Cardiovascular Drugs and Therapy 1991;5:561-576 9 Kluwer Academic Publishers, Boston. Printed in U.S.A.

The Applied Pharmacology of Beta-Adrenoceptor Antagonists (Beta Blockers) In Relation to Clinical Outcomes J.D. Fitzgerald Mere Croft, Cheshire, UK

Summary. Despite the fact that beta blockers were introduced into clinical practice 25 years ago, new beta blockers with differing kinetic and d y n a m i c profiles continue to be developed and marketed. This overview assesses some o f the more extensively studied a g e n t s from the point of view o f proof of utility and the validity o f claims for therapeutic advances. The clinical data suggests that despite the expectations o f improvements based on kinetic and dynamic consideration, none o f the n e w e r agents have been shown unequivocally, either in terms o f efficiency or tolerability, to be an advance over the reference agents, the betal antagonists atenoloi and metroprolol. This m a y be either because such improvements will not occur or because of shortcomings in the design and duration o f comparative studies. There are trends to suggest that celiprolol has lesser effects on bronchial function and that it has a lesser impact on lipoprotein profiles. Approaches are s u g g e s t e d that might enable clinicians to appraise for t h e m s e l v e s the validity of claims for the improved efficiency o f n e w beta blockers.

Cardiovasc Drugs Ther 1991;5:561-576

Key Words. beta blockade, atenolol, metoprolol, celiprolol

T h e literature on the pharmacology and clinical effects of beta antagonists is voluminous. F o r example, the most exhaustive review of the subject comprises a book of more than a thousand pages [1]. Since many scientists and clinicians are unlikely to read such a mass of information, t h e r e remains a need to identify the basic principles upon which a judgement may be formed concerning n e w e r beta antagonists that are intended to be, or are already claimed as, medically defensible improvements over currently available agents. The purpose of this brief overview of the newer beta antagonists is to propose certain criteria upon which claims for improvement may legitimately be based. All new agents s t a r t their life cycle in the pharmacology laboratory, where they are examined in order to determine w h e t h e r the particular molecule possesses certain p r e d e t e r m i n e d criteria. In the case of the first beta blocker (dichloroisoprotenenol), the property of beta blockade was discovered serendipitously and the clinical utility envisioned for it was the treatment of pheochromocytoma [2]. In contrast,

pronethalol was selected against certain defined pharmacologic criteria, in order to determine whether the blockade of cardiac beta adrenoceptors would increase exercise tolerance in angina pectoris and control arrhythmias associated with acute myocardial ischemia [3]. The fortuitous discovery of the antihypertensive effects of beta blockers increased considerably the interest in the field, as did the discovery of beta blockers with intended partial agonism [4] or betal-selective activity [5]. It is a curious fact that the beta antagonist sotalol was patented prior to publication of the antianginal activity of pronethalol, yet it was not commercialized for a further 12 years. This was despite the fact that sotalol has a plasma life in humans of more than 10 hours, is hydrophilic, and is minimally metabolized; these properties, which will be reviewed below, are now claimed for several of the newer agents. Thus the factors that determine w h e t h e r a given agent is eventually m a r k e t e d depend not only on its intrinsic pharmacologic properties, but even more so on the marketing and commercial s t r a t e g y of a particular company. As will emerge, there are several paradoxical aspects concerning the rationale for the selection of particular agents. The first published analysis of the relevance of the pharmacologic profile of a beta blocker in relation to its potential therapeutic activity identified four aspects that required consideration [6]: 1. Specificity and potency at the beta adrenoceptor 2. Relative selectivity for the beta 1 adrenoceptor 3. Partial agonism (intrinsic sympathomimetic activity) 4. Membrane-stabilizing properties In that era, clinical cardiologists were much concerned about the "cardiodepressant" effects of the reference

Address for correspondenceand reprint requests: Dr. J.D. Fitzgerald, Materia Medica, Mere Croft, Chester Road, Mere, Near Knutsford, Cheshire WA16 6LG, UK. 56/

562

Fitzgerald

beta antagonist propranolol. This concern was based on the elegant studies by Vaughan Williams indicating that in concentrations in vitro >10 6 M pronetholol and propranolol resembled quinidine in their electrophysiologic profile [7]. It was speculated that the reduction in left ventricular contractility might be attributed to this action. Subsequently, much clinical investigation established that, in the concentrations achieved in clinical use, the membrane-stabilizing properties of propranolol were not demonstrable in humans [8]. This example is quoted in order to illustrate the ease with which pharmacologic properties of beta antagonists have been, and are, extrapolated to both the clinical effect and also to the clinical benefit. The theme in this review is the need to separate rigorously the three elements of pharmacologic profile, clinical effect, and clinical benefit. In many of the current published symposia, these distinctions appear somewhat blurred. An example serves to illustrate the point. There is a long-standing desire to eliminate the possibility of bronchospasm in susceptible subjects who might require beta~blocking therapy. Some of the newer agents clearly are beta1 selective. In clinical studies, they have much less effect than propranolol on respiratory function in susceptible subjects. A choice of propranolol as comparator is understandable but questionable, since it is 5-10 times more selective for the beta2 receptor than the betal [9]. Having shown the difference in clinical effects between propranolol and the betal-selective agent, speculations are then made that beta1 selectivity is clinically beneficial. This extrapolation should only be made once extensive postmarketing studies show that bronchospasm does not occur when the agent is used in everyday clinical use. The issue is that pharmacodynamic studies and controlled clinical trials establish the profile and efficacy of a new agent but do not necessarily prove the clinical benefit [10]. The relevance of these distinctions is illustrated in the other reviews in this issue. The basic problem is one that is generic to all science and may be stated as: "We never know as muct, as we think we do." The beta-blocking literature provides a good example of this sobering aphorism.

concerning efficacy, effectiveness, and safety of new beta blockers [11].

Pharmacologic properties 1. Are the experiments technically proficient? 2. Have the appropriate comparator agents been used? 3. If the agent has more than one pharmacologic property are the multiple actions all demonstrable within a narrow concentration range, or is there more than one log order difference in the P A 2 or EDs0 for the different properties? 4. Are the actions due to the parent compound or are there active metabolites? If so, are their pharmacologic properties adequately defined? 5. Has allowance been made for technical differences in species, anesthesia, and acute versus chronic administration in coming to a final conclusion concerning the drug? Clinical effects Since cardiac beta1 blockade appears to be mandatory for achieving clinical benefits in hypertension, angina, and cardiac arrhythmias, as well as reduction in sudden death [12], then comparative studies between different beta blockers should be made at equieffective cardiac beta-blocking doses. This need not apply necessarily when mixtures of agents with additional nonadrenoceptor properties are being used. The additional value of including evidence of cadiac beta blockade in clinical protocols is that variations in absorption, plasma levels, and active metabolites can be allowed for by using inhibition of the heart-rate response to endogenous or exogenous agonists as a bioassay procedure. The major areas of pharmacodynamic interest are hemodynamic--peripheral blood flow with or without measurements of arterial compliance, respiratory, and metabolic (which includes effects on hypoglycemia in diabetes and plasma lipid profiles). Effects on exercise tolerance and alterations in the perception of fatigue are becoming more important. Hemodynamic criteria CENTRAL HEMODYNAMICS

Selected Criteria for Assessment Beta Blockers

of

The purpose in listing these criteria is not to provide a definitive or extensive framework for appraisal of new beta blockers, but to stimulate readers to improve and refine their own criteria. Perhaps the inevitable debate about the appropriateness of such criteria would stimulate a group to develop a consensus

1. Were baseline conditions achieved prior to the intervention? (? minimal resting period of 45-60 minutes). 2. Was a placebo injection used? 3. How much cardiac beta blockade was present when the observations were made? 4. Was the patient supine or erect? 5. If exercise was included, was it supine, erect, bicycle, or treadmill?

New Beta-Blockers and Clinical Outcomes

6. If two agents are being compared, were the same subjects used? If not, were the patients comparable? 7. If acute hemodynamics are reported, is the same profile achieved with long-term exposure? 8. In comparative studies on impaired left ventricular function, what was the etiology of the myocardial dysfunction? 9. Was the methodology invasive or noninvasive? How well validated is the methodology, e.g., impedance plethysmography versus dye dilution, in the measurement of cardiac output? PERIPHERAL BLOOD FLOW 1. Which vascular bed was studied? Can the data be validly extrapolated to other beds? 2. Was the agent given systemically or locally? 3. What might be the role of cardiovascular reflexes on the changes described? 4. What methodology was used (Doppler, strain gauge, or dye)? 5. Were the subjects healthy or were they patients? If the latter, could the disease limit the extrapolation of the findings to other situations?

Respiratory 1. Were comparator agents used at equiactive b e t a r blocking doses? 2. Were the subjects healthy or with respiratory problems? 3. Were t h e y known to be intolerant to propranolol? 4. Were the dose/response studies to beta 2 agonists employed? 5. Was bronchospasm induced? 6. How big a patient cohort was used? 7. How clinically relevant is the methodology, e.g., FEV1, vital capacity, airways resistance?

