Effect of metoprolol on myocardial function and energetics ... · ardial failure results in i~ad~~~~t~ s~t~fl~i~ ion of intracardiac pressures and activati~~n Q satnry neurohormonal

ardial failure results in i~ad~~~~t~ s~t~fl~i~ ion of intracardiac pressures and activati~~n Q

satnry neurohormonal s (I-6). It is well an~~otensin converting e inhibitors, which counteract the effects of the renin-angiotensin and, to a lesser degree, the

hetic nervous system (5), are beneficial in myocardial (2,7,8). Since 1975, when ~a;~~stein et al. (9) first

reported ‘,e beneficial effects of beta-adrenergic blocking agents OR heart failure, several investigators (1,104) have

et&atb : Dr. Eric 1. Eichhom, Director, Cardiac Cath-

Laboratory (IIfAZ), University of Texas Southwestern and Dallas veterans Adnlitration Medical Centers. 4500 South Lancaster, Dallas, Texas 75216.

01994 by the American College of Car&logy

f~use~ on the use of these agents to antagonize the e ~~~ro~i~ from ou improved ejection fraction and fun with dilated ~ardiamyopat~y treated with beta-blockers, these agents remain an enigma. It is co~at~rint~itive that a ‘“negative inotrope” would result in improved ventricular performance in patients with myocardial failure. Because our previous trial (10) was an open label study of bucindolol, a nonselective beta-antagonist with weak vasodilatory properties (19), it remains unclear whether such salutary hemodynamic effects would extend to a beta-blocker without direst vasodilatory ~ro~rties and without direct va~odilato~ without beta-antagonism. ‘I’berefore, we sought to determine whether metoprolol, a pure beta,-antagonist, improves myocardial performance and eticiency in patients with dilated cardionqnpathy.

Finally, the metabolic effects of beta-blockers in patients with heart failure havs not been elucidated. It is well known that fatty acid oxidation by the heart uses more oxygen per unit

09351097/94/%7*00

SACC Vol. ‘d, No. 5 Novvcmber I, 9994zl3BB)-2@

Meaq9rolol Placebo P GrQUp (n = 15) Group (n = 9) Value

s

er 1, %990 and April 14, 1993, 25 age 48 f 10 years, range 34, to 75)

nonischemic dilated cardiom thy underwent cardiac catheterization at the Dallas V ans Ad~~i~istrat~o~ -Medical

ventricular ejection fraction were in New York eart to IV. All patients with a cardiomyopathy were ap-

into the study unless they had a complicat- with severe renal (creatinine >2.5 amic oxaloacetic transa

[SGPTj more than three times normal), pufmonary, r ogic or endocrine disease were exchrded. Only one patient was diabetic and was taking

ization, and he was a well ats with previous myocardial

or bypertrophic cardiomyopathy or ~rirna~ valvular disease were excluded, as were patients with a recent (~3 months) history of alcohol abuse. Five (56%) of 9 controi patients and 8 (53%) of 15 metoprolol-

some history of akobd abuse (Table i) bo vs. rn~to~~o~o~~. In addition, two of niue ients had stopped drinking <6 months (but

>3 months) before entry into the study, whereas none of the metopsolol-treated patients had a history of drinking ~6 months before entry (p = NS). All patients with <6 months of clinical heart failure of unclear etiology had endomyocardial biopsy; none had evidence of myocarditis.

ABI medications were allowed, e 3 months of eatry Ah patients converting euzyme inhibitors duri

nous dextrose was given before c erization. Written in- formed consent was obtain

the Human Studies Sub i~~strat~o~ and Uuiver

e to record

Baseline coronary sinus tkemodilution and cardiac output measurements and left heart pressure recordings were performed before initiation of overdrive atrial (ccmnary sinus) pacitrg.

at& hearr rate at baseline and 3 months) ed to eliminate ait~ati~~s in ~o~t~a~ti~i~ and

relaxation due to changes in heart rate from baseline t0 3 months. Ten minutes after initiation of atrial pacing at 10 to 15 beats/mitt above the intriusic heart rate, lef? heart pressure and cardiac output measurements were repeated. This was fool- Iawed by digital ventriculograp y using 15 to 25 ml (total

(60:40) nouionic contrast media (fohexol, aboratories). Ventricu!ography was per-

formed in the 3V’ right amerior oblique projection at 30 frames/s with cineradiographic equipment (Philips nr04el Op-

timus w1200, Eindhoven, The Netherlands) interfaced directly to a digital radiographic computer ( 41OOC, Milpitas) arrd stored as a 512 matrix. The images were stored and processed using previously published techniques (Is).

rements were recorded, loading condi- a~~ta~ni~i~. Intravenous

sodium nitroprusside was initiated at a ose of 0.25 to 0.50 &kg body weight per min and increased by 0.25 @,kg per mm every 2 to 5 min to achieve a reduction of 15 ts 29 mm Hg in aortic end-systolic pressure. Care was used to avoid bypoten- sion. Repeat simultaneous ventriculograms and hemody

1312 ElCHHORN ET AL. BETA-BLOCKERS IN HEART FAILURE

JACC Vol. 24 No. 5 Novembsr 1, 1994:lh-20

namic recordings were performed at one or two time points after aortic end-systolic pressure had been altered. Because feft ventricufography may increase end-diastolic pressure tran- sientfy, there was a delay of at least 10 min between ventricu- fograms to allow the left ventricle to return to equilibrium.

After 3 months of therapy, the same procedure was used with care to match the original atrially paced heart rate used in the baseline study.

All patients underwent selective coronary arteriography at the end of the 50 catheterization to exclude significant co disease as a cause of their cardiomyopathy.

~t~~Q~. The study drug (metoprofot or placebo in B 3:2 randomization) was

twice daily. All patients prolol tolerated titration to f’ulf dosage.

