PRACTICAL THERAPEUTICS Drugs 48 (4) : 549-568 .

1994
0012-6667/94/00 1(}{)549/ S20.OO/0

© AdisInternat ional Umited. All rights reserved.

Choosing the Right ~-Blocker

A Guide to Selection
John R. Hampton
Cardiova scular Medicine, Queen's Medical Centre, Universit y Hospital, No tting ham, England

Contents
Summary . 549
1. Drugs Affecting Adrenoceptors 550
2. The Newer ~ -Blockers . 551
3. ~-Blockers for the Treatment of Hypertension . 552
3.1 The Ma in Hypertension Trials . 553
4. ~-Blockers for the Prevention of Myocardial Infarction 555
4.1 Primary Prevention . . 555
4.2 Secondary Prevention 555
4.3 Early Treatment Trials . 555
4.4 Late Entry Tria ls . . . . 556
4.5 ~- B l ocka d e Afte r Thrombolysis 558
4.6 ~-B lo c kers and Aspirin (Ac etylsalic ylic Ac id ) 559
5. ~- Bl ockers for th e Treatment of Arrhythm ias . . . 559
6. Selection of a ~-Bloc ker fo r the Treatment of Heart Failure . 560
6.1 Possib le Me c hanisms of Action of ~- Bloc kers in the Hea rt . 56 1
6.1.1 ~-Adre nerg ic Receptor Funct ion 561
6.1.2 Toxic Effect of Cate c holamines 561
6.1.3 Wall Motion . 561
6.1.4 Arrhythmias . 561
6.1 .5 Effects on the Renin-Angiotensin System . 562
6.2 Evidence for Individual ~-Blockers . . . . . . . . 562
6.3 Trials of ~-Blockers in Heart Failure . 562
6.3.1 Effect of ~-Blockers on Haemodynamic Paramete rs 562
6.3.2 ~-B lockers and Exerc ise Duration . 563
6.3.3 Effect of ~-Blockers on Mortality in Heart Failure . . 563
6.3.4 ~-Blockers for Ischaemic Heart Disease or Dilated Cardiomyopathy . 564
7. Conclusion . 564

Summary P-Blockers have been in clinic al use for 30 years, and have an accepted role
in (among others) the treatment of high blood pressure, the secondary prevention
of myocardial infarction and the treatment of arrhythmias. Their place in the
treatment of heart failure is currently under investigation. The drugs available in
the 1970s and early 1980s were subjected to intense investigation. A new gener-
ation of P-blockers. including some such as carvedilol and bucindolol, with vas-
odilating properties, is now appearing. As yet these later agents have not been the
subject of large clinical trials.
Clinical practice involves the treatment of individual patients with defined
550 Hampton

dosages of particular drugs. It is, therefore, not acceptable to base practice on

theories derived from the clinical pharmacolog y of a particular drug, on the results
of small trials or on a meta-analysis of results from a number of trials that were
individually inadequate. Clinical practice must follow the results of large-scale
trials in defined populations.
The major trials in hypertension, myocardial infarction, arrhythmias and heart
failure provide the best evidence for the use of individual P-blockers in each of
these clinical situations. In patients with high blood pressure, P-blockers do not
seem to have any particular advantage over other hypotensive agents. In myocar-
dial infarction, relatively late use of a P-blocker undoubtedly reduces fatality,
though the value of early treatment is less clear. p-B1ockers are not powerful
antiarrhythmics, but they do appear to prevent sudden death. Their possible role
in heart failure is perhaps the most interesting current field of P-blocker research.
There are very few comparative studies of p-blockers, and it is difficult to make
precise recommendations. None of the new generation of P-blockers has yet been
used in a trial that is large enough trial for any of them to be accepted for routine
use in preference to older drugs. The use of individual p-blockers, as with any
drug, should follow the results of clinical trials. Propranolol and atenolol have
been studied most intensely in hypertension. For secondary prevention of myo-
cardial infarction, the evidence is best for timoloI. Sotalol is probably the best
antiarrhythmic among the P-blockers. Whether any individual P-blocker is best
for heart failure remains to be seen.

Pronethalol, the first adrenergic /3-receptor
blocking drug in clinical trial sl I ] was shown to re-
lieve angina in 1963.l 2] The hypotensive effect of
pronethalol was not iced in 1964 )3] In the same
year propranolol was shown in a clinical trial to be
a useful agent for the treatment of patients with
high blood pres sure.Ul The antianginal effect of
propranolol was demonstrated in 1965,[5] as was
its effect in reducing fatality after myocardial in-
farction .(6J
Since then, a multitude of /3-blocking drugs
have been developed and this drug class has been
shown to be of use in the treatment of arrhythmias,
portal hypertension, anxiety and tremor, migraine,
schizophrenia, hyperthyroidism and glaucorna.Pl
Subtle, and less subtle, pharmacological differ-
ence s between /3-blockers have been reviewed pre-
viou sly,!S] but practising clinicians find it diffi cult
to decide how to select a particular /3-blocker for a
particular clinical situation.
Thi s review concentrates on the 3 main clinical
conditions in which /3-blocker s are used - hyper-
tension , myocardial infarction and arrh ythmias -
and on cardiac failure , a condition in which interest
in /3-blocker use is increasing. The theme is that a
particular drug can onl y become standard therapy
when its value in a clinical situation has been dem -
onstrated in I or more adequate clinical trial s. Thi s
mean s that new /3-blockers, which have a variety
of ancillary prop erties, must be viewed with cau -
tion until large-scale trial s have been publi shed.
1. Drugs Affecting Adrenoceptors
Adrenoceptors mediate the activity of the endog-
enou s catecholamines noradrenaline (norepineph-
rine) and adrenaline (epinephrine), and some of the
actions of dopamine. Adrenoceptors were initially di-
vided into a and /3: adrenaline was most active at a
sites and isoprenaline (isoproterenol) at /3 sites.
a-Receptors have been subdivided into al and
a 2 on the basis of drug s that stimulate or block
them. nr -Receptors are the postsynaptic receptors
in the smooth muscle of blood vess els. Their stim-
ulation (by methoxamine or phenylephrine) lead s
to vas ocons triction and their blockade (by in-
doramin or prazo sin ) causes vasodilatation. c z-
Receptors are on the pres ynaptic membrane. Their
stimulation lead s to inh ibition of transmitter re-

© Adis Inte rna tiona l Limited. A ll rig hts rese rved . Drug s 48 (4) 1994
Choosing the Right ~-Blocker
lease and vasodilatation. a2-Receptors are stimu-
lated by clonidine and blocked by phentolamine
(which is also an ai-blocker).
~-Receptors are also subdivided: ~l -receptors
predominate in the heart and their stimulation (by
dobutamine and dopamine) causes an increase in
the rate and force of contraction, ~2-receptors are
found in vascular, bronchial and uterine smooth
muscle. Their stimulation [e.g. by salbutamol (al-
buterol) 1causes muscular relaxation. Isoprenaline
stimulates both ~l- and ~2-receptors; adrenaline,
noradrenaline and dopamine are nonselective stim-
ulants of both subclasses of both a- and ~-recep­
tors.
The ~-receptor blockers are either nonselective
and affect both ~ 1- and ~2-receptors (e.g. propran-
olol, oxprenolol, sotalol, pindolol), or are rela-
tively selective for the ~l subclass (atenolol,
metoprolol, acebutolol). Nonetheless, all of these
have some effect on ~2-receptors as well and, al-
though they are sometimes called 'cardioselec-
tive', they are not cardiospecific.
In theory, a ~-receptor agonist or antagonist
could be selected for a particular clinical purpose
from its activity against ~1- or ~2-receptors. Unfor-
tunately, their activity against the receptor type de-
pends on the dosage used and the route of admin-
istration. Thus, a ~2-receptor stimulant such as
salbutamol causes an increased heart rate and a ~I
selective antagonist such as atenolol can still ag-
gravate asthma.
The ~-receptor antagonists can be classified ac-
cording to properties other than their selectivity for
~I- and ~2-receptors . Some (e.g. oxprenolol,
pindolol , acebutolol) have partial agonist activity
(PAA, which is also sometimes described as intrin-
sic sympathomimetic activity or ISA). Some have
membrane-stabilising (local anaesthetic) activity :
propranolol, oxprenolol and acebutolol are exam-
ples of this group. Some are lipid soluble (propran-
olol, oxprenolol, metoprolol, labetalol), while oth-
ers are only water soluble (nadolol) and some are
partially lipid soluble. One of the first developed
group of ~-blockers, labetalol, also has activity
against a-receptors. Individual drugs have other
© Adis International Limited. All rights reserved.
551
specific properties. For example, sotalol has class
III antiarrhythmic activity and the class I antiar-
rhythmic drug propafenone has some ~-blocking
effect.
To a very limited extent these properties may
help in the selection of a ~-blocker. For example,
tremor and anxiety-related symptoms are mediated
by ~2-receptors and, therefore, respond best to a
nonselective ~-antagonist. Migraine seems to re-
spond best to drugs without PAA. Drugs without
local anaesthetic activity are best for topical treat-
ment of glaucoma. These are, however, excep-
tional examples. In general, the selection of a ~­
blocker should be based on the results of clinical
trials and not on theories derived from the pharma-
cological properties of a drug.
2. The Newer ~-Blockers
The drugs considered so far were all available
in the early 1980s. Despite their undoubted value
in the treatment of hypertension and angina and in
the secondary prevention of ischaemic heart dis-
ease, they all - to a greater or lesser extent - cause
unpleasant effects in some people . These effects
include mental and physical lethargy, Raynaud's
phenomenon, worsening of claudication, unpleas-
ant dreams, headache, male impotence and the
masking of hypo glycaemia. There is clearly a need
for ~-blockers with the beneficial effects that seem
to be common to the group, but without what so far
have seemed to be the inevitable unwanted effects .
A new generation of ~-blockers with a variety
of additional properties has now reached the clini-
cal trial stage and some have become available for
use: one review[81 lists 28 such drugs . The most
interesting of these are ~-blockers that have vas-
odilating activity,[9,IOI achieved by a variety of
mechanisms including a-receptor blockade, ~2­
agonism, and a dilator action independent of the (l-
and ~-receptors. Some drugs cause vasodilatation
by a combination of actions (table I).
In addition, some have been developed with
particular aims, such as esmolol, which is given
intravenously and has a very short half-life.
Xamoterol was produced as a partial agonist with
Drugs 48 (4) 1994
552
Table I. Vasodilating P-blockers
Drug Mechanism of
vasodilatation
Labetolol a-Receptor agonist
Medroxolol a-Receptor agonist
Carvedilol a-Receptor agonist
Direct vasodilation
Celiprolol ? a-Receptor agonist
P2-Agonist
Direct vasodilation
Bucindolol ? a-Receptor agonist
P2-Agonist
Dilevalol p2-Agonist
Nebivolol Direct vasodilation
Abbreviations: PAA =partial agonist activity.
a ~-blocking effect at high heart rates and was in-
tended for the treatment of heart failure,!l\] Some
drugs in this group have been found to have a sur-
prising spectrum of activity: for example, car-
vedilol, as well as having a -agonist activity,112] has
been found to be an antioxidant and free radical
scavenger,113] to inhibit vascular smooth muscle
cell proliferationll tl and to reduce infarct size in an
animal model of myocardial infarction.Uvl
While all these ancillary properties may provide
theoretical reasons for selecting a particular ~­
blocker, the lack of importance of different proper-
ties of the earlier drugs (PAA, membrane stabilisa-
tion and so on) must make the clinician cautious.
Comparisons between new and old ~-blockers
have not so far demonstrated that practice supports
theory. Thus, carvedilol, despite its vasodilating
activity, does not seem to be superior to atenolol as
a treatment of high blood pressure.(l6) In a long
term study with repeated invasive measurements,
dilevalol (nonselective ~-blockade plus vasodila-
tation by ~2-receptor agonist) and carvedilol (non-
selective ~-blockade plus ai-receptor blockade)
did not produce a very marked reduction in periph-
eral vascular resistance, and their antihypertensive
effects seemed to be due to a reduction in cardiac
output that was mainly mediated by a reduction in
heart rate.lI?]
Nebivolol (Br-selective blocker plus direct va-
sodilator) has been found to have a similar anti-
© Adis International Limited. All rights reserved.
Hampton
anginal effect to that of atenolol .U'llCeliprolol (~I­
Other properties selective antagonist and ~2-agonist) is more potent
than atenolol and reduces peripheral vascular resis-
tance in hypertension,119] but it seems little better
than its predecessors as an antihypertensive agent
Non-selective
and its adverse effects seem to be similar to those
Selective of propranolol or atenolol. l20]
PAA These comparative studies of new and old ~­
blockers have involved few patients (most examine
Non-selective about 20 patients; a study involving 50 patients is
PAA large in this group). While it remains possible that
Non-selective PAA individual drugs may have minor advantages, there
Selective is little evidence as yet to suggest that the new
drugs constitute a major advance. Certainly none
of them has been exposed to the intensive investi-
gation in very large trials that characterise the ~­
blocker boom of the 1980s. Major alterations in
clinical practice must be based on the sure founda-
tion of unequivocal trial results, and this nearly al-
ways means that many patients must be included.
To select a ~-blocker for the routine management
of patients with hypertension, ischaemic heart dis-
ease, arrhythmias and probably heart failure, we
must inevitably review the result of old, but large,
clinical trials .
3. ~·Blockers for the Treatment
of Hypertension
Perhaps the one thing that is common to all ~­
blockers is the ability to reduce elevated blood
pressure. The precise mechanism by which this oc-
curs is not entirely clear. ~I-Antagonism (preven -
tion of increased rate and force of contraction of
cardiac muscle in response to sympathetic stimu-
lation) is essential, but a characteristic of ~-block­
ade is an increase in peripheral resistance, which
should raise blood pressure.
At least 4 possible mechanisms have been pro-
posed to explain the antihypertensive effect of ~­
blockers and none is of itself entirely convincing.
It has been suggested that there is a central nervous
system effect that diminishes sympathetic outflow,
that the effect depends entirely on a reduction of
cardiac output (which may be a rate effect), that
renin secretion is diminished and that presynaptic
Drugs 48 (4) 1994
Choosing the Right ~-Blocker 553

