REVIEW ARTICLE Drugs 2000 Nov; 60 (5): 997-1016

0012-6667/00/0011-0997/$25.00/0

© Adis International Limited. All rights reserved.

Effects of β-Blockers on
Neurohormonal Activation in Patients
with Congestive Heart Failure
David Baran,1 Evelyn M. Horn,1 Katarzyna Hryniewicz2 and Stuart D. Katz1
1 Columbia Presbyterian Medical Center, Division of Circulatory Physiology, Department of
Medicine, Columbia University College of Physicians and Surgeons, New York, New York, USA
2 Department of Internal Medicine, Warsaw Medical School, Warsaw, Poland

Contents
Abstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 997
1. Experimental Studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 999
2. Clinical Studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1002
2.1 First and Second Generation β-Blockers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1002
2.1.1 Metoprolol . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1002
2.1.2 Bisoprolol . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1006
2.2 Third Generation β-Blockers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1006
2.2.1 Bucindolol . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1006
2.2.2 Carvedilol . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1008
2.2.3 Nipradilol . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1009
2.2.4 Pindolol . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1009
2.2.5 Xamoterol . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1009
3. Comparative Studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1010
4. Possible Mechanisms of Action . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1011
5. Clinical Implications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1012
6. Summary and Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1013

Abstract The effect of β-adrenoceptor antagonists (β-blockers) on neurohormonal ac-

tivation in patients with congestive heart failure has been the subject of study in
numerous small clinical trials. Short term therapy with β-blockers is associated
with a variable acute neurohormonal response which may be determined by the
pharmacology of the agent under study and the baseline characteristics of the
patient population. Long term therapy with β-blockers devoid of intrinsic sym-
pathomimetic activity (partial agonist activity) is associated with evidence of
decreased plasma markers of activation of the sympathetic nervous system, the
renin-angiotensin system, and endothelin-1. β1-Selective and nonselective β-
blockers appear to be associated with evidence of decreased neurohormonal ac-
tivation, with differential effects on β-adrenoceptor density. Agents with partial
agonist activity appear to differ from pure antagonists, with some studies report-
ing evidence of increased neurohormonal activation. The mechanisms by which
β-blockers reduce neurohormonal activation and the clinical relevance of changes
998 Baran et al.

in adrenergic function to their use in the treatment of heart failure require further
investigation.

Congestive heart failure (CHF) is a clinical syn- uted to down-regulation of β1-adrenoceptors, rela-
drome characterised by a cluster of signs (ele- tive up-regulation of β2-adrenoceptors, and un-
vated jugular venous pressures, rales, oedema) and coupling of β-adrenoceptors due to abnormalities
symptoms (dyspnoea on exertion, orthopnoea, par- in G-proteins (guanosine triphosphate regulatory
oxysmal nocturnal dyspnoea, fatigue) related to ab- binding proteins), adenylate cyclase activities, β-
normal cardiac function. While the clinical syndrome adrenoceptor kinase activity, and intracellular sig-
of CHF can be the end result of virtually any type nal transduction pathways.[4-10] Chronic activation
of cardiac disease, the majority of patients have of the sympathetic nervous system contributes to pro-
clinical heart failure attributable to a primary ab- gression of heart failure by promoting ventricular
normality in systolic function due to ischaemic and vascular remodelling, promoting myocellular
heart disease (IHD), hypertensive heart disease, or apoptosis and increasing interstitial fibrosis.[11,12]
idiopathic dilated cardiomyopathy (IDC). In pa- Clinical trials with β-adrenoceptor antagonists (β-
tients with heart failure due to systolic dysfunction, blockers) have demonstrated short term adverse ef-
the pathophysiology of disease progression is de- fects at the initiation of therapy related to deterio-
termined by a complex interaction of haemodyna- ration in haemodynamic function and long term
mic abnormalities and neurohormonal activation, substantial reductions in morbidity and mortality
which induce progressive changes in the structure in association with improved haemodynamics, fa-
and function of the left ventricle.[1,2] The progres- vourable phenotypic myocardial changes and fa-
sive changes in ventricular structure and function
appear to be the key determinants of long term mor-
bidity and mortality (fig. 1). Myocellular injury
Recognition of the importance of ventricular re-
modelling as a determinant of long term morbidity
and mortality has resulted in changes in therapeutic
Haemodynamic Neurohormonal
strategies for patients with heart failure. Long term abnormalities activation
therapeutic goals in these patients are based on re- ↓CO, ↑SVR ↑RAS, ↑SNS
duction of long term morbidity and mortality with
neurohormonal antagonists that slow the progres-
sion of or reverse ventricular and vascular remodel-
Ventricular Vascular
ling. Much of the long term benefit of angiotensin remodelling remodelling
converting enzyme (ACE) inhibitors is derived
from their properties as specific antagonists of the
renin-angiotensin-aldosterone system and conse-
quent favourable effect on ventricular remodelling.[3] Progressive CHF
Death
The concept of therapeutic suppression of neu-
rohormonal activation in heart failure has been ap- Fig. 1. Schematic representation of proposed pathophysiology
plied to the sympathetic nervous system. Heart failure of congestive heart failure. Neurohormonal activation appears
to play a critical role in the progression of pathological changes
is characterised by increased plasma noradrenaline in structure and function of the cardiovascular system. Suppres-
(norepinephrine) levels, decreased myocardial nor- sion of neurohormonal activation is a therapeutic strategy that
has led to numerous trials of β-blockers in the heart failure pop-
adrenaline content, increased sympathetic neural traf- ulation. CHF = congestive heart failure; CO = cardiac output;
fic, and adrenergic hyporesponsiveness. Decreased RAS = renin-angiotensin system; SNS = sympathetic nervous
response to adrenergic stimulation has been attrib- system; SVR = systemic vascular resistance.

 Adis International Limited. All rights reserved. Drugs 2000 Nov; 60 (5)
β-Blockers and Neurohormones in CHF 999

vourable effects on ventricular remodelling.[13-16] adrenaline kinetics and direct recording of sympa-
The rationale for the use of β-blockers in heart fail- thetic neural activity. Although not using a direct
ure and a summary of therapeutic trials with β- measurement of neurohormonal activation, studies
blockers are the subject of recent reviews.[13,17-19] which examined other aspects of the function of
β-Blockers are a heterogeneous class of thera- the β-adrenoceptor complex, such as changes in
peutic agents (fig. 2). Their primary pharmacolog- β-adrenoceptor density, β-adrenoceptor kinase ac-
ical action is attenuation of the effects of endoge- tivity and heart rate variability, are also included
nous catecholamines via competitive antagonism in the review.
of the β-adrenoceptor in cardiovascular and other
tissues. In addition, β-blockers may alter activation 1. Experimental Studies
of the sympathetic nervous system and other neu-
rohormonal systems via direct and indirect mech- The acute effects of β-adrenoceptor blockade
anisms. Additional pharmacological properties with propranolol were assessed in a bovine model
which may influence the clinical effects of β- of heart failure induced by progressive pulmonary
blockers include β1-selectivity, lipophilicity, in- hypertension after ligation of the right pulmonary
trinsic sympathomimetic activity (partial agonist artery.[21] This experimental model produces a pe-
riod of compensated right ventricular hypertrophy
activity), inverse agonism and G-protein interac-
lasting up to 6 weeks, followed by progressive
tions.[10,20] The clinical effect of β-adrenoceptor
heart failure. Administration of propranolol 0.2
blockade is likely to be determined by the net effect
mg/kg was associated with a deterioration in rest-
of the agent on the entire β-adrenoceptor complex,
ing haemodynamics and a substantial increase in
including the receptor, G-protein coupling, adenylate
plasma noradrenaline and adrenaline (epinephrine)
cyclase and β-adrenoceptor kinase activity. Third
levels in animals with right ventricular failure, but
generation β-blockers such as carvedilol have
not animals with compensated right ventricular hy-
unique pharmacological properties (β1- and β2-
pertrophy or control animals. Right ventricular
receptor blockade, α1-receptor blockade, G-pro- myocardial noradrenaline content tended to be
tein interactions) which may provide more com- much lower in animals with heart failure when
plete suppression of sympathetic activation than first compared with those with compensated right ven-
and second generation agents. Accordingly, in this tricular hypertrophy or control animals (0.11 ±
review, newer agents with ancillary properties (third 0.01 vs 1.16 ± 0.33 vs 1.48 ± 0.14 µg/g), although
generation agents) are grouped separately from first no formal statistical comparison is provided. The
and second generation agents. increase in plasma noradrenaline levels in response
This review focuses on experimental and clini- to propranolol may represent a compensatory adre-
cal studies that have specifically investigated the nal-mediated response to maintain circulatory ho-
effects of β-adrenoceptor blockade on neurohor- meostasis, rather than a direct pharmacological ac-
monal activation in heart failure with a primary tion.
emphasis on the sympathetic nervous system, and The short term effects of β-adrenoceptor block-
secondary consideration of the effects on the renin- ade were assessed in a canine model of heart failure
angiotensin system, endothelin-1 and natriuretic induced by rapid right ventricular pacing.[22] Four
peptides when data are available. The majority of weeks of oral propranolol therapy (4 mg/kg/day)
reports have relied on measurement of the level of was associated with increased plasma noradrena-
neurohormones in plasma as an indirect index of line levels when compared with untreated paced
neurohormonal activation. Plasma levels are pro- animals with heart failure and unpaced control an-
vided in the original units reported for each study. imals (1100 ± 210 vs 720 ± 80 vs 270 ± 50 ng/L; p
A few studies have used more direct measures of < 0.05). Myocardial noradrenaline content was sig-
sympathetic activity by isotopic assessment of nor- nificantly decreased in all paced animals with heart

 Adis International Limited. All rights reserved. Drugs 2000 Nov; 60 (5)
1000 Baran et al.