Plecebo-treated patients crossed over to open-label meto- onth study period. We performed a radio- am 6 m~~tbs after randomization (i.e.,

n-label metoprofof) to assess the change in left ventricular ejection feaction after crossover to active drug* we had no knowledge of who received the active drug i tia13 months, the drug was titrated or retitrated in all patients according to the schedule given earlier. Because clinical deterioration has been noted in patients witb~rawn from beta-blockers (12,25), we chose not to cross the active patients over to placebo. In addition, this allowed us to determine whether further changes in left ventricular ejection

d htween 3 and 6 months. sis, Left ventricuf~r voluma was

standard angiqrapbic area cardiac cyle sefeckd was not a premature beat or post-

hat. Simultaneous left ven

nt in the cardiac cycle. ~ntraob~~er and of our digital ventricnfogr~m measure-

To test whetfter ftexformance placebo groups, we prospectively well as tfuee relativefy load-

elastataoc, derived from the end relation (lOZU9); 2) peak positive lirst detitive of left ventric- ufar mure (+dRdt) to end-diastolic volume relation (30); and

trike work (31). ‘Ike indexes were d heart rates before and after 3 months

index (LVSWI) was

LVSw = (MSLVP - MDP) X SVI x 0.01J6,

where MSLVP = mean systolic left ventricular pressure; I+JDp = mean diastolic left ventricular pressure; and SVI = stroke volume index by thermodifution (cardiac jndexfleart rate). f%ute work index = LVSWI x Heart rate.

Arterial elastance (Ea) was determined by a previously validated formula (33,34): = PeslSV, where Pes = end- systolic pressure; and SV =

Assessment oj ~~#~t~li~ ~e~~~~~ce” Left ventricular relaxation was assessed by analyzing changes in the time constant of e.~oncntial isovol~metric pressure decrease (au) uncorrected for afterload (35,36) and the slope of the tau-end-systolic pressure relation and after long-term beta-~dr~ner~ic bloc

The time constant of ed by a previously described

~~a~lber stiffness was

opening to the peak of the (ECG). A no~lin~r eq~at~o~~~tti Products) was used to solve the eq coelficients a, b and k were equation; coefficient k coefficient a = the upward or downward dispf of the @ress~~~-vofMn~c relation. On the basis o

a slope I&, where KN = a measure oImyocardiaf stiffness (39). Detemaimation of myocmdid eficiemy. Myocardial oxygen

consumption (MQC) was determined by the following formula (10):

X (Left ventricular - Coronary sinlas 02 difference);

and myocardial efficiency fly (lo),

Efficiency (%) = (Stroke work X Heart rate/k X MOC),

where k = conversion constant 2.059 (kg-m/ml 0, consumed). Because our determination of mechanical efficiency is caf-

cufated by this ratio of external mechanical work to myocardial oxygen consumption (as just shown), and because total mechanical energy actually represents both external mechanical work and potential ene changes, we also calculated potential energy changes before and after therapy. We calculated potential energy by a calculation based on end-systolic pses- sure (V,), and end-systolic volume (ESV) using the following formula (40):

Potential energy = {[(End-systolic pressure) x (ESV - &)jL?}

X (1.33 X lob4 J/mm Hg-ml).

ed between eat~~ter-

- ko2j/(A02 - Cso2j,

ible arn~u~t of carbon dioxi tent was calculated as

02 content (mlldl) = (02 sat %o) (1.46 ml 82/g Hgb) (Hgb g//dl);

Q content OnmolAiter)

= {[02 content (ml/dl)](lO dl/liter)j/(22.4 ml/mmol),

where sat = saturation. Variability of our tory quotient measurements has been tested by repeated analysis of measurements from seven paLent cardial respiratory quotient taken 30 s apart (1

patients crossed over by &ration to open-label metopro~o~. All patients were then asked to undergo a radionuclide ventricu- logram at 6 months after randomization. One patient did not receive a &month scan because of sudden death before the scan; another patient from the placebo group was intolerant to metoprolol titration because of severe heart failure and did not undergo a 6-month scan. Radionuclide ventriculography was performed and analyzed as previously described (43,44). All radionuclide studies were analyzed in blinded manner.

Statistical analysis. Changes in hemodynamic variables and ventricular function were compared by two-way (group effect and time effect) repeated measures analysis of variance.

rences between groups were then tested by a Scheffi F test. Differences within groups were tested by a Student paired t test only if the repeated measures analysis demonstrated an intrngroup time effect. Baseline characteristics were compared by an unpaired t test. For the peak +dPidt-end-diastolic volume relation, a repeated measures analysis of covariance (with end-diastolic volume as the covariate) was use3 with 1) intergroup (metoprolol vs. placebo) and

kra8roup (time effect adl

examined. F0r end-s~skol~c ePa of each i~d~v~~ua~ patient

eated ~e~$~res an

ho-way repeated meas

Cke~~st~c~~ One of the 25 original ~atiessts cardiac cathetc~~~at~o~l and was cxc~ude~

nstrated in Table 1, bot

relaxation. fteen patients were randomized to re- 9 to receive placebo. All patients toler-

g twice daily without nts during titration, ar! 1 admission for worseni

the two groups are summarized in Table 2. Heart rate was uced by meto~rolo~, thus accounting for a lack of incr ardiac output despite an increase in stroke volume.

ventricular end-diastolic pressure at rest was not r metoprolol. wevea peak systolic and end-systoli were increa by beta-blockade.