blockade inhibits neurotransmitter relea se. Thi s
uncertainty about the mode of action means that
surrogate end-points - including the reduction of
blood pressure - cannot be accepted when the aim
of treating an asymptomatic condition like high
blood pressure is simply to reduce complications
and death .
Thi s means that meta-analysis of publi shed tri-
als cannot help a clinician to select a ~-blocker. A
fundamental assumption of meta-anal ysis is that
all the drug s studied, and all the patients included
in the trials, can contribute equally to answering
questions such as: does ~-bl oc kade prolong sur-
vival in hypertension? As we have seen, the ~­
blockers are different and inevitably the groups of
patients included in the variou s trials were differ-
ent. We must look at indi vidual trials for guidance
in specific therapy and are, thu s, immediately
faced with a need for large comparative studies of
different ~-blockers - and there is none .
3.1 The Main Hypertension Trials
Table II lists the main trials[21- 351 in which fa-
tality was a major end-point, and shows the abbre-
viations by which the trials are usually known.
These trials were of 3 main types: those in which
patients given some form of active treatment were
compared with patients given no treatment, those
in which treatment policies were compared, and
those in which 2 or more antihyperten sive drugs
were compared.
In 2 of the largest trial s, the Hypertension De-
tection and Follow-up Programme (HDFP)[231and
the Multiple Risk Factor Intervention Trial
(MRFIT),[261 the aim was to compare different
treatment strategies - management in a special
clinic or by the patient's usual phy sici an. In
MRFIT the control of high blood pressure was only
one of the interventions studied. An attempt was
made to influence all cardiovascular risk factor s.
In most of the trials the aim was to find out
whether the treatment of hypertension was benefi-
cial. The mean s by which blood pressure was con-
trolled was of secondary importance. This is re-
fle cted by the wide variety of drug s used as
first-line treatment. The need for second-line drug s
when the first proved inadequate did not materiall y
affect the answer to the basic question of the value
of treatment. However, the use of varying drug re-
gimen s makes it extremely difficult to disentangle
the contribution of individual drugs to the overall
trial result .
Table II shows that in the first 2 large studies
[Veterans Admini stration (VA) and HDFPj [21 ,23]

Tab le II. The major clin ical trials designed to evaluate the clinical benefits of reduc ing high blood pressure
Reference Year of No . of p-Blocker
study patients first-line second-line
randomised
Veterans Administration (VA)t21,22J 1967 523
Hypertension Detection and Follow-up Programme (HDFP)123J 1979 10940
Oslo Studyl 24! 1980 785
Australian Nat ional Blood Pressure Stud y l251 1980 3427 +
Multiple Risk Facto r Intervention Trial (MRFIT)1261 1982 12866 +
European Work ing Party on High Blood Pressure in the Elderly (EW PHE) 127] 1985 940
International Prospective Primary Prevention Study in Hypertension (IPPPSH)[28! 1985 6357 + (Randomised)
Med ical Research Council (MRC -1)[29! 1985 17354 + (Randomised)
Coope & Warrende~30! 1986 884 +
Heart Attack Prima ry Prevention in Hypertension Research Group (HAP PHy)(31 ! 1987 6569 + (Randomised)
Systolic Hypertension in the Elderly Program (SHEP)(321 1991 4736 +
Swedish Trial in Old Patie nts with Hypertension (STOP)(331 1991 1627 +
Med ical Research Counc il Trial of Treatment in O lde r Adults (MRC-2)1341 1992 4396 + (Randomised)
Symbols: + denotes p-blocker used ; - deno tes not used.

