Cardiovascular Drugs and Therapy 3: 327-332, 1989 © Kluwer Academic Publishers. Printed in the U.S.A.

COMBINATION THERAPY WITH CALCIUM- ENTRY BLOCKERS AND BETA- ADRENOCEPTOR ANTAGONISTS IN HYPERTENSION

SUMMARY. The use of more than one drug to control blood pressure may be necessary in up to 50% of hypertensive patients seen in cl inical practice. A rational bas is for com- bination therapy includes 1) the use of drugs that act on dif- ferent physiological systems involved in blood-pressure con- trol and 2) us ing a second drug to counteract reflex responses, which may l imit the effect iveness of the first, and, 3)as is less commonly practiced, the use of low doses of two drugs that act on the same or different physiological systems to avoid the side effects encountered with higher doses of s ingle agents.

The hemodynamic effects of calc ium-entry blocking drugs and beta-adrenoceptor blockers are complementary and syn- ergism might be anticipated, particularly wi th the dihydro- pyridines and beta-blockers, s ince the latter prevent the short-term reflex increase in sympathet ic act ivity occurring as a consequence of vasodilation.

Although there are many studies advocat ing the benefits of such combinations, caution is required with combinations of beta-bleckers and verapamil or di l t iazem because of poten- tial cardiac depressant effects result ing from the more com- plex effects of these calc ium-channel blockers on cardiac myo-cytes and conducting t issue. Such problems would be more l ikely to be encountered in patients wi th long-standing hypertension and in whom poor left ventricular function and coro-nary artery disease may be present.

KEY WORDS. combination therapy, beta-blockers, hyperten- sion, calc ium antagonists

Patterns of t reatment or trends in pharmacotherapy of hypertension have changed over the past 30 years with the advent of newer classes of antihypertensive drugs and, within each class, the development of compounds with greater selectivity for target sites and fewer ad- verse responses. The decisions of whom to treat and when to treat have been, to a great extent, influenced by the results of controlled clinical trials such as the Veterans Administration trial [1], the MRC Trial in Mild Hypertension [2], and the trial of the European Working Party on Hypertension in the Elderly (EWPHE) [3], so that drug treatment of mild hyper- tension and drug t reatment of the elderly can be jus- tiffed on the basis of the trial evidence. However, let us not forget that patterns of t reatment have been influ- enced by many other factors that have had much less substantive scientific basis. Firstly, the treatment of the mild hypertensive and the elderly hypertensive had begun long before any trials reported the benefit of

Peter Sever St. Mary's Hospital Medical School, London, England

therapy in these subgroups of hypertension. Clinicians had assumed that the benefits from treatment seen in earlier trials of moderate and severe hypertension and in younger patients could be extrapolated to mild hypertension and the elderly, and many had actively extended their t reatment population to include those categories.

Secondly, the preferred class of antihypertensive agent for first choice of therapy in the early 1970s swung dramatically from diuretics to beta-blockers in the belief that the latter would afford "cardioprotec- tion" and perhaps reduce the incidence of coronary events in the hypertensive. There was no evidence for such a belief at the time, and up to the present most of the trials have failed to show any convincing cardiovas- cular benefit from drug regimes including a beta- blocking drug [2, 4, 5], with the exception of the MAPHY study [6], which recently reported a reduc- tion in total and cardiovascular mortality in hyperten- sive men treated with a metoprolol regime compared with a thiazide diuretic.

Thirdly, patterns of therapy change when more selective agents with a better pharmacologic and physiochemical profile within a particular class of agents are introduced. This explains the increasing use, for example, of atenolol and the decline in the use of propranolol in the U.K. It must be recognized that marketing forces have also had a major influence on such changes in the choice of drugs.

Finally, the pattern of treatment is changing again in the U.K., with a gradual move away from the traditional first-line drugs, the diuretics and the beta- blockers, to newer classes of agent such as the calcium-

Address for correspondence and reprint requests: Peter Sever, Professor of Clinical Pharmacology and Therapeutics, Queen Elizabeth The Queen Mother Wing, St. Mary's Hospital Medical School, London, W.2., England.