OH

HO

OH

CHCH2NH2

Noradrenaline (norepinephine)

OH
CN
OH CH3 OCH2CHCH2NHCH2CH2O
OCH2CHCH2NHCCH2 OCH3
CH3
N
N
H
Bucindolol H
Carvedilol

OH
O OH
OCH2CHCH2NHCH2CH2NHCON O
(C2H5)2NCNH OCH2CHCH2NHC(CH3)3

CCH3

Celiprolol OH
Xamoterol

OH
OH
OCH2CHCH2NHCH(CH3)2
CH3OCH2CH2 OCH2CHCH2NHCH(CH3)2

Metoprolol
Propranolol

OH
F O OH OH O OCH2CHCH2NHCH(CH3)2
CHCH2NHCH2CH

F CH2OCH2CH2OCH(CH3)2

Nebivolol Bisoprolol

Fig. 2. Structure of noradrenaline (norepinephrine) and various β-blockers investigated in patients with heart failure.[20]

 Adis International Limited. All rights reserved. Drugs 2000 Nov; 60 (5)
β-Blockers and Neurohormones in CHF 1001

failure with or without propranolol therapy when ing protein activity or isoprenaline (isoproterenol)
compared with unpaced normal control animals. responsiveness.
Plasma levels of neuropeptide Y, a parasympathetic The effects of β-adrenoceptor blockade on neu-
neuroenzyme marker, in paced animals with heart rohormonal activation have been assessed by 2
failure with or without propranolol therapy did not groups of investigators in rats after experimental
differ from those in control animals. Myocardial myocardial infarction. Latini and colleagues[25]
neuropeptide Y content was decreased in paced an- studied the effects of metoprolol in rats with exper-
imals with heart failure with or without proprano- imental myocardial infarction produced by coro-
lol therapy when compared with control animals. nary artery ligation. Surviving rats were treated with
These findings are concordant with the report of the metoprolol 10 mg/kg/day via osmotic pump for 4
acute effects of propranolol in the right ventricular weeks, beginning 2 months after the coronary liga-
failure model in calves.[21] tion procedure. Metoprolol therapy was associated
The effects of β-adrenergic blockade on neuro- with a reduction in left ventricular end-diastolic
hormonal activation in a genetic model of heart pressure when compared with untreated infarcted
failure, the BIO 53.58 strain of Syrian cardiomyo- controls (6 ± 1 vs 13 ± 3mm Hg; p value not spec-
pathic hamster, have been assessed by 2 groups of ified) but no change in left ventricular mass or
investigators. This experimental model is charac- myocardial collagen deposition. Plasma noradren-
terised by progressive left ventricular dilation, aline levels in metoprolol-treated rats did not differ
from those in untreated infarcted controls (245 ±
without hypertrophy, with frank CHF before 6
37 vs 224 ± 53 ng/L). Warner and colleagues[26]
months of age. Yamada and colleagues[23] studied
treated rats with propranolol 500 mg/L in drinking
the effects of metoprolol (1 mg/kg/day) in this
water for 6 weeks, beginning 21 days after coro-
model. In a comparison with untreated control an-
nary ligation. When compared with no treatment
imals, metoprolol therapy, administered between
in rats with infarction, propranolol therapy signif-
the ages of 11 and 18 weeks, significantly decreased
icantly increased myocardial β-adrenoceptor den-
plasma noradrenaline level (3.39 ± 0.65 vs 6.02 ±
sity (11.9 ± 1.7 vs 9.3 ± 0.6 fmol/mg; p < 0.05) but
1.03 µg/L; p < 0.05), significantly increased myo-
did not change adenylate cyclase activity or iso-
cardial noradrenaline content (10.9 ± 0.9 vs 6.7 ± prenaline responsiveness.
0.8 µg/g; p < 0.01) and significantly increased Doxorubicin-induced myocardial damage is an-
myocardial β-adrenoceptor density (40.0 ± 2.0 vs other well characterised experimental model of
28.8 ± 2.1 fmol/mg; p < 0.01). The plasma nor- heart failure in which the effects of β-adrenergic
adrenaline level, myocardial noradrenaline content antagonism have been studied.[27] When compared
and β-adrenoceptor density measured during me- with no treatment in heart failure rats, metoprolol
toprolol therapy were similar to those of a control therapy (10 mg/kg/day) administered for 3 weeks
hamster strain without cardiomyopathy. The im- starting 1 week after administration of doxorubicin
provements in adrenergic function were accompa- (3 mg/kg 3 times weekly for 5 weeks) significantly
nied by decreased fibrosis in the left ventricle. decreased plasma noradrenaline level (1.0 ± 0.6 vs
Tomita and colleagues[24] administered arotinolol, 2.2 ± 1.3 µg/L; p < 0.01), increased myocardial
a nonselective β-blocker, by mouth for 10 weeks noradrenaline content (495 ± 87 vs 372 ± 92 ng/g;
(ages 14 to 24 weeks) in the same strain of Syrian p < 0.05), and increased myocardial β-adrenocep-
hamsters. In a comparison with untreated animals, tor density (57 ± 8 vs 46 ± 9 fmol/mg; p < 0.01).
arotinolol therapy significantly decreased plasma The effects of 2 weeks’ administration of aten-
noradrenaline level (103 ± 24 vs 821 ± 227 ng/L; olol (0.1 to 10 mg/kg/day) and carvedilol (10 mg/
p < 0.05) without change in myocardial β1-adreno- kg/day) by osmotic pump on myocardial sensitiv-
ceptor density, adenylate cyclase activity, Gs-bind- ity and β-adrenoceptor kinase activity were evalu-

 Adis International Limited. All rights reserved. Drugs 2000 Nov; 60 (5)
1002 Baran et al.

Table I. Effects of second generation β-blockers on plasma noradrenaline (norepinephrine) levels in patients with heart failure
Reference Agent No. of Duration Plasma Comments
patients noradrenaline
Swedberg et al.[30] Metoprolol 11 Short term ↑
Nemanich et al.[31] Metoprolol 10 8wk ↓
Andersson et al.[32] Metoprolol 21 6-12mo ↓ No change myocardial
noradrenaline
Maisel[33] Metoprolol 15 6-12mo ↓ p = 0.14
Santostasi et al.[34] Metoprolol 15 6mo ↓
Ishida et al.[35] Metoprolol 9 6mo ↓ ↑ lymphocyte
β2-adrenoceptor density
Rahman et al.[36] Metoprolol 6 20mo ↓ ↓ SNS nerve activity
Kukin et al.[37] Metoprolol 30 3mo ↓ ↓ combined with doxazosin
Andersson et al.[38] (MDC)a Metoprolol 41 6mo ↓ No different from placebo at
12 months
Eichhorn et al.[39]a Metoprolol 24 3mo ↓ Coronary sinus blood
Paolisso et al.[40]a Metoprolol 10 3mo ↔ Valvular heart disease
Suwa et al.[41] Bisoprolol 18 10wk ↓
a Placebo-controlled study design.
MDC = Metoprolol in Dilated Cardiomyopathy trial; SNS = sympathetic nervous system.

ated in C57/B16 mice by Iaccarino et al.[28] In this Conflicting findings in experimental studies
species, 2 weeks’ administration of isoprenaline may be due to species differences, model-specific
(0.3 to 30 mg/kg/day) is associated with decreased differences or pharmacological differences. The
myocardial β-adrenoceptor density and increased relatively short duration of therapy in all these
expression and activity of β-adrenoceptor kinase. studies limits the ability to extrapolate the experi-
Both atenolol and carvedilol administration were mental findings to clinical studies of human heart
failure.
associated with specific decrease in β-adrenoceptor
kinase expression (protein content and mRNA levels)
2. Clinical Studies
and activity. Atenolol, but not carvedilol, was also
associated with increased myocardial β-adrenoceptor 2.1 First and Second Generation β-Blockers
density. These findings suggest that β-adrenoceptor
The effects of first and second generation β-
blockade modulates adrenergic function by specif-
blockers on plasma norepinephrine levels are pre-
ically decreasing β-adrenoceptor kinase activity in-
sented in table I.
dependently of effects on β-adrenoceptor density.
These findings may be relevant to clinical observa- 2.1.1 Metoprolol
tions on the interaction of β-blocker therapy and Metoprolol is a β1-selective blocker which has
been the frequent subject of investigation in studies
the inotropic effects of phosphodiesterase inhibi-
in patients with CHF. The Metoprolol CR/XL
tors in patients with heart failure. Bohm and col-
Randomised Intervention Trial in Heart Failure
leagues[29] reported that 6 months of β-adrenoceptor
(MERIT-HF) randomised 3991 patients with mild
blockade with metoprolol 150mg daily increased to severe heart failure of ischaemic and nonischae-
the positive inotropic effects of the receptor-inde- mic origin to placebo or metoprolol controlled re-
pendent phosphodiesterase inhibitor, milrinone, in lease/extended release (CR/XL) in a single-blind
15 patients with heart failure, suggesting drug- fashion for 6 to 20 months (mean follow-up period
induced postreceptor enhancement of adrenergic was 1 year).[42] The maximum dosage of metop-
function. rolol was 200 mg/day (mean 159 mg/day). All-

 Adis International Limited. All rights reserved. Drugs 2000 Nov; 60 (5)
β-Blockers and Neurohormones in CHF 1003

