Studies on Starling's Law of the HeartDeterminants of the Relationship Between Left Ventricular

End-Diastolic Pressure and Circumference

By EUGENE BRAUNWALD, M.D., ROBERT L. FRYE, M.D.,

AND JOHN ROSS, JR. , M.D.

With the technical assistance of Robert. Lewis

IN EXPERIMENTS oti the dog heart-lungpreparation which led to the formulation

of the well-known "Law of the Heart,"1

Patterson, Piper and Starling2 observed thatwhen cardiac work was augmented, ventricu-lar volume sometimes increased in the faceof a constant left ventricular end-diastolicpressure. These investigators concluded thatthe strength of myocardial contraction was afunction of end-diastolic fiber length.1-2

Although this view is supported by the ex-periments of Kozawa,3 other investigatorshave concluded that the initial intraventricu-lar tension4"0 or both the end-diastolic fiberlength and the initial tension7"9 are respon-sible for the characteristics of the subsequentcontraction.

In view of these divergent experimentalresults and the importance of a clear under-standing of the operation of the Frank-Starling mechanism,10 a study was under-taken in the dog with a complete circulationof the relationship between the left ventricu-lar end-diastolic fiber length, as reflected inleft ventricular end-diastolie circumference,and left ventricular end-diastolic pressure.The investigations reported herein were de-signed to determine whether this relationshipis always a constant one or whether it maybe modified by alterations in heart rate,temperature, myocardial contractility, cardiacoutput and arterial pressure. These parameterswere chosen for examination because withthe exception of temperature they are fre-quently altered in situ, and it was thereforeof interest to determine whether or not

From the Section of Cardiology, Clinic of Surgery,National Heart Institute, Bethesda, Md.

Received for publication June 9, 1960.

changes in tlie.se factors can influence theend-diastolic pressure-circumference relation-ship. The effects of changes in temperaturewere studied because, as indicated below, theyhelped to elucidate one of the mechanismscontrolling the end-diastolic pressure-circum-ference relationship.

Recent observations have indicated thatthe area beneath the ventricular pressurecurve, i.e., the tension-time index, is thelieinodynamic determinant of myocardial oxy-gen consumption.11 Accordingly, the presentexperiments were also designed to permitexamination of the relationship between thetension-time index and the left ventricularend-diastolic circumference. Preliminary re-ports of portions of this study have beenpresented previously.12-18

MethodsTwenty-seven experiments were performed on

adult mongrel dogs, weighing 13 to 24 Kg. with,-ui average weight of 17.S Kg. Anesthesia, wasprovided by a combination of morphine, chloralose,and urethane in a manner described previously.14

A left thoracotomy was performed, utilizing posi-tive pressure respiration with 100 per cent oxygen.The heart was exposed through a left thoracotomy.and the pericardium was opened widely. Left ven-tricular pressure was measured by means of alarge-bore (1.5 mm. I.D.) eannula, 7 mm. inlength, which was introduced into the left ven-tricular cavity through the ventricular apex andattached directly to a Statham P23D pressuretransducer.* The left ventricular pressure pulserecording was amplified, and usually only thediastolic pressure was recorded, permitting pre-cise measurement of left ventricular end-diastoliepressure.

Left ventricular circumference was recordedcontinuously by means of the mercury-filled re-

*This method was suggested to the authors by Dr.R. .1. Linden.

1254 Circulation Research, Volume VIII, November 1960

by guest on May 17, 2018

http://circres.ahajournals.org/D

ownloaded from

LEFT VBNTR.LCU.IJAR D1AST0L1C LJRESSURE 1255

sistance gages described by Whitney.1" The gagesprovided measurements of the external circum-ference of that portion of left ventricular musclewhich they surrounded, and they have been em-ployed extensively by Rushmer for such measure-ment of left ventricular circumference.10 Thegages were sewn around the base of the left ven-tricle, the right ventricle being excluded by place-ment of the gage within the cavity of the rightventricle next to the interventricular septum.16

At the completion of each experiment, the gagewas excised and calibrated by recording the de-flection resulting from stretching it to variousknown lengths. The physical characteristics andFrequency response of these gages were recentlydescribed in detail by Lawton and Collins;17 inthe present experiments their use was confined tothe range in which they are linear. Aortic pres-sure, an electrocardiographic lead, left ventricularpressure, left ventricular circumference, and in 6experiments systemic blood flow were recorded ona multichannel oscillograph (fig. 1).