327

328 Sever

entry blockers and the converting-enzyme inhibitors. What are the reasons for this change in practice? The disappointing results of the MRC and other trials in failing to reduce coronary risk, taken together with the known adverse effects of beta-blockers and diuretics on blood lipids, has persuaded many that the so-called lipid-neutral drugs (calcium-entry blockers and ACE inhibitors) might be a preferable long-term treatment in the younger hypertensive. Secondly, the very high frequency of adverse drug reactions and side effects seen with the diuretics and beta-blockers, for example, in the MRC trial [2], demands the search for and use of more acceptable drugs in terms of side effects, and thirdly there have been claims that some agents-- notably the calcium-entry blockers and ACE inhibi- tots--have a more appropriate hemodynamic profile in the hypertensive subject and possibly act at a site closer to that of the underlying pathogenetic abnor- mality (the calcium-entry blockers).

Combination Therapy

Fifty to sixty percent of hypertensive patients in general practice can be managed on a single drug, leav- ing 40% to 50% needing more than one drug to control blood pressure. Data from the MRC and EWPHE trials give figures of 20% to 35% of patients with mild hypertension requiring supplemental therapy to achieve adequate control of blood pressure [2], and over 35% of the elderly treated initially with a diuretic require additional therapy [3]. In hospital practice, this latter figure is higher; indeed, in our own clinic greater than 80% of patients are on two or more drugs.

What is the Basis for Combination Therapy?

In the design of appropriate therapeutic regimes, one should consider the following points:

1. The use of combinations of drugs that act upon dif- fering physiological systems in the control of blood pressure, e.g., sympatholytics and diuretics.

2. Where reflex responses to a drug mitigate its blood- pressure-lowering effects, then a second agent may be used to counteract such a response (e.g., vasodilators and beta-blockers).

3. The use of suboptimal doses of two different drugs that act on the same or different physiological sys- tems to avoid side effects of higher doses of a single agent.

Beta-Adrenoceptor Antagonists (Beta-Blockers)

The mode of action of the beta-blockers in hyperten- sion is complex and is still incompletely understood. Most beta-blockers reduce cardiac output and inhibit renin release from the kidney, thereby reducing angiotensin stimulation of aldosterone secretion and also the angiotensin-facilitating effect in the sym- pathetic nervous system [7]. Beta-blockers may also have a peripheral effect on inhibiting prejunctional facilitatory beta-receptors and may reduce norad- renaline overflow at sympathetic nerve endings [8], but the importance of such an action is unknown.

The long-term hemodynamic consequences of beta- blockade are to a certain extent determined by the presence or absence of partial agonist or intrinsic sym- pathomimetic activity (ISA) [9]. For instance, beta- blockers without ISA primarily reduce cardiac output, whereas those with ISA reduce vascular resistance with little or no effect on cardiac output, despite compar- able falls in arterial pressure [6]. A scheme explaining the proposed mechanisms for the fall in arterial pres- sure is shown in Table 1. In the absence of ISA, beta- blockers such as atenolol block postjunctional cardiac betal-receptors and reduce cardiac output. There is a baroreflex-mediated increase in efferent vasocon- strictor sympathetic nerve activity, which partially prevents the fall in blood pressure by increasing vas- cular resistance. Blockade of prejunctional beta-recep- tors, however, abolishes, after a certain time interval, the effect of increased sympathetic drive, thus vascular resistance falls to pretreatment values and blood pressure falls primarily as a result of cardiac output de- crease. On the other hand, beta-blockers with ISA suf- ficiently high to replace basal sympathetic tone (Pin- dolol ®) do not reduce cardiac output, so that reflex changes in vascular resistance do not occur. Eventual- ly prejunctional beta-blockade inhibits noradrenaline overflow and results in a decreased vascular resistance to below pretreatment levels [8]. Further evidence for such a hypothesis comes from studies on plasma nor- adrenaline concentrations, which invariably fall with beta-blockers possessing ISA [9], but remain un- changed or even increase with treatment with pure an- tagonists. The direct pharmacologic effect of blockade of cardiac beta-receptors is to slow sinus node dis- charge, to increase atrioventricular conduction time and to increase the refractory period of the AV node, which is important when these agents are combined with calcium-entry blockers.