cause mortality [relative risk (RR) 0.66, adjusted p ceptor density or receptor affinity with treatment.
value 0.0062), sudden death (RR 0.59, p < 0.001) Similarly, no significant change in noradrenaline
and death from worsening heart failure (RR 0.51, kinetics was detected in this 8-week protocol. The
p < 0.01) were all significantly reduced in patients physiological relevance of lymphocyte β2-adreno-
assigned to metoprolol compared with those as- ceptor density measurements is uncertain, as me-
signed to placebo. Neurohormonal data are not yet toprolol is a β1-selective agent.
published from the MERIT-HF population and Waagstein and colleagues[44] studied, nonblind,
therefore are derived from mostly small noncon- 33 consecutive patients treated with metoprolol
trolled studies, with a few controlled studies also (target 50mg 3 times daily) for a mean of 16 months
available. (range 6 to 20 months). 16 of the patients were in
New York Heart Association (NYHA) class IV
Noncontrolled Studies CHF, and 15 were in class III. Dosages of metop-
Swedberg and colleagues[30] reported on the acute rolol were slowly titrated to the target. Endomyo-
effects of intravenous metoprolol administration in cardial biopsies were performed at the start of the
11 patients with IDC. Arterial and coronary sinus
protocol and at completion, for measurement of
plasma noradrenaline levels were determined be-
myocardial β-adrenoceptor density. Myocardial β-
fore and after administration of metoprolol 15mg.
adrenoceptor density increased significantly from
Arterial and coronary sinus plasma noradrenaline
pretreatment values with metoprolol (30.3 ± 2.9 to
levels increased in response to β-adrenoceptor
49 ± 7.1 fmol/mg protein; p < 0.0.05).
blockade (3.8 ± 1.6 to 5.0 ± 2.9 nmol/L and 5.0 ±
Andersson et al.[32] investigated the acute and
3.6 to 8.2 ±6.0 nmol/L, respectively; p < 0.05). As
chronic effects of intravenous metoprolol in pa-
observed in experimental models, acute increases
tients with heart failure on haemodynamics, plas-
in plasma noradrenaline in response to β-adreno-
ceptor blockade may be attributable to a reflex re- ma noradrenaline and myocardial noradrenaline
sponse to short term deterioration in haemodyna- extraction. Short term non-blind administration of
mic function. Haber and colleagues[43] reported on intravenous metoprolol (2.5 to 15mg, mean 13mg)
the acute effect of non-blind intravenous adminis- was associated with evidence of deterioration of
tration of propranolol and metoprolol in 24 pa- resting haemodynamic function (decreased ejec-
tients with heart failure. Acute deterioration in left tion fraction and cardiac index) and activation of
ventricular systolic function as determined by fall the sympathetic nervous system (increased arterial
in systolic pressure, resting cardiac index and end- noradrenaline level and increased myocardial re-
systolic elastance was similar for both agents. lease of noradrenaline). Long term administration
Nemanich et al.[31] reported findings in 10 pa- of metoprolol (mean dosage 127 mg/day) for 6 to
tients with CHF attributable to IHD (6 patients) 12 months was assessed in 21 patients (15 non-
and IDC (4 patients). Metoprolol was administered ischaemic, 6 ischaemic) in a non-blind study.[45]
non-blind and titrated to a maximum dosage of 50mg Arterial and coronary sinus plasma noradrenaline
twice daily. Plasma catecholamines, noradrenaline and adrenaline levels were determined at rest and
kinetics and peripheral blood lymphocyte β2- during supine bicycle ergometry. In contrast to the
adrenoceptor density were measured before treat- acute study, arterial noradrenaline levels were sig-
ment and after 8 weeks of therapy. Plasma nor- nificantly reduced during long term metoprolol
adrenaline and adrenaline levels were elevated at therapy when compared with pretreatment values
baseline but decreased significantly with treatment (3.72 ± 2.4 to 2.19 ± 3.5 nmol/L; p < 0.02). Myo-
(613 ± 706 ng/L baseline to 303 ± 142 ng/L, and cardial noradrenaline extraction (arterial – coro-
71 ± 50 to 40 ± 21 ng/L, respectively; p < 0.05 for nary sinus noradrenaline level) trended to reduc-
noradrenaline, p < 0.02 for adrenaline). There was tion at rest when compared with pretreatment
no significant change in lymphocyte β2-adreno- values (–0.28 ± 0.37 vs –0.13 ± 0.09 nmol/L; p =

 Adis International Limited. All rights reserved. Drugs 2000 Nov; 60 (5)
1004 Baran et al.

0.14). This comparison was statistically significant after 6 months of therapy (272 ng/L). Changes in
during exercise (–1.13 ± 1.5 vs –0.27 ± 0.48; p < plasma noradrenaline level after 1 month of ther-
0.05), suggesting that metoprolol therapy de- apy were inversely correlated to change in left ven-
creases stress-induced myocardial production of tricular ejection fraction after 1 month of therapy.
noradrenaline. Plasma adrenaline levels did not Ishida and colleagues[35] studied the effects of 6
change from baseline during metoprolol therapy at months of metoprolol therapy (mean dosage 46
rest or during exercise. mg/day) in 9 patients with NYHA Class II-III heart
Heilbrunn and colleagues[46] studied 15 patients failure symptoms due to IDC. When compared with
with idiopathic CHF, 3 in NYHA Class III, 7 in pretreatment baseline values, non-blind metoprolol
NYHA Class II, and 5 in NYHA Class I. Me- therapy was associated with a significant decrease
toprolol was titrated over 8 weeks to a mean dose in plasma noradrenaline level (0.381 ± 0.068 vs
of 103mg (range 75 to 150mg total daily). Myocar- 0.243 ± 0.095 µg/L; p < 0.05) and significant increase
dial β-adrenoceptor density significantly increased in lymphocyte β2-adrenoceptor density. Changes
after 6 months of metoprolol therapy when com- in β2-adrenoceptor density correlated positively
pared with pretreatment values (39 ± 7 to 80 ± 12 with left ventricular ejection fraction (r = 0.81, p <
fmol/mg protein; p < 0.005) to levels comparable 0.0001) and inversely with plasma noradrenaline
to previously reported normal values,[47] indicating level (r = –0.73, p < 0.001). The increased β2-
that metoprolol treatment could normalise (i.e. up- adrenoceptor density is unlikely to be directly re-
regulate) β-adrenoceptor number in CHF. The in-
lated to the pharmacological action of metoprolol,
crease in β-receptor density was accompanied by a
a β1-selective agent, but may represent a secondary
functional improvement in the positive inotropic
effect related to changes in plasma noradrenaline
response to administration of a synthetic β-agonist,
level or improved haemodynamic status.
dobutamine.
Rahman et al.[36] studied the effects of metoprolol
Maisel[33] reported the results from a non-blind
treatment on sympathetic activity in a prospective
trial of 6 to 12 months of metoprolol therapy (mean
non-blind, noncontrolled trial of 6 patients with
daily dose 76mg) in 15 patients with NYHA Class
heart failure. Muscle sympathetic nerve activity in
II-IV heart failure symptoms, due to IHD in 10
patients and IDC in 5. A nonrandomised, age- and the peroneal nerve, plasma noradrenaline and atrial
symptom-matched control group of 15 patients natriuretic factor levels were measured before and
with contraindications to metoprolol therapy was after an average of 20 months of metoprolol ther-
also studied. Plasma noradrenaline levels tended to apy (mean dosage of metoprolol of 45.8 mg/day).
decrease from baseline in patients treated with Plasma noradrenaline and atrial natriuretic peptide
metoprolol but not in untreated patients (baseline levels decreased during therapy when compared with
550 ± 90 vs 370 ± 70 ng/L metoprolol-treated and pretreatment values (2.9 ± 1.1 to 1.7 ± 0.5 nmol/L;
585 ± 100 ng/L untreated patients, p = 0.14). p = 0.07 and 59.7 ± 14.6 to 31.6 ± 10.7 pmol/L; p
Plasma adrenaline level did not change over the < 0.05; respectively). Sympathetic nerve activity de-
course of the study. creased by 50% when compared with pretreatment
Santostasi and colleagues[34] reported the findings values, even after adjustment for lower heart rate
of a non-blind, prospective study of the effects of 6 on metoprolol (p < 0.005). In addition, a 62% re-
months of metoprolol therapy (30 to 150 mg/day, duction in resistance to blood flow in the calf cir-
mean 78 mg/day) in 15 patients with NYHA Class culation was reported, consistent with vasodilation
II-III heart failure symptoms due to IDC. Plasma in the skeletal muscle bed served by the peroneal
noradrenaline levels decreased significantly from nerve. These data provide direct evidence that de-
baseline by 28% after 1 month of metoprolol ther- creased plasma noradrenaline levels in response to
apy (265±115 vs 183 ± 84 ng/L; p < 0.05) but not metoprolol are associated with decreased sympa-

 Adis International Limited. All rights reserved. Drugs 2000 Nov; 60 (5)
β-Blockers and Neurohormones in CHF 1005

thetic activity, but unfortunately lack a suitable During submaximal supine exercise, arterial
control group. plasma noradrenaline levels were significantly de-
Kukin and co-workers[37] studied the effect of creased in patients treated with metoprolol when
combined α1- and β1-blockade with doxazosin (4 compared with placebo at 12 months (9.16 ± 5.8 vs
mg/day) and metoprolol (100 mg/day) versus me- 11.8 ± 6.9 nmol/L; p < 0.04). Exercise arterial plas-
toprolol (100 mg/day) alone in 30 patients with ma adrenaline level was significantly lower in the
CHF NYHA Class II-IV symptoms, due to IHD in metoprolol group when compared with placebo at
10 patients and IDC in 20. After 3 months of ther- 6 months (0.99 ± 0.84 vs 1.48 ± 1.3 nmol/L; p =
apy, plasma noradrenaline levels decreased signif- 0.02), but this difference was no longer significant
icantly from baseline in patients treated with at 1 year. There were no significant differences in
metoprolol alone (567 ± 108 to 302 ± 41 ng/L; p < myocardial noradrenaline or adrenaline extraction
0.02) and patients treated with metoprolol plus during exercise at either 6 months or 1 year. Inter-
doxazosin (786 ± 91 to 434 ± 42 ng/L; p < 0.01), pretation of these findings is limited, since the num-
with no significant difference between the 2 treat- bers of patients were small, and for various reasons,
ment groups. only limited numbers of patients completed the ex-
ercise portion of the study.
Intermediate and Long Term Eichhorn and colleagues[39] studied 24 men with
Placebo-Controlled Studies
dilated cardiomyopathy, treated for 3 months, in a
The Metoprolol in Dilated Cardiomyopathy double-blind placebo-controlled trial (3 : 2 metop-
(MDC) trial randomised 383 patients with IDC to rolol 100 mg/day to placebo randomisation ratio).
treatment with metoprolol or placebo in a double- Arterial and coronary sinus catecholamine levels
blind fashion for 1 year.[48] The maximum dose was were determined before treatment and after 3 months
metoprolol 50mg 3 times a day, in a forced titration of blinded therapy. Coronary sinus noradrenaline
pattern. In a neurohormonal substudy of MDC, ar- levels were significantly lower in patients treated
terial and coronary sinus noradrenaline and adren- with metoprolol than in those receiving placebo
aline levels and coronary sinus blood flow were (569 ± 426 vs 422 ± 357 ng/L pretreatment and 730
measured at study entry and at 6 and 12 months in ± 445 vs 1114 ± 72 ng/L post-treatment, respectively;
41 patients.[38] Of these 41 patients, 14 completed a p < 0.03). Coronary sinus adrenaline showed a non-
12-month submaximal exercise test. After 6 months, significant trend to reduction during metoprolol
arterial plasma noradrenaline levels significantly therapy (p = 0.09). After completion of the blinded
decreased from baseline in patients treated with study, all the patients were switched to open-label
metoprolol when compared with the placebo group metoprolol and followed for an additional 3
(2.94 ± 1.7 to 1.72 ± 2.3 nmol/L vs 2.76 ± 3.3 to months.[49] Coronary sinus noradrenaline level did
3.2 ± 2.3 nmol/L; p < 0.03). After 12 months, arte- not correlate with the magnitude of change in left
rial plasma noradrenaline levels in the 2 treatment ventricular ejection fraction.
group were no longer different (p = 0.16). No sig- Paolisso and colleagues[40] studied 10 patients
nificant difference between metoprolol and placebo with CHF due to either valvular or nonischaemic
in arterial plasma adrenaline levels was noted at causes in a double-blind, crossover, randomised trial,
either 6 or 12 months. All groups demonstrated net with a target dose of metoprolol of 25mg twice
release of noradrenaline across the coronary circu- daily for 3 months. Patients treated with metop-
lation at rest, with no significant differences be- rolol and placebo demonstrated no significant dif-
tween patients treated with metoprolol and pla- ference in plasma noradrenaline levels (5.11 ± 0.38
cebo. Both groups demonstrated modest extraction vs 5.38 ± 0.16 nmol/L) or plasma adrenaline level
of adrenaline across the coronary circulation at rest (388.9 ± 54.9 vs 358.4 ± 58.7 nmol/L). The results
which tended to decrease nonsignificantly over time. of this trial are difficult to interpret because the