Metabolic LIPIDS

1. Were the plasma lipid levels in the normal range prior to the study? 2. Had the subjects been on prior diuretic therapy? 3. Were the relevant lipid fractions measured? 4. Was the duration of exposure sufficient to ensure a sustained change in the profile? 5. Was the magnitude of the change, both in qualitative and quantitative terms, relevant to permit extrapolation to concern about cardiovascular risk factors? DIABETES MELLITUS 1. Have comparative studies on effects on insulininduced hypoglycemia been carried out?

563

2. How clinically relevant are the changes observed likely to be? 3. Have there been any long-term studies in diabetes or have the trials protocols excluded diabetics from the evaluation? OTHER BIOCHEMICAL P A R A M E T E R S

1. Is there any effect on plasma potassium or uric acid and is the effect different to the reference beta blocker?

Exercise and fatigue 1. Was the study single or multiple dose? 2. Was the exercise supine, erect; submaximal, maximal, or prolonged to exhaustion? 3. Were the subjects physically trained or unfit? 4. How was the fatigue measured, and how valid and sensitive was the methodology? In regard to the clinical evaluation of drugs in the major beta-blocking indications, well-established criteria have been laid down by the various regulatory authorities, and these guidelines will not be enlarged upon here.

Pharmacological Profiles of the Newer Beta Antagonists: Selectivity Space does not permit an exhaustive analysis of all the beta-blocking agents that are or might become available for clinical use within the next few years. The agents selected for analytical comment represent a range of pharmacologic profiles and have already had widespread clinical evaluation. The review will focus on the pharmacologic evidence for the classification of the agent and the possible impact such pharmacologic properties may have on clinical effects and clinical outcomes. Beta blockers may be subclassified according to their pharmacologic, physiochemical, and kinetic profiles. A range of subclassifications has been proposed by different experts in the field [1,13,14]. In view of the large n u m b e r of beta-blocking agents, any classification will be a compromise between a n u m b e r of requirements. It seems sensible to focus on those factors likely to influence the clinician in the choice of agent. F r o m the applied pharmacology point of view, it is necessary to distinguish between agents causing a marked change in central hemodynamics, peripheral blood flow, and lesser effects on respiratory and metabolic function. The kinetic profile is relevant in terms of active metabolites and the frequency/quantity of drug required for daily dosing. F r o m the list of 28 compounds given in the introductory table, 13 com-

564

Fitzgerald

Table 1. Comparative classification of the newer beta blockers

Beta blocker subclass

Reference beta blocker

Newer beta blocker

1. Nonselective

Propranolol Sotalol Nadolol Timol Atenolol Metoprolol

Flestolol

2. Betal-selective

No PAA

3. Beta2-selective No PAA 4. Nonselective With PAA

5. N o n s e l e c t i v e + alpha

blockade

6. Betal-selective + alpha blockade 7. Nonselective with beta2-selective agonism 8. Nonselective + nonadrenoceptor vasodilation

9. Betal-selective with nonadrenoceptor vasodilation

Butoxamine IPS 339 Alprenolol Oxprenolol Pindolol Labetalol

Betaxolol Bisoprotol ICI 118551 Carteolol Esmolol Prenalterol Carvedilol Medroxalol Amosulalol Arotinolol

Comments

Major use in glaucoma Highly lipophilic but little hepatic metabolism or membrane stabilizing properties More potent and betaL-selective Extensively metabolized. Withdrawn because of animal toxicity 30 times more potent than propranolol (dose 5-25 mg) PAA = Pindolol Short-acting parenteral prodrug (withdrawn) Withdrawn

Withdrawn due to heptotoxicity in patients

Dilevalol Nipradilol

Acts possibly via nitrate mechanisms, more potent than propranolol

Teratolol BW A 575c YM 16151-1

Renal vasodilator (? DA1 agonism) ACE inhibitor Calcium-channel blocker (1/7potency of nifedipide) Withdrawn 1/3as potent as atenolol Minimal effects on TPR but mechanism is disputed TI~ = 2-3 hours Some betae stimulant effect as well as unknown relaxing mechanism 10 times more potent than atenolol. Effective dose 5 mg once daily. V. lipophilic with little CNS effects. Vascular relaxation associated with non-beta-blocking L-isomer

Prizidolol Bevantolol

Celiprolol Nebivolol

pounds have been selected for more detailed considera t i o n on t h e b a s i s of a p o t e n t i a l l y e n h a n c e d technical o r c o m m e r c i a l profile. E m p h a s i s will b e p l a c e d on t h o s e c o m p o u n d s in w h i c h clinical s t u d i e s i l l u s t r a t e a p a r t i c u l a r principle. I n t h e c r i t i c a l a n a l y s i s t h a t follows, c e r t a i n p h a r m a c o l o g i c p r o p e r t i e s a r e s e l e c t e d as b e i n g t h e d e t e r m i n i n g f a c t o r s t h a t c h a r a c t e r i z e the a g e n t f r o m t h e a s p e c t of a p p l i e d p h a r m a c o l o g y . F r e q u e n t l y t h e specific a g e n t h a s a c l u s t e r of p h a r m a c o logic p r o p e r t i e s t h a t m a y i n t e r a c t , a n d it is t h e i n t e r a c t i o n t h a t p o s s i b l y d e t e r m i n e s t h e n e t effect. The k i n e t i c s a n d m e t a b o l i s m m a y p l a y a p a r t in t h e hemo-

d y n a m i c a n d a n t i h y p e r t e n s i v e profile o b s e r v e d d u r i n g chronic a d m i n i s t r a t i o n . F o r e x a m p l e , esmolol and acebutolol m a y s h a r e e q u i v a l e n t b e t a - b l o c k i n g and p a r t i a l - a g o n i s t p r o p e r t i e s , b u t t h e b r i e f d u r a t i o n of action of t h e f o r m e r r e s u l t s in a n a b s e n c e of s u s t a i n e d h y p o t e n s i v e a c t i v i t y . Since t h e p r i m a r y p u r p o s e of this r e v i e w is to p r o v i d e clinicians w i t h a critical f r a m e w o r k w i t h i n w h i c h to p o s e t h e a p p r o p r i a t e questions, it s e e m s l e g i t i m a t e to s e p a r a t e t h e v a r i o u s compounds by the dominant pharmacologic property or p r o p e r t i e s t h a t p r o m p t e d initial s e l e c t i o n of t h e compound for d e v e l o p m e n t ( T a b l e 1).

New Beta-Blockers and Clinical Outcomes

Relative beta t selectivity Within the group of compounds being considered (see Table 1), there is a range of betaJbetae selectivity ratios. There is no clear agreement as to what is the minimal ratio upon which the claim of "beta 1 selectivity" is based. The most betal-selective compound published is CGP 20172A (B1/B 2 ratio 10,000:1) [15]. This agent is not available for clinical study in humans by the oral route, though CGP 20172A, given parenterally, has been evaluated briefly in volunteers [Imhof, personal communication]. The most betal-selective antagonist that is widely available for clinical use is bisoprolol. The accurate determination of betal/beta 2 ratios in humans is notoriously difficult [16]. Recently Wetlstein et al. [17] combined the techniques of ex vivo radioreceptor binding on plasma samples taken from volunteers and compared the betal/beta e ratios as determined by ligand binding with simultaneously determined inhibition of exercise tachycardia using the latter as a type of bioassay of cardiac blockade at equiactive cardiac blocking doses. The corresponding ratios in ligand binding studies on plasma from the same human subjects gave b e t a J b e t a 2 ratios of 35:1 for atenolol and 75:1 for bisoprolol. The effects of bisoprolol on respiratory function was possibly less than atenolol or metoprolol in nonasthmatic lung disease in doses up to 20 mg daily, but nevertheless it caused decreases in vital capacity and F E V 1 in asthmatic subjects following a single dose of 10 mg daily [18]. The effects of bisoprolol were easily r e v e r s e d by using the beta e stimulant terbutaline [19]. It is reasonable to pose the question whether the marked increase in the b e t a J b e t a 2 selectivity ratio confers clinically relevant improvements over standard beta 1 selective therapy. There is clear evidence that, in comparison with placebo, bisoprolol is the most betas-selective agent without additional pharmacologic actions and is efficacious as an antiarrhythmic, antihypertensive, and antianginal agent [20,21]. In a well-designed large multicenter comparison of bisoprolol 5 and 10 mg with atenolol 50 mg in 249 mild hypertensive subjects, the investigators concluded "the 3 treatments were similar in efficacy and tolerance" [22]. Thus bisoprolol is clearly a useful agent in that it is potent, highly beta~ selective, and efficacious in the major clinical indications. Its long plasma halflife, lack of m a r k e d metabolic clearance,and low lipophilicity make it attractive within the current framework of perceived desirable properties of beta blockers. Clinical experience suggests that, apart from absence of effects on plasma lipids [23], it does not appear to offer major clinical improvement in hy-

565

pertension in t e r m s of the increased number of responders or predictable improved tolerability. Presumably if the most betal-selective blocker has not been shown to be demonstrably b e t t e r than atenolol, then other beta 1blockers with no additional properties are unlikely to offer significant clinical improvements.