0th stroke and minute work were increased by metoprolol, ereas potential energy did not change (Tables 2 and 3).

espite this increase in mechanical work, coronary flow and myocardial oxygen consumption decreased. This resulted in an increase in myocardial efficiency in the metoprolol group only (from 8.5 2 5.6% to 18.5 4 10.9%, p = 0.01 vs. pretherapy, p = 0.10 vs. placebo),

Despite improvement in ventricular performance and myocardial efficiency, coronary sinus uorepi~ephrine did not decrease. However, compared with the placebo group, prolol attenuated progressive increases in coronary sinus

nephrine and epinephrine. toiis ~ss~ss~~~~, At matched paced heart rates, meto- increased end-systolic and reduced left ventricular en

diastolic pressures (Table 3). Despite these changes, end- systolic volume decreased, and ejection fraction increased, in the metoprolol group (Table 3, Fig. 1). Stroke work increased

1314 EtcHHoRN ET AL. JAW Vol. 24. No. 5 BETA-BLOCKERS IN HEART FAILURE November II, 1994d3lO-2O

Table 2.. Baselim Left Ventricular Characteristics Before and After Metoprolol or Placebo

Before After p Value Treatment Treatment (metoprolol vs. placebo)

Heart rate (mitt-‘) Placebo Metoprolol

Cardiac output (liters/min) Placebo Metoprolol

Stroke volume (ml) Phi&o Metoprolol

L

systolic pressure (mm Hg)

Metoprolol End-soytolic pressure (mm Hk)

Placebo Metoprolol

Stroke work index (g-m/mz) PlaCebo Metoprolol

Minute work index (kg-m/m’) Placebo Metopmlol

Coronary sinus blood flow (ml/mitt) Placeho Metoprolol

Myocardial oxygen difference (vol %I) PI W&O Metoprolol

Myocardisl ox-ytyt consumption (ml/min) Plaaho

Comnaty sinus tmwpinephrine PlXt?bo

ol nus epinepttrine (pgiml)

Plaeeho Metopmlol

Respiratory quotient Placebo Metopmlol

4.2 2 1.4 4.4 ?I 1.2 4.12 1.2 4.8 z i.c

52 2 20 59 ‘c 19

56 + 22 74 ?I 2.W

28.6 2 5.7 26.8 f 7.8 19.1 t 9.4$ 16.8 L 7.3

WJ*8 IO2 + 6 105 t 21 124 + 2Qt

61 26 6027 62 2 14 77 * 19t

17.0 s+ 5.0 20.3 2 IO.9 18.7 ?z 7.8 32.9 t 17.5’

1.4 r 0.4 1.5 t 0.6

1.6 2 0.7 2.1 ?I 0.W

146 ? 102 201 2 136 178 r 88 115 + 71

12.2 r 1.7 11.5 t 2.3 12.4 + 1.3 11.7 c 1.2t

16.8 r 9.2 22.1 ?C 12.5 22.3 + 11.3 13.1 t 1.6

1.9 + 3.8 9.2 c 6.1 85 c 5.6 18.5 2 10.Qt

730244% 1.114 9 727 569 C% 426 433 t 357

866f8a 65 + 55

135 lr 174 46236

0.88 I 0.25 0.70 2 0.27

0.68 t 0.15 0.78 2 0.25

0.014

0.76

0.19

U.91

0.017

0.023

0.25

0.026

0.95

0.035

0.10

0.025

0.09

0.054

‘p < 0.003 versus before treatment. $p < 0.05 versus before treatment. $ p c 0.05 versus before treatment with plaabo. Data presented are mean value + SD. LVEDP = left ventricular end-diastolic pressure.

only in the metoprolol group despite the finding that preload, reflected by end-diastolic volume, decreased.

Lcft ventricular ejection fraction increased in 13 of 15 patients randomized to metoprolol treatment by 3 months of therapy. One of the remaining two patients had an improvement in ejection fraction by 6 months compared with baseline, Thus, 14 (93%) of 15 patients had some improvement in ejection fraction by 6 months. By 3 months of metoprolol therapy, 10 (67%) of 15 patients experienced an increase of

0.05 in ejection fr~&_u~, e (4u%) olt 15 had an increase of 0.10; and 3 (20%) of 15 had an increase of 0.20. By 6 months of metoprolol therapy, 11 (79%) of 14 patients assessed experienced an increase of 0.05 in ejection fraction; 10 (71%) of 14 had an increase of O.lQ and 6 (43%) of 14 had an increase of 0.20.

Although end-systolic elastance did not increase in the placebo and metoprolol groups, the y intercept of the end-systolic pressure-volume relation was increased in the metoprolol group (Table 3). This suggests that a parallel

JACC Vol. 24, No. 5 Nnvemher 1, 1994: E3tO-20

ElCMHORN ET AL. BETA-BLOCKERS IN UEART FAILURE

1315

%b8e 3. Left Ventriczmlar Characteristics During Atrial Pacing Before and After Study Drug Therapy

Measuremenr Before

Treatmenr After

Treatment p Value

(mctoprolol vs. placeho)

Heari Raw and Prcssurcs

Wcart raw (rniil ‘i Placebo

Metoprobl LVEDP (mm Hg)

Placeho

Meroprolol End-systolic pressure (mm Hg)

Placebo

Metopr&i Aortic end-diastolic pressure (mm Hg)

PlWL4X~

Mcroprolot Potential energy (.I)

PlitCCllU

Meq~rolol

EDVt (ml/m:)

Placcho Mewprolol

ESV! (mum')

Plxrbo Mctoprolol

LVEF Placebo Metoprolol

Ees (mm &/ml) Placeho Metoprolol

Ees y inlcrcepr (mm Hg) Placebo Metoproiol

Ea (mm H&‘/ml) Placebo Metoprolol

Stroke work index (g-m/m’) Placebo Mctoprolol

Peak +dP/dl (mm l-I@) Placebo Metuprolol

PRSW (g-m/ml) Placebo Metoprolol

95 I5 8 97 f 9 942 13 935 13

26’+ .- - 59 . . 25.8 c 9.5 19.2 5 lU.6* 12.5 rr 7.4;