To prevent coagulation an initial dose of 500mg. of Mepesulfate and approximately 250 mg.per hour thereafter were administered. In 6 ex-periments bilateral cervical vagotomy was carriedout. In 24 experiments heart rate was held con-stant or varied by means of a stimulating electrodesewn to the right atrium or right ventricle.

Tn 21 experiments central aortic pressure wasmeasured by means of a catheter introduced viathe femoral artery; in these experiments cardiacoutput was not recorded. Left ventricular end-diastolic pressure and circumference were alteredby rapid infusions of 100 to 200 ml. of whole bloodinto the femoral vein or by bleeding similaramounts from the femoral artery. The blood wasobtained from donor dogs and was thoroughlymixed with that of the experimental animal priorto the experimental period.

In 6 experiments a preparation similar to onedescribed previously was utilized;18 this prepara-tion permitted independent control of cardiac out-put and arterial pressure (fig. 2). The aorta wasdivided distal to the left subclavian artery andwide-bore cannulas were inserted into each seg-ment. The blood flow was then diverted throughan air-filled Starling resistance and a Shipley-Wilson recording rotameter10 into an open reser-voir. The blood delivered into the reservoir wasreturned to the cannulated descending aorta andcommon carotid arteries by means of a roller-pump. The brachioeephalic and left subclavianarteries were then ligated. The recording rotameterthus measured total left ventricular output withthe exception of coronary blood flow. Left ven-tricular output could be varied stepwise by modi-fying the output of the pump, which determined

Circulation Research, Volume VIII, November I960

LEFTVENTRICLE

LEfTVENTRICULAR

CIRCUM

CENTRAL

AORTIC

PRESSUREmm Hg

Figure 1Typical tracing. From above downward: leftventricular pressure, left ventricular circumfer-ence, aortic pressure and electrocardiogram.

the rate of blood flow into the arterial tree andthereby the return of blood to the heart. Arterialpressure was controlled by varying the degree ofinflation of the Starling resistance. Hypothermiaand rewarming of the dog were accomplished byrefrigeration or heating of a water bath surround-ing the reservoir. The blood temperature was mon-itored by means of a thermistor probe insertedthrough a jugular vein into the superior vena cava.

Each point plotted on figures 3 and 5 to 9 rep-resents the average of 6 consecutive values of leftventricular end-diastolic pressure aJid of circum-ference. Bach experimental point represents theconditions which existed when the dog was in asteady or equilibrium state at least 1 minute afterany given hemodynamic intervention. When afteran intervention the left ventricular end-diastolicpressure rose at every level of end-diastolic cir-cumference which was examined, the curve or linerelating end-diastolic pressure to end-diastoliccircumference was shifted toward the pressureaxis. The tension-time index in mm. Hg secondsper miBute was calculated from the area underthe systolic portion of the left ventricular pres-sure curve, as described previously.11

ResultsHeart Bate

The effect of changes in heart rate on therelationship between left A'entricular end-diastolic pressure and end-diastolic circum-ference was examined in 13 dogs. Vagotomyhad been performed in 3 of these dogs, andheart rate was controlled by means of anelectric pacemaker sewn to the right atrium

by guest on May 17, 2018

http://circres.ahajournals.org/D

ownloaded from

LEFT VENTRICULAR DIASTOLIC PRESSURE 1257

L VEND DIASTOLIC

PRESSURE

•

• •

A

A

-

1 1 l _

A

A

AA •

A

A R-R >32sec.• R-R < 32 sec.

-

C7I

Mm

PR

ES

S

o

o

OIA

S-

azLL

>

12 5 130 13 5L V E N D DIASTOLIC CIRCUMFERENCE, CM

Figure 4Results of a typical experiment illustrating therelationship between left ventricular end-diastolicpressure and circumference for individual beatsin an experiment in ivhich atrial fibrillation hadbeen produced. R-R refers to the time intervalbetween consecutive Rl oaves of the electrocardio-gram.

rate described above. This relationship wasthen restudied at an identical heart rate afterthe dog had been cooled to between 28 and30 C. The blood cooling was accomplished byinserting the reservoir (fig. 2) into an icebath. The blood temperature was monitoredby means of a thermistor probe insertedthrough a jugular vein into the superior venacava. It was consistently observed that at thelower temperature the left ventricular end-diastolic pressure was higher for any givenend-diastolic circumference (fig. 5).

