Journal of Clinical InvestigationVol. 45, No. 10, 1966

Studies on Digitalis. XVI. Effects on MyocardialOxygen Consumption *

JAMES WV. COVELL, EUGENEBRAUNWALD,tJOHNRoss, JR., ANDEDMUNDH. SONNENBLICK

(From the Cardiology Branch, National Heart Institute, Bethesda, Md.)

The effects of cardiac glycosides on myocardialoxygen consumption (MVo2) have been investi-gated extensively, and although the results of thesestudies have not been entirely uniform, it is nowgenerally considered that these agents are "the onlydrugs which increase the force of contraction ofthe myocardium without at the same time increas-ing oxygen consumption" (1). The finding inprevious investigations that digitalis does not in-crease MVo2 (2-8) would not appear to be com-patible with observations that the glycosides in-crease the velocity of myocardial fiber shortening(9, 10), since recent studies have suggested thatthe latter variable is an important determinant ofMVo2 (11, 12). These considerations promptedan examination of the effect of acetylstrophanthidinon MVo2 in a preparation that allowed control ofhemodynamic variables other than contractionvelocity that can significantly influence MVo2.

Methods

Experiments were performed on eight dogs weighing15.5 to 17.2 kg and anesthetized with morphine, 2.8 mgper kg; chloralose, 96 mg per kg; and urethane, 620 mgper kg. The chest was entered through a sternal splittingincision, and ventilation was maintained with a Harvardrespiratory pump, using 100%o 02. Heparin (500 U perkg) was administered. The venae cavae were then can-nulated, and the venous return was diverted into a reser-voir, from which bypass of the right side of the heart wasachieved with an occlusive roller pump, as described indetail previously (12). The pump supplied blood to thepulmonary artery through a cannula inserted via theright ventricular outflow tract; the right ventricle wastherefore empty and performing no external work.Fresh blood obtained from donor dogs and exchangedwith that of the experimental animal was used to primethe reservoir and circuit. Mean systemic arterial pres-sure was regulated by a reservoir connected to a com-

* Submitted for publication April 25, 1966; acceptedJune 16, 1966.

tAddress requests for reprints to Dr. Eugene Braun-wald, Cardiology Branch, National Heart Institute,Bethesda, Md. 20014.

pressed air circuit and attached by large bore cannulaeto the femoral arteries. The sino-atrial node wascrushed, and heart rate was maintained constant by elec-trical stimulation 1 of the right atrium or ventricle.Stroke volume could therefore be controlled by main-taining the output of the pump constant.

A flow transducer was placed around the ascendingaorta, and instantaneous aortic blood flow was mea-sured with a gated-sine wave electromagnetic flowmeter.2Left ventricular pressure was measured with a StathamP23Db pressure transducer attached directly to a large-bore metal cannula that was inserted through the the ven-tricular apex. Central aortic pressure was measuredthrough a similar cannula inserted into the aortic archthrough the left subclavian artery. The first derivative(dp/dt) of the left ventricular pressure pulse was deter-mined with an analogue differentiating circuit 3 and wasrecorded continuously together with pressure, flow, andthe electrocardiogram on a direct-writing multichanneloscillograph. Experimental data were also transcribedon FM tape 4 and replayed for analysis of phasic data.

Coronary blood flow was measured during a steadystate by timed collection of the right ventricular drainage,this effluent representing total coronary blood flow minusthe left ventricular thebesian vein drainage. Arterial andcoronary venous blood samples were obtained simul-taneously and analyzed in duplicate for 02 by the methodof Van Slyke and Neill. MVo2 was then calculated asthe product of coronary blood flow and the coronary ar-teriovenous 02 difference. MVo2 calculated in thismanner represents the contributions both of the function-ing left ventricle and of the empty nonworking rightventricle. The latter is undoubtedly very small and hasbeen ignored in this study.