Relative beta2 selectivity The only agent that has been available for reasonably extensive human study is ICI 118551, which has a betae/beta 1 selectivity ratio of >300:1 [22]. It has been widely used as a "pharmacological tool" in order to determine the presence and function of betae adrenoceptors, especially in the h e a r t [15,23]. Two features of ICI 118551 should be recognized by those using it for any in vivo study. Firstly it undergoes extensive rapid hepatic biotransformation to form at least five metabolites, three of which are active and fortuitously are also beta 2 selective, but less so than the parent compound. The active metabolites are selectivity

betae/betal 1. 2. 3. 4.

1 1 3 3

oxo derivative hydroxy derivative oxo derivative hydroxy derivative

ratio >66 >60 >48 >40

Adrenoselectivity was assessed in the anesthetized dog measuring shifts in the dose/response curve to isoprenaline for half-rate and perfusion pressure in the autoperfused hind limb. [M. Collis, personal communication]. Studies of ICI 118551 given orally to human volunteers and patients show that it is clearly v e r y beta 2 selective, but it has not been possible to determine how much of the observed effect is due to the most selective parent compound or to the lesser betaeselective metabolites. The second practical feature about ICI 118551 is that it induces hepatic tumors in rats in prolonged toxicity studies. The drug is no longer available for human administration. Limited clinical studies have been carried out to determine the effects of ICI 118551 in hypertensive patients. In two studies of the effects of ICI 118551 in essential hypertension, no reduction in blood pressure was observed when it was dosed with 50 or 100 mg three times daily [24,25]. In the study by Dahlof et al. [24], ICI 118551 reduced heart rate and cardiac output significantly but did not alter mean blood pressure. In a more recent study by Vincent et al. [26], ICI 118551, 50 mg t h r e e times daily, was compared with propranolol 80 mg t h r e e times daily and placebo

566

Fitzgerald

in 19 hypertensive subjects. Systolic and diastolic blood pressures were significantly reduced after 7 days treatment. Plasma renin levels were lowered by both drugs, but ICI 118551 did not alter noradrenaline levels. Isoprenaline challenge showed that the dose of ICI 118551 did not block betal adrenoceptor responses. It is difficult to reconcile these conflicting findings. ICI 118551 causes a significant reduction in heart rate in all the studies and a fall in cardiac output in the single study where this was measured indirectly. In the negative study [26], a placebo was not used, though the subjects were known responders to other beta blockers. Thus the question as to w h e t h e r selective beta 2 blockade reduces blood pressure remains controversial. F r o m the point of view of understanding hypertensive mechanisms, this is unsatisfactory. If betae-selective blockade contributes to the hypotensive effect of beta blockers, then it could be anticipated that: 1. Nonselective beta blockers would cause a g r e a t e r fall in blood pressure than betarselective blockers because of the additional effect of betas-blockade. This has n e v e r been demonstrated. 2. If beta 2 blockade was contributing significantly, then adding ICI 118551 to atenolol, for example, should cause a g r e a t e r fall in blood pressure. Unfortunately this experiment has n e v e r been reported. The subsequent withdrawal of ICI 118551 on toxicity grounds suggests that these issues will not now be resolved. It may well be that one of the beta2-selective metabolites of ICI 118551 is free of neoplastic hepatic effects and would be an acceptable substitute.

Partial A g o n i s t Activity (PAA): Intrinsic S y m p a t h o m i m e t i c Activity (ISA) Introduction The first beta blocker to be discovered was dichloroisoproternol. It was synthesized as part of a program intended to develop a bronchodilator with a longer duration of action than isoprenaline. During pharmacologic evaluation, Moran and Perkins [27] observed that it antagonized some of the effects of isoprenaline, even though, given by itself, it relaxed precontracted bronchial smooth muscle. This stimulant p r o p e r t y was recognized by Black [28], who in searching for a novel approach to antagonize the effects of catecholamines upon the ischemic heart, set out to find an agent free

of intrinsic sympathomimetic activity (ISA). This led to the development of pronethalol, which was believed initially to be devoid of ISA but subsequently was found to have significant stimulant activity. Its successor, propranolol, was shown to have no ISA. Shortly afterwards, alprenolol was developed and was the first agent in which a degree of ISA was specifically introduced into the molecule in order to ameliorate the perceived unwanted hemodynamic effects of propranolol [44]. This is in contrast to the situation with both oxprenolol and pindolol, where the presence of ISA was not an action primarily sought. Ironically, the first complete paper on the pharmacology of pindolol states that it does not have any ISA [29]. Thus, apart from alprenolol, the concept that ISA, or as it is now termed, partial agonist activity, may be of therapeutic benefit has arisen serendipitously. Subsequently there has been controversy in the literature as to the therapeutic relevance of this property [3036]. In the section that follows, an attempt will be made to provide some guidance as to how to approach and evaluate the conflicting views that continue to be expressed in the voluminous literature on this aspect of beta-blocker action.

Concepts and detinitions The conceptual basis of partial agonism has been elegantly summarized by Ariens [37]. He distinguishes between the ability of agents to bind to specific cellular components and their ability to activate this component, i.e., the receptor, a f t e r binding has occurred. Substances that activate specific receptors are termed agonists and those that have affinity but do not activate are termed antagonists. The terms affinity and intrinsic activity, in relation to receptors, have analogy with enzyme kinetics, where the binding constant Km expresses affinity and the substrate turnover rate, Vmax, is analogous to intrinsic activity. The important feature is that compounds with intermediate intrinsic activity behave as partial agonists, implying that even in concentrations that fully saturate the specific receptor, the biological effect is less than that obtainable with a full agonist. The biological effects of beta-adrenoreceptor partial agonists are not predictable if based upon solely theoretical considerations. Partial agonists can differ from full agonists in the following ways: 1. Tissue responses: A full agonist such as isoprenaline elicits a maximal biological response in all tissues, whereas a partial agonist may, for example, exhibit a g r e a t e r stimulant effect in the atrium than the ventricle [38]. 2. Time course: Dichloroisopretenolol has a much

New Beta-Blockers and Clinical Outcomes

567

more gradual onset of effect than isoprenaline in the guinea-pig papillary muscle [38] 3. Species: In the kitten atrium, pindolol is almost a full agonist, whereas in the guinea-pig atrium it has almost no stimulant action. Interestingly, the stimulant action of pindolol in the kitten atrium is only partially blocked by propranolol [39]. 4. Separation of action: Certain partial agonists, for example, pindolol, oxprenolol, and alprenolol, exhibit marked beta-blocking antagonism in low concentrations that do not show any simultaneous stimulant effect. The stimulant effect becomes apparent as the concentration is increased [40,41].

Table 2. Classification of selected beta adrenoceptor partial agonists according to beta receptor selectivity and degree of partial agonism

Thus in attempting to extrapolate to the clinical relevance of partial agonist properties in a beta blocker, caution is necessary, because of the range of pharmacologic differences that different partial agonists may exhibit. A further variable is whether the partial agonism is

3. Beta~ selective PAA.

1. Nonselective, for example, pindolol or oxprenolol 2. Beta I selective, for example, acebutolol, epananolol, or xamoterolol 3. Beta e selective, for example, dilevalol or celiprolol Clearly in reading the conclusions and recommendations contained in reviews on the clinical and therapeutic relevance of partial agonism, it is important to evaluate how much attention has been paid to these pharmacologic issues. F o r example, a recent review of the potential benefits of beta blockers with partial agonism concludes that with: "The beta blockers with a higher level of ISA, such as pindolol, blood pressure falls in parallel with a reduction in total peripheral resistance to below the pre-treatment level" [33]. This statement is true for pindolol but not true for epanolol, which, although having only slightly less PAA than pindolol, has minimal effects on blood pressure, presumably because the partial agonism is beta I selective [35]. Another reviewer concludes, "In no instance has I.S.A. been shown to be anything more than an advantage, potential or demonstrable" [36]. It is not immediately obvious what such a conclusion means, but it cannot be sustained in the face of the fact that xamoterol, a potent beta 1 blocker with more partial agonism than pindolol, does not lower blood pressure and can, at rest, increase blood pressure. These examples are given only to emphasize to the reader that great care is needed in making generalized extrapolations f r o m pharmacologic properties of partial agonists to their clinical value, without the intervening phase of careful clinical experimentation. Two of the more important variables that must be

Selectivity Subclass

Degree of partial agonism

1. Nonselective partial agonists

Low

2. Beta1 selective PAA.

Medium High Medium High High

Examples Penbutolol Alprenolol Oxprenolol Pindolol Carteolol Prenalterol Practolol Epanolol Celiprolol Xamoterol Dilevalol (? Celiprolol)

Low = <10%; medium = >10, <25%; high = -<25%. % is expressed based on isoprenaline 100. considered before making generalizations are the degree of partial agonism and w h e t h e r the compound is selective for one of the beta adrenoceptors. Examples of the various subclasses of beta blockers with partial agonism are given in Table 2. Such a table illustrates the difficulties in attempting to provide simple generalizations concerning the clinical relevance of partial agonism. The final and most important confounding factor, as far as partial agonism is concerned, is the effect of the level of endogenous sympathetic tone present prior to drug administration. In high-tone sympathetic situations, for example, submaximal exercise, partial agonists are indistinguishable from full antagonists, despite their partial-agonist properties. In contrast, at rest or in catecholamine-depleted states, such as the Shay-Drager syndrome, they show stimulant effects. A consequence of the issues described above is that any review of the clinical relevance of partial agonism is unlikely to be clear and concise because of the different degrees of stimulant activity and the differing types of partial agonism possessed by the agents used clinically. Central and peripheral hemodynamics Since beta blockers are used mainly for chronic therapy, it is more rational to consider the long-term hemodynamic effects of beta blockers with partial agonism. A key issue is that there is little difference in the degree of blood pressure reduction between nonselective beta blockers with and without partial agonism. H o w e v e r the hemodynamic profile differs in that, for example, pindolol increases resting heart rate and cardiac output, accompanied by a 30% fall in total peripheral resistance (TPR), whereas proprano-