© Ad is Internat ional Umite d . All rights rese rved. Drugs 48 (4) 1994
554
P-blockers were not used at all. In the OsI0[24] and
Australian[25] studies and in MRFIT[26] and Sys-
tolic Hypertension in the Elderly Program
(SHEP)[32] a P-blocker was added if diuretic treat-
ment proved insufficient. Propranolol was used in
the Oslo Study and in MRFIT, and either propran-
olol or pindolol in the Australian Study. Atenolol
was the second line drug in SHEP. In the Coope and
Warrender Study and in the Swedish Trial in Old
Patients with Hypertension (STOP) trial[33] a P-
blocker (respectively, atenolol and a choice be-
tween atenolol, metoprolol or propranolol) was
used as first-line therapy.
Only 4 trials compared drug regimens. In the
International Prospective Primary Prevention
Study in Hypertension (lPPPSH)[28] the objective
was to determine whether the inclusion of a P-
blocker (oxprenolol) in a treatment regimen had
any advantage. A wide variety of other drugs was
permitted to control blood pressure. The aim of the
2 Medical Research Council (MRC) trials was to
compare the effects of a diuretic, a P-blocker (pro-
pranolol in MRC- 1,[29] atenolol in MRC-2)[34]and
placebo. In the Heart Attack Primary Prevention in
Hypertension Research Group (HAPPHY)
trialpll a diuretic was compared with a P-blocker,
but either atenolol or metoprolol could be used, and
there was no intention to compare these 2 drugs .
Added to all these variations on the theme of
hypertension control is, of course, the type of pa-
tient included. The VA121] and Os10[ 241 studies in-
cluded only men. The European Working Party on
High Blood Pressure in the Elderly (EWPHE)[27]
and SHEP[32] trials involved 'elderly' patients, but
their definition of elderly was over 60 years .
STOp[33] and MRC-2 l34] were both for 'elderly'
patients, but included only patients aged over 70 or
65 years, respectively. There were also consider-
able differences between the trials in the level of
blood pressure required for randomisation.
With all these differences between the trials it is
scarcely surprising that the main difficulty has
been to determine whether or not treatment of hy-
pertension as such is beneficial. Different trials
© Adis International Limited. All rights reserved.
Hampton
have given different indications of the reduction in
death, myocardial infarction and stroke.
The STOP trial led to the greatest reduction in
fatality, but the participating centres could use any
I of 4 treatment regimens and it is not at all clear
how many patients were treated with each. The
MRC trials suggested that no mortality benefit re-
sulted from active treatment. Strokes were signifi-
cantly reduced by active treatment in EWPHE,
SHEP, MRC-l and MRC-2 , but not in the Austra-
lian Study. In EWPHE, SHEP and MRC-2 there
was also a significant reduction in coronary events.
It is, however, difficult to establish how these re-
sults related to the drugs used because of the inter-
play between first- and second-line agents .
In the trials comparing P-blockers and diuretics,
such evidence as there is suggests that diuretics are
superior. For example, in MRC-2, diuretic treat-
ment was associated with a lower rate of myocar-
dial infarction (4.2 vs 7.2% in the P-blocker group)
stroke (4.2 vs 5.2%) and vascular death (6.1 vs
8.6%). In IPPPSH,[28] the addition of oxprenolol
to the treatment regimen led to better blood pres-
sure control, but the rates of total death, sudden
death, myocardial infarction and stroke were the
same in the groups given oxprenolol or placebo. It
is perhaps unfortunate that the only trials specific-
ally designed to investigate the place of P-blockers
in a treatment regimen chose to use a drug (ox-
prenolol) which seems to have been least success-
ful of all the P-blockers in reducing death after
myocardial infarction (see section 4.3 and 4.4).
Thus, despite the enormous amount of effort
that has been devoted to the study of blood pressure
control, there is essentially no evidence that helps
the clinician to select a P-blocker. All P-blockers
reduce blood pressure, but in the absence of a direct
comparative study nothing points to the superiority
of any particular drug. The apparent advantage of
the diuretics compared with the P-blockers is a sur-
prise that has not so far been explained.
In practice, clinicians will probably base their
choice of a diuretic or a P-blocker on the particular
unwanted effects that they wish to avoid.l 35]
Drugs 48 (4) 1994
Choosing the Right ~-Blocker
4. ~-Blockers for the Prevention of
Myocardial Infarction
4.1 Primary Prevention
Primary prevention means preventing a first
myocardial infarction in a patient with no previous
evidence of myocardial disease . No true primary
prevention trial of a ~-blocker has been carried out,
and such evidence as there is about the role of ~­
blockers in primary prevention of myocardial in-
farction is derived from the blood pressure trials .
In EWPHE,[27] SHEP,[321 STOp[33] and MRC-
341
2[ - all trials involving elderly patient s - the
reduction in myocardial infarction associated with
active treatment was, respectively, 17, 27, 13 and
19%. Of these trials, only STOP and MRC-2 in-
volved the routine use of a ~-blocker. In MRC-2,
propranolol treatment did not cause any reduction
in myocardial infarction. In HAPPHY, fatal plus
nonfatal myocardial infarction occurred more of-
ten in the ~-blocker than in the diuretic group, but
there was no placebo group to provide a compara-
tor.
Thus, while there is evidence that reducing
blood pressure prevents heart attacks , there is little
evidence that ~-blockers themselves specifically
prevent a heart attack . Further, there is no evidence
to suggest that any single ~-blocker has any partic-
ular advantage.
4.2 Secondary Prevention
Secondary prevention of myocardial infarction
means preventing a further episode in a patient
who has already experienced I or more infarcts .
The first evidence that ~-blockers might prevent
death following myocardial infarction came from
a small , poorly controlled study of propranolol by
Snow in 1965)61 In the 1970s and 1980s there was
an explosion of post-infarction trials, some 65 of
which have been subjected to a detailed meta-
analysis[ 361 that suggested that ~-blockers might
reduce death by about 25%. Such an analysis does
not, however, help the clinician to decide which
patient to treat at what stage of the illness, with
which drug or in what dose. In the absence of direct
© Adis International Limited. All rights reserved.
555
comparative studies of different ~-blockers (there
have been none) we have to look at individual trials
to help decide which patients to treat and what drug
to use. Here, I shall restrict myself to the studies
with more than 100 patients per treatment group .
Following a myocardial infarction, a ~-blocker
could be given as soon as possible, either intrave-
nously or orally, or treatment could be delayed un-
til the acute stage of the illness (and, therefore ,
most of the complic ations) are over. Initially, it was
thought that the effect of a ~-blocker would be pri-
marily antiarrhythmic, but these drugs are not
powerful antiarrhythmic agents. It seems likely
that any effect they have is mediated by some other
mechanism which has yet to be identified . Divid-
ing the studie s into those in which ~-b]ockers were
given 'early' treatment (begun within 2 days of the
acute episode) and those in which treatment was
started 'late' (2 or more days after myocardial in-
farction) is perforce arbitrary, but it does make
clinical sense and corresponds to the decision a
clin ician must make. I shall, therefore, divide the
trials into 2 groups on the basis of when treatment
was begun. That is, the division is between treat-
ment being given as soon as possible[37-44] and
given 'late' )45-571
Figures I and 2 show the percentage change in
fatality associated with active treatment, together
with the 95 % confidence intervals for these
changes, in the 'early' and 'late' trials, respectively.
A wide variety of ~-blockers has been included in
these studies, but it is important to note that they
do not include any of the new generation of ~­
blockers with ancillary properties such as vasodi-
latation.
4.3 Early Treatment Trials
Among the early treatment trials, relatively few
patients were included and the confidence intervals
for the trial results are so wide as to preclude much
chance of demonstrating a clear effect (fig. I) . The
first trial to demonstrate a significant reduction in
mortality with a ~-blocker involved 1395 patients
and intravenous metoprolol.I'Sl Mortality was
5.7% in the metoprolol group and 8.9% in the pla-
Drugs 48 (4) 1994
556 Hampton

The case for using ~-blockers as soon as possi-

Propr anolo (l37) :
ble after the onset of a myocardial infarction is,
Pracl olo l!381
therefore, hardl y con vincing, but it is best for aten-
Alpreno loll391 I
0101 and metoproloI.
Ox prenolol!401

Atenolo il" 1
4.4 Late Entry Trials
Propra nolol 14l1

Metoprolol l• 21
Figure 2 shows the percentage changes in fatal-
Meloprolol l' 3)

Ate no lol l" 1 m. ity with the associated 95 % confidence intervals

resulting from ~-blocker treatment begun more
than 48 hours after the onset of myocardial infarc-
- 100% 0% 100%
Decrease in mortality Increase in mortality
tion. There have been 13 trials[45-57] with mortality
as a main end-point which included more than 200
Fig. 1. Results of the main trials in which ~-blockers were pre- patients.
scribed within 48 hours of the onset of myocardial infarction . The
bars represent the 95% confidence interval for the percentage The pattern of result s is more clear than in the
reduction in deaths; the figure within the bars gives the actual earl y entry trial s. In only 2 of the late trial s was
trial result. active treatment associated with an increased fatal-
ity and in both the 95 % confidence interval s
included zero effect, indicating a non significant
cebo group. A second study of intravenous
difference from placebo. Four trials - with
metoprolol, Metoprolol in Acute Myocardial In- practolol,[46] timolol,l48] propranololl'Pl and
farction (MIAMI)[43] found a non significant re- acebutololl-" ! - demon strated that ~-blockers
duction in death s. However, the 2 metoprolol stud- cau sed a significant reduction in fatalit y. The re-
ies are clearl y not comparable because, although mainder showed a non significant bene fit in asso ci-
metoprolol was used in the same way, a lower risk ation with ~-blocker treatment.
patient group was included in the second study. The
death rate in MIAMI was 4.3 % in the metoprolol
group and 4.9% in the placebo group. Furthermore,
Aipreno lo(l' 5' ~"J._ _' "
the studies differed in duration, at 90 and 15 days,
Pracloloil' 6j
respectively. Propranolol 1"1
The largest of all the ~-blocker postinfarction Timo lol '4Bj
Proprano lol l• 9 ]
trials was the First International Study of Infarct Oxprenolol lSOi
Survival (ISIS-I)[44] which recruited 16027 pa- Oxp renolol l" )
Sola lol 152i
tient s. Active treatment was intravenous followed Proprano lol lS>i
by oral atenolol, but the main trial end-point was
Oxp renolol l" l
after onl y 7 days of treatment. Again, a low risk Melop rolo1lSS)
group of patients was included. The death rate in Praclolol 1561
Acebula loil 57)
the atenolol and placebo groups , respectively, was
3.9 and 4.6%. Thi s 15% redu ction in fatality asso- - 100% 0% 100%
Decrease in mortality Increase in mortality
ciated with atenolol treatment j ust achieved con-
ventional level s of statistical sig nifica nce. The Fig. 2. Results of the main trials in which ~-blocker treatment
main benefit seemed to be in the first day or two , was begun more than 48 hours after the onset of myocardial
infarction. The bars represent the 95% confidence interval for
possibly because of a reduction in myocardi al rup- the percentage reduction in deaths ; the figure within the bars
ture . gives the actual trial result.