In two hearts the effects of acetylstrophanthidin werestudied at high initial left ventricular end-diastolic pres-sures. Since pulmonary edema and hemorrhage occurat high pressures in the right heart bypass preparation,a disposable bubble oxygenator 5 ventilated with 97% 02,3% C02 was substituted for the lungs. Blood from the

1 American Electronic Laboratories, Colmar, Pa., model104A.

2 Biotronex Laboratory, Silver Spring, Md., modelBL-310.

3 Designed by Electronic Gear, Inc., Valley Stream,N. Y., utilizing a Philbrick P65AU operational amplifier.

4 Model 3900 tape recorder, Sanborn, Cambridge, Mass.5 Travenol Laboratories, Morton Grove, Ill.

1535

COVELL, BRAUNWALD,ROSS, AND SONNENBLICK

venae cavae was drained into the oxygenator and thenpumped directly into the left atrium through a cannulainserted via a right pulmonary vein. The hila of bothlungs and the main pulmonary artery were then ligated.

The experimental design in both preparations was thesame. During the control period, i.e., before the adminis-tration of acetylstrophanthidin, the relationship betweenthe left ventricular end-diastolic pressure (LVEDP),stroke volume, and stroke work was determined by in-creasing the output of the pump in a stepwise fashion,with heart rate and mean aortic pressure maintainedconstant. The output of the pump was then set so that theLVEDP was relatively low in the right heart bypasspreparations and relatively high in the other two ex-periments. Coronary blood flow was measured, and ar-terial and coronary venous blood samples were obtainedin duplicate during a steady state. Then, while pumpoutput, mean aortic pressure, and heart rate were main-tained constant, an average of 0.26 cat U per kg (0.19 to0.40 cat U per kg) acetylstrophanthidin was injected intothe reservoir. Ten to 20 minutes later, at the time ofthe maximal hemodynamic response, MVo2 was againdetermined, and the left ventricular function curve wasthen repeated.

The diastolic pressure-volume curve of the left ven-tricle was determined at the end of each experiment bya method described in detail previously (13). In brief,the heart was arrested with 25% KCl, the mitral andaortic valves were occluded, and the cavity was filled with2-ml increments of fluid. The left ventricular end-diastolic volume was then calculated directly from theresulting pressure-volume curve and the recordedLVEDP.

Stroke work in gram-meters was calculated as [(MAP- LVEDP) (SV) (1.36) 1 /100, where MAP= mean bloodpressure, LVEDP= left ventricular end-diastolic pres-sure (in millimeters Hg), and SV=stroke volume (inmilliliters). Stroke power (in gram-meters per second)was obtained by dividing stroke work by the duration ofejection, and mean ventricular ejection rate (in millilitersper second) was calculated by dividing the stroke volumeby the duration of ejection. The maximal rate of leftventricular ejection was determined from the peak de-flection of the electron'agnetic flowmeter tracing. Thisvalue for maximal flow velocity in the ascending aortadoes not take into account that very small fraction ofblood which enters the coronary circulation during earlysystole. The tension-time index (TTI) in millimetersHg per second was determined by planimetric integra-tion of the ventricular pressure pulse during ejection(14). External cardiac efficiency was calculated as theratio (in per cent) between minute work and the workequivalent of the 02 consumption per minute (15).

Total tangential wall tension at the internal equatorwas calculated from the LaPlace relation: T = Pirr',where T = wall tension in grams, P = intraventricularpressure in gram-centimeters squared, and r = internalleft ventricular radius. r was calculated from the end-diastolic volume and the aortic flow rate by assuming aspherical left ventricular model and solving the equation

V = 4/37rr' for r. The tensions calculated from theLaPlace relationship were then plotted at 10-msec inter-vals, and the area (in gram-seconds) under the tension-time curve from the onset of contraction to peak ten-sion was determined by planimetric integration.

Results

Acetylstrophanthidin exerted a positive inotropiceffect in each study. As shown in the representa-tive experiment reproduced in Figure 1, it ele-vated the relationship between LVEDPand strokevolume (Figure lA), stroke work (1B), strokepower (1C), mean ejection rate (iD), peakaortic flow rate (lE), and the peak left ventriculardp/dt (IF). The effects of the administration ofthis drug at a constant stroke volume, aortic pres-

4, .8

.24 8 12

LVEDPRcm H20 LVEDPRcm 20.

FIG. 1. EFFECTS OF ACETYLSTROPHANTHIDIN (ACS),0.24 CAT U PER KG, ON MYOCARDIALPERFORMANCECURVESIN DOG NO. 5 (HEART WEIGHT= 104 G). Heart rate wasmaintained at 176 per minute throughout. MER= meansystolic ejection rate; Ao = aortic; LV dp/dt = peakrate of rise of left ventricular pressure. LVEDPR=

left ventricular end-diastolic pressure.