In one of these 5 experiments the relation-ship between left ventricular end-diastolicpressure and end-diastolic circumference re-mained constant when the heart rate was per-mitted to slow spontaneously from 130/min.to 80/min. as the temperature was loweredfrom 35.0 C. to 28.4 C. However, when theheart rate was re-elevated to 130/min. at 28.4C, the left ventricular end-diastolic pressurewas higher at any given end-diastolic circum-ference than it had been at 35.0 C.

In a sixth experiment the relationship be-tween left ventricular end-diastolic pressureand end-diastolic circumference remained con-stant when the heart rate was permitted toslow from 177 to 75/min. as the temperaturewas lowered from 34.8 to 28.0 C.

Circulation Research, Volume VIII, November 1960

S.I #

• 34 5 CO 3 0 O°CA 345°C

O 17.0 17.5 18.0

L.V. END DIASTOLIC CIR CUMFERENCE.CM.

Figure 5Results of a typical experiment illustrating therelationshi2) between left ventricular end-diastolicpressure and circumference studied first at atemperature of 34.5 C, then at a temperatureof 30.0 C, and finally again at a temperature of34.5 G.

Acute "Heart Failure"In 6 dogs left ventricular contractility be-

came impaired as the experiment was con-tinued for several hours. This state of im-paired contractility was characterized by pro-gressive dilatation, i.e., an increased left ven-tricular end-diastolic circumference was re-quired in order to maintain any given sys-temic arterial blood pressure. It was ob-served consistently that this form of myo-cardial failure was accompanied by a lowerleft ventricular end-diastolic pressure for anygiven level of end-diastolic circumference. Arepresentative experiment is illustrated infigure 6.

Alterations in Arterial Pressure and Cardiac OutputIn the 6 experiments in which systemic flow

and arterial pressure could be controlled in-dependently, the effect of alterations in ar-terial pressure and of cardiac output on therelationship between left ventricular end-dia-stolic pressure and end-diastolic circumferencewas studied. Cardiac output was first aug-mented at each of various levels of meanaortic pressure, and aortic pressure was thenprogressively elevated at a constant cardiacoutput. The dogs were vagotomized in 3 ofthe 6 experiments, but this maneuver did not

by guest on May 17, 2018

http://circres.ahajournals.org/D

ownloaded from

1258 BRAUNWiVLD, FRYE, ROSS

L V END OiaSTOUC CIRCUMFERENCE.CM.

Figure 6Results of a typical experiment illustrating therelationship between left ventricular end-diastolicpressure and circumference, first at the beginningof the experiment ("control") and after myocardialcontractility declined several hours later ("fail-ure").

affect the results. It was observed in eachexperiment that the relationship between leftventricular end-diastolie pressure and end-diastolic circumference was not modified eitherby changes of aortic pressure or of cardiacoutput, i.e., changes in aortic pressure andcardiac output did not result in displacementof the points relating end-diastolic pressure toend-diasto'.ic circumference. The results of arepresentative experiment are shown in figure7.

In 4 experiments left ventricular end-dia-stolic circumference was held constant whileaortic pressure and cardiac output were variedin opposite directions. End-diastolic pressureremained relatively constant and showed nosystematic change (fig. 8).

Relationship Between Tension-Time Index and LeftVentricular End-Diastolic Circumference

In 3 experiments aortic pressure and cardiacoutput were varied in opposite directions.The magnitude of these changes was suchthat both left ventricular end-diastolic cir-cumference and end-diastolic pressure changedin the same direction as the cardiac output.In each instance the changes in the end-dia-stolie circumference were directionalh' oppo-

15

0

5

n

1—

-

0

/ •

-

1

0/

f

1

• jo

m

1

1

0 • y

i

1

0

1

* //•£

'»mI

S.L

1

-

-

—

-

*4

117.0 17.5 18.0

L.V. END-DIASTOLIC CIRCUMFERENCE, CM.