 Adis International Limited. All rights reserved. Drugs 2000 Nov; 60 (5)
1006 Baran et al.

cause of heart failure was related to valvular heart grees of sympathetic activation) are most affected
disease in 50% of the patients with no reported in- by bisoprolol use. In addition, the Fourier analysis
formation on left ventricular ejection fraction. was able to document an increase in the high fre-
quency heart rate variability band with bisoprolol
2.1.2 Bisoprolol treatment, which is thought to be an index of in-
Bisoprolol is a second generation, highly β1- creased parasympathetic tone.
selective receptor antagonist which has been re- In summary, clinical studies of first and second
cently reported to reduce long term mortality in generation β1-selective blockers have yielded con-
patients with CHF.[50] There are few published data flicting findings. The dosage and duration of treat-
on the effects of bisoprolol on indices of neuro- ment and differences in patient populations may
hormonal activation. account for some of the discordant findings. Most
Suwa and colleagues[41] compared the effects of studies reported decreased plasma levels of nor-
therapy with bisoprolol (5 mg/day for 10 weeks) adrenaline and increases in β-adrenoceptor den-
and diltiazem (60 to 120 mg/day) in a randomised sity.
study of 18 patients with NYHA Class III and IV
heart failure due to IDC. Five patients dropped out 2.2 Third Generation β-Blockers
of the study because of worsening heart failure
symptoms. In the remaining patients, plasma nor- The effects of third generation β-blockers on
adrenaline level decreased significantly from pre- plasma norepinephrine levels are presented in table II.
treatment values in 8 patients treated with bi-
2.2.1 Bucindolol
soprolol (487 ± 317 to 296 ± 177 ng/L; p < 0.05)
Bucindolol is a third generation, nonselective
and also tended to decrease in 5 patients treated with
β-blocker with weak vasodilating effects related to
diltiazem.
an undetermined mechanism. The effects of long
Additional studies reported the effects of bisop-
term bucindolol therapy were evaluated in the
rolol on heart rate variability, an index which pro-
Beta-Blocker Evaluation Survival Trial (BEST).
vides information on relative activation of the sym-
Preliminary reports of the results of this trial were
pathetic and parasympathetic nervous systems.
presented at the 1999 American Heart Association
The effects of bisoprolol therapy on heart rate
Scientific Sessions.[65] 2798 patients with ischaemic
variability were recorded in 2 reports from a sub-
and nonischaemic chronic heart failure were treated
study of 63 patients from the Cardiac Insufficiency
with bucindolol or matching placebo for an average
Bisoprolol Study (CIBIS).[51,52] The CIBIS study
of 2 years. Although plasma noradrenaline level de-
was a double-blind, placebo-controlled study of
creased during bucindolol therapy (by 20%), bu-
patients with NYHA Class III and IV heart failure,
cindolol therapy was associated with a modest,
with a mean follow-up of 1.9 ± 0.8 years.[53] All
nonsignificant (p = 0.109) 10% reduction in all-
patients in the Holter substudy had a 24-hour re-
cause mortality when compared with placebo, a
cording before randomisation, and a repeat after 2
nonsignificant increase in left ventricular ejection
months of blinded treatment. Nine patients were
fraction, and statistically significant 16% reduction
not included in the analyses because of inadequate
in risk for hospitalisation for worsening of heart
Holter recordings. All were in sinus rhythm. With
failure (p < 0.001).
bisoprolol, the average R-R interval significantly
increased (representing slower heart rates). While Noncontrolled Studies
the scatterplot length did not change, the scatter- Eichhorn and co-workers[54] examined the effects
plot width increased at the 50th, 75th and 90th per- of 3 months of bucindolol therapy in 15 patients
centiles of R-R length. This was confirmed by Fast with CHF (12 idiopathic and 3 ischaemic). Bu-
Fourier analysis. The interpretation of this finding cindolol was titrated in open-label fashion from
is that higher heart rates (associated with higher de- 12.5mg twice daily to 100mg twice daily. Plasma

 Adis International Limited. All rights reserved. Drugs 2000 Nov; 60 (5)
β-Blockers and Neurohormones in CHF 1007

Table II. Effects of third generation β-blockers on plasma noradrenaline (norepinephrine) levels in patients with heart failure
Reference Agent No. of patients Duration Plasma noradrenaline Comments
Eichhorn et al.[54] Bucindolol 15 3mo ↔ PRA ↓
Gilbert et al.[55]a Bucindolol 24 3mo ↓
Woodley et al.[56]a Bucindolol 51 3mo ↓ Significant only in idiopathic
group
Olsen et al.[57]a Carvedilol 60 3mo Arterial ↔
CS ↓
Krum et al.[58]a Carvedilol 56 14wk Trend ↓ Not significant vs placebo
Krum et al.[59] Carvedilol 15 14wk Trend ↓ Endothelin-1 ↓
Yoshikawa et al.[60] Nipradolol 18 6mo ↓ ↑ Lymphocyte
β2-adrenoceptor density
Virk & Davies[61] Xamoterol 30 Acute ↔
Sato et al.[62] Xamoterol 10 Acute ↔
McMurray et al.[63]a Xamoterol 51 10 days ↔ Postmyocardial infarction
Erlemeier et al.[64] Xamoterol 30 3mo ↔
a Placebo-controlled study design.
CS = coronary sinus; NS = not significant; PRA = plasma renin activity; ↔ indicates no change;↑ indicates increase; ↓ indicates decrease.

noradrenaline and renin levels were measured be- including 1 month of up-titration. Plasma nora-
fore and after 3 months of long term therapy. In drenaline level significantly decreased in patients
addition 123I metaiodobenzylguanidine (MIBG) receiving bucindolol when compared with patients
scanning was performed at baseline and study end. receiving placebo (423 ± 79 to 212 ± 101 ng/L vs
This isotope shares pharmacokinetic properties 362 ± 71 to 456 ± 272 ng/L; p = 0.01).
with noradrenaline, and thus its uptake reflects re- Woodley and colleagues[56] studied 51 patients
gional variations in noradrenaline uptake. Some stratified for ischaemic versus idiopathic cardiomy-
patients demonstrated a decrease in plasma nor- opathy, in a double-blind, placebo-controlled trial
adrenaline levels from baseline, but this was not a with bucindolol. This paper extended the work by
consistent finding, and no significant difference was the same group of investigators, discussed above,[55]
detected in the group (403 ± 231 to 408 ± 218 ng/L). with an identical study protocol for administration
In addition, quantitative analysis of the MIBG scans of bucindolol. The data from some patients are in-
showed no differences from baseline, consistent with cluded in both reports. 50 of the 51 patients toler-
the lack of change in plasma noradrenaline level. ated the initial bucindolol challenge dosages and
Plasma renin levels declined with bucindolol ther- 49 patients completed the 12-week protocol (22
apy (11.6 ± 13.4 to 4.3 ± 4.1 µg/L/h; p < 0.05), nonischaemic and 27 ischaemic). Plasma noradrena-
although the significance of this finding in patients line levels were similar at baseline in patients with
treated with ACE inhibitors is uncertain. idiopathic cardiomyopathy and those with ischaemic
cardiomyopathy. When the patients were grouped
Controlled Studies together without regard to the cause of cardiomy-
Gilbert and colleagues[55] assessed the effects of opathy, plasma noradrenaline levels tended to de-
bucindolol therapy in a randomised, double-blind, crease in those treated with bucindolol when com-
placebo-controlled study of 24 patients with IDC. pared with placebo (427 ± 55 to 297 ± 32 ng/L vs
23 patients who tolerated an initial challenge dose 409 ± 52 to 415 ± 73 ng/L; p = 0.09). When groups
of bucindolol 3.125mg were randomised to bu- were analysed on the basis of the cause of cardio-
cindolol versus placebo in a 3 : 2 ratio with weekly myopathy, plasma noradrenaline levels decreased
up-titration to a target dosage of 100mg twice significantly only in the group with idiopathic car-
daily. The total duration of therapy was 3 months, diomyopathy (idiopathic 461 ± 82 to 211 ± 39 ng/L

 Adis International Limited. All rights reserved. Drugs 2000 Nov; 60 (5)
1008 Baran et al.