Calcium Antagonists and B-Blockers 329

Table 1. Proposed mechanism o/beta-adrenoceptor blockers in the treatment of hypertension

Non-ISA ISA

Postjunctional beta-adrenoceptor blockade Baroreflex Prejunctional beta-adrenoceptor blockade

Cardiac output $ =

Vascular resistance ~ = Vascular resistance ~ $

Arterial pressure $ 8

After Man in 't Veld and Schalekamp [6]. ISA = intrinsic sympathomimetic activity.

Calcium-Entry Blockers

Turning to the calcium antagonists or, more correctly, the calcium-entry blockers (CEB), these are a hetero- genous group of chemical entities that have in common an action to inhibit transmembrane calcium influx.

The verapamil group of compounds are chemically related to the beta-blockers, although they do not possess such activity [10]. The benzothiazepine deri- vatives, such as diltiazam, structurally, but not phar- macologically, resemble the benzodiazepines. The dihydropyridines consist of an extensive group of calcium-entry blockers, and within this group a num- ber of 1,4-dihydropyridine analogues have been ex- amined to delineate their structure-activity relation- ships. Finally, flunarazine and cinnarizine are di- phenylalkylamines and are again structurally quite dissimilar from the other groups; it has been noted that structure-activity relationships within this group follow those of calmodulin inhibitors.

The International Society and Federation of Car- diology working group on classification of calcium an- tagonists for cardiovascular disease has proposed a classification based on specificity of agents for cal- cium-channel blocking activity [11]. Group A agents are highly specific for voltage-dependent calcium channels and include verapamil (V), nifedipine (N) (and related dihydropyridines), and diltiazem (D). Within this category radioligand binding studies may identify different binding sites for the three agents (V, N, and D). The dihydropyridine binding site has been characterized using radioligands such as [3H]-nimo- dipine. Unlabeled dydropyridines compete for the radioligand binding site, and the characterization of the displacement curves is consistent with a single dihydropyridine receptor (N) [10].

The characteristics of binding of verapamil and

diltiazem, however, are different and suggest either different or overlapping binding sites. For example, verapamil exhibits a biphasic displacement curve for nimodipine binding, which has been interpreted as either evidence for the selectivity of this group of calcium antagonists for subpopulations of calcium channels, which are labeled nonselectively by 1,4- dihydropyridines, or, alternatively, that the class II drugs are selective for interconvertable states of a single channel [12].

Diltiazem, on the other hand, stimulates the bind- ing of tritiated nimodipine and may clearly alter the pharmacologic profile of class I and class II drugs [12].

It is proposed that Group-B channel blockers com- prise less specific agents that in binding studies may or may not compete for dihydropyridine binding sites. These agents include the combined Na+/Ca ÷ blockers (bepridil, tiapamil, and fendiline); cinnarazine and flunarizine, which have quite distinct binding sites from verapamil, dihydropyridine, and diltiazem; and prenylamine and perhexiline.

Lastly nonspecific agents, Group C, are drugs such as chloropromazine and beta-blockers, which at high doses possess weak calcium-channel blocking activity.

The depolarization in vascular smooth muscle is de- pendent on the inward movement of calcium ions. Two mechanisms are responsible for contraction in vascular smooth-muscle cells. Electromechanical coupling is mediated by voltage-sensitive calcium channels where, in response to depolarization, extracellular calcium moves down its electrochemical gradient into the cell to initiate the contractile process. Phar- macomechanical coupling involves agonist-induced contraction that occurs without depolarization of the membrane, and vasoconstriction results from the release of intracellular calcium from sarcoplasmic

330 Sever

reticulum. Subsequently, this receptor-mediated ef- fect also results in an increase in the influx of ex- tracellular calcium [13].

The increase in cytosolic calcium induced by either mechanism results in enhanced binding of calcium to calmodulin and the activation of myosin light-chain kinase; the resultant phosphorylation of the light chain of myosin promotes interaction between actin and myosin and contraction.

The calcium-channel blockers interfere with mobili- zation of calcium and reduce the elevation of in- tracellular calcium occurring by either mechanism, but the blockade of voltage-dependent calcium chan- nels in vascular smooth-muscle cells occurs at much lower concentrations than those that are required to in- terfere with the receptor-mediated mechanism. Of note is the fact that calcium-channel blockers relax arterial smooth muscle, but have little effect on most venous beds.