1536

EFFECTS OF DIGITALIS ON MYOCARDIALOXYGENCONSUMPTION

11CO

E

.,

I-IE

2

er

cr

gi3:

60

30

100

70

A

/-

'oIl

.660-

20

80

40

CONTROL ACS

160

noI_

3:

9

100

.: .~~

2 10

O.

,, 1700

1300

900

32

24a

E

CLu4

16

-I4

..........L~1IH

CONTROL ACS

41.10,mC>cnmL'iI.-

I.-cn

V)

m

'E-iC>

I--cn-1C.mmuj

F

360

280-

200-

120

40

a - Dog no 7 1a- Dog no 8

40 0- RHB Avg.- ISD

30 I

20

10

0CONTROL ACS

FIG. 2. EFFECrS OF ACETYLSTROPHANTHIDIN. Thedata shown as circles represent the average values (± 1SD) in the six right heart bypass (RHB) preparations.The squares and triangles represent the data obtained inexperiments no. 7 and 8 in which the LVEDPRwas ini-tially high (Table I). MVo2=minute myocardial oxy-gen consumption.

sure, and heart rate are shown in Table I and Fig-ure 2, and representative recordings from bothpreparations are shown in Figure 3.

The average of duplicate determinations ofMVo2 is shown in Table I; the standard deviationof the differences between duplicate determinationswas + 0.26 ml per minute. In the six experi-ments with the right heart bypass preparation inwhich the LVEDPwas low, it was observed thatMVo2 rose by an average of 2.56 ml per minute,representing an increase of 34% (Figure 2A).Since mean aortic pressure and stroke volumewere maintained constant, left ventricular strokework also remained essentially unchanged. Ac-cordingly, the ratio of external left ventricular min-ute work to the work equivalent of the MVo2, i.e.,the external efficiency, decreased in all experi-ments, by an average of 24.4%o of the control value.

*

*It4

0(4

¢

04

t:

Po %

n oo-so0_ >oom4v-tOo o oi- -4 4w~ ~ ~ -__ _ _ 4mN"

CONN C4 %O 0 ~0 0 t- 0 (01 t- C4

\0U)-nV0-0-o00--0V-S.- _; 4 .; . _, 4

- 4)i . . .. . ......i- 4--C l0 4c 0C40t ~s- -_O e*s~ 0c o

ff__ _ 0-0 _4-voQ_ N

'4 02X_.

Inw 1- nt_Wooo)e 0 _ o oc o oto-0- inCtnL m .0 C t u),tin- W)0o

"I

.,tlI"!¢S

ho __4

04tk C4 444 414 4(

0 a ) 0U) It0 0q04 0 000 Q% to0oo U)OOoo oo4Umozoomo-_ C4 444____ 4'N_ __ 4 m~

o444--o4444404444o4o4

8-o

t4 N C

.> 0 -t 0t0qllom0~b ~J$ W)tt- 00 o' tD0 000t-C, -0r0o I O0

ol0 M t- 0 Uto U-)(Vst-),j M0 r0-0 etO44 m CM00o~owu _ _ __ _____--4~14

Xt4 N o0000 )4 '0 0U-44- 00 o 0

a. - )0-4-0- - -V-0 -

~0

10

04

En

n 4)

<0

+00000.0000.U)44 -00U)

OerOM0 0- V,. U.. ..)

r- u+\0v\4Xee-oN\0t4 -.. - C r* M-u-otC4't- s- 4, C e

W Wn__ ________CC4- 0 0

toot4

U

a-Oa,- -N4 UOm ON

N 9 -4 -4 N

00- 00 r0-U) 0-0 t InU

U) 0s + + U)

- cto 0

NOhvO o 0s -- t-0 - +

CdU) cn) ) U) U) °U U U U U UD U Q

+

_ U u U) 00

1537

ii 44 U,a.Cd

. U)'4i

0-cs o

4(40

IT)'

4-I 06>044)>

0~.