© Ad is International Lim ite d . All rig hts reserve d . Drug s 48 (4) 1994
Choosing the Right ~-Blocker
Total patient population
(no. 01 screenings)
Total no . of patients In tr ial
No. 01
patients
In different
grou ps '----,----' <---,------' '---,------' L..--,----'
No. 01 deaths
% mortality 10% 11% 16% 16%
Total
mortality
Fig. 3. Results of the Norwegian trial of timolol following myo-
cardial infarction.14B) Patients aged 20 to 75 years were given
timolol 10mg twice daily from 7 to 28 days over 17 months.
Figure 2 makes it clear that no ancill ary prop-
erty of the ~-blocking drugs is important for pre-
ve nting death after myocardial infarction .
Practol ol (withdrawn because of specific adverse
effects) is a cardioselective agent with PAA. Pro-
pranolol and timolol are both non selective.
Acebutolol is a non selective blocker with moder-
ate PAA. Sotalol, with class III anti arrh ythmic ac-
tivity, did not cause a significant reduction in fatal-
ity. Although oxprenolol seems to have been the
lea st success ful drug among tho se studied, thi s is
clearly not due to its PAA, otherwise acebutolol
would have been equally ineffective. Pure selectiv-
ity without PAA (e.g. metoprolol) clearl y does not
carry any particular benefit. The only conclusion
that can be dr awn from this group of trial s is that
prevention of de ath after myo cardi al infarction is
a cl ass effect common to all ~-blockers . Th e prob-
lem is then to decide how to se lect whic h dru g to
use.
© Adis Internationa l Lim ited. All rig hts reserved .
557
The most clear-cut result wa s obtained in the
Norwe gian Timolol Tri al 1481 and fig ure 3 di spl ays
the results of thi s tri al in a format whi ch permits
both intention-to-treat and treated pati ent analysis .
There was a rel ati vel y high fatality ra te (16.2%) in
the placebo gr oup. Thi s wa s the same whether
tre atment was continued or withdrawn. In the
group wh o were giv en and continued to take
timolol, the fatality rate wa s 10%. In tho se with-
drawn fro m tim olol, the fatality rate wa s II %. By
intention-to-treat analysi s, the fatality rate on
timolol wa s 10.4 %, giving a 36 % relative reduc-
tion in fat ality.
Propranolol wa s significantly effective in the
Beta-Blocker Heart Attack Trial (BHAT),1 491 in
which the pla cebo fatality rate was 9.8%. This in-
dicates that ~-blockade is effecti ve in relatively
lo w risk as well as rel ati vely high risk patients. The
aim of the ace buto lol studyl571was in fact to select
high risk patients, but thi s proved un succe ssful, the
fatality in the placebo group being II %.
It is perhaps important to note that the drug most
obvious ly succ essful in the earl y entry trials, aten-
0101, has not been used in a late tre atment trial.
Furthermore, there are no trials of the new gener-
ation of ~-bl ockers with va sodilator effec ts. Th e
choice of the age nt which should be used for sec -
ondar y pre vention following myocardial infarc-
tion , therefor e, rests bet ween timolol and propran-
olol.
In the BHAT Study, the use of propranolol wa s
varied according to blood concentrations of the
drug, which is clearly not a practical proposition
for routine clinical care. In the propranolol study
of Hansteen,l5 31 a fixed dose was used and a sig-
nificant reduction was seen in sudden death, but
not in tot al mortality. The do sage regimen in the
timolol study wa s si mple and ea sily reproducible
(5mg twi ce daily increasing to IOmg twi ce daily ).
~-Blockers clearly do prevent death after myo-
cardial infarct ion, and the evidence is be st for a
relatively late start of treatment. Th e effect may
well be co m mo n to all drugs in the ~-blocker
gro up, but the results are mo st clear-cut and the
do sage regimen is best established for tim olol.
Drugs 48 (4) 1994
558
4.5 P-Blockade After Thrombolysis
All the postinfarction trials, without exception,
predated the widespread use of thrombolytic
agents. Since the value of thrombolysis has been
demonstrated, only 2 studies have specifically in-
vestigated the value of ~-blockers in patients with
acute myocardial infarction who have been treated
with a thrombolytic agent.
Phase 2 of the Thrombolysis in Myocardial In-
farction (TIMI) trial was primarily designed to
investigate the value of early percutaneous trans-
luminal coronary angioplasty (PTCA) after throrn-
bolysis.(58) However, a subgroup of 1390 of the
total 3262 patients in the trial were randomly as-
signed to receive intravenous metoprolol15mg im-
mediately followed by oral metoprolol, or to re-
ceive oral metoprolol from day 6. The death rates
at 6 days, in hospital, and at 42 days were similar
in the 2 groups. However, in the first 6 days there
were fewer nonfatal reinfarctions among the pa-
tients given intravenous metoprolol (16 vs 31 in the
group not given metoprolol until day 6), fewer
ischaemic episodes (I07 compared with 147), and
the difference between groups in terms of these
combined end-points was significant.
In the second study,[59] the effect of intravenous
followed by oral atenolol was compared with pla-
cebo treatment in 194 patients with acute myocar-
dial infarction who were treated with a thrombo-
lytic agent. There was no significant difference
between the groups in coronary artery patency,
ejection fraction, measures of infarct size, arrhyth-
mias or clinical events, except that in the atenolol
group there was an increased incidence of pulmo-
nary oedema.
The main thrombolytic trials do little to support
the conclusion that the effects of thrombolysis and
~-blockade are additive. This is because no attempt
was made specifically to study the value of ~­
blockers.
In the second International Study ofInfarct Sur-
vival (ISIS-2),[60] intravenous ~-blockers were
recommended and a record was made of whether
the use of these drugs in each patient was planned.
No record was kept, however, of whether the pa-
© Adis International Limited. All rights reserved.
Hampton
tients actually received this planned treatment.
Subset analysis showed that in the group given
streptokinase, the vascular fatality rate at 5 weeks
was 8.4 % among 559 patients in whom the use of
~-blockers was planned, compared with 9.3% in
the group of 8035 patients in whom the use of ~­
blockers was not planned. Comparable rate s for the
group given placebo instead of streptokinase were
9.7 and 12.1%. Thus, as might be expected, a ~­
blocker seemed to reduce further the lower fatality
rate associated with streptokinase treatment, but
not to the same extent as occurred in the patients
who had not been given streptokinase.
In the Italian Study of streptokinase, Gruppo
Italiano per 10 Studio Della Streptochinasi
Nell'Infarto Miocardico (GISSI-l),[61] ~-blockers
were only given to 8% of patients during the hos-
pital stay and the later use of these drugs is not
recorded . In the Anglo-Scandinavian Study of
Early Thrombolysis (ASSET),[62] a study of re-
combinant tissue plasminogen activator (rtPA, al-
teplase), over 40% were being treated with a ~­
blocker by 6 months, but the early use of these
drugs was not recorded. In GISSI-2,!63] a compar-
ison of streptokinase and rtPA, intravenous aten -
0101 was given in over 40 % of patients, and in the
GUSTO (Global Utilization of Streptokinase or
rTPA for Occluded Coronary Arteries) comparison
of these agents,[64] ~-blockers were given intrave-
nously to 46 % of the patients and oral ~-blockers
to 71 %. In none of these studies, however, was a
comparison made between patients who did or did
not receive a ~-blocker.
In the ISIS-4 study of magnesium, nitrates and
ACE inh ibitors in acute myocardial infarction,
there was a marked variation in clinical practice
between the participating centres (Collins, per-
sonal communication). In the patients treated in the
UK, only 2% were given an intravenous ~-blocker,
compared with 7% in the rest of Europe, 8% in
Canada and 25% in the US.
The beneficial effects of ~-blockers and throm-
bolytic agents in acute myocardial infarction pre-
sumably result from totally different mechanisms,
and an additive effect would, therefore, be ex-
Drugs 48 (4) 1994
Choosing the Right ~-Blocker
pected. On the other hand, there must be a lower
limit to the fatality rate that can be achieved. Such
evidence as there is consistent with a small addi-
tional benefit in the short term when intravenous
~-blockers are combined with thrombolytics, but
there is essentially no evidence of their long term
effect in patients given a thrombolytic agent at the
time of the acute infarction. Furthermore, there is
no evidence of the superiority of any one ~­
blocker, though most of our knowledge is based on
the use of atenolol and metoprolol.
4.6 ~-Blockers and Aspirin
(Acetylsalicylic Acid)
ISIS-2[60] was designed also to study the use of
aspirin. The planned use of ~-blockers appeared to
give a small additional benefit to aspirin treatment.
In patients given aspirin, the fatality rate was re-
duced from 9.5 to 8.0% by the planned addition of
a ~-blocker, and in patients not given aspirin it was
reduced from 11.9 to 10.1 %. It must be empha-
sised, however, that these results were obtained by
subset analysis at the end of the study and, there-
fore , cannot be accepted as final proof of the addi-
tive effect of ~-blockers to either streptokinase or
aspirin.
The long term postinfarction studies of ~-block­
ers were conducted before the use of aspirin was
widespread. It is not known if ~-blockers and as-
pirin have an additive effect over the long term.
5. ~-Blockers for the Treatment
of Arrhythmias
~-blockers are not powerful antiarrhythmic
drugs, but they are the only drugs that reduce fa-
tality from sudden death which is presumably usu-
ally due to the most important arrhythmia, ventric-
ular fibrillation. Thus, while there is convincing
evidence that ~-blockers reduce the incidence of
sudden death following myocardial infarction, the
Cardiac Arrhythmia Suppression Trial (CAST)[65]
showed that the proarrhythmic effect of the class
IC antiarrhythmic drugs flecainide and encainide
outweighed any antiarrhythmic benefit there might
have been. Sudden death was reduced in the pro-
© Adis International Limited. All rights reserved.
559
pranolol study of Hansteen,[ 531 where it was a de-
fined end-point and this was a consistent finding
of subgroups in the other trials. For example, in a
3-year postinfarction study with metoprolol.l'vl
14.7 % sudden deaths occurred in the placebo
group compared with 5.8% in the patients given
metoprolol.
There is reasonable physiological evidence to
support the concept that ~-blockers might prevent
ventricular fibrillation . Sympathetic stimulation
reduces and ~-blockade increases the ventricular
fibrillation in open-chest dogs.l 67]
Although ~-blockers are not usually considered
first line drugs for treating arrhythmias such as sus-
tained ventricular tachycardia, simple metoprolol
treatment has been shown in a randomised trial to
be superior to the best antiarrhythmic therapy that
could be determined by serial electrophysiological
testing.l 68]
Most of the large studies of ~-blockers as anti-
arrhythmic agents have been conducted in the set-
ting of acute myocardial infarction and the findings
have not been consistent. Intravenous atenolol has
been found to reduce R on T extrasystoles and ven-
tricular arrhythmias.l 69,70] Metoprolol has been
found to have no effect on arrhythmias on the fir st
day of a myocardial infarction, but to have a sup-
pressive effect later.[7)] However, in another study
of metoprolol.Fel the drug was found to have no
effect on recorded arrhythmias even though it was
associated with a reduction in documented ventric-
ular fibrillation . Propranolol and atenolol have
been found to have no effect on arrhythmias de-
tected by Holter monitoring at I and 6 weeks fol-
lowing myocardial infarctionJ73] Oxprenolol has
also been found to have little effect on postinfarc-
tion arrhythmiasJ74]
Of all the ~-blockers , the one that seems to have
most antiarrhythmic effect is sotalol, which is dis-
tinguished by having class III antiarrhythmic ac-
tivity as well as being a nonselective ~-blocker
without PAA. Sotalol increases the duration of the
action potential and prolongs the absolute and rel-
ative refractory periods of the atria, ventricles,
atrioventricular (AY) node and accessory path-
Drugs 48 (4) 1994
560
ways.[75] Sotalol has been shown to be as effective
as quinidine in maintaining sinus rhythm after car-
dioversion of atrial fibrillationl/vl and is recom -
mended for the prophylaxis of paroxysmal atrial
fibrillation.l 77 ] In an electrophysiological study it
has been shown to prevent the induction of arrhyth -
mias in patients with pre-excitation syndromes and
it is useful in preventing paroxysmal tachycardia
in such patients.l 78l
In the Electrophysiological Study versus Elec-
trocardiographic Monitoring (ESYEM) trial,l79]
imipramine, mexilitine, pirmenol, procainamide,
propafenone, quinidine and sotalol were compared
in patients with ventricular extrasystoles. Serial
testing was performed either by electrophysiolo-
gical techniques or by Holter monitoring and pa-
tients were treated with the first drug found to be
effective. Sotalol was more effective than the other
6 drugs in preventing death and recurrence of ar-
rhythmias .
The main competitor of sotalol as a broad-
spectrum antiarrhythmic agent is amiodarone.
However, amiodarone has undesirable effects on
the liver, thyroid and lung and is, therefore, usually
used as a drug of last resort. Sotalol is now the
preferred drug for the prevention of paroxysmal
ventricular tachycardia.l'vl but it cannot be seen as
a totally safe drug. It prolongs the QT interval and,
therefore, has the ability to induce torsade de
pointes ventricular tachycardia, just as can
amiodarone.U'U In 853 patients collected from a
series of studies, sotalol was found to cause serious
proarrhythmic events (usually torsade de pointes)
in 30 patients (3.5%).l 82l This proarrhythmic pro-
pensity may explain why sotalol is no more effec-
tive in reducing death following myocardial infarc-
tion than any other ~-blocker.
Thus , the prevention of ventricular fibrillation
following myocardial infarction is probably a class
effect, common to all ~-blockers, though this has
been most convincingly demonstrated with pro-
pranolol. Otherwise, sotalol is probably the best
~-blocker to use as an antiarrhythmic agent.
© Adi s International Limited. All rights reserved.
Hampton
6. Selection of a ~-Blocker for the
Treatment of Heart Failure
An improvement in heart failure with propran-
olol and alprenolol was first described by Waag-
stein in 1975,[831 but even after nearly 20 years,
~-blockers are by no means a standard way oftreat-
ing heart failure. However, evidence is increasing
that they have a useful role. With the introduction
of ~-blockers with vasodilating properties and with
the realisation that heart failure is common and car-
ries a very poor prognosis, it seems increasingly
likely that ~-blocker treatment for heart failure will
become common, if not routine .
~-blockers are probably now in the position that
ACE inhibitors were 4 or 5 years ago and the cli-
nician will need to review the evidence of the value
of ~-blockade for heart failure by the same stand-
ards as are used to judge their role in hypertension,
myocardial infarction or the treatment of arrhyth-
mias. Theoretical reasons why ~-blockers should
be useful in heart failure are not sufficient to adopt
them into clinical practice, nor are small studies -
and particularly so if they were not randomised and
double-blind.
It is essential to remember that haemodynamic
parameters such as heart rate, blood pressure, car-
diac output and ejection fraction correlate poorly
with a patients' symptoms. Inducing haemodyna-
mic improvement may well not improve exercise
time or quality of life. Similarly, symptoms and
survival are not well correlated, as was shown by
the ability of drugs such as milrinone f84] to im-
prove exercise ability, but to cause an excess of
deaths .
Haemodynamic studies, which are usually short
term, are relatively easy to perform in patients with
heart failure . A group of20 or 30 patients is usually
sufficient to give a significant result. Exercise test-
ing is much more difficult. Patients vary consider-
ably over a period of weeks or months , with spon-
taneous improvement and with changes in
concomitant medication. Heart failure carries a
poor prognosis and it must be expected that in a
3-month study several patients will die, making the
statistical handling of data very difficult. 200 to
Drugs 48 (4) 1994
Choosing the Right I3-Blocker
300 patients are likely to be needed for an exercise
study to give a clear result. If we assume that a
treatment such as a ~-blocker might reduce fatality
in heart failure by 25%, the number of patients
needed in a trial will depend on their clinical state
(the worse the risk, the fewer patients needed) but
roughly 2000 to 3000 patients will have to be in-
cluded.
Determining whether ~-blockers really do have
a beneficial effect in improving symptoms or pro-
longing survival has been made even more difficult
by the clear demonstration that ACE inhibitors do
both. Thi s means that any P-blocker study now has
to be designed on the basis of all patients either
being on an ACE inhibitor or at least having been
shown to be intolerant of ACE inhibitor. The
chances of a ~-blocker leading to further improve-
ment will be much less than the chances that they
might help a patient who is not on an ACE inhibi-
tor.
All the publi shed studie s of P-blockers in heart
failure must be regarded with the following prin-
ciples in mind. Were the stud ies designed well
enough and large enough to give a credible an-
swer? Were the patients treated with ACE inhibi -
tors ?
6.1 Possible Mechanisms of Action of
~-Blockers in the Heart
Activation of the sympathetic system and an in-
crease in circulating catecholamines is a proper
and useful circulatory response to heart failure. It
is a common clinical experience that the introduc-
tion of a ~-blocker (e.g. for the secondary preven-
tion of myocardial infarction) may induce heart
failure . However, we now accept that activation of
the renin-angiotensin system - also entirely appro-
priate in mild or preclinical failure - can reach the
point where vasoconstriction and sodium retention
are totally inappropriate. The assumption is that
the sympathetic system may also become inappro-
priately overactivated with excess vasoconstric-
tion and that treatment with a ~-blocker would pro-
tect the circulation from these excessive effects.
There are many possible ways in which exces s
© Adis International Limited. All right s reserved.
561
sympathetic activation could be harmful, but it is
not clear which, if indeed any, are important.
6.1.113 -Adrenergic Receptor Function
P 1- Receptors are downregulated in the failing
heart and the response to p-agonist drugs is atten-
uated .18S] Associated with this is a depletion of
myocardial noradrenaline (norepinephrine). In
theory, long term treatment with a P-blocker can
allow upregulation of receptors and restore the
heart to an appropriate response to ~-stimulation .
Preci sely why this should be benefi cial if the
receptors remain pharmacologically blocked is not
clear. Furthermore, it might be thought that with-
drawal of P-blockers after prolonged treatment
would lead to at least a temporary clinical improve-
ment and this does not seem to be the case.!86]
6. 1.2 Toxic Effect of Catecho/amines
In animal models , high concentrations of cate-
cholamines cause necrosis of myocardial cells and
P-blockers are protective. Whether this applies to
the failing human heart we do not know.!87] Plasma
noradrenaline levels are a marker of the severity of
heart failure and sympathetic activation can be de-
tected before clinical heart failure is apparent.
Bucindolol treatment has been associated with a
marked fall in plasma noradrenaline,188] but
whether this has a primary effect (by reducing
spillover from synaptic clefts or by increasing nor-
adrenaline clearance) or a secondary effect due to
improvements in heart failure is not clear.
6. 1.3 Wall Motion
It has been proposed that P-blockers improve
either or both systolic and diastolic ventricular
function. P-Blockers might be expected to have an
anti-ischaemic effect and, therefore, improve sys-
tolic function and possibly diastolic function as
well. However, the most powerful agent for im-
proving myocardial relaxation is isoprenaline, so
it would be surprising if P-blockers with their neg-
ative lusitropic (impairing relaxation) effect im-
proved heart failure by this rnechanism.l'''"
6. 1.4 Arrhythmias
Sudden death is common in patients with heart
failure (reported in various series to be between 4
Drugs 48 (4) 1994
562
and 86% of deaths) and significant arrhythmias
also occur (25 to 71% of patients in various studies
have been shown to have ventricular tachycar-
dia).f 871 The postinfarction trials showed that at
least in this clinical situation P-blockers reduced
sudden death. Subgroup analysis of some of these
trials, particularly the BHAT propranolol study,
have shown that the greatest reductions in fatality
were among patients with mild or moderate heart
failure.f 90) It is not clear whether this is a result of
preventing ventricular fibrillation or whether mor-
tality reduction comes from some other mechanism
involving improvement in heart failure .
6.1.5 Effects on the Renin-Angiotensin System
Stimulation of the renin-angiotensin system in
heart failure is in part due to increased sympathetic
tone, and this might be inhibited by P-blockers.
6.2 Evidence for Individual ~-Blockers
There are, then, plenty of mechanisms by which
P-blockers could improve heart failure . There is
nothing that can be gleaned from such theories that
would suggest one P-blocker might be better than
another and it is here that an additional important
benefit might be obtained from some non-f-block-
ing activity such as vasodilatation.
6.3 Trials of ~-Blockers in Heart Failure
Table III lists the P-blocker studies which were
double-blind and placebo-controlledJ90-IOI )
There have been a variety of noncomparative stud-
ies, but these can now be regarded as of mainly
historical interest. Publications involving 10 pa-
tients or fewer and those currently available only
in abstract form have been excluded. Metoprolol
has been the most frequently studied drug, presum-
ably because the initial findings of Waagstein.f 83)
Acebutolol and labetalol are represented once
each, but then interest has clearly shifted to the
vasodilating P-blockers. There have been 4 studies
of bucindolol, and I each with carvedilol and
nebivolol, although more studies with each of these
have been presented in abstracts.
© Adis International Limited. All rights reserved.
Hampton
The overall pattern of these results seems to be
that (where reported) symptoms and ejection frac-
tion have improved. There seems to be a tendency
suggesting that long term treatment is more effec-
tive than short term treatment. So far as it goes, all
these drug studies seem to have similar effects.
Bucindolol does not stand out as more effective
than metoprolol. The exception is a single study of
alprenolol that suggests that this drug caused harm
rather than benefit. Comparative studies of differ-
ent P-blockers are, however, almost nonexistent.
6.3.1 Effect of!3-Blockers on
Haemodynamic Parameters
Improvement in ejection fraction seems an al-
most universal finding after P-blocker ther-
apy,[102,104) but a few individual studies are worth
considering. Metoprolol (in a noncomparative
study, not included in table 1)[105) has been shown
in patients with dilated cardiomyopathy to increase
ejection fraction , decrease left ventricular diastolic
dimensions, reduce mitral regurgitation, decrease
pulmonary capillary wedge pressure and (surpris-
ingly) systemic vascular resistance. This was ac-
companied by a rise in blood pressure. Carvedilol,
in a short term noncomparative study,(106) de-
creased heart rate, blood pressure and systemic
vascular resistance and increased ejection fraction .
Bucindololl'i''l increased cardiac index, decreased
pulmonary capillary wedge pressure and increased
ejection fraction, but in another study(95) decreased
ejection fraction . Nebivolol increased stroke vol-
ume and, although there were increases in systemic
vascular resistance and pulmonary capillary wedge
pressure, these were not significantly different
from zero change.f 102)
In a comparison of metoprolol and car-
vedilol,(107) metoprolol increased mean pulmo-
nary artery pressure and slightly decreased cardiac
index, but had little effect on other parameters ex-
cept heart rate . Carvedilol, on the other hand, had
less effect on heart rate, but caused a marked re-
duction in pulmonary capillary wedge pressure and
systemic vascular resistance, and a smaller reduc-
tion in mean pulmonary and systemic arterial pres-
sures .
Drugs 48 (4) 1994
Choosing the Right ~-Blocker 563