Overall, it seems that of the calcium-entry blockers, dihydropyridines, which are the most potent inhibitors of slow-channel calcium transport, possess the greatest potency in terms of effects on the vasculature. However, the potency of blockers of slow-channel transport in heart muscle does not always parallel their potency in vascular smooth muscle [14].

The primary hemodynamic effect of the calcium- channel blocking drugs used in the treatment of hyper- tension results from their action on vascular smooth muscle, effecting a lowering of peripheral vascular resistance. However, as demonstrated by Fleckenstein [15], the blockade of the slow inward calcium current in myocardial cells results in complex actions on the heart [16], which are beyond the scope of this paper. Suffice it to say that, in the heart, membrane de- polarization in atria and ventricular myocytes and in atria and ventricular conducting tissue results from the fast inward flux of sodium and the second slow calcium influx. On the other hand, in both SA and AV nodes depolarization is essentially dependent on the slow calcium channel. The calcium-channel blockers may thus, in addition to producing a negative inotropic response, also exert negative chronotropic and dromo- tropic responses [13]. These two last effects may be more dependent upon whether the particular agent delays the recovery of the slow calcium channel, i.e., the process of regaining the capacity to transport calcium.

Dihydropyridines in isolated tissues reduce the slow inward calcium current in a dose-dependent manner, do not affect the rate of recovery of the slow calcium channel, and show little dependence on the frequency of stimulation. Thus, at doses commonly used in clini- cal practice, nifedipine, and probably other 1,4-

dihydropyridines, have little effect on AV-nodal con- duction. Moreover, although negative inotropic effects of dihydropyridines can be demonstrated in vitro [17], because the threshold for effects on vascular smooth muscle is so much lower than for prominent effects on the heart, cardiac output is rarely affected by conven- tional therapeutic doses of this class of drugs.

The pharmacologic properties of verapamil demon- strate that, in addition to blocking the slow calcium channel, in contrast to dihydropyridines, verapamil also decreases the rate of recovery of the channel [18], and its action is enhanced as the frequency of stimula- tion increases. Thus, AV conduction is slowed and the rate of the sinus-node pacemaker is depressed [13]. Verapamil is less potent than the dihydropyridines on the vasculature and has been described as less vasoselective, thus, in therapeutic doses that produce vasodilatation, cardiac actions are more likely to occur, including negative inotropic, chronotropic, and dromotropic effects.

Diltiazem is also less vasoselective than the di- hydropyridines, with actions on the vasculature and SA and AV nodes occurring, in vitro, with roughly equivalent doses [19]. Like verapamil, but probably to a lesser extent, diltiazem's actions are also frequency dependent. Thus, with doses used in clinical practice in the treatment of hypertension, the negative chrono- tropic effects are encountered.

The hemodynamic responses when these drugs are used in the hypertensive patient are obviously depend- ent upon their respective selectivity for the vascula- ture. With the selective agent nifedipine, the dilata- tion of arterial resistance vessels is accompanied by reflex sympathetic stimulation, tachycardia, and a positive inotropic effect. On the other hand, with verapamil and diltiazem the doses required to produce vasodilatation also produce cardiac effects, and the reflexly mediated sympathetic response is offset by the direct negative inotropic and chronotropic effects, with little or no change in heart rate and cardiac output. In some instances with diltiazem, acute intravenous ad- ministration causes an increase in heart rate that is not seen with longer term oral administration [13].

The Principles Underlying the Combination of Calcium-Entry Blockers and Beta-Blockers in Hypertension

A rational basis for the use of combined calcium-entry blockers and beta-blockers can be claimed for the following reasons. Hemodynamically the actions of the

Calcium Antagonists and B-Blockers 331

two classes of agents are complementary, with cal- cium-entry blockers having a predominant action on vascular resistance and beta-blockers on cardiac out- put. Synergism with the dihydropyridines might be an- ticipated in view of the fact that short-term treatment is associated with reflexly mediated increases in sym- pathetic activity, causing an increase in heart rate, cardiac output, and renin release. Although the heart- rate increase with long-term treatment with the dihy- dropyridine calcium antagonists is less dramatic, the reflexly mediated rise in renin is maintained, but not so great. There is thus a logic in preventing the cardiac and renal renin responses to calcium-entry blockade by combining the dihydropyridines with a beta-blocker.