_ _

<u

(0

0

0 ;'-'0*oN

04C.CO

II 0

o

C4

11 x

0.4.U1

0 U

V.~E

.04Cd0 11

= *^0

=m EaX >n

0

@<e

U 9 in

4)z 0 a.0. 0

'-0

.0> >1441~0 <(Q

S04.3 00

>d ) 4s> >

W. ,-

41 be

el

COVELL, BRAUNWALD,ROSS, AND SONNENBLICK

A Dog 45CONTROL

Ao Flow r -mlI. /sec L t$ F

0I

ECG

200JAo. Pres. ........200 .:::.... :.. :tt '-

mm.Hg

LV. Press tt;i-i+mm. Hg

LV ED Press 0?f > t-:cm. 120

LV d P/dt 2800,mmHg/se~c

L VEDP 7.2Ao. Flow 110Lv dp/dt 1720MV02 (mI/mmn/100g.) 7.75

ACS

4 -4

4''.4 4S

1.4165

401910.22

B. Dog 8CONTROL

1~~~~3

I :~, -+

..,

.'.

_ _ ~

,_

-+-~~~~~~05;sec.'t'... - --...

1550.38

ACS

ta

J. ~ *.Ltz-itt

t -t t.

152-L~2900 [._._F ,6 itet+|FF.39t

FIG. 3. REPRESENTATIVETRACINGSOBTAINEDFROMTWOEXPERIMENTS.

The peak aortic flow rate, peak rate of rise ofleft ventricular pressure, mean systolic ejectionrate, and stroke power increased significantly, byaverages of 36%, 82%, 11%o, and 12%o of control,respectively (Figure 2B, C, D, G).

The LVEDP was low during the control pe-riod (average = 4.1 cm H2O) and decreasedslightly, by an average of 2.3 cm H2O, after theadministration of acetylstrophanthidin (Figure2H). Similarly, left ventricular end-diastolicvolume decreased slightly (Figure 2I). Sinceaortic pressure was held constant and left ventricu-lar end-diastolic volume declined somewhat, peakleft ventricular tension decreased slightly, from anaverage of 870 g to an average of 710 g (Figure2E). Since the duration of ejection also tendedto decrease, the integrated systolic tension fellsomewhat more than the peak tension, from anaverage of 110.9 g-sec to an average of 65.8 g-sec(Figure 2F).

The results of the two experiments (Table I, no.7 and 8) in which the control LVEDPand left ven-tricular end-diastolic volume were markedly ele-vated differed from those observed in the right heartbypass preparation. Thus, MVo2 either remainedconstant after acetylstrophanthidin (no. 7) or de-clined slightly (no. 8) (Figure 2A). In con-

trast to the right heart bypass experiments, ad-ministration of acetylstrophanthidin now resultedin a marked decrease in LVEDPand left ventricu-lar end-diastolic volume (Figure 2H and I), andas a consequence, peak left ventricular tension andintegrated systolic tension declined to. a far greaterextent than in the right heart bypass preparations(Figure 2E and F). On the other hand, the in-creases in peak aortic flow, left ventricular dp/dt,and mean systolic ejection rate were comparableto those observed in the six experiments carriedout in the right heart bypass preparation.

In experiment no. 8 the administration of acetyl-strophanthidin lowered LVEDPfrom 28.6 to 16.3cm of H2O. While the effect of the drug persisted,the LVEDP was re-elevated to 27.2 cm H20 byincreasing stroke volume (Figure 4). This in-tervention allowed examination of the effects ofacetylstrophanthidin on MVo2 at comparable val-ues of peak and integrated tension. Under theseconditions MVo2 rose significantly above control,from 7.69 to 10.6 ml per minute (Figure 4).

DiscussionNumerous investigators have studied the effects

of cardiac glycosides on MVo2, but the resultshave been conflicting. In most experiments the

1538

EFFECTS OF DIGITALIS ON MYOCARDIALOXYGENCONSUMPTION

I

to .C

,° E 9 _

E

L2100

LLo

1700i7OO400 _

F Q _

200 -

24

> - 20

16

35

a _g

25-

27V

00 X

>E-J

DOGNO8

0-

22F

17

CONTROL

FIG. 4. SUMMARYOF RESUX

Initially the effects of ACS vinotropic effect of the drug v

left ventricular end-diastolic pre

Then, while the ACS was still(preload) was increased andturned close to control levels.sion. IST = integrated systovolume.