Table III. Double-blind studies of ~-blockers in heart failure

Author and year Design No. of Aetiology Drug Effects on
patients symptoms ETT EF
Ikram (1984)1911 C 28 0 Acebutolol ±
Currie (1984)[921 C 10 Oil Metoprolol ±
Engelmeier (1985)1931 PIC 25 0 Metoprolol + + +
Anderson (1985)[941 P 50 0 Metoprolol + + +
Pollock (1990)1951 P 19 Oil Bucindolol + + +
Leung (1990)[961 C 12 0 Labetolol + +
Gilbert (1990)1881 P 24 0 Bucindolol +
Woodley (1991)1971 P 50 Oil Bucindolol + ± +
Fisher (1991)1961 P 35 I Metoprolol + +
Paolisso (1992)[991 C 10 ON Metoprolol + + +
Olsen (1992)II00J P 54 0 Carvedilol + +
MOC (1993)[101 1 P 380 0 Metoprolol + +
Wisenbaugh (1993)[1021 P 24 Oil Nebivolol +
Abbreviations and symbols: C = crossover; 0 = dilated cardiomyopathy; EF = ejection fraction; ETT = exercise tolerance; I = ischaemic;
MOC = Metoprolol in Dilated Cardiomyopathy Study; P = parallel group; V = valve disease; + denotes has an effect; ± denotes may have an
effect; - denotes has no effect.