568

Fitzgerald

lol causes, for the equivalent fall in mean arterial pressure, little change in heart rate, stroke volume, or cardiac output [42]. This profile is in marked contrast to the acute hemodynamic response to propranolol, which is characterized by 25% fall in cardiac output and a large rise in total peripheral resistance. The change in hemodynamic profile between acute and chronic treatment has been attributed to changes in vagal tone and venous return [42]. An important difference is observed between betas-selective blockers with moderate (epanolol) or high (xamoterol) beta 1 partial agonism. The latter drug does not lower blood pressure and the former causes only a minimal, i.e., 6/4 mmHg, fall in blood pressure [35]. Furthermore prenalterol, a nonselective agent with partial agonist activity equivalent to xamoterol, does not reduce blood pressure either. The conclusion may be drawn that a) there is a threshold for cardiac stimulant activity above which no reduction in blood pressure is observed [43]; b) bet~2 stimulation plays an important role in the blood pressure reduction observed with both pindolol and dilevalol; and c) despite a cardiac beta-blocking efficacy equivalent to atenolol and propranolol, the presence of beta 2 stimulation in pindolol and dilevalol does not enhance the hypotensive response. Furthermore, it does not enhance the antianginal activity. This remains something of a paradox. Therefore the only advantages of the presence of partial agonism for drugs that are given chronically may be in improved tolerability. The relationship between improved tolerability and the presence of partial agonism is controversial. The clinical situations in which the differing hemodynamic profiles of partial agonist beta blockers might be expected to result in improved tolerability are 1. The incidence of heart failure 2. Cold peripheries and intermittent claudication 3. Fatigue Heart Failure. Analysis of the published literature on the incidence of heart failure associated with betablocking therapy sugests that it is less than 1% of treated patients [44]. In reality, the true incidence is unknown and should be stratified according to the degree of ventricular dysfunction prior to betablocking therapy. The observations that prolonged beta-blocking therapy can improve some forms of heart failure (see article by Persson and Erhardt in this issue) and can cause the regression of left ventricular hypertrophy secondary to essential hypertension provides a further basis for the confounding of epidemiologic studies on heart failure and beta-blocking therapy. Despite the attractive theoretical basis for

anticipating that partial agonist beta blockers would be less likely to cause clinical heart failure, there are reports that both pindolol and xamoterol do cause it [45,46]. Therefore there is no evidence to support statements made by reviewers to the effect " . . . in patients with or predisposed to heart failure or peripheral vascular disease, beta blockers with a significant degree of I.S.A. may offer a safer alternative to those which lack this property" [33]. On theoretical grounds, xamoterol has the most suitable hemodynamic profile for treating patients with heart failure. Despite this, a significant number of patients deteriorated when treated with this agent, suggesting that vigilance and experience is required in using beta blockers in heart failure, whatever the hemodynamic profile. Beta blockers can be beneficial or harmful, probably for reasons we do not understand, and partial agonism should not provide an aura of apparent improved safety. Cold Peripheries and Claudication. Partial agonists reduce skin blood flow less than nonselective full antagonists [47]. Marshall [48] compared a range of beta blockers with and without partial agonism and concluded that cold peripheries were more likely in the absence of partial agonist activity. Switching patients receiving atenolol, who complained of cold extremities, to pindolol led to symptomatic improvement [49], though in a study of 758 patients Feleke et al. [50] could not distinguish pindolol from propranolol or timolol. In a recent comparison between metoprolol and epanolol (which causes little change in total peripheral resistance) involving 211 patients, no difference could be shown in the incidence of cold extremities [51]. Similarly dilevalol, which possesses vasodilator betae stimulating properties, did not cause significantly less cold extremities than metoprolol [52]. In an open uncontrolled comparison of dilevalol (N = 117) and propranolol (N = 60), neither agent is reported as causing cold extremities [53]; this must be unusual for propranolol. It is relevant that in a comparison of selective and nonselective beta blockers without partial agonism, no increase in the frequency of Raynaud's phenomenon was observed, which suggests that there may be differing mechanisms underlying cold extremities induced by beta blockers and Raynaud's phenomenon [54]. Fatigue. This symptom is often the most troublesome for individual patients. Its cause is unknown and cannot be definitely attributed to the hemodynamic profile of a specific beta blocker. If it were due to the fall in cardiac output and increase in total peripheral resistance associated with beta blockers without par-

New Beta-Blockers and Clinical Outcomes

tial agonism, then pindolol, which has the theoretically more benign hemodynamic profile, should cause less fatigue. N o difference in fatigue has been detected [55]. Similarly no difference in the incidence of fatigue could be shown between epanolol and metoprolol, despite their differing hemodynamic profiles [51]. RESPIRATORY EFFECTS. There have frequently been suggestions that partial agonists would be less likely to induce bronchospasm in asthmatic subjects [33,34,56]. Clinical experience indicates that partial agonists, w h e t h e r nonselective, beta 1 selective or beta 2 selective, m a y all induce bronchospasm in susceptible subjects. A recent survey of spontaneous reports to the Committee on Safety of Medicines records that 747 reports were received of bronchospasm associated with beta-blocking therapy associated with 34 deaths. E v e n with topically applied eye-drop preparations, there have been 66 reports of bronchospasm and three deaths [46]. The availability of effective alternative antihypertensive and antianginal agents leads to the conclusion that beta blockers as a class should be avoided in patients with a current or past history of asthma [57]. It is noteworthy that even the newer beta 1selective antagonists, such as bucindolol [58] and celiprolol [54], which have both partial agonism and nonadrenoceptor smooth muscle inhibitory properties, can cause adverse bronchial effects. PLASMA LIPIDS. Prolonged therapy with propranolol alters plasma lipoprotein profiles in a direction similar to that associated with increased risk of coronary artery disease [60]. Propranolol does not alter total cholesterol or L D L cholesterol, but increases total V L D L cholesterol and triglycerides by an average of 20% and reduces H D L cholesterol by about 11%. Beta 1selective agents, atenolol and metoprolol, have less pronounced effects, but the trends are similar [61]. In contrast, long-term pindolol therapy has much less effects. Plasma triglycerides are elevated about 8-12% in six studies and decreased by 7% in nine other studies. H D L cholesterol was raised by an average of 8% in six long-term studies and reduced 4% in 10 other studies [62]. Total cholesterol and LDL cholesterol were unaffected. The conclusion of extensive reviews in this field is that beta blockers with partial agonism are less likely to have adverse effects on lipoprotein profiles than those without it. The clinical importance of these lipoprotein changes is not established. A recent analysis of outcomes in the BHAT trial concludes that the alterations in H D L ( - 6%) and increase in triglycerides ( + 17%) associated with prolonged propranolol t h e r a p y had little effect in modi-

569

fying the 20% reduction in morbidity and mortality [63]. Thus the lesser effect of partial agonism on lipoprotein profiles is a positive characteristic, but its overall clinical importance is uncertain. In patients with established lipoprotein abnormalities prior to therapy, it is prudent to use an agent that does not exacerbate these abnormalities. Vasodilatory Beta Blockers. The immediate hemodynamic response to an injection of propranolol, the reference beta blocker, includes a rise in peripheral vascular resistance, which can be attenuated by prior administration of an alpha-adrenergic blocking drug [42]. Prolonged t h e r a p y with propranolol is associated with a small increase or little change in total peripheral resistance (TPR) in comparison with control, yet it is recognized that the basic hemodynamic abnormality in essential hypertension is increased TPR. It is therefore rational to a t t e m p t to improve the hemodynamic profile of the reference beta blockers by modifying their pharmacologic properties so that t h e r e is an additional reduction in TPR. The most obvious means to achieve this end is to give a vascular smooth muscle vasodilator at the same time. This can be done by adding the agent separately, mixed in the same capsule, or by discovering agents with the dual actions in the same molecules. The addition of vasodilator drugs, such as hydralazine, minoxidil, diazoxide, alpha blockers, or calcium-channel inhibitors, results in an additional fall in blood pressure in hypertensive subjects already receiving beta blockers. In many clinical trials involving the addition of vasodilators, the hypertensive patients had not responded with an appropriate fall in blood pressure to beta-blocker monotherapy, and the vasodilator had been added in doses appropriate to achieve a satisfactory fall in blood pressure. It is established that adding a vasodilator to beta blockade induced by any of the subclasses of beta blockers will enhance the antihypertensive effect [64,65]. It is a logical step, therefore, to attempt to create molecules that possess the total effects of cardiac beta blockade and relaxation of vascular smooth muscle. This relaxation should be distinguished from the secondary reduction in vascular tone that accompanies chronic beta blockade in hypertension [42]. A number of different approaches have been used to bring about vasodilation. BETA2 ADRENOCEPTOR STIMULATION. The nonselective beta blocker dilevalol possesses additional vasodilator properties due to beta 2 adrenoceptor stimulation. Thus dilevalol resembles pindolol, except that the partial agonism is more selective for beta e adrenoceptors with minimal direct cardiac stimulation [66].