The potential disadvantages of the combination result from additive and potentially detrimental nega- tive inotropic, chronotropic, and dromotropic effects inherent in both classes of drugs.

In studies of dihydropyridine/beta-blocker in com- bination, conduction problems rarely occur, although occasional bradyarrhythmias and heart failure have been reported. The constellation of side effects encoun- tered is similar to that seen with either drug when used as monotherapy [20]. When verapamil or diltiazem is combined with a beta-blocker, conduction defects and heart failure are likely to occur more commonly and are explicable by the more complex actions of these drugs on cardiac myocytes and conducting tissue. Although it must be recognized that most reported adverse reac- tions of these combinations have occurred in patients with coronary artery disease and not hypertension, there is no reason to suppose that these changes cannot be extrapolated to the hypertensive patient, par- ticularly those patients with long-standing hyperten- sion in whom poor left ventricular and associated cor- onary artery disease may be present.

The impression that the common side effects of calcium-entry blockade, namely, ankle edema, flush- ing, and headaches, occur less frequently with the ad- dition of a beta-blocker was not confirmed in a recent review [20], and one would want to see more evidence on this important aspect of combined therapy.

Finally, there is increasing interest and awareness of the importance of trophic influences on myocardial and vascular smooth-muscle cells in the hypertensive subject with left ventricular hypertrophy and vascular smooth-muscle hypertrophy, the maintenance of which in the treated hypertensive is probably related to cardiovascular morbidity. Both noradrenaline and angiotensin II may exert atrophic influence on the muscle cells of the heart and blood vessels, thus beta- receptor antagonism, by preventing the actions of noradrenaline on the heart and inhibiting renin release

with consequent reduction of the effect of angiotensin II on blood vessels, may have additional potential benefit.

Pharmacokinetics and Dynamics of Combinations of Beta-Blockers and Calcium

Whatever logic there may be for combining these agents with complementary blood-pressure-lowering properties in the treatment of the hypertensive in- dividual, great care is needed in the design of appropri- ate fixed-dose combinations of drugs if their phar- macokinetic profiles are essentially very different. Although the duration of blood-pressure lowering for many antihypertensive drugs does not follow their plasma-level concentrations, one would require from trials of fixed-dose combination therapy (particularly in once-a-day dosage) evidence for sustained anti- hypertensive efficacy throughout the 24-hour period, for example, as has been shown for the combination of nifedipine with acebutalol [21].

To conclude on the subject of clinical trials of com- bination therapy, it is important that the same strict requirements necessary for the evaluation of efficacy and tolerability of single drugs used in the treatment of the hypertensive subject should be applied to the in- vestigation of drug combinations. Attention must be drawn to the design and execution of such trials, irre- spective of the rational basis for a particular combina- tion, for without good trial data no definitive state- ments can be made regarding selected combination therapy. The literature is replete with poorly designed studies that were often open and did not have a placebo control. Frequently, the blood-pressure responses to one drug, e.g., the calcium antagonist or the beta- blocker, are studied first, followed by the combination of drugs. Such a design takes no account of the repeat- ability of measurement or of the order effect, and added benefits for the two drugs in combination cannot be confirmed from such studies.

However, despite the cautionary note, there is, in my opinion, a clear rational pharmacologic basis for the use of combinations of dihydropyridines and beta- adrenoceptor blocking drugs in the treatment of hyper- tension, on grounds of complementary or synergistic actions. Some concern remains in my mind over the use of verapamil-like drugs in such combinations [22], and potentially also of diltiazem and beta-blockers, because of potential additive effects on the heart, par- ticularly in patients with coexisting coronary artery

332 Sever

disease. Good clinical trial data is necessary on the ef-

ficacy and tolerability of combination therapy, and

perhaps more evidence will be gained for the possibility

that very small doses of combination drugs, when used

in the treatment of mild hypertension, might be far

more acceptable to the patient than larger doses of

single drugs with their attendant high incidence of ad-

verse responses.