6.3.2 ~-Blockers and Exercise Duration
Table III shows that in the majority of studies
examining this factor, exercise time was improved
during ~-blocker therapy. There are, however, ex-
ceptions, with alprenolol causing a reduced exer-
cise time albeit in a study of short duration.l 91]
Although there is a possibility of publication bias,
metoprolol seems consistently effective in improv-
ing exercise duration . What is perhaps surprising
is that the results with bucindolol are not consis-
tent, with improvement seen in 2 studies,195,97] but
not in a third .188] In a single study of nebivolol.l 102]
exercise time was not increased .
As I have already pointed out, all these studies
are very small and, given the variability of exercise
testing, it would not be surprising for the individ-
ual studies to give misleading results. However,
there is nothing in these publications to suggest a
particular benefit from anyone drug compared
with another.
6.3.3 Effect of~-Blockers on Mortality in
Heart Failure
Only 2 trials, the Metoprolol in Dilated Cardio-
myopathy (MDC) study(lOI) and the study of
xamoterol in severe heart failure ,1108] are any-
where near large enough to give information about
the effect of a ~-blocker on mortality in patients
with heart failure .
The MDC studyllOl] was relatively small (383
patients) and was designed to study a combined
end-point of fatal plus nonfatal events. Patients
could be in relatively mild heart failure (NYHA
class II to III, ejection fraction up to 40%), but
about 80% were simultaneously treated with an
ACE inhibitor. Metoprolol was initially given as
5mg twice daily and increased slowly to 100 to
150mg daily. The trial failed to achieve a statisti-
cally significant reduction in its defined primary
end-points of all-cause fatality plus a need for car-
diac transplantation. There was a 34% reduction in
this combination [95% confidence interval (CI) -6
to +62%; p = 0.06) with 23 deaths and 2 patients
needing transplantation in the metoprolol group,
compared with 19 deaths and 19 needing trans-
plantation in the placebo group. The 'need for
transplantation' as a surrogate for death adds a new
complexity to the interpretation of clinical trials .
Still, the results of this trial cannot be claimed to
be sufficiently promising to necessitate a wide-