(12 ? 4 59 2 9

62 +_ 13 72 +. 17t

70 It 5 77 2 l-l

79z IS 78-t 14

1.40 _k I.08 1.25 z Oh8

1.17 2 1.t.l I.26 2 1.26

Volumes and Syswlic Propertics _I_- I__l_-j.p

177 2 Sh 159 * 63 137 z 36$ lilb i 37t

I.54 t 5.t 134 tl so IOX 2 36:k 80 2 37$

0.14 2 0.07 0.17 + 0.08 0.22 + O.lO$ 0.33 2 0.13$

0.63 2 0.48 0.53 +- 0.19 0.81 2 0.4x 0.84 -t 0.39

-82 2 154 -48 rT 52 -84 2 llh -11 _f 34t

2.9 r. 2.2 2.6 f 1.7 1.X 2 0.8 1.7 z 0.8

16.4 -c 5.4 17.1 2 6.2 I0.Y -c 7.11 27.0 z 1S.l-l

774 5 101 WY z 5s x75 z 2.x 1.073 2 3sg

0.12 f O.OH 0.U6 2 0.04 0.18 + 0.16 0.35 + 0.24t

0.11

0.15

O.&i3

0.85

0.57

IL84

0.44

u.05

0.66

0.49

U.83

0.1 1

0.04

0.08

Diastolic Properties

Peak -dP/dt (mm Hg/s) Placebo Meloprolol

Tau (ms) Plncebo Mctoprolol

R (mslmm ftg) Placebo Metoprolol

787 2 76 808 -c 104 922 + 2.54’ 1,106 2 3311 0.097

123 of: 25 104 + 32/I x9 + 258 7R 2 2411 0.41

3.5 ? 1.7 2,s -c I.511 1.5 -c 1.0s 0.8 -t 0.81 OSH

KN Placebo Metoproiol

9.8 f 5.0 1 I .o z! 8.0

3.7 2 3.0$ 3.4 _+ 2.6 (X7 k

Placebo 0.031 ? 0.021 0.042 2 0.036

Metoprolol 0.031 + 0.035 0.031 +_ 0.040 0.64 a

Placebo 4.46 + 15.80 -1.38 + 17.57 Metoprolol -5.16 2 25.11 0.13 = 10.68 0.37

‘p < 0.1, #p < 0.05 versus placebo pretreatment, ‘rp < 0.05, §p < 0.01, lip < 0.1 versus pretreatment. Data presented arc mean value ?r SD. a = diastolic

pressure-volume shift coefficient; Ea = arterial elastance; Ees = end-systolic elastance; EDVI (ESVI) = end-diastotc (end-systolic) volume index; k = constant of

chamber stiffness; KN = myocardial stiffness; LVEDP = left ventricular end-diastolic pressure; LVEF = left ventricular ejection fraction; Peak +dP/dt (-dP/dt) = peak positive (negative) fust derivative of left ventricular pressure; PRSW = preload recruitable stroke work; R = end-systolic pressure relation; Tau = time Constant of

exponential isovolumetric pressure decrease.

1316 EICHHORN ET AL. BETA-BLOCKERS IN HEART FAILURE

JACC Vol. 24, No. 5 November 1, 1994:1310-20

ra are pup mean value and error.

ard shift) occurred in the ~~~-§~t~lic ressure-volume relation consistent with either an increase in

or a reduction in ventricular volume, or

Peak +dP/dt at matched paced heart rates increased with bta-blockade (p = 0.023), an effect that is not explained by an increase in preload (Pig. 2). The slow of the stroke work-end diastolic volume relation demonstrated an increase in preload

recruitable stroke work in th oprolo! group only (Table 3). Taken together, these data st that lon~~~rrn mcto~ro!~~! therapy may increase contractility in patients with dilated

ic n~~s$mg~t. Although isovolumic re!~~tion im- ith metoprolo! (as rc~~ctcd by a decrease in of ~x~n~nti~l i~vo!umetri~ decrease staff,

the ta~~n~~tolic pressure relation [R] and an increase in c&o group likew!se had a small

us, the metoprolol group did not lacebo group. The reduction in

entia! isovolumetric decrease (tau) p&ably would have been more

prominent had end-systolic pressure not increased in this (3637). ddiastolic volume and pressure decreased in the meto- group (Table 3). However, myocardia! stiffness and

chamber sti!Tness did not change with therapy. In addition, the diastolic pressure-volume relation was not shifted upward or downward with administration of metoprolo! or placebo, as

lack of change in coeWcient “a.” Thus, metoprolol ificantly improve diastolic function ~rn~red with

tics and neu~ho~Q~~1 tiv~tio~. umption decreased despite increase

in !eftventricu!ar minute work (Table 2), suggesting improved myocardia! efficiency with beta-blockade.

Myocardia! respiratory quotient tended to increase in the

metoprolol-treated patients, whereas it tended to decrease in the placebo-treated patients (Table 2). These data suggest that beta-blockade alters substrate utilization by incr~sing carbohydrate utilization (increased myocardia! respiratory quotient). By comparison, placebo-treated patients had a progressive increase in noncarbohydrate utilization (reduction in respiratory quotient). Although the intragroup changes in respiratory quotient did not reach statistical significance (p = 0.36 for the metoprolo! group vs. pretherapy and p = 0.12 for

group vs. pretherapy), the intergrou at metoprolo! at least may inhibi he respiratory quotient (p = 0.054).