02 uptake of fresh heart sliiof heart muscle and heardid not appear to be affistances (16-19). Lee ob.cally contracting, isolated c

ouabain first increased conting Vo2, but that Vo2 thertropic effect was reachedas the muscle went intoRohde and Ogawa in 191thidin increased MVo2 in tand that this increase was I

mentation of the product ofsure and heart rate inducPeters and Visscher (23)(24) reported that cardiaincrease MlVo2 in the canition in which end-diastolic

stant, although in some experiments no changein MVo2 occurred.

On the other hand, several studies support thecontention that digitalis does not affect MVo2,or even decreases it. Thus, Gremels described adecline in MVo2 and an increase in efficiency afterdigitalis administration in the isolated heart-lungpreparation (2, 3). Gollwitzer-Meier and Krugernoted no change in MVo2 in fresh heart-lungpreparations and a fall in MVo2 in grossly failingpreparations (4). Olson, Roush, and Liang re-ported that acetylstrophanthidin did not alterMVo2 in the failing intact dog heart (5), and Bingand his colleagues (6, 7) observed no alterationin MVo2 when strophanthidin or lanatoside C wasadministered to normal subjects or to patients withheart failure. The problem was most recently in-vestigated by Sarnoff and his collaborators in theisolated support canine heart (8). These workersobserved that acetylstrophanthidin produced amarked increase in myocardial contractility with-out a change in either MVo2 or external efficiency,

ACS ACS +PRELOAD although they appreciated that the reduction of

,TS IN EXPERIMENT NO. 8. ventricular end-diastolic volume produced by thewere determined while the drug would be expected to diminish the MVo2.was permitted to decrease Although considerable data indicate that digi-essure and volume (EDV). talis may exert a positive inotropic effect without1acting, ventricular inflows . .. .LVEDPR and EDV re- increasing MVo2, our finding of significant aug-

PT=peak systolic ten- mentation of MVo2 induced by acetylstrophanthi-olic tension. SV= stroke din contrasts sharply with the results of most of

the experiments reviewed above. The presentresults are in accord, however, with the findings

ces and of homogenates in earlier studies from this laboratory in which thet muscle mitochondria effects of three other inotropic interventions, i.e.,ected by digitalis sub- paired electrical stimulation, norepinephrine, andserved in the isometri- calcium were found to augment MVo2 substan-at papillary muscle that tially (11, 12).tractility without affect- This disparity of results may be understood byi rose as the peak ino- analyzing the effects of an inotropic interventionand continued to rise on MVo2, relative to its action on the mechanicalcontracture (20, 21). factors of contraction that are known to affect the

2 found that strophan- MVo2. In the right heart bypass preparation, thethe isovolumic cat heart stroke volume, aortic pressure, and heart rate wereproportional to the aug- all held constant; moreover, since in this nonfail-f ventricular pulse pres- ing preparation the LVEDP and left ventricular-ed by the drug (22). end-diastolic volume were initially in a low physi-and Moe and Visscher ological range, these two variables could not de-

Lc glycosides tended to dine a great deal after administration of acetyl-ine heart-lung prepara- strophanthidin. Accordingly, both the peak and

volume was held con- integrated left ventricular systolic tensions de-

1539

COVELL, BRAUNWALD,ROSS, AND SONNENBLICK

creased relatively little after acetylstrophanthidinhad been administered. If the drug had exertedno other hemodynamic effect, then the decline intension, albeit small, would have been expected toresult in a slightly lower MVo. (14, 25, 2o).However, acetylstrophanthidin also markedly in-creased the velocity of myocardial fiber shorteningas reflected in the augmentation of the peak rateof left ventricular ejection, the peak rate of leftventricular pressure rise, and the mean systolicejection rate.