© Adis International Limited. All rights reserved. Drugs 48 (4) 1994

564
spread change in the clinical management of heart
failure.
In the xamoterol study,' 108] 516 patients with
severe heart failure (NYHA III to IV) despite treat-
ment with diuretics and ACE inhibitors were
randomised in a double-blind fashion to treatment
with xamoterol 200mg twice daily or placebo.
There was some suggestion of an improvement in
symptoms, but not in signs , and there was no im-
provement in exercise duration in the xamoterol
group. Xamoterol reduced heart rate and blood
pressure , but did not effect arrhythmias. On an in-
tention-to-treat analysi s there were 39 deaths in the
xamoterol group (9.1%) and 6 (3.7%) in the pla-
cebo group. This difference was statistically differ-
ent and led to the premature discontinuation of the
trial, and the withdrawal of the drug.
Xamoterol is an unusual ~-blocker in being a
partial agonist at low heart rates. It could be argued,
therefore, that the increase in mortality seen in this
study would be particular to xamoterol and un-
likely to be seen with any other ~-blocker. In the
absence of further trials with other ~-blocking
drugs, however, this is not necessarily a safe as-
sumption to make.
6.3.4 ~-Blockers for Ischaemic Heart Disease or
Dilated Cardiomyopathy
The neuroendocrine responses to heart failure
are the same whatever the underlying cause. It
might, therefore , be assumed that a treatment effec-
tive in patients with ischaemic disease would also
be effective in those with dilated cardiomyopathy.
This is true in the case of ACE inhibitors. On the
other hand, the pathology of the heart and presum-
ably its metabolism is different in the 2 circumstan-
ces with ischaemic hearts having a mixture of nor-
mal (albeit partly ischaemic) muscle and fibrosis,
and cardiomyopathies having a generalised abnor-
mality. The problem is further complicated by the
simple fact that many patients clinically diagnosed
with cardiomyopathy have underlying (or coexist-
ing) coronary disease. Detailed investigations in-
cluding angiography might well not be practicable
in a study large enough to show whether or not
~-blockers affect survival.
© Adis International Limited. All rig hts reserved.
Hampton
Given the undoubted beneficial effect on sur-
vival following myocardial infarction, and espe-
cially so in patients with heart failure, it might be
expected that ~-blockers would be particularly
valuable when heart failure results from ischaemic
disease. However, most published work has been
in patients with dilated cardiomyopathy.
Table III shows that improvement seems
equally likely, whatever the underlying diagnosis .
However, in the only study that set out specifically
to compare patients with ischaemic disease and
cardiomyopathy, improvement was seen only in
the latter.l97 ]
7. Conclusion
Treating an individual patient necessitates the
selection of an individual drug and using it in a
particular dosage. The drug and its dosage should
be selected on the basis of clinical trial results, not
as a result of speculation based on basic science
and clinical pharmacology. This means that clinical
practice will have to be based on individual good
clinical trials . A meta-analysis of trials with differ-
ent drugs involving different types of patients is not
adequate. Ideally, a drug or class of drugs, should
be shown to be effective in comparison with pla-
cebo, and then individual members of this class of
drug should be compared. This has happened with
thrombolytic agents, but not with ~-blockers .
In hypertension, the reduction of blood pressure
by ~-blockers is undoubtedly a class effect. The
evidence of benefit in terms of stroke and death are
best for propranolol, and perhaps after that with
atenolol. Such evidence as there is, however, sug-
gests that diuretics may be preferable to ~-block­
ers.
Post-infarction, the reduction of fatality is also
a class effect. The evidence for immediate or intra-
venous treatment is not overwhelming, but it is best
for atenolol and metoprolol. In the long term,
timolol and propranolol have given the most clear-
cut results and because of the simple dosage regi-
men used in the Norwegian Trial,!481 timolol re-
mains the preferred drug.
Drugs 48 (4) 1994
Choosing the Right ~-Blocker
~-Blockers are not powerful antiarrhythmic
agents, but they may be safer than the standard
antiarrhythmic drugs. They also have the distinct
benefit of reducing sudden death (presumably via
decreased ventricular fibrillation), which antiar-
rhythmic agents do not. For the treatment of symp-
tomatic arrhythmias, sotalol is probably the most
effective, but it is undoubtedly proarrhythmic and
may, therefore, have no particular advantage over
the standard antiarrhythmic agents or indeed over
other ~-blockers .
The role of ~-blockers in patients with heart
failure remains to be established. The fact that
metoprolol has been most used probably represents
the particular interest and background of the most
active groups in this field. There seems good evi-
dence for an improvement in haemodynamic pa-
rameters, reasonable evidence for an improvement
in exercise time, but as yet no convincing evidence
for a reduction in mortality.
The new generation of ~-blockers, especially
those with vasodilating properties, prompts and in-
deed necessitates a complete re-evaluation of the
clinical approach to ~-blockade. Wecannot assume
simply on the basis of clinical pharmacology and
studies of small numbers of patients that these
drugs represent a significant advance. They need
to be subjected to the same intensity of investiga-
tion that was applied to the first generation of ~­
blockers in the 1970s and 1980s. There must be
large studies with these drugs as monotherapy and
comparisons with standard therapy. When ethics
require it (as with ACE inhibitors in heart failure),
studies will have to be performed in addition to
existing optimal therapy. Until the results of such
trials are available, clinicians will be best advised
to limit themselves to the older drugs with their
long track records.
References
I. Black JW, Crowther AF, Shanks RG, et al. A new adrenergic
beta-receptor antagonist. Lancet 1964; I : 1080
2. Black JW. Stephenson JS . Pharmacology of a new adrenergic
beta-receptor-blocking compound (nethalide). Lancet 1962;
2: 311
3. Prichard BNC. Hypotensive action of pronethalol. BMJ 1964;
I: 1227-8
© Adis International Limited. All rights reselVed.
565
4. Prichard BNC , Gillam GM . Use of propranolol in treatment of
hypertension. Drugs 1983; 25 Suppl. 2: 5-11
5. Gillam PMS, Prichard BNe. Use of propranolol in angina pec-
toris. BMJ 1965; 2: 337-9
6. Snow PJD. Effect of propranolol in myocardi al infarction . Lan-
cet 1965; 2: 551-3
7. Turner P. Which ancillary properties of beta-adrenoceptor
blocking drugs influence their therapeutic or adverse effects ?:
a review. J R Soc Med 1991; 84: 672-4
8. Yedinak Ke. Formulary considerations in selection of beta-
blockers . PharmacoEconomics 1993; 4 (2): 104-21
9. McAreavey D, Vermeulen R, Robertson JIS . Newer beta block-
ers and the treatment of hyperten sion. Cardiovasc Drugs Ther
1991; 5: 577-88
10. Prichard BNe. Beta-blocking agents with vasodilating action .
J Cardiovasc Pharmacol 1992; 19 Suppl. I: S I-S4
II. German and Austrian Xamoterol Study Group. Double-blind
placebo-controlled comparison of digoxin and xamoterol in
chronic heart failure . Lancet 1988; I : 489-93
12. McTavish D, Campoli-Richards D, Sorkin EM. Carvedilol : a
review of its pharmacodynamic and pharmokinetic proper-
ties, and therapeutic efficacy . Drugs 1993; 45 (2); 232-58
13. Yue T-L, Cheng H-Y, Lysko PG, et al. Carvedilol, a new vaso-
dilator and beta adrenoceptor antagonist. is an antioxidant and
free radical scavenger. J Pharmacol Exp Ther 1992; 263 (I):
92-8
14. Sung CoP, Arleth AJ, Ohlstein EH. Carvedilol inhibits vascular
smooth muscle cell proliferation . J Cardiovasc Pharmacol
1993; 21: 221-7
15. Hamburger SA, Barone FC, Feuerstein GZ, et al. Carvedilol
(Kredex) reduces infarct size in a canine model of acute myo-
cardial infarction . Pharmacology 1991; 43 : 113-20
16. Young PH, on behalf of the INT-CAR-07 (UK) Study Group . A
comparison of carvedilol with atenolol in the treatment of
mild-to-moderate essential hypertension. J Cardiovasc Phar-
macoI1992; 19 Suppl. I : S82-5
17. Omvik P, Lund-Johansen P. Long-term hemodynamic effects at
rest and during exercise of newer antihypertensive agents and
salt restriction in essential hyperten sion : review of epanolol,
doxazosin, amlodipine, feldopine, diltiazem, lisinopril,
dilevalol, carvedilol, and ketanserin. Cardiovasc Drugs Ther
1993; 7: 193-206
18. Ruf G. Trenk D, Jahnchen E, et al. Determination of the anti-
ischemic activity of nebivolol in comparison with atenolol.
Int J Cardiol 1994; 43: 279-85
19. Frishman WHo F1enbaum W, Schoenberger J. et al. Celiprolol
in systemic hypertension. Am J Cardiol 1989; 63: 839-42
20. Lamon KD. Clinical safety and efficacy of celiprolol. Am Heart
J 1991; 121: 683-7
21. Veterans Administration Cooperative Study Group on Antihy -
pertensive Agents . Effects of treatment on morbidity in hy-
pertension. II: results in patients with diastolic blood pressure
averaging 90 through 114 mmHg. JAMA 1970; 213: 1143-52
22. Veterans Administration Cooperative Study Group on Antihy-
pertensive Agents. Effects of treatment on morbidity in hy-
pertension: results in patients with diastolic blood pressure s
averaging 115 though 129 mmHg . JAM A 1979; 202 : 1028-34
23. Hypertension Detection and Follow-up Program Cooperative
Group. Five-year findings of the Hypertension Detection and
Follow-up Program, I: reduction in mortality of persons with
high blood pressure, including mild hypertension. JAMA
1979;242:2562-71
24. Helgeland A. The Oslo Study -treatment of mild hypertension :
a five year controlled drug trial. Am J Med 1980; 69: 725-32
Drugs 48 (4) 1994
566
25. The Australian Therapeutic Trial in Mild Hypertension . Report
by the Management Committee . Lancet 1980; I: 1261-7
26. Grimm RH, Cohen JD, Smith W McF, et al. and the Multiple
Risk Factor Intervention Trial Research Group. Hypertension
management in the Multiple Risk Factor Intervention Trial
(MRFIT): six-year intervention results for men in special in-
tervention and usual care groups. Arch Intern Med 1985; 145:
1191-9
27. European Working Party on High Blood Pressure in the Elderly.
Mortality and morbidity results from the European Working
Party on High Blood Pressure in the Elderly Trial. Lancet
1985; I: 1349-54
28. The IPPPSH Collaborative Group. Cardiovascular risk and risk
factors in a randomized trial of treatment based on the beta-
blocker oxprenolol : the International Prospective Primary
Prevention Study in Hypertension (IPPPSH) . J Hypertens
1985; 3: 379-92
29. Medical Research Council Working Party. MRC trial of treat-
ment of mild hypertension : principal results. BMJ 1985; 291:
97-104
30. Coope J, Warrender TS. Randomised trial of treatment of hy-
pertension in the elderly in primary care. BMJ 1985; 293:
1145-51
31. Wilhelmsen L, Berglund G, Elmfeldt D. Beta blockers versus
diuretics in hypertensive men: main results from the HAP-
PHY Trial. J Hypertens 1987; 5: 561-72
32. SHEP Cooperative Research Group. Prevention of stroke by
antihypertensive drug treatm ent in older persons with iso-
lated systolic hyperten sion : final results of the Systolic Hy-
pertension in the Elderly Program (SHEP) . JAMA 1991;
265 : 3255-64
33. Dahlof B, Lindholm LH, Hansson L, et al. Morbidity and mor-
tality in the Swedish Trial in Old Patients with Hypertension
(STOP-Hypertension). Lancet 1991; 338: 1281-5
34. MRC Working Party. Medical Research Council trial of treat-
ment of hypertension in older adults: principal results. BMJ
1992;304:405-16
35. Report of Medical Research Council. Working Party on Mild to
Moderate Hypertension : adverse reactions to bendrofluazide
and propranolol for the treatment of mild hypertension . Lan-
cet 1981; 2: 539-43
36. Yusuf S, Peto R, Lewis J, et al. Beta blockade during and after
myocardial infarction: an overview of the randomized trials.
Prog Cardiovasc Dis 1985; 27: 335-71
37. Norris RM, Caughey DE, Scott PJ. Trial of propranolol in acute
myocardial infarction. BMJ 1968; I: 398-400
38. Barber JM, Boyle D McC, Chaturvedi NC, et al. Practolol in
acute myocardial infarction. Acta Med Scand 1976; 199
Suppl. 587: 213-9
39. Andersen MP, Bechsgaard P, Frederiksen J, et al. Effect of al-
prenolol on mortality among patients with definite or sus-
pected acu te myoc ardial infarction: preliminary results .
Lancet 1979;2: 865-7
40. Wilcox RG, Rowley JM, Hampton JR, et al. Randomised pla-