Changes in coronary sinus norepinephrine closely core- lated with changes in my~~dia! res~irato~ ~~Qtie~t (Pig. 3). These data suggest that a relation may exist between neurohormonal activation and substrate utilization in patients with heart failure. Despite this correlation, no relation was seen

h ~~~~~~~~~~~~~~~~§ and

m, the left ventricular -f 0.80 at bassline to

0.32 * 0.13 at 3 months (p = 0.0013 vs, baseline) to 0.41 + 0.13 at 6 months Qp = 0.017, 6 months vs. 3 months). En the seven

ients who had all three assessments, ejection 0.13 2 0.0s lo 0.17 t 0.09 at 3 manths (p =

nd increased to 0.28 + 0.09 3 months after crossover to metoprolol (p = 0.011 vs. 3 month class improved over the 6 months of fdlow- metoprolol-treated patients. Functional class w baseline, 2.1 1- 0.7 at 3 months = NH vs. baseline) and impaQved to 1.5 2 0.7 by b mm of therapy (p < 0.05 vs. baseline and vs. 3 months). Functional class did not change in the placebo-treated group over the 3 months of double-blind

therapy.

The use of beta-adrenergic blockade to treat patients with heart failure remains an enigma, Improvements in ventricular function in the presence of a “negative” inotropic agent is counterintuitive and remains controversial despite a consider- able number of published reports (1,9-18) that demonstrate a

salutary effect of these agents in these patients. Our study demonstrates improved left ventricular function and a reduction in left ventricular chamber size in patients with dilated cardiomyopathy randomized to metoproiol therapy. Additian-

ent in three relatively k~ad independent indexes of performance, the end-systolic pressure-volume relation, peak +dP/dt-end-diastolic volume relation and slope of the stroke work-end-diastolic volume relation suggests that left ventricular contractility may have increased, although the long-term changes in heart size precludes our ability to state this definitively.

as expressed by left

seen in the placebo-treated patients. Such a reduction in ~d-d~asto~~c pressure may represent one beneficial effect of eta-b~~c~~~~ ager~t~ OII these patients when they exercise 16,i ‘9,46). As heart rate increases, beta-blockers

increase in end-di ic pressure with exerAse. As with OUP cindolo! (IO), there was an intragroup

(reduction in the tau-end-systolic no intergroup eikct was seen be-

i~~~Q~~~~~t in the placebo group ic pressure in the mctopmlol gmp,

ement in relaxation times (36 I-k improveverncnt, metoprol id not

significan$J reduce COrOfla sinus nrorepincphrine. This may have been due to the fact B metopmlol reduces the cl

. By contrast, metoprolol retard the progressive increase in B1pore nephrine seen in the

1acebo group (p = O.K.?5 vs. placebo) nd alsO reduced rest

cart rate (p = 0.014 vs. placebo), a marker of receptor blockade and adrcnergic antagonism.

Although myocardial respiratory quotient did not signifi- cantly increase with beta-blockade, metoprolol blunted the progressive decrease in respiratory quotient seen in the placebo group (p = 0.054 vs. placebo). Thus, metoprolol may have resulted in preferential carbohydrate utilization. We believe that this is the first study to suggest an in vivo relation between substrate utilization and adrenergic activation in pat&s with heart failure. This finding is not surprising in view of previous animal studies demonstrating increased lipoiysis (21,24) and

reduced insulin-induced glucose transport (48) in the presence of catecholamines. In addition, previous data in patients with heart failure have shown e ted insulin-mediated suppres- sion of hepatic glucose out increased glucose and lactate uptake and a decline in basa fatty acids with the long-term administration of metoproloS(49,§0). Because congestive heart failure is a state of increas drenergic stimulation, with elevated sympathetic nerve tr and increased plasma norepinephrine levels (2-6), lipolysis and glycogenolysis may be stimulated (21,241. Beta-blockers may act to reduce effective postsynaptic norepinephrine and shift the balance of substrate selection in favor of carbohydrate.

Several experimental observations suggest that hatv acid metabolism by the heart is less efficient than carbohydrate metabolism, a factor which may be hemodynamically relevant in the setting of catecholamine excess and limited oxygen delivery (20,24). Although fatty acid utikZ3tiOla in th@ prf2Xnck:

Of this stress may serve to decrease myocardial efficiency, the lack of a direct relation between changes in respiratory quO- tient and changes in efficiency in this study suggest that this is

1318 EICHHORN ET AL. BETA-BLOCKERS IN HEART FAILURE

3ACC Vol. 23, No. 5 Novemhrr 1. 199J:131&-20

either not a factor or not the only factor resulting in improved efficiency. Further investigation into the mechanism(s) irirr;r&y beta-blockade improves myocardial efficiency is wai- ranted using more sophisticated methods, suck as magnetic resonance spectroscopy, proton emission tomography and molecular techniques.

tions, The main limitation of this study was randomization to achieve identical patients in

the metoprolol and placebo groups. Despite this limitation, the metoprolol group clearly responded more favorably to therapy both clinically and in terms of left ventricular systolic function. Even the more impaired placebo group was capable of re- sponding to therapy, as manifested by improved ejection fraction after crossover to open-label metoprolol. In addition

line differences, there was some s~n~~neons improvement in ejectian fraction in four of the nine placebo-tre~~ted patients, which reduces the statistical si~nifi~~~n~~ of our find-

, However, spontaneaus iniprove~~~n& in ejection fraction in a substantial pa tion of placebo-treated patients with dilated cardiomyop is not unique to our trial and has been seen in other large randomized trials (17,lS). That we reached

cance despite these baseline differences and ous improvement in the placebo group attests

to the very consistent improvement seen with metoprolol. Another limitation of this study is our limited ability to

measure myocardial metabolism in vivo. Myocardial respiratory quotient has a relatively large standard error and may be insensitive to small changes in substrate utilization. We thus believe that our results are suggestive but require further investigation.

Nitroprusside may result in geometric alterations as ventricular volumes decrease, and could introduce an error into the volume calculations. In addition, in certain circumstances of large volumes constrained by the ~r~~ardium, application of nit ‘de (or drugs that reduce ventricular volumes over time) ult in a downward shift of the diastolic pressure- vh.me rehtion as the constraint is relieved (51). l!n circum-

where there was a clear downward shift of a pressure- loop with nitroprusside, the pressure-volume diastolic

points before that application of nitroprusside were used. itroprusside may also result in some baroreflex-

nges in contractility (52). However, because tients with congestive heart failure have high neurosympa- tic tone at baseline (2,4-6,X!), baroreflex stimulation from

nitroprwide should be of minimal significance. In addition, the Small range of pressure reduction (15 to 25 mm Hg) induced by nitroprusside may hap: Z:,ted the accuracy of

the end-systolic pressure volume relation, especially at curvilinear (53,512).