vs ischaemic 397 ± 76 to 374 ± 34 ng/L; p < 0.05). 3-week run-in period was used to assess tolerability
There were no significant changes in plasma nor- of the medication. 49 of the 56 patients who com-
adrenaline level in the placebo group, regardless of pleted a 3-week tolerability phase of low dose car-
the cause of cardiomyopathy. vedilol were randomised to either placebo or con-
tinued up-titration of carvedilol to 25mg twice
2.2.2 Carvedilol daily in a blinded fashion. Plasma noradrenaline
Carvedilol has a β1 : β2 selectivity profile that levels tended to decrease with carvedilol but were
is slightly β1-selective, but for clinical purposes it not statistically different from that with placebo
is considered a third generation, nonselective β- (715 ± 135 to 588 ± 39 ng/L vs 580 ± 61 to 727 ± 133
blocker. Carvedilol is also an α-blocker with vas- ng/L; NS). Plasma adrenaline level significantly
odilating properties and may also mediate some of decreased with carvedilol therapy versus placebo
its antiadrenergic effects through interactions with
(80 ± 28 to 49 ± 8 ng/L vs 56 ± 10 to 109 ± 28 ng/L;
G-proteins.[66] A meta-analysis of published clini-
p < 0.02). Serum aldosterone level tended to de-
cal trials of carvedilol therapy of 6 to 12 months’
crease with carvedilol compared with placebo (31
duration demonstrated reductions in morbidity and
± 5 to 23 ± 4 ng/dl vs 25 ± 5 to 44 ± 12 ng/dl; p =
mortality.[67-70] However, mortality was not the
0.065).
primary end-point of any of these clinical trials.
In another report by Krum and colleagues,[59]
Olsen et al.[57] reported the findings of a pros-
plasma endothelin-1 and plasma noradrenaline and
pective, randomised study of carvedilol in 60 pa-
aldosterone levels were measured before and after
tients with NYHA Class II and III CHF (43 idio-
carvedilol therapy in 15 patients (10 nonischaemic,
pathic and 17 ischaemic in origin). Patients who
5 ischaemic). Plasma endothelin-1 levels decreased
tolerated a dosage of carvedilol 3.125mg twice
significantly with carvedilol therapy compared
daily for 1 week were randomised to either carvedi-
lol or placebo in a 3 : 2 allocation, with up-titration with the increases seen in the placebo group (7.6 ±
to 25 to 50mg twice a day over 3 months. 36 pa- 1.9 to 5.5 ± 1.4 ng/L vs 4.2 ± 1.8 to 6.4 ± 1.6 ng/L;
tients received carvedilol and 24 were assigned to p < 0.05). Plasma noradrenaline levels also demon-
placebo. Right heart catheterisation, arterial and strated a substantial decrease with carvedilol (1078
coronary sinus plasma noradrenaline and adrena- ± 364 to 639 ± 176 ng/L vs 547 ± 80 to 923 ± 153
line levels and right ventricular endomyocardial bi- ng/L) but without statistical significance, possibly
opsy were obtained before and after carvedilol ther- because of a β-error in a small patient population.
apy. Arterial noradrenaline level did not change, but Serum aldosterone levels significantly decreased
coronary sinus plasma noradrenaline level signifi- in patients treated with carvedilol when compared
cantly decreased during carvedilol therapy when with placebo (37 ± 11 to 18 ± 8 µg/L vs 19 ± 5 to
compared with placebo (665 ± 351 to 510 ± 343 ng/L 57 ± 14 µg/L; p < 0.05). Plasma endothelin-1 levels
vs 524 ± 295 to 748 ± 756 ng/L; p < 0.05). There correlated significantly with the plasma levels of
were no differences among treatment groups in noradrenaline and aldosterone. In addition, an in-
adrenaline levels in arterial and coronary sinus verse relation between endothelin-1 level and symp-
blood. There were no differences in β-adrenocep- tomatic improvement (reduction in NYHA func-
tor density or changes in β-receptor subtype ratios tional class) as well as submaximal exercise
from endomyocardial right ventricular biopsy tis- capacity was noted.
sue. Goldsmith and colleagues[71] studied the effects
Krum et al.[58] described the results of a double- of carvedilol on parasympathetic function, as as-
blind placebo-controlled study of carvedilol in a sessed by heart rate variability analysis in 10 pa-
group of 56 patients with CHF (38 described as tients with CHF. When compared with pretreatment
nonischaemic and 17 ischaemic) mostly in NYHA values, carvedilol therapy was associated with a sig-
Class III (35 patients) or IV (6 patients). A non-blind nificant increase in high frequency power, a correlate

 Adis International Limited. All rights reserved. Drugs 2000 Nov; 60 (5)
β-Blockers and Neurohormones in CHF 1009

of parasympathetic activity (geometric means 26 study. Interpretation of these data is confounded by

vs 64 ms2; p < 0.01). the inclusion of data from patients treated with car-
teolol, an agent with partial agonist activity. As
2.2.3 Nipradilol described in the following sections, agents with
Nipradilol is a nonselective β-blocker with vas- partial agonist activity may have effects on plasma
odilating activity which may be related to an endo- noradrenaline level which differ from pure antag-
thelium-dependent mechanism. Yoshikawa and onists.
co-workers[60] studied the effects of nipradilol on
neurohormonal activation in 18 patients with IDC, 2.2.4 Pindolol
in a nonrandomised, controlled cohort study. Pindolol, a nonselective β-blocker with partial
When compared with pretreatment values, 6 agonist activity, was administered to 10 patients
months of nipradilol therapy (mean daily dosage with IDC in an acute haemodynamic study.[73] Rest-
5.8 mg/day) tended to decrease plasma noradrena- ing haemodynamics and catecholamine and aldos-
line level (0.53 ± 0.44 to 0.36 ± 0.22 µg/L; NS) and terone levels were determined before and 2 hours
plasma atrial natriuretic factor level (152 ± 99 to after administration of a single oral dose of pin-
45 ± 42 ng/L; NS) and significantly increased lym- dolol 5mg on one day and a single oral dose of
phocyte β-adrenoceptor density (6.7 ± 1.8 to 9.8 ± pindolol 10mg 24 hours later. There was no signif-
1.4 fmol/ mg protein; p < 0.05). No significant icant change from pretreatment values in plasma
changes in neurohormonal levels or β2-adrenocep- noradrenaline level after the 5mg dose, but there
tor density occurred in the control group. There were was a trend towards increasing noradrenaline after
no significant differences reported in plasma levels the single 10mg dose (455 ± 474 vs 572 ± 450 ng/L;
of adrenaline, renin, aldosterone or arginine-vaso- p < 0.10). Plasma adrenaline was significantly in-
pressin during the study period in either treated or creased from baseline with the 5mg dose (83 ± 60
control patients. vs 254 ± 342 ng/L; p < 0.05), but not with the 10mg
The findings from this group of investigators dose (133 ± 16 vs 100 ± 78 ng/L; p < 0.10). There
were extended by an additional study in 6 patients were no significant changes in serum aldosterone
(5 idiopathic and 1 ischaemic cardiomyopathy) with either dose of pindolol.
treated with carteolol (mean dosage 7.3 mg/day), a
nonselective β-blocker with partial agonist activ- 2.2.5 Xamoterol
ity.[72] Neurohumoral levels before and after 2 weeks Xamoterol is a mixed β-agonist/antagonist, with
and 6 months of therapy, in response to nipradilol approximately 43% of the agonist activity of the
or carteolol, are reported as pooled data. When full agonist isoprenaline. The ratio of β-adrenocep-
compared with pretreatment values, plasma nor- tor antagonism to β-adrenoceptor agonism depends
adrenaline level significantly decreased after 2 on sympathetic tone.[74] Xamoterol provides mild
weeks (769 ± 509 to 409 ± 315 ng/L; p = 0.01) and stimulation when sympathetic tone is low, and pro-
6 months (400 ± 368 ng/L; p = 0.0066). Plasma tects from overstimulation when it is high. While
atrial natriuretic factor level significantly decreased xamoterol binds both β1- and β2-adrenoceptors in
from pretreatment values after 2 weeks (156 ± 85 animals, with β1 affinity 10 to 100 times greater
to 86 ± 67 ng/L; p < 0.01), but not after 6 months than β2 affinity, no evidence of β2 stimulation or
(106 ± 135 ng/L). Lymphocyte β2-adrenoceptor den- appreciable antagonism has been identified in hu-
sity increased significantly from pretreatment val- mans. However, the effects of xamoterol on β2-
ues after 2 weeks of therapy (7.6 ± 3.3 to 10.2 ± adrenoceptors in human heart failure, which is
4.4 fmol/mg protein; p= 0 .05), but not after 6 characterised in part by a relative increase in myo-
months (7.8 ± 3.0 fmol/mg protein). There were no cardial β2-adrenoceptor density, have not been de-
significant changes in the adrenaline, renin, aldo- termined. Clinical trials in heart failure have dem-
sterone or arginine-vasopressin levels during the onstrated improved haemodynamics and exercise

 Adis International Limited. All rights reserved. Drugs 2000 Nov; 60 (5)
1010 Baran et al.

capacity in short term trials, but increased mortal- In summary, the majority of clinical trials with
ity during long term therapy.[75] third generation nonselective β-blockers demonstrate
The acute effects of xamoterol on neurohormo- a decrease in plasma levels of noradrenaline. In con-
nal activation have been reported in 2 studies. Virk trast to β1-selective agents, no increase in myocar-
and Davies[61] investigated the acute effects of in- dial β-adrenoceptor density was observed, although
travenous xamoterol (0.2 mg/kg) in 30 patients lymphocyte β-adrenoceptor (β2-adrenoceptor) den-
with NYHA Class II-III heart failure of mixed causes. sity increased. It is important to note that many of
Administration of xamoterol did not change plasma these trials included a tolerability phase, in which
noradrenaline level from baseline values at rest (450 individuals unable to tolerate initiation of low dose
± 46 vs 499 ± 54 ng/L) or during maximal exercise β-blockers were excluded from randomisation. The
(1671 ± 198 vs 1727 ± 201 ng/L). Sato and col- exclusion of these patients confounds comparison of
leagues[62] administered xamoterol 0.15 mg/kg to the findings with those from earlier trials of first
10 patients with mild to moderate left ventricular and second generation agents. The neurohormonal
dysfunction (mean left ventricular ejection fraction response to agents with partial agonist activity is
0.52) secondary to mixed valvular, ischaemic hy- distinct from that of pure receptor antagonists.
pertrophic and idiopathic causes. Six matched pa- Failure to suppress neurohormonal activation dur-
tients received saline control injection. Plasma nor- ing long term therapy may have contributed to the
adrenaline level did not change from pretreatment adverse effects on mortality reported with xam-
values in patients receiving xamoterol (234 ± 87 vs oterol.[75]
210 ± 109 ng/L) or saline. Plasma noradrenaline
level significantly increased from pretreatment 3. Comparative Studies
values at peak exercise in patients treated with
Few direct comparisons of β-blockers are avail-
xamoterol (754 ± 279 vs 1068 ± 551 ng/L; p < 0.05)
able. A small intermediate term clinical trial sug-
but not in patients treated with saline.
gested that the clinical effects of metoprolol and car-
The effects of short term therapy with xamoterol vedilol are distinct in patients with heart failure.[76]
have been reported by 2 groups of investigators. A long term mortality trial comparing metoprolol and
McMurray and colleagues[63] investigated the ef- carvedilol is currently in progress (COMET).[20]
fects of xamoterol (400 mg/day for 10 days) on Newton and Parker[77] compared the acute effects
neurohormonal activation after myocardial infarc- of intravenous metoprolol and propranolol on car-
tion in a randomised, double-blind study. Of the 51 diac sympathetic activity in 18 patients with NYHA
patients in the study, 16 required diuretics for the Class III-IV symptoms, secondary to IHD in 12
treatment of heart failure. Plasma levels of nor- patients and IDC in 6. Systemic and coronary cir-
adrenaline and atrial natriuretic factor and plasma culation steady-state noradrenaline kinetics were
renin activity did not differ in patients treated with assessed with tritium-labelled noradrenaline be-
xamoterol or placebo. Erlemeier and colleagues[64] fore and after a short term intravenous infusion of
investigated the effects of xamoterol (400 mg/day either propranolol (n = 9) or metoprolol (n = 9)
for 3 months) on plasma noradrenaline level at rest titrated to produce a 15% decline in heart rate, rate
and during submaximal bicycle exercise in a ran- of change of left ventricular pressure (dP/dT) or
domised, placebo-controlled trial of 30 patients mean arterial blood pressure, or a 5mm Hg increase
with NYHA Class II-III heart failure secondary to in left ventricular end-diastolic pressure. A mean
IHD. When compared with pretreatment values, dose of propranolol 2mg (range 1 to 3.5mg) or
xamoterol therapy did not change plasma nor- metoprolol 3.6mg (range 2 to 6mg) was adminis-
adrenaline levels at rest (491 ± 318 vs 488 ± 226 tered. Total body noradrenaline spillover was not
ng/L) or during exercise (758 ± 336 vs 675 ± 325 affected by acute β-adrenoceptor blockade with ei-
ng/L). ther agent. Cardiac noradrenaline spillover was