570

Fitzgerald

When given intravenously, it causes an acute fall in peripheral vascular resistance [67]. It is noteworthy that dilevalol is the R i 9R isomer of labetalol. ALPHA ADRENOCEPTOR BLOCKADE. The first beta blocker to possess the additional property of alpha blockade was labetalol. It is a mixture of four enantiomers, one of which is dilevalol. It is a nonselective beta blocker and the mixture shows no partial agonism. Its alpha-blocking action is mainly at the alpha 1 adrenoceptor and is much less potent at this receptor than at the beta r e c e p t o r [69]. The recently described agent arotinolol resembles labetalol, except that it is modestly selective for the beta1 receptor [68]. A further compound, amosulolol, differs from labetalol in that it has a higher affinity for alpha I adrenoceptors than for beta receptors. It is a nonselective beta blocker without partial agonism [70]. Perhaps the most widely studied combined alpha beta blocker is carvedilol. This is a nonselective beta blocker without partial agonism and possesses strong vasodilator properties. These are due to antagonism at the alpha i adrenoceptor [71], though initially it had been thought that carvedilol inhibited vascular smooth muscle by a nonadrenoceptor mechanism. In high concentrations in vitro it has a weak calcium antagonist action [72]. NONADRENOCEPTOR VASODILATORS. Celiprolol is a potent betai-selective blocker with nonselective partial agonism. P a r t of its vasodilator action is therefore attributed to a beta 2 agonist activity [73]. In addition, it possesses vasodilator activity, which is not blocked by propranolol. This is probably the same property that results in a relaxation of bronchial smooth muscle [74]. Nipradilol is a potent nonselective beta blocker without partial agonist properties. It possesses marked vasodilator actions, which may be due to two effects: Firstly, a nitratelike vasodilation with a greater effect on venous than arterial smooth muscle [75]; secondly, it antagonizes alpha adrenoceptors [76]. There are insufficient published data on human studies to comment upon its pharmacologic profile in clinical doses. Nebivolol is a racemic mixture of D-nebivolol (SRRR) and L-nebivolol (RSSS). The molecule has a number of asymmetric centers, and there are 10 potential isomers, of which nebivolol comprises a 50: 50 mixture of two isomers. In anaesthetized dogs, in the dose range of 0.025-0.04 mg/kg IV, it causes a small significant increase in cardiac output and stroke volume, accompanied b y a reduction in calculated systemic vascular resistance. When the D-isomer of nebi-

volol, which is the beta-blocking isomer, is given alone, this atypical hemodynamic profile is not observed. The D-isomer is a potent highly selective beta 1 adrenoceptor antagonist whose hemodynamic profile resembles that of atenolol [77]. Apparently, the presence of the non-beta-blocking L-isomer is responsible for the modification of the hemodynamic response [78]. DL nebivolol also differs from classical beta I antagonists in that it causes an immediate and sustained fall in blood pressure when given intraperitoneally to spontaneous hypertensive rats. This is not observed with the betal-blocking D-isomer or with atenolol. Nebivolol has no affinity for alpha adrenoceptors, calcium or potassium channels, nor does it apepar to modify sympathetic tone. Comparative subacute hemodynamic studies in animals and in human volunteers or hypertensive subjects and patients with acute myocardial infarction suggests that it has much less unwanted effects on ventricular function than propranolol and atenolol [79]. The therapeutic utility of such differences has yet to be established. ANTIHYPERTENSIVE EFFICACY OF VASODILATOR BETA BLOCKERS. Some studies with labetalol suggest that when it has been given in doses appropriate to each patient, it causes a g r e a t e r fall in blood pressure than propranolol [80,81]. Apart from this finding, the currently available vasodilator beta blockers have not shown a consistent improvement in efficacy over the reference beta antagonists [52,82-84]. The reasons for not showing enhanced efficacy might include: . Kinetics. There is a convention that all hypotensive drugs should be effective when given once daily. The kinetics of carvedilol make once-daily dosing of 50-100 mg rational, but this is less so for dilevalol and celiprolol (200-600 mg daily). It may be that the wish to utilize a narrow dose range has obscured the potential additional hypotensive effect that might be attained with vasodilator beta blockers if they were titrated on an individual patient basis. . Dynamics. Some compounds may be lacking in sufficient vasodilator potency at the effective cardiac beta-blocking dose. T h e r e is an inescapable problem in attempting to reconcile a narrow dose range with optimization of dual pharmacologic actions in a fixed ratio molecule. This is in contrast to the uniform success observed in studies where the vasodilator agent is administered separately in hypertensive patients receiving beta blockers [64,65]. The efficacy of vasodilator beta blockers has not been shown to be g r e a t e r than standard betablocker monotherapy in treating stable angina pec-

New Beta-Blockers and Clinical Outcomes

toris [85]. This is in contrast to the improved efficacy observed when, for example, verapamil is added to propranolol in treating anginal patients [86]. TOLERABILITY OF VASODILATOR BETA BLOCKERS. On the assumption that the unwanted effects of reference beta antagonists can be ameliorated by reducing the degree of beta blockade, it might be anticipated that vasodilator beta blockers would be b e t t e r tolerated. In general there is no consistent evidence that this is so. Evidence is emerging that celiprolol may be better tolerated in asthmatic subjects [85]. However, in an overview discussion of celiprolol, Dorow stated in relation to celiprolol, "In m y opinion, at this point in time, asthmatic subjects should not be treated with any beta blocker" [87]. If there should be overriding clinical indications for a beta blocker to be used in an asthmatic subject, consideration would be given to the use of bisoprolol or celiprolol in low doses under close supervision. The other unwanted effects of beta blockade, such as fatigue and cold extremities, may still be observed with vasodilator beta blockers. Therefore the issue is reduced to a quantitative r a t h e r than a qualitative difference. There are claims that dilevalol, celiprolol [88], and carvedilol [89] are b e t t e r tolerated than the reference beta antagonists. This may be so, but there are serious methodologic problems in the reports upon which these claims are based. The performance criteria for designing appropriate questionnaires and applying them correctly in order to obtain data on side effects and quality of life have recently been reviewed [90-92]. Comparative studies of differing subclasses of beta blockers designed to prove a lesser incidence of side effects should meet the criteria laid down, and to date most of the studies do not. At present only tentative conclusions should be drawn about the claims for overall improvement and tolerability with these agents. Data from postmarketing surveillance studies using open study designs with unvalidated questionnaires should be regarded only as appropriate for problem identification r a t h e r than for quantitative statements.

Metabolic Effects of Beta Blockers The effects of beta blockers on lipoprotein patterns have been described above. Carbohydrate and fat metabolism are influenced by adrenoceptor activation, though the responses vary, depending upon the species and the nutritional status. Briefly, beta-receptor activation causes an increase in plasma glucose, free fatty acid (FFA), and lactate [93]. The rise in plasma

571

glucose is due to alpha-receptor-mediated hepatic glycogenolysis and simultaneous inhibition of insulin release, as well as to a beta-receptor-mediated hepatic glycogenolysis. Thus combined alpha and beta blockade is required to inhibit adrenergic-mediated hyperglycemia. Lipolysis is under dual betas- and beta ereceptor control [94], whilst lactate release is due to betae-mediated muscle glycogenolysis. T h e r e are a range of physiologic controls of intermediary carbohydrate and lipid metabolism, so the presence of nonselective beta blockade does not affect these pathways under normal conditions. When hypoglycemia is induced by intravenous insulin, nonselective beta blockade prolongs the hypoglycemia, because the normal counter-regulatory mechanisms initiated by the reflex release of adrenaline are blocked. Thus propranolol causes delay in the return of plasma glucose due to enhanced glucose uptake without an increase in glucose mobilization [95]. Betal-selective blockade also delays the r e t u r n of plasma glucose to normal after insulin-induced hypoglycemia, but it is less than that with nonselective blockers. F r e e fatty acid levels are also depressed significantly by propranolol and much less by beta1 blockade. Of the counter-regulatory hormones, growth hormone and prolactin are not affected by propranolol, but adrenaline levels are significantly higher, whereas atenolol does not affect them. In the studies specifically on the effects of atenolol on insulin-induced hypoglycemia in Type 1 diabetes, it was concluded that selective beta I antagonists can be used with safety in the t r e a t m e n t of insulin-dependent hypertensives [96]. In contrast, the study of propranolol combined with thiazides in non-insulin-dependent diabetics showed a substantial rise of 30 mg/dl of glucose levels in 9 of 20 fasting diabetics [97]. The effects of either agent used alone in these patients were not reported. In a recent review by Houston [98], it is stated that all beta blockers induced abnormal glucose tolerance, quoting a review of his own and a paper on microalbuminuria. The balance of opinion suggests that 1. In nondiabetic patients, selective and nonselective beta blockade does not affect glucose homeostatis. 2. Betal-selective blockade does not affect glucose homeostasis significantly in T y p e 1 or Type 2 diabetics, though non-selective beta blockade with propranolol can impair glucose tolerance. In diabetic patients who have survived myocardial infarction, it was observed, 1 y e a r after myocardial infarction that there was a 7% mortality in those treated by beta blockade versus a 17% mortality for diabetics not treated with beta blockers [99]. This was a retro-

572

Fitzgerald

Table 3. Matrix of factors for evaluation when considering newer beta blockers Indication b hypertension

Therapeutic efficacya

Concomitant disease

Tolerability

C.V. risk factors

Reference beta antagonist

Atenolol

3

3

4

3

Newer beta antagonist Alternative nonbeta-blocking medical therapy Nondrug therapy

Celiprololc

3

3

3-4

Captopril

3

4

Diet/lifestyle modification

2

5

Effect on morbidity mortality

Cost/benefit

4

3 Post AMI hypertensives ?

? ?