The comparison of left ejection fraction at 6 and 3 months involves use of two diierent techniques: angiographic I& ve former

and radionuclide ventriculography. The tric assumptions to assess ejection fraction

(26Xk the latter makes no geometric assumptions (43,44). mite these differences, these two techniques have been shown to correlate well with each other (44). In addition, comparison of

36 studies of patients with dilated cardiomyopathy by radionu- elide and angiographic lteclmiqnes in our in~tit~~ti~~ have shown excellent correk&r! (r = 0.78, SEE previous studies (55) have demons tion fraction with atrial pacing, because the ejection fraction at 3 months was determined with atrial pacing and that at 6 aortas was determined at rest, some slight overestimation of ejection fraction at 3 months may have occurred. ever, this makes the improvement seen from 3 to 6 mouths much more significant.

Measurement of left ven ar performance in vivo a lo~8”t~~ln i~te~~~~t~~~~ is of left ventricular function is iW

ventricular p~rrorn~at~~~ tt intrinsic ~~yo~a~d~a~ fun& that c~~~~n~~~er p~rformi~n~~ parallel leftward shift in IThtiOfl, 2) iill increase in I the setting of reduce pressure, 3) an upward s volume relation, and 4) an incre stroke work. Although long-term changes in ventricular size and 8eorn~t~ could expla’ our data, it is like6y that such changes in the ventricle ect improvement in contractile performi~nce. indeed, ~ed~~~tio~ in ricuiar volumes by itself suggests that cardiac function has

Because previous studies (57) e shown that the peak ~d~~dt-end-diastolic volume relation is more sensitive than the end-systolic pressure-volume relation for detecting ~~~~~~8~s in ~~~tro~i~ state, it is not im~o~sisteflt to have a parallel shift in elastance with an increase in the d&-end-diastolic volume relation. In addition, the upward shift in the peak ~d~/dt-end-diastolic volume relation in this case cannot be due to a change in aortic pressure because aortic end-diastolic pressure (aortic valve opening pressure) was unchanged after ~~dnlinistration of meto~roto~.

We recognize that coronary sinus thermodilution determination of coronary sinus blood flow is limited by changes in catheter position within the coronary sinus, patterns in coronary venous drainage and coronary sinus refux (58,59). We attempted to minimize these problems by meticulous exami- nation of catheter position during each procedure. In addition, the presence of a placebo-treated group in this study helps to assess the significance of any random measurement errors.

Finally, it is clear that the analysis of myocardial energetics is complex. Myocardial efficiency can be defined as the relation between myocardial o.xygen consumption and its covariate, the pressure-volume area (total mechanical energy) (40). Total mechanical energy may further be defined as the sum of external mechanical work plus potential energy produced during a cardiac cycle. Because external mechanical work (stroke work) increased, potential energy was unchanged, and myocardiai oxygen consumption decreased, it is likely that myocardial efficiency increased. This increase is most likely due to an increase in shortening work as opposed to pressure work because potential energy did not increase substantially and

JACC Vol. 24, No. 5 November 1, 1994:n3BO-?0

ment seen the

ears to nmme left ventris-

erits further investigation

We express our appreciation to Donald Haagen. RCPT, Janie McCoulskey, CCRN. Willie Gary, Jenelle Durbin, RN, Graciela Fields and all the cardiology fellows for their expert help during the performance of these studies. We are grateful to Richard C. Risser. MS for help with statistical analyses and to Paul A. Sobotka, MD, William Henrich, MD and Mark Feldman, MD for critical evaluation of the manuscript.

1. Eichhorn W. Do beta-blockers have a role in parienls with congestive hear! failure? CardioI Clin 1994;12:133-42.

2. Packer M, Lee WH, Kessler PD, Gottlieb SS, Bernstein JL. Kukin ML. Role of neurohormonal mechanisms in determining survival in patients with severe chronic heart failure. Circulation 1987;75 Suppl IV:IV-80-92.

3. Bristow MR. Kantrowitz NE, Ginsburg R, Fowler MB. P-Adrenergic function in heart muscle disease and heart failure. J Mol Cell Cardiol 198317 Suppl2:41-52.

4. Cohn JN, Levine TB, Olivari MT, et al. Plasma norepinephrine as 3 guide to prognosis in patients with chronic congestive heart failure. N Engl J Med 1984;311:819-23.

5. Francis GS, Benedict C, Johnstone DE, et al., for the SOLVD Investigators. Comparison of neuroendocrine activation in patients with left ventricular dysfunction with and without congestive heart failure. Circulation 1990;82: 1724-9.

6. Swedberg K, Eneroth P, Kjekshus J, Wilhelmsen L, for the CONSENSUS Trial Study Group. Hormones regulating cardiovascular function in patients with severe congestive heart faih!re and their relation to mortality. Circula- tion 1990;82:1730-6.

7. The SOLVD Investigators. Effect of enalapril on survival in patients with reduced left ventricular ejection fractions and congestive heart failure. N Engl J Med 1991;325:293-302.

8. The CONSENSUS Trial Study Group. Effects of enalapril on mortality in severe congestive heart failure: results of the Cooperative North Scandana- vian Enalapril Survival Study (CONSENSUS). N Engl J Med l987;31@ 142935.

9. Waagstein F, Hjalmarson A, Vamauskas E, Wallentin 1. Effect of chronic beta-adrenergic receptor blockade in congestive cardiomyopathy. Br Heart J 1975;37:1022-36.