 Adis International Limited. All rights reserved. Drugs 2000 Nov; 60 (5)
β-Blockers and Neurohormones in CHF 1011

significantly reduced after propranolol treatment Gilbert and colleagues[76] published an analysis
(277± 55 to 262 ± 53 pmol/min; p< 0 .05) and of the findings from 2 separate prospective, placebo-
significantly increased after metoprolol treatment controlled studies of metoprolol and carvedilol
(233 ± 57 to 296 ± 82 pmol/min; p < 0.05). Plasma limited to the subgroup of patients with IDC. Some
noradrenaline levels tended to increase in response of the findings were previously published, as sum-
to both propranolol and metoprolol (2.5 ± 0.7 to marised above.[76] The dosage used in the studies
2.9 ± 0.8 and 2.3 ± 0.6 to 3.5 ± 0.8 nmol/L, respec- of each of these agents (metoprolol 125 to 150 mg/
tively). Propranolol treatment did not induce day, carvedilol 50 to 100 mg/day) produced similar
changes in adrenaline level, whereas metoprolol decreases in resting heart rate, suggesting an equiv-
was associated with a significant increase in coro- alent degree of β-adrenoceptor blockade. Plasma
nary sinus adrenaline level (0.4 ± 0.3 to 0.7 ± 0.4 noradrenaline levels from coronary sinus blood de-
nmol/L; p < 0.05) and a borderline significant in- creased during therapy with carvedilol when com-
crease in arterial plasma adrenaline levels (0.6 ± pared with placebo (655 ± 351 to 510 ± 343 ng/L
0.2 to 0.9 ± 0.4 nmol/L; p = 0.06). vs 524 ± 295 to 748 ± 756 ng/L; p < 0.05) but
Yamada and colleagues[78] investigated the ef- tended to increase from baseline during therapy
fects of β-adrenoceptor blockade with metoprolol with metoprolol (533 ± 327 to 610 ± 490 ng/L). A
(mean dosage 31 mg/day) and bisoprolol (mean comparison of the change from baseline in coro-
dosage 2.4 mg/day) on serial lymphocyte β2- nary sinus plasma noradrenaline levels for the 2
adrenoceptor density measurements in 12 patients agents showed borderline statistical significance
with IDC. Pretreatment lymphocyte β2-adrenocep- (–155 ± 219 ng/L for carvedilol vs 77 ± 488 ng/L
tor density was decreased in patients with heart for metoprolol; p = 0.07). The transmyocardial gra-
failure when compared with normal volunteers (886 dient for noradrenaline decreased during therapy
± 84 vs 1657 ± 108 binding sites/cell; p < 0.001). with carvedilol when compared with placebo (262
Therapy with both agents was associated with sig- ± 358 to 110 ± 346 ng/L vs 135 ± 162 to 320 ± 400
nificant increases in lymphocyte β-adrenoceptor ng/L; p < 0.05) but did not change during therapy
density. In the 6 patients treated with metoprolol, with metoprolol. Right ventricular endomyocard-
β2-adrenoceptor density increased from 27% over ial β-adrenoceptor density increased during ther-
pretreatment values after 1 week of therapy to 158% apy with metoprolol compared with placebo (47.8
over pretreatment values after 2 years of therapy ± 21.4 to 75.0 ± 40.5 fmol/mg vs 41.9 ± 23.3 to
(both p < 0.01 vs pretreatment values). In the 6 40.0 ± 16.3 fmol/mg; p < 0.05) but did not change
patients treated with bisoprolol, β2-adrenoceptor during therapy with carvedilol. Post-treatment β-
density increased from 31% over pretreatment val- adrenoceptor density was significantly greater in
ues after 1 week of therapy to 72% over pretreat- patients treated with metoprolol than in those
ment values after 2 years of therapy (both p < 0.01 treated with carvedilol (75.0 ± 40.5 vs 28.7 ± 21.5
vs pretreatment values). A statistical comparison fmol/mg; p < 0.05). Although the study design is
of the effects of the 2 treatments was not reported. not a true randomised comparison of metoprolol
In both treatment groups serial measurements re- and carvedilol, these findings demonstrate differ-
vealed significant increases in β2-adrenoceptor ent effects of these 2 agents on the β-adrenoceptor
density after 3 months of therapy and a slow con- system, particularly in regard to the effect on β-
tinuous increase in β2-adrenoceptor density over adrenoceptor density. Differences in the pharma-
the 2-year course of the study. The mechanism of cological properties of these 2 agents (β1-selectivity
change in β2-adrenoceptor density in response to and G-protein binding) or differences in the pla-
these agents is uncertain, as both metoprolol and cebo groups for the 2 studies may account for these
bisoprolol are β1-selective blockers. findings.

 Adis International Limited. All rights reserved. Drugs 2000 Nov; 60 (5)
1012 Baran et al.

4. Possible Mechanisms of Action trend towards increased mortality in patients treated

with moxonidine when compared with placebo (p =
The primary mechanism by which β-blockers 0.20; MOXCON trial, presented at the Heart Fail-
alter neurohormonal activation cannot be determined ure Society of America Scientific Sessions, Sep-
from available experimental and clinical studies. It tember 1999.
has been proposed that the effect of these agents on
neurohormonal activation is entirely due to reflex 5. Clinical Implications
mechanisms secondary to drug-induced changes in
cardiovascular function. Thus, early increases in The clinical implications of decreased activity
neurohormonal activation are attributable to acute of the sympathetic nervous system in patients re-
haemodynamic deterioration, and late decreases in ceiving β-blockers, in whom the primary effector
neurohormonal activation are attributable to chronic mechanism of the sympathetic neurotransmission
improvements in haemodynamic function. How- system is blocked, is uncertain. Clinical studies have
ever, several studies demonstrated acute decreases not fully assessed the effects of β-blocker therapy
in neurohormonal activation which occurred be- on the β-adrenoceptor complex inclusive of G-pro-
fore any haemodynamic improvement would be teins, adenylate cyclase activity and β-adrenoceptor
expected. In these studies, a specific pharmacolog- kinase activity. Differences among agents in their
ical action has been postulated. In addition to reflex acute and chronic effects on indices of neurohor-
haemodynamic mechanisms, β-blockers could de- monal activation may be related to basic differ-
crease neurohormonal activation by altering neu- ences in pharmacological properties. β1-Selectivity,
ronal uptake of noradrenaline or by central inhibi- lipophilicity, inverse agonist activity, partial ago-
tion of sympathetic output. Experimental studies nist activity, G-protein binding properties, and va-
demonstrating increased myocardial noradrenaline sodilator effects may all directly or indirectly mod-
content in response to β-adrenergic receptor block- ulate the effects on neurohormonal activation. In
ade suggest that increased neuronal uptake may addition, ancillary pharmacological properties such
play a role. as membrane stabilisation effects and antioxidant
In addition to direct and/or indirect effects of effects may contribute to the long term clinical ef-
plasma neurohormonal levels, β-blockers alter the fects of these agents.[79] The efficacy of a newer,
β-adrenergic effector mechanism by increasing β- third generation agent, carvedilol, may be further
adrenoceptor density (first and second generation enhanced by down-regulation of β-adrenoceptors,
agents) and partially restoring functional coupling and more comprehensive antiadrenergic effects by
of the β-adrenoceptor with its intracellular signal virtue of multiple receptor blockade and G-protein
transduction pathways (third generation agents). interactions. Cardiac noradrenaline levels are lower
Whether changes in expression of components of with bucindolol and carvedilol than with metoprolol
the β-adrenoceptor complex are merely part of the for these reasons. However, bucindolol therapy was
restoration of normal myocardial phenotype or ac- associated with less favourable effect on mortality
count for efficacy of β-blockers has not been estab- (BEST trial) than therapy with metoprolol, bis-
lished. oprolol or carvedilol in placebo-controlled trials.
Although β-blockers probably mediate their ef- Differences in pharmacology and haemodyna-
fects primarily by inhibition of sympathetic adren- mic and neurohormonal effects may not be critical
ergic stimulation, other unidentified factors may be to the long term clinical benefits of β-blocker ther-
important. Interestingly, a placebo-controlled clin- apy, as clinical trials with carvedilol, metoprolol
ical trial of moxonidine, a centrally acting α-agonist and bisoprolol, agents with distinct pharmacology,
which inhibits sympathetic outflow, lowers plasma are associated with substantial reductions in long
noradrenaline levels and causes peripheral vasodi- term morbidity and mortality during long term
lation, was prematurely terminated because of a therapy.[50,70] The ongoing COMET trial, which is

 Adis International Limited. All rights reserved. Drugs 2000 Nov; 60 (5)
β-Blockers and Neurohormones in CHF 1013

1. First and second generation and third generation

Post-treatment change in plasma noradrenaline (ng/L)