4

4

?

?

3

?

?

?

aEfficacy includes kinetic and dynamic factors. bA separate matrix can be created for each indication. eA numerical scale of 0-5 could be used to indicate increasing positive weighting. The scaling is subjective and specific to the individual practitioner, but the matrix could provide the basis for comparingeach doctor's appraisal. The author has provided his evaluative matrix solely for illustrative purposes.

spective survey, but it p r o m p t s the need for caution in extrapolating the significance of a biochemical test, i.e., glucose tolerance, to clinical outcome. It seems reasonable to conclude that as far as n e w e r b e t a blockers are concerned, they are unlikely to improve on the reference b e t a 1 blockers in relationship to glucose homeostasis in normal and diabetic subjects. The effects of carvedilol, celiprolol, and dilevalol on lipoprotein patterns have been reported. Celiprolol t h e r a p y is associated with a modest rise or no change in plasma triglycerides, accompanied by a rise in H D L concentration and a reduction in the L D L / H D L cholesterol ratio. These observations have been confirmed in t h r e e longer t e r m studies [100-102]. They are attributed to the combined effect of a beta2 partial agonism and alpha blockade. Comparable data on dilevalol and carvedilol have not been obtained in a literature search. Global Assessment Beta Blockers

of New

Any critical overview of the n e w e r b e t a antagonists ought to address a matrix of interacting factors, some of which are listed in Table 3. The t h e m e in this review has been to define the needs for i m p r o v e m e n t in the therapeutic profile of b e t a blockers and then to examine the p r e s e n t published data t h a t might support claims for i m p r o v e m e n t over the reference beta antagonists. The practicing clinician will also want to know how different classes of antihypertensive, antianginal, or antiarrhythmic agents compare with beta

blocker subclasses. E x p e r t s hold widely differing views on the c o m p a r a t i v e merits of beta blockers in, for example, hypertension. In a recent critical review, Houston [103] proposed that beta blockers w e r e contraindicated in essential hypertension, except for patients with concomitant angina and cardiac arrhythmias. In noncomplicated hypertension, he proposes that only b e t a blockers with partial agonism should merit consideration as an alternative to non-betablocking, antihypertensive therapy. In contrast, other investigators view b e t a blockers as first-line t h e r a p y in both hypertension and angina, with a restricted utility in cardiac arrhythmias. The t a r g e t s that the n e w e r beta-blocking agents should achieve in o r d e r to displace the reference compounds include: 1. A g r e a t e r n u m b e r of responders on monotherapy. 2. A reduced incidence of beta-blocker-associated unwanted effects. 3. Safer use in the presence of concomitant diseases, especially asthma, h e a r t failure, and atrioventricular block. 4. Preservation of positive effects on morbidity and mortality in patients with coronary a r t e r y disease. The acquisition of clinical evidence to support such improvements demands an evaluation p r o g r a m of considerable b r e a d t h and longevity. There is currently a tendency to use s u r r o g a t e endpoints, especially hemodynamic studies and non-placebo-controlled comparative studies, in o r d e r to provide convincing evidence

New Beta-Blockers and Clinical Outcomes

to s u p p o r t c l a i m s of i m p r o v e m e n t o v e r s t a n d a r d b e t a blockers. A s t i m e p a s s e s , c o m p a n i e s who h a v e initia t e d a p p r o p r i a t e clinical e v a l u a t i o n p r o g r a m s will r e a p t h e b e n e f i t s . T h u s bisoprolol, celiprolol, c a r v e d i lol, and d i l e v a l o l h a v e all u n d e r g o n e e x t e n s i v e evaluation. U n f o r t u n a t e l y dilevalol has b e e n w i t h d r a w n from t h e m a r k e t d u e to s p o r a d i c h e p a t i c d a m a g e , which w a s a p p a r e n t l y n o t d e t e c t e d in t h e P h a s e I I I studies. T h e r e m a i n i n g c o m p o u n d s p o s s e s s a t t r a c t i v e kinetic a n d d y n a m i c p r o p e r t i e s . T h e r e a r e s t r o n g hints t h a t celiprolol m a y be r e l a t i v e l y less bronchospastic in a s t h m a t i c t h a n r e f e r e n c e b e t a b l o c k e r s , and it has l e s s e r e f f e c t s on l i p o p r o t e i n p a t t e r n s . The n e e d to d e s i g n g o o d c o m p a r a t i v e t o l e r a b i l i t y s t u d i e s a n d also to e x a m i n e t h e i r effects on p r o g n o s i s in c o r o n a r y a r t e r y d i s e a s e r e m a i n s . This is e s p e c i a l l y important when considering the possible relationship b e t w e e n n o v e l a n c i l l a r y p h a r m a c o l o g i c p r o p e r t i e s and clinical o u t c o m e s . T h e r e a r e t h r e e e x a m p l e s to illust r a t e this. F i r s t l y , t h e i m p o r t a n c e of r e d u c i n g h e a r t r a t e in p o s t - i n f a r c t i o n p a t i e n t s is c o n s i d e r e d desirable, and t h e p r e s e n c e of p a r t i a l a g o n i s m m a y nullify t h a t action a t r e s t [12,104]. Secondly, t h e u n d e s i r a b l e changes in l i p o p r o t e i n p a t t e r n s a s s o c i a t e d w i t h propranolol h a v e n o t a l t e r e d t h e benefit o b s e r v e d in p o s t i n f a r c t i o n p a t i e n t s , including d i a b e t i c s [42,99]. F i n a l l y , t h e m a j o r goal of f u t u r e a n t i h y p e r t e n s i v e t h e r a p y is to r e d u c e t h e incidence of c o r o n a r y a r t e r y disease. T h i s h a s n o t y e t b e e n a c h i e v e d [105]. The t a s k facing t h o s e w h o w i s h to i n t r o d u c e i m p r o v e d b e t a b l o c k e r s b a s e d on v a l i d d a t a is a d a u n t i n g one. T h e r e a r e e x a m p l e s w h e r e t h e t a s k has p r o v e d too g r e a t and t h e a g e n t , d e s p i t e e x t e n s i v e e v a l u a t i o n , h a s not b e e n b r o u g h t to m a r k e t .

References

1. Cruickshank JM, Prichard BNC. Beta blockers in clinical practice. London: Churchill Livingstone, 1987. 2. Riddell DA, Schull LG, Frist TH, et al. Experiences with pheochromocytoma in 21 patients: Use of dichloroisoproterenol hydrochloride for cardiac arrhythmias. A n n Surg 1963;157:980-988. 3. Black JW, Stephenson JS. Pharmacology of a new adrenergic beta receptor blocking compound (Nethalide). Lancet 1962;2:311-313. 4. Ablad B, Brogard M, Ek L. Pharmacological properties of H56/28 beta adrenergic receptor antagonist. Acta Pharmacol Toxicol 1967;25 (Suppl. 2):9-40. 5. Dunlop D, Shanks RG. Selective blockade ofadrenoceptive beta receptors in the heart. Br J Pharmacol Chemother 1968;32:201-218. 6. Fitzgerald JD. Perspectives in adrenergic beta receptor blockade. Clin Pharmacol Therap 1969;10:292-306. 7. Vaughan Williams EM. Mode of action of beta receptor

