NR. Continued improvement inhypertension management in England. Results from the Health Survey forEngland 2006. Hypertension 2009;53:480–486.

P.S. Sever and F.H. Messerli2506b



ownloaded from


Health & Medicine

Optimal%20management%20 hta%202011%20ehj