Figure 7Results of a typical experiment illustrating therelationship between left ventricular end-diastolicpressure and circumference. Open circles representdata obtained as cardiac output tva>s increasedfrom 1,020 to 2,330 ml./min. at a mean aorticpressure of 62 mm. Mg. Triangles represent dataobtained as cardiac output was increased from900 to 2J280 ml./min. at a mean aortic pressureof 93 mm. Hg. Squares represent data obtainedas cardiac output was increased from 560 to 1,700ml./min. at a mean aortic pressure of 121 mm.Hg. Open triangles represent data obtained ascardiac output teas increased from 900 to 1,220ml./min. at a mean aortic pressure of 136 mm.Hg. Closed circles represent data obtained at aconstant cardiac output of 1,300 ml./min. as meanaortic pressure iras increased. Heart rate tensconstant at 146/min. throughout.

site to the changes in the tension-time index.The results of a representative experimentare reproduced in figure 9.

DiscussionThe experiments reported herein were per-

formed in the anesthetized dog with an openchest after some operative interventions. Al-though it is realized that the results of ex-periments performed under these conditionsare not always identical to those obtained in

Circulation Research, Volume VIII, November I960

by guest on May 17, 2018

http://circres.ahajournals.org/D

ownloaded from

LEFT VENTRICULAR DIASTOLIC PRESSURE 1259

EDCCM.

EDP IMM.HG I

'Oh

Figure 8Results of a typical experiment in which meanaortic pressure and stroke volume mere alteredin opposite directions, while end-diastolic circum-ference (EDC) ivas held constant. EBP refersto cnd-dinstolic pressure. Heart rate was constanttit J-Ki/viiii.

the intact organism, the manipulations werenecessary to provide precise control of eachhemodynamic variable.

It was found that the elevation of heartrate above a critical level was associated witha higher left ventricular end-diastolic pres-sure for any given end-diastolic circumfer-ence. These observations on the effects ofchanges in heart rate are in agreement withthe views of Katz,10 as well as with the find-ings of Ullrich, Riecker and Kramer21 on theventricle contracting isometrically. Theseworkers demonstrated a progressive elevationof end-diastolic pressure at any given end-diastolic volume as heart rate was increased.Our findings are also consonant with those ofBuckley, Ogden, Linton and Sidky who dem-onstrated an augmentation of impedance toventricular filling when the heart rate wasincreased.22'23 As has been indicated byMitchell, Linden and Sarnoff,24 the inertia andviscosity of the myocardium during ventricu-lar filling are capable of modifying the pres-sure-length relationships of ventricular fibers.It is clear that tachycardia abbreviates dia-stole significantly,25 and it would seem pos-sible, as suggested by Meek,2" that at veryrapid heart rates ventricular relaxation maynot be complete at the onset of ventricularcontraction; this could result in a change inthe ventricular pressure-volume relationshipwith a greater end-diastolic pressure at anygiven end-diastolic circumference. The ob-

Circulation Research, Volume Vlll, November 1960

20

STROKEVOLUME 10ML.

0

MEAN l 5 °AORTIC , 0 Q

PRESS. l 0 °MM.HG. 50

0

4000

iiSb.-3000

SEC. 2000

18.0

EOC '7.5C M- 17.0

16.5

20

EOPMM. HG.

10

n

, 1

, 1

i 1

'I

II

1 I

1

1 1

1 I

1 1

1 1

- ^ ^ S.L.*5 -

2 CARDIACOUTPUT

I L./MIN.

Figure 9Results of a typical experiment in which meanaortic pressure and stroke volume were alteredin opposite directions. T.T.I, refers to the tension-time index, EDC to end-diastolic circumferenceand EDP to end-diastolic pressure.

servatious of Wiggers,27 that premature ven-tricular contractions which were induced be-fore or during the early ventricular fillingphase are accompanied by an elevated ven-tricular end-diastolic pressure, are also con-sistent with this concept.