573

antagonists on cardiac muscle. A m J Cardiol 1966;18: 399-405. 8. Rahn KH, Hawlina A, Kersting F, et al. Studies on the antihypertensive action of the optical isomers of propranolol in man. Naunyn-Schmiedebergs Arch Pharmacol 1974;286:319-323. 9. Wilson AG, Brooke OG, Lloyd HJ, et al. Mechanism of action of beta adrenergic receptor blocking agents in angina pectoris: Comparison of action of propranolol with dexpropranolol and practolol. Br Med J 1969;4:399-401. 10. Fitzgerald JD, O'Donnell SR. The antagonism by propranolol and alpha methyl propranolol (ICI 77602) of vascular and cardiac responses to isoprenaline in anaesthetised dogs. Clin Exp Pharmacol Physiol 1978;5:585. 11. Reidenberg M. The quality of clinical research. Clin Pharmacol Ther 1990;47:664-670. 12. Fitzgerald JD. By what means might beta blockers prolong life after acute myocardial infarction? E u r Heart J 1987;8:945-951. 13. 0pie LH. Qualities of an ideal beta adrenoceptor antagonist in comparison of existing agents with a new cardioselective hydrophyllic vasodilator beta adrenoceptor antagonist, celiprolol. A m J Cardiol 1988;61:8c-13c. 14. Prichard BNC, Tomlinson B. The additional properties of beta adrenoceptor blocking drugs. J Cardiovasc Pharmacol 1986;8(Suppl. 4):sl-s15. 15. Lemoine H, Ehle, B, Kaumann AJ. On the participation of B 1 and B2 adrenoceptors in the relaxation of calf trachea caused by ( - ) noradrenaline, ( - ) adrenaline and ( + ) fenoterol. Naunyn-Schmiedebergs Arch Pharmacol 1985; 329:R81. 16. McDevitt DG. The assessment of beta adrenoceptor blocking drugs in man. E u r J Clin Pharmacol 1977;4: 413-425. 17. Wellstein A, Palm D, Beltz GG, et al. Reduction of exercise tachycardia in man after propranolol, atenolol and bisoprolol in comparison to beta adrenoceptor occupancy. E u r Heart J 1987;8(Suppl. M):3-8. 18. Lammers JWJ, Folgering HTM, an Herwaarden CLA. Respiratory tolerance of bisoprolol and metoprolol in asthmatic subjects. J Cardiovasc Pharmacol 1986;8(Suppl. 11):69-73. 19. Chatterjee SS. The cardioselective and hypotensive effects of bisoprolol in hypertensive asthmatics. J Cardiovasc Pharmacol 1986;8(Suppl. 11):74-77. 20. Lithell J, Selinus I, Hosie J, et al. Efficacy and safety of bisoprolol and atenolol in patients with mild to moderate hypertension: A double-blind parallel group international multicentre study. E u r Heart J 1987;8(Suppl. M):55-64. 21. Maltz MB, Dymond DS, Nathan AW, et al. A comparison of once daily bisoprolol (5 or 10 mg.) and atenolol 100mg. in the treatment of angina pectoris. E u r Heart J 1987; 8(Suppl. M):37-42. 22. Bilski AJ, Haliday SE, Fitzgerald JD, et al. The pharmacology of a selective adrenoceptor antagonist (ICI 118551). J Cardiovasc Pharmacol 1983;5:430-439. 23. O'Donnell SR, Wanstall JC. Evidence that ICI 118551 is a potent, highly selective B2 adrenoceptor antagonist and can be used to characterize beta adrenoceptor populations in tissues. Life Sci 1980;27:671-677. 24. Dahlof B, Andren L, Spensson A, et al. Anti-hypertensive mechanism of beta adrenoceptor antagonists: The role of beta-2-blockade. J Hypertens 1983;l(Suppl. 2):112-115. 25. Robb OJ, Petrie JC, Webster J, et al. ICI 118551 does not

574

26. 27. 28.

29.

30. 31. 32. 33. 34. 35. 36. 37. 38.

39. 40.

41. 42.

43.

44.

Fitzgerald

reduce blood pressure in hypertensive patients responsive to atenolol and propranolol. B r J Clin Pharmacol 1985; 19:541p-542p. Vincent HH, Man in'T Veld A. Boomsa F, et al. Is B; antagonism essential for the antihypertensive action of beta blockers? Hypertension 1987;9:198-203. Moran NC, Perkins ME. Adrenergic blockade of the mammalian heart by a dichloro analogue of isoproteronol. J Pharmacol 1958;125:223-237. Black JW. The predictive value of animal tests in relation to drugs affecting the cardiovascular system in man. In: Wolstenholm G, Porter R, eds. Dr~g responses in man. London: J. & A. Churchill Ltd., 1967:113, 121-122. Giudicelli JF, Schmitt H, Boissier JR. Studies on D,L-4(2-hydroxy-3-isopropylamino propoxy)-indole (LB 46). A new potent beta adrenergic blocking drug. J Pharmacol E x p Ther 1969;168:116-126. Shanks RG. Review of a review: Partial agonist activity of beta adrenoceptor blocking drugs. E u r Heart J 1981; 2:253-255. Reale A, Motolese M. Partial agonist activity of beta adrenergic blocking agent and cardiac performance: A review. E u r Heart J 1981;2:245-251. Gibson DG. Pharmacodynamic properties of beta adrenergic blocking drugs in man. Drugs 1974;7:8-38. Weinberger NH. Potential benefit of combination therapy with diuretics and beta blockers having intrinsic sympathomimetic activity. E u r Heart J 1990;11:560-565. Taylor SH. Intrinsic sympathomimetic activity: Clinical fact or fiction? A m J Cardiol 1983;52:16d-26d. Cruickshank JM. Measurement and cardiovascular relevance of partial agonism activity (PAA) involving B1 and B2 adrenoceptors. P h a r m a c Ther 1990;46:199-242. Waller DG. Beta adrenoceptor partial agonists: A renaissance in cardiovascular therapy? B r J Clin Pharmacol 1990;30:157-171. Ariens EJ. Intrinsic activity: Partial agonists and partial antagonists. J Cardiovasc Pharmacol 1983;5:$8-$15. Kaumann AJ, Blinks JR. Stimulant and depressant effects of beta adrenoceptor blocking agents on isolated heart muscle: A positive inotropic effect not mediated through adrenoceptors. Naunyn-Schmiedebergs Arch Pharmacol 1980;311:205-218. Kaumann AJ, Lobnig BM. Mode of action of (-)-pindolol on feline and human myocardium. B r J Pharmacol 1986;89:207-218. Kaumann AJ, Blinks JR. Adrenoceptor blocking agents as partial agonists in isolated heart muscle. Dissociation between blockade and stimulation. N a u n y n Schmiedebergs Archiv Pharmacol 1980;311:237-248. Bilski A, Wale J. Two types of intrinsic sympathomimetic activity with beta adrenoceptor blocking drugs. B r J Pharmacol 1976;56:362p. Man in'T Velt AJ, Van den Neiracker AH, Schalekamp MA. Do beta blockers really increase peripheral vascular resistance? Review of the literature and new observations under basal conditions. A m J Hypertens 1988;1:91-96. Leonetti G, Sampieri L, Cuspidi C, et al. Does B 1 selective agonist activity interfere with the anti-hypertensive efficacy of B 1 selective blocking agents? J Hypertens 1985; 3(Suppl. 3):$243-$245. Conway J. Beta adrenergic blockade and hypertension. In: Oliver MF, Cardiology, 3. London: Butterworth 1975: 376-403.

45. Krupp P, Fanchamps A. Pindolol: Experience gained in ten years of safety monitoring. A m Heart J 1982;104:486-490. 46. Committee on Safety of Medicines. Current problems. 1990;28:1-2. 47. Warren DJ, McSorey P, Niak RB. Effects of beta adrenergic blocking drugs on peripheral blood flow (abstr.) VIII Congress of Cardiology, Tokyo 1978. 48. Marshall AJ, Roberts CJC, Barrit DW. Raynaud's phenomenon as a side effect of beta blockers in hypertension. B r Med J 1976;1:1489-1499. 49. Ohlsson O, Lindell SE. Effects of pindolol and prazozin on hand blood flow in patients with cold extremities and on treatment with beta blockers. A c t a Med Scand 1981; 201:217-219. 50. Feleke E, Lyngstan O, Rastam L, et al. Complaints of cold extremities among patients on anti-hypertensive treatment. Acta Med Scand 1983;213:381-385. 51. Jerm S. The effects of epanolol on quality of life. Drugs 1989;38(Suppl. 2):71-74. 52. Materson BJ, Lucas CP, Valchakis ND, et al. Dilevaiol compared with metoprolol for treatment of hypertension: A multi-centre trial. Dellvalol--the beta blocking vasodilator. Satellite Symposium to the Third European Meeting on Hypertension. 1987, Milan, Italy, abstract 18-19. 53. Schoenberger J, Frishman W, Gorwit J, et al. Comparison of the side effect profile of dflevalol with propranolol. J Cardiovasc Pharmacol 1988;ll(Suppl. 2):$39. 54. Coffman JD, Rasmussen HN. Effects of beta adrenoceptor blocking drugs in patients with Raynaud's phenomenon. Circulation 1985;72:466-470. 55. Greminger P, Vetter H. Boerlin HJ. A comparative study between 100 mg. atenolol and 20 mg. pindolol slow release in essential hypertension. Drugs 1983;25(Suppl. 2):37-41. 56. Louis WJ, McNeil JJ. Beta adrenoceptor blocking drugs: The relevance of intrinsic sympathomimetic activity. B r J Ctin Pharmacol 1982;13(Suppl. 2):317S-320S. 57. Tattersfield AE. Beta adrenoceptor antagonists and respiratory disease. J Cardiovasc Pharmacol 1986;8(Suppl. 4):$35-$39. 58. Ruffin RE, Crockett A J, Alpers JH. The effect ofbucindolol on the airway function of asthmatics. E u r J Clin Pharmacol 1986;30:559-565. 59. Hoffman W, Hoffmann H. Results of the Austrian celiprolol post-marketing surveillance study. J Cardiovasc Pharmacol 1986;8(Suppl. 4):$88-$90. 60. Northcote R J, Todd IC, Ballantine D. Beta blockers and lipoproteins: A review of current knowledge. Scot Med J 1986;31:220-228. 61. van Brummelen P. The relevance of intrinsic sympathomimetic activity for beta blocker induced changes in plasma lipids. J Cardiovasc Pharmacol 1983;5(Suppl. 1): $51-$55. 62. Roberts WC. Recent studies on the effects of beta blockers on blood lipid levels. A m Heart J 1990;112:709-714. 63. Byington RP, Worthy J, Craven T, et al. Propranolol induced lipid changes and their prognostic significance after myocardial infarction: The beta blocker heart attack trial experience. A m J Cardiol 1990;65:1287-1291. 64. Zacest R. Gilimore E, Koch-Weser J. The treatment of essential hypertension with combined v.asodilation and beta adrenergic blockade. N E n g l J M e d 1972;286:617-622. 65. Pedersen OL, Christensen CK, Mikelsen E. Relationship between the anti-hypertensive effect and steady state plasma concentration of nifedipine given alone or in combi-

N e w Beta-Blockers and Clinical Outcomes

66. 67.