Data have previously been presented to supportthe position that the velocity of myocardial fibershortening accompanying a basic improvement incontractile state is an important determinant ofMVo2 (11, 12). It would thus appear that theaugmentation of MVo2 observed in the presentstudy is related, at least in part, to this increasein the velocity of myocardial fiber shortening re-

sulting from the increase in the contractile stateafter acetylstrophanthidin. The digitalis sub-stance also augmented the extent of myocardialshortening. If we assume the left ventricle tohave a spherical shape, the left ventricular internalcircumference decreased by an average of 2.11 cm,

representing 22.9% of the end-diastolic circum-ference, during each systole in the control period,and by 2.78 cm, representing 33.5%c of the end-diastolic circumference during each systole afterthe acetylstrophanthidin. Therefore, it is possiblethat this increase in the extent of fiber shorteningalso played a role in the observed augmentation ofAMVO2. Nevertheless, it is clear from this andprevious studies (11, 12) that whenever an in-crease in the contractile state of the heart is in-duced, as characterized by an augmentation of theintrinsic velocity of muscle shortening, an aug-

mentation of MVo2 occurs even though there are

no changes in external work performance or ten-sion generation.

The major difference between the right heartbypass preparation and the isolated heart or heart-lung preparations employed by earlier investigatorswho studied the effects of digitalis on MVo2 there-fore appears to lie in the fact that the formerpreparation was nonfailing whereas the latter were

in varying degrees of failure and their LVEDPand left ventricular end-diastolic volumes were ofnecessity markedly elevated. For example, in thestudy by Sarnoff and co-workers (8), the control

value of LVEDP averaged 19.5 cm of H)0 be-fore the acetylstrophanthidin and was markedlydecreased, to an average of 8.6 cm H20, after thedrug. In contrast, in the present study the con-trol value of LVEDPaveraged only 4.1 cm H20and declined by only 2.3 cm HO. Striking de-creases in ventricular end-diastolic volume werealso induced by digitalis in the failing hearts stud-ied by Gollwitzer-Meier and Kruger (4). In theface of a constant aortic pressure these strikingdecreases in ventricular end-diastolic pressure andvolume must have been associated with large re-ductions of ventricular wall tension. Since myo-cardial wall tension is generally accepted to bean important determinant of MVo2 (14, 25, 26),the marked decreases in tension occurring inearlier experiments would have been expected toreduce MVo2 and to oppose the augmentation ofMVo2 resulting from the increased velocity ofmyocardial fiber shortening and Vmax (maximalvelocity).

The results of several of the earlier investiga-tions are in accord with this hypothesis. As men-tioned earlier, digitalis glycosides increased MT;o2in experiments in which ventricular end-diastolicvolume was held constant (22-24). Further-more, the findings in two of our experiments, inwhich the level of LVEDP and left ventricularend-diastolic volume before the administration ofacetylstrophanthidin were purposely elevated, alsosupport the view that the fall in myocardial ten-sion occurring in the failing heart may mask theincrease in MVo2 produced by digitalis. In thesetwo experiments (Table I, no. 7 and 8) acetyl-strophanthidin augmented the peak aortic flowrate, mean ejection rate, and left ventricular dp/dtto about the same extent as in the right heart by-pass preparation. However, the absolute reduc-tions in left ventricular tensions, both peak andintegrated, were far greater than in the rightheart bypass preparation, and MVo2 failed to rise.Finally, in experiment no. 8, in which LVEDPand left ventricular end-diastolic volume were re-elevated to control levels after the administrationof acetylstrophanthidin, thus preventing the de-cline in left ventricular tension produced by thedrug, MVo2 was observed to rise to values ex-ceeding control (Figure 4).

The results of the present investigation are rele-vant to a consideration of the effects of digitalis on

1540

EFFECTS OF DIGITALIS ON MYOCARDIALOXYGENCONSUMPTION1

MV02 in man. Cardiac glycosides have beenshown to elevate systemic vascular resistance (27)and arterial pressure (28) and to increase- thevelocity of myocardial contraction, as reflected inaugmentations of the rate of rise of intraventricularpressure (29), mean systolic ejection rate (30),and the rate of movement of radiopaque markerson the ventricles (10). These hemodynamicchanges would tend to increase MVo2 and in thepresence of coronary artery disease without heartfailure could lead to myocardial ischemia. Onthe other hand, in the presence of heart failure,the reduction in heart size induced by digitaliswould tend to reduce myocardial wall tension andthus prevent the augmentation of MVo2.