cebo-controlled trial comparing oxprenolol with disopyram-
ide phosphate in immediate treatment of suspected
myocardial infarction . Lancet 1980; 2: 765-9
41. Wilcox RG, Roland JM, Banks DC, et al. Randomised trial
comparing propranolo l with atenolol in immediate treatment
of suspected myocardial infarction. BMJ 1980; I : 885-8
42. Hjalmarson A, Elmfeldt D, Herlitz J, et al. Effect on mortality
of metoprolol in acute myocardial infarction . Lancet 198I; 2:
823-7
© Adis International Limited. All rights reserved.
Hampton
43. The MIAMI Trial Research Group. Metoprolol in acute myo-
cardial infarction (MIAMI) : a randomised placebo-controlled
international trial. Eur Heart J 1985; 6: 199-226
44. ISIS-I (First International Study of Infarct Survival) Collabo-
rative Group. Randomised trial of intravenous atenolol among
16027 cases of suspected acute myocardial infarction: ISIS- I.
Lancet 1986;2:57-66
45. Wilhelmsson C, Vedin JA, Wilhelmsen I, et al. Reduction of
sudden deaths after myocardial infarction by treatment with
alprenolol. Lancet 1974; 92: 1157-60
46. A Multicent re International Study. Improvement in prognosis
of myocardial infarction by long-term beta-adrenoreceptor
blockade using practolol. BMJ 1975; 3: 735-40
47. Baber NS, Evans DW, Howitt G, et al. Multicentre post-infarc-
tion trial of propranolol in 49 hospitals in the United King-
dom, Italy and Yugoslavia. Br Heart J 1980; 44: 96-100
48. The Norwegian Multicenter Study Group. Timolol-induced re-
duction in mortality and reinfarction in patients surviving
acute myocardial infarction. N Engl J Med 1981; 304: 801-7
49. Beta-Blocker Heart Attack Trial Research Group. A randomized
trial of propranolol in patients with acute myocardial infarc-
tion. JAMA 1982; 247: 1707-14
50. Coronary Prevention Research Group. An early intervention
secondary prevention study with oxprenolol following myo-
cardial infarction. Eur Heart J 1981; 2: 389-93
51. Taylor SH, Silke B, Ebbutt A, et al. A long-term prevention
study with oxprenolol in coronary heart disease. N Engl J Med
1982; 307: 1293-130 I
52. Julian DG, Prescott RJ, Jackson FS, et al. Controlled trial of
sotalol for one year after myocardial infarction. Lancet 1982;
I : 1142-7
53. Hansteen V, Moinichen E, Lorentsen, et al. One year's treatment
with propranolol after myocardial infarction: preliminary re-
port of Norwegian multicentre trial. BMJ 1982; 284: 155-60
54. The European Infarction Study Group. A secondary prevention
study with slow release oxprenolol after myocardial infarc-
tion: morbidity and mortality. Eur Heart J 1984; 5; 189-202
55. Olsson G, Rehnqvist N, Sjorgren A, et al. Long-term treatment
with metoprolol after myocardial infarction: effect on 3 year
mortality and morbidity. J Am Coli Cardiol 1984; 5: 1428-37
56. Australian and Swedish Pindolol Study Group: the effect of
pindolol on the two years mortality after complicated myo-
cardial infarction. Eur Heart J 1983; 4: 367-375
57. Poissel J-P, Leizorovicz A, Picolet H, et al., for the APSI Inves-
tigators. Secondary prevention after high-risk acute myocar-
dial infarction with low dose acebutolol. Am J Cardiol 1990;
66: 251-60
58. The TIMI Study Group. Comparison of invasive and conserva-
tive strategies after treatment with intravenous tissue plasmin-
ogen activator in acute myocardial infarction: results of the
Thromboly sis in Myocardial Infarction (TIMI) Phase II Trial.
N Engl J Med 1989; 320: 618-27
59. Van de Werf F, Janssens L, Brzostek T, et al. Short term effects
of early intravenous treatment with a beta-adrenergic block-
ing agent or a specific bradycardiac agent in patients with
acute myocardial infarction receiving thrombolytic therapy. J
Am Coli Cardiol 1993; 22: 407-16
60. ISIS-2 (Second International Study of Infarct Survival) Collab-
orative Group. Randomised trial of intravenous streptokinase,
oral aspirin, both, or neither among 17 187 cases of suspected
acute myocardial infarction: ISIS-2. Lancet 1988; 2: 349-60
61. Gruppo Italiano per 10Studio Della Streptochinasi Nell'lnfarto
Miocardico (GISSI). Effectiveness of intravenous thrombo-
Drugs 48 (4) 1994
Choosing the Right ~-Blocker
lytic treatment in acute myocardial infarction. Lancet 1986;
I: 397-401
62. Wilcox RG, Von der Lippe G, Olsson CG, et aI., for the Anglo-
Scandinavian Study of Early Thromb olysis. Effects of al-
teplase in acute myocard ial infarction: 6-month results from
the ASSET study. Lancet 1990; 335: 1175-8
63. G1SS1-2: a factorial random ised trial of alteplase versus strep-
tokinase and heparin versus no heparin among 12 490 pa-
tients with acute myocardial infarction . Lancet 1990; 336:
65-71
64. The GUSTO Investigators. An international randomized trial
compar ing four thrombolytic strategies for acute myocardial
infarction . N Engl J Med 1993; 329: 673-82
65. Cardiac Arrhythmia Suppression Trial (CAST) Investigators.
Prelimin ary report: effect of encainide and f1ecainide on mor-
tality in a randomi zed trial of arrhythmia suppression after
myocardial infarcti on. N Engl J Med 1989; 321: 406-12
66. Olsson G, Rehnq vist N, Sjogren A, et al. Long-term treatment
with metoprolol after myocardial infarction: effect on 3 year
mortality and morbidity. J Am Coli Cardiol 1985; 5: 1428-37
67 . Bigger JT, Coromilas J. How do beta-blockers protect after
myocardial infarction? Ann Intern Med 1984; 101 (2): 256-8
68. Steinbeck G, Andresen D, Bach P,et al. A comparison of electro-
physiologically guided antiarrhythmic drug therapy with bela-
blocker therapy in patients with symptomatic, sustained
ventricularlachyarrhythmias. N Engl J Med 1992; 327: 987-92
69. Yusuf S, Sleight P, Rossi P, et al. Reduction in infarct size,
arrhythmias and chest pain by early intravenous beta block-
ade in suspected acute myocardial infarction . Circulat ion
1983; 67 Suppl.l : 1-32-1-35
70. Rossi PR, Yusuf S, Ramsdale D, et al. Reduction of ventricular
arrhythmia s by early intravenous atenolol in suspected acute
myocardial infarcti on. BMJ 1983; 286: 506-10
71. Murray DP, Murray RG, Littler WA. The effects of metoprolol
given early in acute myocardial infarction on ventricul ar ar-
rhythmias . Eur Heart J 1986; 7: 217-22
72. Ryden L, Ariniego R, Arnman K, et al. A doub le-blind trial of
metoprolol in acute myocard ial infarction : effects on ventric-
ular tachyarrhythrn ias. N Engl J Med 1983; 308: 614-8
73. Roland JM, Wilcox RG, Banks DC, et al. Effect of beta blocker s
on arrhythmi as during six weeks after suspected myocardial
infarction. BMJ 1979; 2: 518-21
74. Bethge K-P, Andresen D, Boissel J-P, et aI., for the European
Infarct ion Study Group . Effect of oxprenolol on ventricular
arrhythmias: the European Infarction Study experience. J Am
Coli Cardiol 1985; 6: 963-72
75. Leibowitz D. Sotalol : a novel beta-blocker with class III anti-
arrhythmic activity. J Clin Pharmacol 1993; 33: 508-12
76. Edvardsson N. Compa rison of class I and class III action in
atrial fibrillation. Eur Heart J 1993; 14 Suppl. H: 62-6
77. Murgatroyd FD, Camm AJ. Atrial arrhythmias. Lancet 1993 ;
341: 1317-22
78. SCOll Millar RN. Efficacy of sotalol in controllin g reentrant
supra ventricular tachycardias. Cardiovasc Drugs Ther 1990 ;
4: 625-30
79. Mason JW, for the Electrophy siologic Study versus Electroc ar-
diogr aphic Monitoring Investigator s. A comparison of seven
antiarrhythmic drugs in patients with ventricular
tachy arrhythmia s. N Engl J Med 1993; 329: 452-8
80. Shenasa M, Borg grefe M, Haverkamp W, et al. Ventricular
tachycardia. Lancet 1993; 342: 1512-9
81. Ben-David J, Zipes DP. Torsades de pointes and proarrhythm ia.
Lancet 1993; 341: 1578-82
© Ad is International Limited. All right s reserved.
567
82. Hohn loser SH, Zabel M, van de Loo A, et al. Efficacy and safety
of sotalol in patients with complex ventricular arrhythmia s.
Int J CardioI1992; 37: 283-91
83. Waagstein F, Hjalrnarson A, Varnauskas E, et al. Effect of
chronic beta-adrenergic receptor blockade in congestive car-
diomyopath y. Br Heart J 1975; 37: 1022-36
84. Packer M, Carver JR, Rodeheffer RJ, et aI., for the PROMISE
Study Research Group. Effect of oral milrinone on mortality
in severe chronic heart failure. N Engl J Med 1991; 325:
1468-75
85. Bristow MR, Ginsburg R, Minobe W, et al. Decreased catechol-
amine sensitivity and beta-adrenergic-receptor density in fail-
ing human hearts. N Engl J Med 1982; 307: 205-11
86. Swedberg K, Hjalmarson A, Waagstein F, et al. Adverse effects
of beta-blo ckade withdrawal in patients with congestive car-
diomyopathy. Br Heart J 1980 ; 44: 134-42
87. Meinert z T, Hofmann T, Zehender M, et al. Beta-blocking
agents vs. antiarrhythmic interventions in heart failure com-
plicated by arrhythmias. J Cardiovas c Pharmacol 1990; 16
Suppl. 5: S-151- S-157
88. Gilbert EM, Anderson JL, Deitchman D, et al. Long-term beta-
blocker vasodilator therapy improves cardiac function in idio-
pathic dilated cardiomyopathy: a double-blind, randomized
study of bucindolol versus placebo. Am J Med 1990; 88: 223-9
89. Clarkson P, Wheeldon NM, McDonald TM. Left ventricular
diastolic dysfunction. Q J Med 1994; 87: 142-8
90. Chadda K, Goldstein S, Byrington R, et al. Effect of propranolol
after acute myocardial infarction in patients with congestive
heart failure. Circulation 1986; 73: 503-10
91. Ikram H, Fitzpatrick D. Double-blind trial of chronic oral bela
blockade in congestive cardiomyopathy. Lancet 1981; 2: 490-2
92. Currie PJ, Kelly MJ, McKenzie A, et al. Oral beta-adrenergic
blockade with metoprolol in chron ic severe dilated cardiomy-
opathy. J Am Coli Cardiol 1984; 3: 203-9
93. Engelmeier RS, O'Co nnel lB , Walsh R, et al. Improvement in
symptoms and exercise tolerance by metoprolol in patients
with dilated cardi omyopath y: a double-blind , randomi zed,
placebo-c ontro lled trial. Circulation 1985; 72 (3): 536-46
94. Anderson JL, Lutz JR, Gilbert EM, et al. A randomized trial of
low-dose beta-blockade therapy for idiopathic dilated cardio-
myopathy. Am J Cardiol 1984; 55: 471-5
95. Pollock SG , Lystash J, Tedesco C, et al. Usefulness of
bucindolol in conge stive heart failure. Am J Cardiol 1990 ; 66:
603-7
96 . Leung WH, Lau CPo Wong CK, et al. Improvement in exerci se
performance and hemodynamics by labetolol in patients with
idiopathic dilated cardiom yopathy. Am Heart J 1990; 119:
884-90
97. Woodley SL, Gilbert EM, Anderson JL, et al. Beta blockade
with bucindolol in heart failure caused by ischemic versus
idiopath ic dilated cardi omyopath y. Circulation 1991; 84:
2426-4 1
98. Fisher ML, Gottlieb SS, Hamilton B, et al. Beneficial effects of
metoprolol in congestive heart failure associated with coro-
nary artery disease: a randomized trial [abstract]. Circulation
1991; 84 Suppl. II: 11312
99. Paolisso G, Gambardella A, Marrazzo G, et al. Metabolic and
cardiovascular benefits deriving from beta-adrenergic block-
ade in chronic congestive heart failure. Am Heart J 1992; 123:
103-10
100. Olsen SL, Yanowitz FG, Gilbert AM, et al. Beta-blocke r related
improvement in submaximal exercise tolerance in heart fail-
ure from idiopathic dilated cardiomyopathy [abstract]. J Am
Coli CardioI1 992; 19: 146
Drug s 48 (4) 1994
568
101. Waagstein F, Bristow MR , Swedberg K, et aI., for the
Metoprolol in Dilated Cardiomyopathy (MDC) Trial Study
Group. Beneficial effects of metoprolol in idiopathic dilated
cardiomyopathy. Lancet 1993 ; 342 : 1441-56
102. Wisenbaugh T, Katz I, Davis J, et al. Long-term (3 month) ef-
fects of a new beta-blocker (nebivolol) on cardiac perfor-
mance in dilated cardiomyopathy. J Am Coli Cardiol 1993 ;
21 : 1094-1100
103. Waagstein F. Beta blockers in heart failure . Cardiology 1993 ;
82 Suppl. 3: 13-8
104. Doughty RN, MacMahon S, Sharpe N. Bet a-blockers in heart
failure : promising or proved? J Am Coli Cardiol 1994; 23 :
814-21
105. Waagstein F, Caidahl K, Wallen tin I, et al. Long-term beta-
blockade in dilated cardiomyopathy: effects of short- and
long-term metoprolol treatment followed by withdrawal and
readministration of metoprolol. Circulation 1989; 80 : 551-63
© Adls International limited. All rights reserved .
Hampton
106. DasGupta P, Broadhurst P, Lahiri A. The effects of intravenous
. carvedilol, a new multiple action vasodilatory beta-blocker,
in congestive heart failure. J Cardiovasc Pharmacol 1991 ; 18
Suppl. 4 : S12-6
107. Bristow MR, Olsen S, Larrabee P, et al. The acute effects of
carvedilol and metoprolol in patients with dilated cardiomy-
opathy . [abstract] J Am Coli Cardiol1993; 21: 314A
108. The Xamoterol in Severe Heart Failure Study Group.
Xamoterol in severe heart failure. Lancet 1990 ; 336 : 1-6
Correspondence and reprints: JR .Hampton, Professor of Car-
diology, Cardiovascular Medicine, Queen's Medical Centre,
University Hospital, Nottingham NG7 2UH, England.
Drugs 48 (4) 1994