The effects of hypothermia on the relation-ship between left ventricular end-diastolicpressure and end-diastolic circumference addfurther support to the hypothesis that incom-plete ventricular relaxation may modify thisrelationship. At any given heart rate, theduration of the contractile state is prolongedand diastole is abbreviated as the heart iscooled.2" It is also well established that theduration of isometric relaxation is significant-ly prolonged under hypothermia.28' 29 In viewof the abbreviation of diastole which takesplace in the presence of tachycardia at a con-stant temperature, as well as in hypothermiaat a constant heart, rate, it was of interest

by guest on May 17, 2018

http://circres.ahajournals.org/D

ownloaded from

1260 BRAUNWALD, FRYE, ROSS

to note that under both of these experimentalconditions the left ventricular end-diastolicpressure increased at any given end-diastoliccircumference. However, the left ventricularend-diastolic pressure-end-diastolic circumfer-ence relationship was not altered in the 2 ex-periments on hypothermia in which the heartrate was permitted to slow and in which theabbreviation of diastole was prevented. How-ever, it is also possible that factors other thanthe mechanical one resulting from changes inthe duration of diastole may modify the leftventricular end-diastolic pressure-end-diastol-ic circumference relationship when changes inheart rate or temperature are induced.

A number of the earlier investigators indi-cated that the resistance to ventricular ejec-tion may modify the ventricular filling pres-sure-fiber length relationship. In addition toPatterson, Piper and Starling,2 Mansfeld andHecht30 also observed that an increase inaortic pressure results in an augmentation ofventricular end-diastolic volume without anychange in end-diastolic pressure. Kiese andGaran found that for any given increment ofdiastolic volume both left and right atrialpressures increased more when aortic pres-sure Avas elevated than when the stroke vol-ume was increased.31 In contrast, Gollwitzer-Meier and Kriiger reported that a smallerelevation of atrial pressure occurred whenany given increment in diastolic volume wasbrought about by elevating aortic pressurethan when stroke volume was increased.32

However, in the present study no alterationof the left ventricular end-diastolic pressure-end-diastolic circumference relationship wasnoted when both aortic pressure and cardiacoutput were varied over wide ranges. Whencardiac output was altered by varying theoutput of the pump, the perfusion pressureto which the carotid sinuses were subjectedwas also changed. Since it has been shownby Mitchell, Linden and Sarnoff that stimu-lation of the vagi and of the cardiac sympa-thetics does not modify the ventricular end-diastolic pressure-length relationship,24 itwould appear unlikely that these alterationsin carotid sinus perfusion pressure modifiedthe results reported herein.

The difference in our results from those ob-tained by others may be related to the factthat the earlier workers measured the com-bined volume of both ventricles, while in-creased resistance to ejection was appliedonly to the left ventricle. The augmented cor-onary flow which this maneuver undoubtedlyproduced may well have resulted in a substan-tial, but unrecognized, increase in the strokeand end-diastolic volumes of both ventricles.In contrast, in the present study the circum-ference of only the left ventricle was meas-ured. It should also be noted that the earlierinvestigators worked with heart-lung prepa-rations rather than with animals with com-plete circulations. The heart-lung preparationalways exhibits varying degrees of heart fail-ure; the application of data derived undersuch conditions to the heart supported by itshumoral and neurogenic stimuli has beenquestioned,0'14 and the alterations in the leftventricular end-diastolic pressure-volume re-lationships which the earlier investigatorsobserved may perhaps be related to progres-sive failure of the heart. The results obtainedfrom the preparation employed for the ex-periments reported herein cannot be con-sidered to be representative of those obtainedfrom an intact, unanesthetized dog. However,the hearts were stable for periods which ex-ceeded 6 hours during which times they werecapable of performing without dilatationlevels of external work similar to thoseachieved by the resting intact dog.

In the present investigation an increasedleft ventricular end-diastolic circumferenceat any given end-diastolic pressure was dem-onstrated in the presence of a depression ofmyocardial contractility which occurred inthe course of a single experiment. Similarconclusions have been derived from experi-ments in which skeletal muscle33 and the frogventricle34' 35 were subject to repeated stress.Kabat and Visscher36 also reported that whenthe tortoise ventricle fails spontaneously orwhen it is asphyxiated, it develops an in-creased fiber length for any given filling pres-sure. Our findings, however, are not consonantwith those of Buckley and Zeig, who observedthat when the dog's left ventricle fails acutely,