10. Eichhorn EJ, Willard JE, Alvarez L: et al. Effect of beta-adrenergic blockade

on myocardial function and energetics in congestive heart failure: improvements in hemodynamic. contractile, and diastolic performance with bu;in- dolol. Circulation 1990;82:473-83.

P 1. Wisenbaugh T, Katz I. Davis J, et al. Long-term (3 month) effects of a new brta-blocker (nebivolol) on cardiac performance in dilated cardiomyopathy. J Am Coil Cardiol 1993;21:1094-1100.

12. Waagstein F, Caidahl K, Wallentin I, Bergb C, Hjalmarson A. Long term P-blockade in dilated cardiomvooathv: effects of short- and Ion+term metoprolol treatment followed by withdrawal and readministration of=&tii prolol. Circulation 1989;80:551-63.

13. Heilbrunn SM, Shah P, Bristow MR, Valantine HA, Ginsburg R, Fowler MB. Increased P-receptor density arid improved bemodynamic response to catecholamine stimulation during long-term metoprold therapy in heart failure from dilated cardiomvopdthv. Circulation P989:79:483-90.

14.

15.

IO.

II.

18.

19.

20.

21.

22.

23.

24.

25.

26.

24.

28.

29.

30.

31.

32,

33

Woodley SL. Gilbert EM. And&son JL, et al. P-Blockade with bucindolol in heart failure due to iscbemic versus idiopathic dilated cardiomyopathy.

Circulation 1991;84:2426-41,

Gilbert EM, Anderson JL. Deitchman D. et al. Chronic /$blocker. V3SodihtOr therapy improves cardiac function in idiopathic dilated cardio- Iryopathy: a double-blind, randomized study of bucindolol versus placebo. Am J Mcd l9YO$M:223-Y.

Engelmcicr RS, O’Connell JB, Walsh R. Rad N, ScanIon PJ, Gunnar RM. Improvement in symptoms and exercise tolerance by metoprolol in patients with dilated cardiomyopathy: a double-blind, randomized. placebo. controlled trial. Circulation lYH5;72:536-46.

Waagstein F. Bristow MR, Swedberg K. et al.. for the Metoprolol in Dilated Cardiomyopathy (MDC) Trial Study Group. Beneficial effects of metoprolol in idiopathic dilated cardiomyopathy. Lancet 1993;342:1441-6,

Bristow MR, O’Connell JB, Gilbert EM, et al., for the Bucindolol Investi- gators. Dose-response of chronic P-blocker treatment in heart failure from either idiopathic dilated or ischemic cardiomyopathy. Circulation 1994:89: 1632-42.

Hershberger RE, Wynn JR, Sundberg L. Bristow MR. Mechanism of action of bucindolol in human vemricular myocardium. J Cardiovasc Pharmacol l990$5:Y59-67.

Vik-Mo H. Mjos OD. Influence of free fatty acids on myocardial oxygen consumption and iscbemic injury. Am J CdrdioI 1981:48:361-5.

Mjos OD. Effect of inhibition of lipolysis on myocardial oxygen consumption in the presence of isoprotrrenol. J Clin Invest 1971;50:1869-73.

Bing RJ. Siegal A, Ungar :. Gilbert M. Metabolism of the human heart. II. Studies on ht, ketone and amino acid metabolism. Am J Mcd 1954:16:504- IS.

Kjekshus JK, Mjos OD. Effect WI inhibition of lipolysis on myocardial oxygen consumption in the presence of isoproterenol. J Clin Invest 1972;51:1767-76.

Simonsen S, Kjcksbus JK. The effect of free fatty acids on myocardial oxygen consumption during atrial pacing and catecholamine infusion in man. Circulation 1978358:484-Y 1.

Swedberg K, Hjalmarson A, Waagstein F. Wallentin I. Adverse effects of beta-blockade withdrawal in patients with congestive cardiomyopathy. Br Heart J i980;44:134-42.

Kennedy JW, Trenholme SE, Kasser IS. Left ventricular volume and mass from single-plane cineangiocardiograms: a comparison of anteroposterior and right anterior oblique methods. Am Heart J 1970;80:343-52.

Sandier H, Dodge HT. The use of single plane angiocardiograms for the calculation of left ventricular volume in man. Am Heart J 1968;75:325-34.

Grossman W, Braunwald E, Mann T, McLaurin LP, Green LH. Contractile state of the left ventricle in man 3s evaluated from end-systolic pressure- volume relations. Circulatioa lY74;563845-52.

Aroney CN, Herrmrnu HC, Semigran MJ, Dee GW, Boucher CA, Fifer MA. Linearity of the left ventricular end-systolic pressure-volume relalion in patients with severe heart failure. J Am Coil Cardiol 1989:14:127-34.

Little WC. The left ventricular dP/dt,,,,-end diastolic volume relatio!! ifi dogs, Circ Res 1985;56:808-15. Glower DD, Spra(t JA. Snow ND, et al. Linearity of tbc Frank-StarlinE relatiollship in the intact heart: the concept of preload recruitable stroke work. Circulation 3985;71:994-3009. Ross J, BraunwaId E. The study of left ventricular function in man by increasing resistance to ventricular ejection with angiotenain. Circulation

196429~739-49. Sunagawa K, Maughan WL, Burkholf D, Sagawa K. Left ventricular interaction w&h arterial load studied in isolated canine ventricle. Am J

Physiol 1983;245:H773-80.

1320 EICHHORN ET AL. BETA-BLOCKERS IN W&ART FAILURE

JACC Vol. 24, NV. 5 November I, t9Y4t310-20

34. Sunagawa K, Maughan WL, Sagawa K. Optimal arterial resistance for the maximal stroke work studied in isolated canine left ventricle. Circ Res 1985;56:586-95.