SummaryThere has been considerable dispute concerning

the effects of digitalis on myocardial 02 consump-tion (MVo2) and efficiency. Analysis of previ-ous data suggested that the interpretation of re-sults was complicated by the changes in circulatorydynamics induced by the drug. Accordingly, theeffects of acetylstrophanthidin (average dose =0.26 cat U per kg) were studied in six nonfailing,canine, right heart bypass preparations in whichheart rate, stroke volume, and mean aortic pressurewere held constant. MVo2 increased in all ex-periments, by an average of 2.56 ml per minute,whereas calculated cardiac efficiency declined byan average of 24.4% of control. Even thoughmean arterial pressure was held constant, the gly-coside reduced the integrated systolic tension byan average of 40%, and the peak systolic tensionby an average of 18%o, chiefly as a consequence of asmall decline in left ventricular end-diastolic vol-ume. However, the velocity of myocardial fibershortening increased considerably, the peak leftventricular ejection rate rising an average of 36%and the peak ventricular dp/dt increasing by anaverage of 82%. Acetylstrophanthidin did notalter MVo2 in two hearts which were studied in anidentical manner, but in which left ventricular end-diastolic pressure was initially elevated. How-ever, in those experiments the large fall in end-diastolic volume resulted in a marked fall in sys-tolic tension.

We conclude that digitalis tends to increaseMVo2but that its strongly positive inotropic effectfrequently results in a reduction of ventricular wall

tension that tends to oppose and to mask thiseffect.

References1. Modell, W. The pharmacologic basis of the use of

digitalis in congestive heart failure. Physiol. andPharmacol. for Phycns. 1966, 1, 1.

2. Gremels, H. Zur Physiologie und Pharmakologie derEnergetik des Saugetierherzens. Naunyn-Schmied-eberg's Arch. exp. Path. Pharmak. 1933, 169, 689.

3. Gremels, H. tber den Einfluss von Digitalis Gly-kosidin auf die energetischen Vorgange am Sauge-tierherzen. Naunyn-Schmiedeberg's Arch. exp.Path. Pharmak. 1937, 186, 625.

4. Gollwitzer-Meier, K., and E. Kruger. Herzener-getik und Strophanthinwirkung bei verschiedenenFormen der experimentellen Herzinsuffizienz.Pfluigers Arch. ges. Physiol. 1936, 238, 251.

5. Olson, R. E., G. Roush, and M. M. L. Liang. Effectof acetyl strophanthidin upon the myocardial me-tabolism and cardiac work of normal dogs anddogs with congestive heart failure (abstract).Circulation 1955, 12, 755.

6. Bing, R. J., F. M. Maraist, J. F. Dammann, Jr., A.Draper, Jr., R. Heimbecker, R. Daley, R. Gerard,and P. Calazel. Effect of strophanthus on coronaryblood flow and cardiac oxygen consumption ofnormal and failing human hearts. Circulation1950, 2, 513.

7. Blain, J. M., E. E. Eddleman, A. Siegel, and R. J.Bing. Studies on myocardial metabolism. V.The effects of lanatoside C on the metabolism ofthe human heart. J. clin. Invest. 1956, 35, 314.

8. Sarnoff, S. J., J. P. Gilmore, A. G. Wallace, N. S.Skinner, Jr., J. H. Mitchell, and W. M. Daggett.Effect of acetyl strophanthidin therapy on cardiacdynamics, oxygen consumption and efficiency inthe isolated heart with and without hypoxia.Amer. J. Med. 1964, 37, 3.

9. Spann, J. F., Jr., E. H. Sonnenblick, T. Cooper,C. A. Chidsey, V. L. Willman, and E. Braunwald.Studies on digitalis. XIV. Influence of cardiacnorepinephrine stores on the response of heartmuscle to digitalis. Circulat. Res. In press.

10. Sonnenblick, E. H., J. F. Williams, Jr., G. Glick,D. T. Mason, and E. Braunwald. Studies on digi-talis. XV. Effects of cardiac glycosides on myo-cardial force-velocity relations in the non failinghuman heart. Circulation. In press.

11. Ross, J., Jr., E. H. Sonnenblick, G. A. Kaiser, P. L.Frommer, and E. Braunwald. Electroaugmenta-tion of ventricular performance and oxygen con-sumption by repetitive application of paired electri-cal stimuli. Circulat. Res. 1965, 16, 332.

12. Sonnenblick, E. H., J. Ross, Jr., J. W. Covell, G. A.Kaiser, and E. Braunwald. Velocity of contractionas a determinant of myocardial oxygen consump-tion. Amer. J. Physiol. 1965, 209, 919.