100
β-blockers appear to be associated with evidence
0 of decreased neurohormonal activation, with dif-
ferential effects on β-adrenoceptor density. Agents
with partial agonist activity appear to differ from
−100
pure antagonists, with some studies reporting evi-
dence of increased neurohormonal activation. The
−200
decrease in plasma noradrenaline level during in-
termediate and long term therapy correlates closely
−300 with the pretreatment plasma noradrenaline level
(fig. 3). Whether decreased plasma noradrenaline
−400 level during long term therapy is attributable to de-
r = 0.759 creased noradrenaline release (as opposed to in-
−500 creased neuronal uptake) is uncertain, as only one
250 500 750 1000 1250 small study has directly demonstrated decreased
Pretreatment plasma noradrenaline (ng/L) sympathetic nervous system activity during long
Fig. 3. Change in plasma noradrenaline (norepinephrine) level term therapy. Additional studies are needed to de-
during long term β-blocker therapy is closely correlated with termine the underlying mechanisms of changes in
pretreatment plasma noradrenaline levels. Data for this figure
neurohormonal activation during β-blocker ther-
were derived from reviewed studies that provided plasma nor-
adrenaline levels before and after therapy with β-blockers (ta- apy.
bles I and II). Data from the study by Paolisso et al.[34] are
excluded, as this small study included many patients with val-
vular heart disease as the cause of heart failure and did not
References
1. Packer M. Pathophysiology of chronic heart failure. [review].
report ejection fraction.
Lancet 1992; 340 (8811): 88-92
2. Katz SD. Mechanisms and implications of endothelial dysfunc-
tion in congestive heart failure. Curr Opin Cardiol 1997; 12
(3): 259-64
a prospective, randomised comparison between car- 3. Pfeffer MA. Angiotensin-converting enzyme inhibition in con-
vedilol and metoprolol, may provide important data gestive heart failure: benefit and perspective. Am Heart J 1993;
to determine whether differences in basic pharma- 126 (3 Pt 2): 789-93
4. Cohn JN, Levine TB, Olivari MT, et al. Plasma norepinephrine
cology are associated with differences in patient out- as a guide to prognosis in patients with chronic congestive
comes; however, the dosage of metoprolol used in heart failure. N Engl J Med 1984; 311 (13): 819-23
this trial is only half that used in MERIT-HF. 5. Chidsey CA, Braunwald E, Morrow AG. Catecholamine excre-
tion and cardiac stores of norepinephrine in congestive heart
failure. Am J Med 1965; 39: 442-51
6. Summary and Conclusions 6. Oren RM, Roach PJ, Schobel HP, et al. Sympathetic responses
of patients with congestive heart failure to cold pressor stim-
In summary, current knowledge of the effects of ulus. Am J Cardiol 1991; 67 (11): 993-1001
7. Bristow MR, Ginsburg R, Minobe W, et al. Decreased catechol-
β-blockers on neurohormonal activation in pa- amine sensitivity and beta-adrenergic-receptor density in fail-
tients with CHF is largely derived from studies of ing human hearts. N Engl J Med 1982; 307 (4): 205-11
small patient populations. Short term therapy with 8. Bristow MR, Ginsburg R, Umans V, et al. Beta 1- and beta
2-adrenergic-receptor subpopulations in nonfailing and failing
β-blockers is associated with a variable response human ventricular myocardium: coupling of both receptor sub-
which may be determined by the pharmacology of types to muscle contraction and selective beta 1-receptor
the agent under study and the baseline characteristics down-regulation in heart failure. Circ Res 1986; 59 (3): 297-309
9. Horn EM, Corwin SJ, Steinberg SF, et al. Reduced lymphocyte
of the patient population. Long term therapy with stimulatory guanine nucleotide regulatory protein and beta-
β-blockers (devoid of partial agonist activity) is adrenergic receptors in congestive heart failure and reversal
associated with evidence of decreased plasma mark- with angiotensin converting enzyme inhibitor therapy. Circu-
lation 1988; 78 (6): 1373-9
ers of activation of the sympathetic nervous sys- 10. Bristow MR. Mechanism of action of beta-blocking agents in
tem, the renin-angiotensin system, and endothelin- heart failure. Am J Cardiol 1997; 80 (11A): 26L-40L

 Adis International Limited. All rights reserved. Drugs 2000 Nov; 60 (5)
1014 Baran et al.

11. Mann DL, Kent RL, Parsons B, et al. Adrenergic effects on the induced cardiac damage in rats. J Cardiovasc Pharmacol
biology of the adult mammalian cardiocyte. Circulation 1992; 1991; 17 (4): 656-61
85 (2): 790-804 28. Iaccarino G, Tomhave ED, Lefkowitz RJ, et al. Reciprocal in
12. Colucci WS. The effects of norepinephrine on myocardial biol- vivo regulation of myocardial G protein-coupled receptor ki-
ogy: implications for the therapy of heart failure [in process nase expression by beta-adrenergic receptor stimulation and
citation]. Clin Cardiol 1998; 21 12 Suppl. 1: I20-4 blockade. Circulation 1998; 98 (17): 1783-9
13. Packer M. Beta-blockade in heart failure. Basic concepts and 29. Bohm M, Deutsch HJ, Hartmann D, et al. Improvement of
clinical results. Am J Hypertens 1998; 11 (1 Pt 2): 23S-37S postreceptor events by metoprolol treatment in patients with
14. Hall SA, Cigarroa CG, Marcoux L, et al. Time course of im- chronic heart failure. J Am Coll Cardiol 1997; 30 (4): 992-6
provement in left ventricular function, mass and geometry in 30. Swedberg K, Hjalmarson A, Holmberg S. Effects of work and
patients with congestive heart failure treated with beta- adren- acute beta-receptor blockade on myocardial noradrenaline re-
ergic blockade. J Am Coll Cardiol 1995; 25 (5): 1154-61 lease in congestive cardiomyopathy. Clin Cardiol 1979; 2 (6):
15. Doughty RN, Whalley GA, Gamble G, et al. Left ventricular 424-30
remodeling with carvedilol in patients with congestive heart 31. Nemanich JW, Veith RC, Abrass IB, et al. Effects of metoprolol
failure due to ischemic heart disease: Australia-New Zealand on rest and exercise cardiac function and plasma catechola-
Heart Failure Research Collaborative Group. J Am Coll Car- mines in chronic congestive heart failure secondary to isch-
diol 1997; 29 (5): 1060-6 emic or idiopathic cardiomyopathy. Am J Cardiol 1990; 66
16. Andersson B, Stromblad SO, Lomsky M, et al. Heart rate de- (10): 843-8
pendency of cardiac performance in heart failure patients 32. Andersson B, Lomsky M, Waagstein F. The link between acute
treated with metoprolol [see comments]. Eur Heart J 1999; 20 haemodynamic adrenergic beta-blockade and long-term effects
(8): 575-83 in patients with heart failure. A study on diastolic function,
17. Eichhorn EJ, Bristow MR. Practical guidelines for initiation of heart rate and myocardial metabolism following intravenous
beta-adrenergic blockade in patients with chronic heart failure metoprolol. Eur Heart J 1993; 14 (10): 1375-85
[editorial]. Am J Cardiol 1997; 79 (6): 794-8 33. Maisel AS. Beneficial effects of metoprolol treatment in con-
18. Slatton ML, Eichhorn EJ. Beta-blocker therapy for heart failure. gestive heart failure. Reversal of sympathetic-induced alter-
Curr Opin Cardiol 1996; 11 (3): 263-8 ations of immunologic function. Circulation 1994; 90 (4):
19. Doughty RN, Rodgers A, Sharpe N, et al. Effects of beta- 1774-80
blocker therapy on mortality in patients with heart failure. A 34. Santostasi G, Fraccarollo D, Dorigo P, et al. Early reduction in
systematic overview of randomized controlled trials. Eur
plasma norepinephrine during beta-blocking therapy with
Heart J 1997; 18 (4): 560-5
metoprolol in chronic heart failure [in process citation]. J Card
20. Bristow MR, Roden RL, Lowes BD, et al. The role of third-
Fail 1998; 4 (3): 177-84
generation beta-blocking agents in chronic heart failure. Clin
35. Ishida S, Makino N, Masutomo K, et al. Effect of metoprolol
Cardiol 1998; 21 12 Suppl. 1: I3-I13
on the beta-adrenoceptor density of lymphocytes in patients
21. Vogel JH, Chidsey CA. Cardiac adrenergic activity in experi-
with dilated cardiomyopathy. Am Heart J 1993; 125 (5 Pt 1):
mental heart failure assessed with beta receptor blockade. Am
1311-5
J Cardiol 1969; 24 (2): 198-208
36. Rahman MA, Hara K, Daly PA, et al. Reductions in muscle
22. Levett JM, Marinelli CC, Lund DD, et al. Effects of beta-block-
sympathetic nerve activity after long-term metoprolol for di-
ade on neurohumoral responses and neurochemical markers
lated cardiomyopathy: preliminary observations. Br Heart J
in pacing-induced heart failure. Am J Physiol 1994; 266 (2 Pt
2): H468-75 1995; 74 (4): 431-6
37. Kukin ML, Kalman J, Mannino M, et al. Combined alpha-beta
23. Yamada S, Ohkura T, Yamadera T, et al. Abnormality in plasma
catecholamines and myocardial adrenoceptors in cardiomyo- blockade (doxazosin plus metoprolol) compared with beta
pathic BIO 53.58 Syrian hamsters and improvement by me- blockade alone in chronic congestive heart failure. Am J Car-
toprolol treatment. J Pharmacol Exp Ther 1997; 283 (3): diol 1996; 77 (7): 486-91
1389-95 38. Andersson B, Hamm C, Persson S, et al. Improved exercise
24. Tomita T, Murakami T, Iwase T, et al. Chronic dynamic exer- hemodynamic status in dilated cardiomyopathy after beta-ad-
cise improves a functional abnormality of the G stimulatory renergic blockade treatment. J Am Coll Cardiol 1994; 23 (6):
protein in cardiomyopathic BIO 53.58 Syrian hamsters. Cir- 1397-404
culation 1994; 89 (2): 836-45 39. Eichhorn EJ, Heesch CM, Barnett JH, et al. Effect of metoprolol
25. Latini R, Masson S, Jeremic G, et al. Comparative efficacy of on myocardial function and energetics in patients with non-
a DA2/alpha2 agonist and a beta-blocker in reducing adren- ischemic dilated cardiomyopathy: a randomized, double-
ergic drive and cardiac fibrosis in an experimental model of blind, placebo-controlled study. J Am Coll Cardiol 1994; 24
left ventricular dysfunction after coronary artery occlusion. J (5): 1310-20
Cardiovasc Pharmacol 1998; 31 (4): 601-8 40. Paolisso G, Gambardella A, Marrazzo G, et al. Metabolic and
26. Warner AL, Bellah KL, Raya TE, et al. Effects of beta-adrenergic cardiovascular benefits deriving from beta-adrenergic block-
blockade on papillary muscle function and the beta-adrenergic ade in chronic congestive heart failure. Am Heart J 1992; 123
receptor system in noninfarcted myocardium in compensated (1): 103-10
ischemic left ventricular dysfunction. Circulation 1992; 86 (5): 41. Suwa M, Ito T, Otake Y, et al. Comparison of the therapeutic
1584-95 effects of the beta-blocking agent bisoprolol and the calcium-
27. Fujita N, Hiroe M, Ohta Y, et al. Chronic effects of metoprolol blocking agent diltiazem in patients with heart failure due to
on myocardial beta-adrenergic receptors in doxorubicin- dilated cardiomyopathy. Jpn Circ J 1996; 60 (10): 767-73