35. Raya TE, Gay RG, iancaster L Aguirre M, Moffett C, Goldman S. Serial changes in left ventricular relaxation and chamber stiffness after large myocardiai infarction in rats. CirLvlation 1988,77:1424-31.

36. Eichhorn W, Willard JE, Alvarez L, et al. Are contraction and relaxation coupled in patients with and without congestive heart failure? Circulation 1992;8532132-9.

37. Cheng CP, Little WC. Enhanced load sensitivity of left ventricular relaxation and early diastolic filling in congestive heart failure [abstract]. Circulation 1993:88 Surml I:[-527

38.

39.

40%

41,

42.

43.

44,

45.

46.

47.

Mjlilsky I. &essrnent of diastolic funclion: suggesled methods and future colwiderations. Circulation 1984;69:836-41.

Mirsky I, ipod s A. Clinical aswssment of diastolic function. Prog Cardiiaac Dis ly9a;32:291-=318.

Sy~a H, Futrki S, Gato Y. Energetic+ of the heart. In: Hori 1.1, Baan J. Ycllin EL, editors. Cardiac Mechanics and Function in mal and Disea.~!d Heart. Tokyo: SpringereVerlalg, 1989zl57-63.

Guyion AC. Medics1 Pbysiol~~~ 6th cd. Philadelphia: Eumlcrs, 1981:.504- (a,901-2.

Klocke RA. Carbon dioxide transport. In: Fishman Al? Farhi LE, Tcnney SM. Griper SR, (editors). The Respirator System. ‘Baltimore: Williams t Wilkins, IpX7:173-97.

Konatam MA, Wynnc J, Holman 81, Brown EJ, Neil1 JM, Kozlowski J. Use of equilibrium @ted) radiunuclide ventriculography to quantitate left ventricular output in patients with and without left sided valvular rcgurgita-

Circulation 19N!:f#7g-RS.

mer GJ. Lewis SE, Hillis LD, et al. Nongcometric determination of left wentricular volumes from equilibrium blood pool scans. Am J Cardiol I9t#45:293-3OO.

Katz A. Cellular mechanisms in congestive heart failure. Am J Cardiol 19&62:3A-8A.

hto H, Hori M, Ozaki H, et al. Exercise.induced upward shift of diastolic left ventricular presurc.volume relation in patients with dilated cardiomy opthy: effects of &adrenoccptor blockade. Circulation 1993$&2215-23.

&son G. Daleskog M. Hjemdahl P, Rehnqvist N. Unchanged peripheral sympathetic activity following withdrawal of chronic mctoprolol treatment: a study of noradrenaline concentrations and kinetics in plasma. Br J Clin Pharmac lPRJ:IH:S73-9.

48. Bostrom M. Nie 2, Goertz G, Hen&son J, Wa~lberg-~enriksso~ Indirect effect of catecholamines on development of insulin resistance skeletal muscle from diabetic rats. Diabetes l98438:906-30.

45. Paolisso G. Gambardella A, Marrazzo 6. et al. Metabolic and cardiovascular benefits deriving from P-adrenergic blockade in chronic congestive heari failure. Am Heart J 1992;123:103-10.

50. Andersson B, 5lomstrom-~~mdqv~st C. cdner T, Waagstein F. Escrcise hemodynamics and myocardial ln~tab~~~srn during long-term bcta- adrenergic blockade in severe heart failure. J Am Coil Cardiol 199P;16: 1059-66.

51. Gilbert JC, Glantz SA. Determinants diastolic pressurc+olume relation. Circ

52. Eichhorn EJ. Reply. Circulation 1991;

53. Mirsliy 1. Tajimi T, Peterson KL. Tbe dn&pment of the cntirc end-systolic pressure-volume and cjcction fraction-a~lerl~ad relations: a new conrept of systolic ~l~~~~~~di~~~ stiffness. Circ~~;~t~~~~~ ~~~7~7~343-~~.

54. van dcr Veldt ET, Burkhoff D, Stecndijk P, Karsdon J. Sngaw,~ K. Haan J. Nonlincilri~ and losd sensitivity of end+ystolic ~rc~~~~re~vo~~i~~c r&ion of canine left ventricle in vivo. Circulati0a ~‘~~~~~~:31~-2?.

55. Eichhnrn Konsmm MA, pavnc DD, %d!em

graf~i~lg. Am J Cardi ~4g9~~3:~~7-9~. , ~~ontg~)mc~ DG. Gross MD. ~~~cl~~~l~isn~s for left

ventricular systolic dysfunction in aortir regurgitation: importance for prc- dieting the fuectiotml response 10 aortic valve rcptac~~~ie~t. J Am Coil Cardiol 3991;173887-97.

Little WC, Cheng C-P. Mumma M. lgurashi Y. Vinten~~ohanse~ J. Johnston WE. Comparison of measures of left ~~entricM~~ir contractile ~~~~rlnal~ce derived from p~ss~rc-volume loops in conscious dogs. Circu~~~tion 198% 1378-87.

Bagger JP. Coroamy sinus blood I%JW d~tc~l~fli~~~oo hy tbc tbe~~no~~lMt~1~~ technique: influcacc of catheter position aad respira&. C~~~~~ov~sc Res l9X5:19:27-31.

Mathey DG, Chatterjce K. Tyberg JV. Likven J, 8:nndage Coronary sinus reflur: a source of error in the measurement of thcrmodilu- tion coronary sinus Row. Circulation l97~5~77g-~.

Cohn JN. Johnson G. Ziescbe S. et al. A comparison of enalaprd with Ilpdmlsrinc-isusorbid~ dmltrale in the treatmum of chronic congersrive bear1 failure. N Engl J Med 19Yl;PS:303-IO.

Documents

Effect of metoprolol on myocardial function and energetics ... · ardial failure results in i~ad~~~~t~ s~t~fl~i~ ion of intracardiac pressures and activati~~n Q satnry neurohormonal