References

1. Veterans Administration Cooperative Study Group on Hypertensive Agents. JAMA 1970;213:1143-1152.

2. Medical Research Council Working Party. MRC trial of treatment of mild hypertension. Br Med J 1985;291:97- 104.

3. The European Working Party. Mortality and Morbidity results from the European Working Party on High Blood Pressure in the Elderly trial. Lancet 1985;1:1349-1354.

4. IPPPSH Collaborative Group. Cardiovascular risk and risk factors in a randomised trial of treatment based on the beta- blocker oxprenolol. J Hyperten 1985;3:379-92.

5. HAPPHY Trial Research Group. Beta-blockers versus diuretics in hypertensive men: Main results from the HAPPILY Trial. J Hyperten 1987;5:561-74.

6. MAPHY Study. Primary prevention with metoprolol in patients with hypertension. JAMA 1988;259:1976-82.

7. Zanchetti A, Bartorelli C. Central nervous mechanisms in arterial hypertension: Experimental and clinical evidence. In: Genest J, Koiw E, Kuchel O, eds. Hypertension. New York: McGraw-Hill, 1977:59-76.

8. Starke K. Regulation of noradrenaline release by presynaptic receptor systems. Rev Physiol Biochem Pharmacol 1977; 77:1-124.

9. Man in't Veld AJ, Schalekamp MADH. How intrinsic sym- pathomimetic activity modulates the haemodynamic re- sponses to beta-adrenoceptor antagonists. A clue to the nature of their antihypertensive mechanism. Br J Clin Phar- macol 1982;13:245S-257S.

10. van Zwieten PA. Differentiation of calcium entry blockers into calcium channel blockers and calcium overload block-

ers. Eur Neurol 1986;25(Suppl 1):57-67. 11. International Society and Federation of Cardiology: Work-

ing group on classification of calcium antagonists for car- diovascular disease. Am J Cardiol 1987;60:630-32.

12. Glossman H, Ferry DR, Lubbecke F, Mewes R, Hofmann F. Calcium channels: Direct identification with radioligand binding studies. Trends Pharmacol Sci 1982;Nov:431- 437.

13. Needleman P, Corr PB, Johnson EM. Drugs used for the treatment of angina: Organic nitrates, calcium channel blockers, and ~-adrenargic antagonists. In: Goodman Gil- man A, Goodman LS, Pall TW, Murad F, eds. The phar- macological basis o[ therapeutics, 7th ed. New York: Mac- millan, 1985:806-826.

14. Nayler WG, Horowitz JD. Calcium antagonists: A new class of drugs. Pharmac Ther 1983;20:203-262.

15. Fleckenstein A. Pharmacology and electrophysiology of calcium antagonists. In: Zanchetti A, Krikler DM, eds. Calcium antagonism in cardiovascular therapy: experience with verapamil Excerpta Medica, 1980:10-29.

16. Braunwald E. Mechanism of action of calcium-channel- blocking agents. N Eng J Med 1982;307:26:1618-1627.

17. Godfraind T, Egleme C, Finet M, Janmin P. The actions of nifedipine and nisoldipine on the contractile activity of human coronary arteries and human cardiac tissue in vitro. Pharmacol Toxicol 1987;61:79-84.

18. Elara T, Kaufmann R. The voltage and time-dependent ef- fects of ( -)-verapamil on the slow inward current in isolated cat ventricular myocardium. J Pharmacol Exp Ther 1978; 207:49-55.

19. Taira N. Differences in cardiovascular profile among calcium antagonists. Am J Cardiol 1987;59:24B-29B.

20. Brouwer RML, Follath F, Buhler FR. Review of the car- diovascular adversity of the calcium antagonist ~-blocker combination: Implications for antihypertensive therapy. J Cardiovasc Pharmacol 1985;7(Suppl 4):$38-$44.

21. Lejeune Ph, Gunselmann W, Hennies L, Heb K, Rittgerodt K, Winn K, Gfrerer G, Schreiber U. Effects of BAY15240, a fixed combination of low dose nifedipine and acebutalol on hypertension: Comparison with standard dose nifedipine. Eur J Pharmacol 1985;28:17-21.

22. Dargie H, Cleland J, Findlay I, Murray G, McInnes G. Com- bination of verapamil and beta-blockers in systemic hyper- tension. Am J Cardiol 1986;57:80D-82D.