 Adis International Limited. All rights reserved. Drugs 2000 Nov; 60 (5)
β-Blockers and Neurohormones in CHF 1015

42. Anonymous. Effect of metoprolol CR/XL in chronic heart fail- ure: a double-blind randomized study [see comments]. J Am
ure: Metoprolol CR/XL randomised intervention trial in con- Coll Cardiol 1995; 25 (6): 1225-31
gestive heart failure (MERIT-HF) [see comments]. Lancet 58. Krum H, Sackner-Bernstein JD, Goldsmith RL, et al. Double-
1999; 353 (9169): 2001-7 blind, placebo-controlled study of the long-term efficacy of
43. Haber HL, Simek CL, Gimple LW, et al. Why do patients with carvedilol in patients with severe chronic heart failure. Cir-
congestive heart failure tolerate the initiation of beta-blocker culation 1995; 92 (6): 1499-506
therapy? Circulation 1993; 88 (4 Pt 1): 1610-9 59. Krum H, Gu A, Wilshire-Clement M, et al. Changes in plasma
44. Waagstein F, Caidahl K, Wallentin I, et al. Long-term beta- endothelin-1 levels reflect clinical response to beta-blockade
blockade in dilated cardiomyopathy. Effects of short- and in chronic heart failure. Am Heart J 1996; 131 (2): 337-41
long-term metoprolol treatment followed by withdrawal and 60. Yoshikawa T, Handa S, Akaishi M, et al. Effect of a new beta-
readministration of metoprolol [see comments]. Circulation blocker, nipradilol, on cardiac function and neurohumoral
1989; 80 (3): 551-63 factors in idiopathic dilated cardiomyopathy. Jpn Circ J 1996;
45. Andersson B, Blomstrom-Lundqvist C, Hedner T, et al. Exer- 60 (5): 285-92
cise hemodynamics and myocardial metabolism during long- 61. Virk SJ, Davies MK. Effects of xamoterol on resting and exer-
term beta- adrenergic blockade in severe heart failure [see cise haemodynamics in patients with chronic heart failure. Br
comments]. J Am Coll Cardiol 1991; 18 (4): 1059-66 J Clin Pharmacol 1989; 28 Suppl. 1: 15S-22S
46. Heilbrunn SM, Shah P, Bristow MR, et al. Increased beta-re- 62. Sato H, Inoue M, Matsuyama T, et al. Hemodynamic effects of
ceptor density and improved hemodynamic response to cate- the beta 1-adrenoceptor partial agonist xamoterol in relation
cholamine stimulation during long-term metoprolol therapy to plasma norepinephrine levels during exercise in patients
in heart failure from dilated cardiomyopathy. Circulation with left ventricular dysfunction. Circulation 1987; 75 (1):
1989; 79 (3): 483-90 213-20
47. Fowler MB, Laser JA, Hopkins GL, et al. Assessment of the 63. McMurray JJ, Lang CC, MacLean D, et al. Neuroendocrine
beta-adrenergic receptor pathway in the intact failing human changes post myocardial infarction: effects of xamoterol. Am
heart: progressive receptor down-regulation and subsensitiv- Heart J 1990; 120 (1): 56-62
ity to agonist response. Circulation 1986; 74 (6): 1290-302 64. Erlemeier HH, Kupper W, Bleifeld W. Exercise capacity, ar-
48. Waagstein F, Bristow MR, Swedberg K, et al. Beneficial effects rhythmias, humoral and chemical parameters during long-
of metoprolol in idiopathic dilated cardiomyopathy. Metop- term therapy with xamoterol. Int J Cardiol 1990; 28 (2):
rolol in dilated cardiomyopathy (MDC) trial study group [see 197-208
comments]. Lancet 1993; 342 (8885): 1441-6 65. Domanski MJ, BEST Investigators. Beta-blocker Evaluation of
49. Eichhorn EJ, Heesch CM, Risser RC, et al. Predictors of sys- Survival Trial (BEST). J Am Coll Cardiol 2000; 35 Suppl. A:
tolic and diastolic improvement in patients with dilated car- 202-3
diomyopathy treated with metoprolol. J Am Coll Cardiol 66. Frishman WH. Carvedilol. N Engl J Med 1998; 339 (24): 1759-65
1995; 25 (1): 154-62 67. Bristow MR, Gilbert EM, Abraham WT, et al. Carvedilol pro-
50. The Cardiac Insufficiency Bisoprolol Study II (CIBIS-II): a duces dose-related improvements in left ventricular function
randomised trial. CIBIS-II investigators and committees. and survival in subjects with chronic heart failure. MOCHA
Lancet 1999; 353: 9-13 Investigators [comment] [see comments]. Circulation 1996;
51. Copie X, Pousset F, Lechat P, et al. Effects of beta-blockade 94 (11): 2807-16
with bisoprolol on heart rate variability in advanced heart 68. Colucci WS, Packer M, Bristow MR, et al. Carvedilol inhibits
failure: analysis of scatterplots of R-R intervals at selected clinical progression in patients with mild symptoms of heart
heart rates. Am Heart J 1996; 132 (2 Pt 1): 369-75 failure. US Carvedilol Heart Failure Study Group. Circula-
52. Pousset F, Copie X, Lechat P, et al. Effects of bisoprolol on tion 1996; 94 (11): 2800-6
heart rate variability in heart failure. Am J Cardiol 1996; 77 69. Packer M, Colucci WS, Sackner-Bernstein JD, et al. Double-
(8): 612-7 blind, placebo-controlled study of the effects of carvedilol in
53. Anonymous. A randomized trial of beta-blockade in heart fail- patients with moderate to severe heart failure. The PRECISE
ure. The Cardiac Insufficiency Bisoprolol Study (CIBIS). Trial. Prospective randomized evaluation of carvedilol on
CIBIS Investigators and committees [see comments]. Circu- symptoms and exercise [see comments]. Circulation 1996; 94
lation 1994; 90 (4): 1765-73 (11): 2793-9
54. Eichhorn EJ, McGhie AL, Bedotto JB, et al. Effects of 70. Packer M, Bristow MR, Cohn JN, et al. The effect of carvedilol
bucindolol on neurohormonal activation in congestive heart on morbidity and mortality in patients with chronic heart fail-
failure. Am J Cardiol 1991; 67 (1): 67-73 ure. N Engl J Med 1996; 334 (21): 1349-55
55. Gilbert EM, Anderson JL, Deitchman D, et al. Long-term beta- 71. Goldsmith RL, Bigger JT, Bloomfield DM, et al. Long-term
blocker vasodilator therapy improves cardiac function in id- carvedilol therapy increases parasympathetic nervous system
iopathic dilated cardiomyopathy: a double-blind, randomized activity in chronic congestive heart failure. Am J Cardiol
study of bucindolol versus placebo. Am J Med 1990; 88 (3): 1997; 80 (8): 1101-4
223-9 72. Yoshikawa T, Handa S, Anzai T, et al. Early reduction of neu-
56. Woodley SL, Gilbert EM, Anderson JL, et al. Beta-blockade rohumoral factors plays a key role in mediating the efficacy
with bucindolol in heart failure caused by ischemic versus of beta-blocker therapy for congestive heart failure. Am Heart
idiopathic dilated cardiomyopathy. Circulation 1991; 84 (6): J 1996; 131 (2): 329-36
2426-41 73. Binkley PF, Lewe R, Lima J, et al. Neurohumoral profile in
57. Olsen SL, Gilbert EM, Renlund DG, et al. Carvedilol improves congestive heart failure: response to beta-blockade. J Lab
left ventricular function and symptoms in chronic heart fail- Clin Med 1988; 111 (4): 393-8

 Adis International Limited. All rights reserved. Drugs 2000 Nov; 60 (5)
1016 Baran et al.

74. Snow HM. The pharmacology of xamoterol: a basis for modu- 78. Yamada S, Ohkura T, Uchida S, et al. A sustained increase in
lation of the autonomic control of the heart. Br J Clin Phar- beta-adrenoceptors during long-term therapy with metoprolol
macol 1989; 28 Suppl. 1: 3S-13S and bisoprolol in patients with heart failure from idiopathic
75. Anonymous. Xamoterol in severe heart failure. The Xamoterol dilated cardiomyopathy. Life Sci 1996; 58 (20): 1737-44
in severe heart failure study group [published erratum appears 79. Ruffolo Jr RR, Feuerstein GZ. Neurohormonal activation, oxy-
gen free radicals, and apoptosis in the pathogenesis of con-
in Lancet 1990 Sep 15; 336 (8716): 698] [see comments].
gestive heart failure. J Cardiovasc Pharmacol 1998; 32 Suppl.
Lancet 1990; 336 (8706): 1-6
1: S22-30
76. Gilbert EM, Abraham WT, Olsen S, et al. Comparative hemo-
dynamic, left ventricular functional, and antiadrenergic ef-
fects of chronic treatment with metoprolol versus carvedilol
in the failing heart. Circulation 1996; 94 (11): 2817-25 Correspondence and offprints: Dr Stuart D. Katz, MD, Co-
77. Newton GE, Parker JD. Acute effects of beta 1-selective and lumbia Presbyterian Medical Center, Division of Circula-
nonselective beta-adrenergic receptor blockade on cardiac tory Physiology, Room MHB5-435, 177 Fort Washington
sympathetic activity in congestive heart failure. Circulation Avenue, New York, NY 10032, USA.
1996; 94 (3): 353-8 E-mail: [email protected]

 Adis International Limited. All rights reserved. Drugs 2000 Nov; 60 (5)