68. 69.

70. 71.

72. 73. 74. 75. 76. 77.

78.

79. 80.

81. 82. 83.

84.

nation with a beta adrenoceptor blocking agent. E u r J Clin Pharmacol 1980;18:287-293. Baum T, Sybertz EJ. Anti-hypertensive properties of an isomer of labetalol and other vasodilator beta adrenoceptor blockers. Fed Proc 1983;42:176-181. Lund~lohnsen P. Haemodynamic effects of beta blocking compounds possessing vasodilating activity: A review of labetalol, prizidilol and delivalol. J Cardiovasc Pharmacol 1988;ll(Suppl. 2):512-517. Louis WJ, McNeil JJ. Labetalol. In: Doyle AE, ed. Handbook of hypertension, Vol. 5. Amsterdam: Elsevier, 1984:225-245. Hashimoto H, Asano M, Hayashi T, et al. Haemodynamic effects of arotinolol in anesthetised dogs and its affinity for adrenoceptors in vitro. A r c h I n t P h a r m a c o d y n 1984; 267:23-34. Takenaka T. Amosulalol. In: Scriabine A, ed. Cardiovascular drugs. New York: Raven Press, 1987:117-134. Nichols AJ, Sulpizio AC, Ashton DJ, et al. I n vitro pharmacologic profile of the novel beta adrenoceptor antagonist and vasodilator, carvedilol. Pharmacology 1989;39:327336. Abshagen U. A new molecule with vasodilating and beta adrenoceptor blocking properties. J Cardiovasc Pharmacol 1987;10(Suppl. 11):$23-$32. Pruss TP, Khandwala A, Wolf PS, et al. Celiprolol: A new beta adrenoceptor antagonist with novel ancillary properties. J Cardiovasc P h a r m a c o l 1986;8(Suppl. 4):$29-$32. Gordon R J, Wolf PS, Pruss TP, et al. Further studies in the bronchodilator properties of celiprolol HCL, a cardioselective beta blocker. Pharmacologist 1983;25:121-124. Shirasawa Y, Fugaii M, Nakamura M. Venodilating action of nipradilol (K-351) in pithed rats pre-treated with dihydroergotamine. J p n J Pharmacol 1985;39:77-82. Uchida Y, Nakamura M, Tsuruta T, et al. NipradiloL In: Scriabine A, ed. N e w cardiovascular drugs. New York: Raven Press, 1987:95-115. Van de Water A, Xhonneux R, Reneman RS, et al. Cardiovascular effects of DL-nebivolol and its enantiomers--a comparison with those of atenolol. E u r J Pharmacol 1988;156:95-103. Van de Water A, Janssens W, Van Neuten J, et al. Pharmacological and hemodynamic profile of nebivolol, a chemically novel, potent and selective B 1 adrenergic antagonist. J Cardiovasc Pharmacol 1988;11:552-563. DeCre J, Geukens H, Franken P, et al. Non-invasive cardiac haemodynamics of nebivolol in men. Acta Antwerpiensia 1989;6:2-21. Hunyor SN, Bauer GE, Ross M, et al. Labetalol and propranolol in mild hypertensives: Comparison of blood pressure and plasma volume effects. A u s t N Z J Med 1980;10:162-166. Kubik MM, Coote JH. Propranolol versus metoprolol versus labetalol: A comparative study in essential hypertension. E u r J Clin Pharmacol 1984;26:1-6. Taylor S, Beattie A, Silke B. Celiprolol in the treatment of hypertension: A comparison with propranolol. J Cardiovasc Pharmacol 1986;8(Suppl. 4):127-131. Stumpe K, Kolloch R, Mathieu M, et al. A comparison of celiprolol and atenolol in the treatment of hypertension: A placebo controlled double-blind study. B r J Clin Pract 1985;39(Suppl. 40):73-75. DeCesaris, Ranieri G, Chiarappa R, et al. Comparison of the.anti-hypertensive efficacy of carvedilol, a new vasodi-

85.

86.

87. 88. 89. 90. 91.

92. 93. 94.

95.

96.

97.

98. 99. 100. 101.

102.

575

lating beta blocker, versus atenolol. A double-blind study. Drugs 1988;36(Suppl. 6):102-105. Riddell JG, Shanks RG, Brogden RM. Celiprolol: A preliminary review of its pharmacodynamic and pharmacokinetic properties and its therapeutic use in hypertension and angina pectoris. Drugs 1987;34:438-458. Winniford MD, Huxley RL, Hillus LD. Randomized, double-blind comparison of propranolol alone and a propranolol-verapamil combination in patients with severe angina. J A m Coll Cardiol 1983;1:492-498. Dorow P. General discussion. A symposium: Hypertension into the nineties. A m J Cardiol 1988;61:57C. Norris RJ. Anti-hypertensive therapy with celiprolol: A new cardioselective beta blocker. A m J Cardiol 1988; 61:23C-26C. Schnurr E, Widmann L, Glocke M. Efficacy and safety of carvedilol in the treatment of hypertension. J Cardiovasc Pharmacol 10(Suppl. 11):S101-S107. Bulpitt CJ, Fletcher AE. The measurement of quality of life in hypertensive patients: A practical approach. B r J Clin Pharmacol 1990;30:353-364. Fletcher AE, Bulpitt CJ, Hawkins CM, et al. Quality of life on anti-hypertensive therapy: A randomized double-blind controlled trial of captopril and atenolol. J Hypertens 1990;8:463-466. Croog SH, Levine S, Testa N, et al. The effects of antihypertensive therapy on the quality of life. N E n g l J Med 1986;314:1657-1664. Himms-Hagen J. Sympathetic regulation of metabolism. Pharmacol Rev 1967;19:367-461. Harms HH, Gooren L, Spoelstera A, et al. Blockade of isoprenaline induced changes in plasma free fatty acids, immuno-reactive insulin levels and plasma renin activity in healthy human subjects by propranolol, pindolol, practolol, atenolol, metoprolol and asabutanol. B r J Clin Pharmacol 1978;5:19-26. Lauridsen UB, Christensen NJ, Lyngsoe J. Effects of nonselective and beta-l-selective blockade on glucose metabolism and hormone responses during insulin induced hypoglycaemia in normal man. J Clin Endocrinol Metab 1983;56:876-882. Kolendorf K, Aerenlund Jensen H, Hoist J J, et al. Effect of acute selective beta-l-adrenoceptor blockade on humeral and cardiovascular responses to insulin induced hypoglycaemia in insulin dependent diabetic patients. Scand J Clin Lab Invest 1982;42:69-74. Fuh MM, Sheuw HH, Shen DC, et al. Metabolic effects of diuretic and beta blocker treatment of hypertension in patients with non-insulin-dependent diabetes mellitus. J Hypertens 1990;3:387-390. Houston MC. Treatment of hypertension in diabetes melhtus. A m Heart J 1989;115:820-829. Kjekshus J, Gilpin E, Call G, et al. Diabetic patients and beta blockers after acute myocardial infarction. E u r Heart J 1990;11:43-50. Hitzenberger G. Effects of beta adrenoceptor blocking agent celiprolol on blood lipids. B r J Clin Pract 1985; 39(Suppl. 40):85-87. Pristautz H, Stradner F, Wirkung von celiprolol und metoprolol auf die serumlipide bei patienten mit verschiedenen formen von hyperlipoproteinamie. Weiner Medizin Woch 1986;136:443-450. Fogari R, Soppi A, Pazotti C, et al. Plasma lipids during chronic anti-hypertensive therapy with different beta

576

Fitzgerald

blockers. J Cardiovasc Pharmacol 1989;14(Suppl. 7): $28-$32. 103. Houston MC. New insights and new approaches for the treatment of essential hypertension: Selection of therapy based on coronary disease risk factors, haemodynamic profiles, quality of life and sub-sets of hypertension. A m Heart J 1989;117:911-951.

104. Hjalmarson A, Gilpin EA, Kjekshus J, et al. Influence of heart rate on mortality after acute myocardial infarction. A m J Cardiol 1990;65:547-553. 105. MacMahon S. The effects of anti-hypertensive drug treatment on the incidence of stroke and of coronary heart disease. Clin Exp Hyper-Theory and Pract 1989;All (5,6):807-823.