13. Ross, J., Jr., J. W. Covell, E. H. Sonnenblick, and E.Braunwald. Contractile state of the heart char-

1541

COVELL, BRAUNWALD,ROSS, AND SONNENBLICK

acterized by force-velocity relations in variablyafterloaded and isovolumic beats. Circulat. Res.1966, 18, 149.

14. Sarnoff, S. J., E. Braunwald, G. H. Welch, Jr., R. B.Case, W. N. Stainsby, and R. Macruz. Hemody-namic determinants of oxygen consumption of theheart with special reference to the tension-timeindex. Amer. J. Physiol. 1958, 192, 148.

15. Gregg, D. E. Coronary Circulation in Health andDisease. Philadelphia, Lea & Febiger, 1950, p. 151.

16. Salomen, K., and 0. Reisser. Zur Frage des Ein-flusses von Digitoxin und Strophanthin auf oxyda-tive Vorgange in Versuchen am Modell sowie, am

atmenden iuberlebenden Herzmuskelgewebe. Nau-nyn-Schmiedeberg's Arch. exp. Path. Pharmak.1935, 177, 450.

17. Wollenberger, A. Metabolic action of cardiac gly-cosides. I. Influence on respiration of heart muscleand brain cortex. J. Pharmacol. exp. Ther. 1947,91, 39.

18. Rothlin, E., and 0. Schoelly. Einfluss herzwirksamerGlykoside auf die Atmung des Herzens, gemessen

mit der Warburg-Apparutur. Helv. physiol. phar-macol. Acta 1950, 8, C 69.

19. Langemann, H., T. M. Brody, and J. A. Bain. Invitro effects of ouabain on slices and mitochondrialpreparations from heart and brain. J. Pharmacol.exp. Ther. 1953, 108, 274.

20. Lee, K. S. A new technique for the simultaneousrecording of oxygen consumption and contractionof muscle: the effect of ouabain on cat papillarymuscle. J. Pharmacol. exp. Ther. 1953, 109, 304.

21. Lee, K. S., D. H. Yu, D. I. Lee, and R. Burstein.The influence of potassium and calcium on theeffect of ouabain on cat papillary muscles. J.Pharmacol. exp. Ther. 1961, 132, 139.

22. Rohde, E., and S. Ogawa. Gaswechsel und Tatigheitdes Herzens unter dem Einfluss von Giften und

Nervenreizung. Naunyn-Schmiedeberg's Arch.exp. Path. Pharmak. 1912, 69, 200.

23. Peters, H. C., and M. B. Visscher. The energymetabolism of the heart in failure and the influenceof drugs upon it. Amer. Heart J. 1936, 11, 273.

24. Moe, G. K., and M. B. Visscher. Studies on thenative glycosides of digitalis lanata with particularreference to their effects upon cardiac efficiencyand their toxicity. J. Pharmacol. exp. Ther. 1938,64, 65.

25. Rodbard, S., F. Williams, and C. Williams. Thespherical dynamics of the heart (myocardial ten-sion, oxygen consumption, coronary blood flow andefficiency). Amer. Heart J. 1959, 57, 348.

26. Monroe, R. G., and G. N. French. Left ventricularpressure-volume relationships and myocardial oxy-gen consumption in the isolated heart. Circulat.Res. 1961, 9, 362.

27. Braunwald, E., R. D. Bloodwell, L. I. Goldberg, andA. G. Morrow. Studies on digitalis. IV. Obser-vations in man on the effects of digitalis prepara-tions on the contractility of the non-failing heartand on total vascular resistance. J. clin. Invest.1961, 40, 52.

28. Mason, D. T., and E. Braunwald. Studies on digi-talis. X. Effects of ouabain on forearm vascularresistance and venous tone in normal subjects andin patients in heart failure. J. clin. Invest. 1964,43, 532.

29. Mason, D. T., and E. Braunwald. Studies on digi-talis. IX. Effects of ouabain on the nonfailinghuman heart. J. clin. Invest. 1963, 42, 1105.

30. Weissler, A. M., W. G. Gamel, H. E. Grode, S.Cohen, and C. D. Schoenfeld. The effect of digi-talis on ventricular ejection in normal humansubjects. Circulation 1964, 29, 721.

1542