Circulation Research, Volume VIII, November lS6O

by guest on May 17, 2018

http://circres.ahajournals.org/D

ownloaded from

LEFT VENTRICULAR DIASTOLIC PRESSURE 1261

its compliance during diastole falls, while itsimpedance to filling is augmented.37 If, asin the acute experiments reported byothers34""30'38 and those presented herein, anincrease in ventricular extensibility occurs inclinical heart failure, the patient could actu-ally benefit therefrom since the altered leftventricular end-diastolic pressure-volume re-lationship would tend to lower the ventricularfilling pressure and, therefore, the atrial pres-sure which would occur at any given ven-tricular diastolic volume. Many of the clini-cal features of heart failure result fromelevated atrial pressures, and any processwhich lowers these pressures would tend toameliorate those manifestations. It has beenshown that chronic heart failure may be as-sociated with changes either in the struc-ture39 or quantity40 of the contractile pro-teins, and it is also of interest that the physi-cal properties of the ventricle are modified byacute depression of myocardial contractility.

In an extension of earlier studies on therelationship between ventricular end-diastolicvolume and the strength of ventricular con-traction,2 Starling and Visscher found thatunder all conditions which they examined,myocardial oxygen consumption (VO2) wasa function of diastolic volume.41 They alsoconfirmed the findings of Anrep42 that whenthe heart was functioning well, its diastolicvolume and therefore its VO2 depended noton arterial pressure or on cardiac output buton the product of these 2 parameters, i.e.,the external work. Thus, Starling and Visscherrelated myocardial VO2 not only to diastolicfiber length, but also to the external workperformed, provided cardiac function was wellmaintained. However, in contrast with thisview a number of investigators31'32> 43> 44 haveshown that a given increase in the work ofthe heart produced by raising aortic pressurewhile maintaining cardiac output constant re-sults in a greater increment of VOo than asimilar increase in work achieved by raisingcardiac output, i.e., the external mechanicalefficiency rises significantly when cardiac out-put is increased but shows little change oreven declines when aortic pressure is aug-mented.

These observations have recently been ex-tended with the demonstration that myocardi-al V~O2 is not determined by the external workof the heart per se but that the primary hemo-dynamic determinant is the area beneath thesystolic pressure curve, i.e., the tension-timeindex.11 In other studiesis it was shown thatVO» is not dependent on ventricular end-diastolic pressure. Indeed, significant changesin V~O2 were observed at a constant left ven-tricular end-diastolic pressure, and it wassuggested that if large changes in the ven-tricular end-diastolic pressure-volume rela-tionship do not occur as a result of alterationsin arterial pressure and flow, the observeddata were not consonant with the view thatmyocardial VO2 is determined by the end-diastolic volume.18 Thus, the finding, that (1)large changes both in arterial pressure andcardiac output do not modify the left ven-tricular end-diastolic pressure-end-diastoliccircumference relationship (fig. 7) and (2)myocardial vO2 is independent of left ven-tricular end-diastolic pressure,18 would notappear to be compatible with classic views.41

Finally, since the tension-time index is de-termined primarily by the arterial pressureand is modified only slightly by the strokevolume,11 while the left ventricular end-dia-stolic circumference is a function both ofarterial pressure and stroke volume, it mightbe predicted that there would be no consistentrelationship between the tension-time-indexand the end-diastolic circumference. This viewis supported by those experiments in whicharterial pressure and cardiac output were ma-nipulated so that the tension-time index andthe end-diastolic circumference actually variedin opposite directions (fig. 9). Thus, whenthese observations are viewed in the light ofthe constant relationship between the tension-time index and myocardial V"O2,

n they alsoappear to be inconsistent with the belief thatmyoeardial VO2 is a function of end-diastolicvolume.

SummaryThe relationship between left ventricular

end-diastolic pressure and circumference wasstudied in 27 open-chest dogs, utilizing a mer-cury-resistance gage to measure left ventricu-

Circvlation Research, Volume VIII, November I960

by guest on May 17, 2018

http://circres.ahajournals.org/D

ownloaded from

1262 BRAUNWALD, FRYE, ROSS

lar end-diastolic circumference. In 6 experi-ments aortic pressure, cardiac output andheart rate were varied independently. Tachy-cardia above a rate critical for each heartelevated left ventricular eud-diastolic pres-sure for any given end-diastolic circumfer-ence. Hypothermia at a constant heart ratehad a similar effect. The altered left ventricu-lar end-diastolic pressure-end-diastolic cir-cumference relationships resulting from tachy-cardia and hypothermia are believed to berelated to the incomplete ventricular relaxa-tion which occurs as the duration of diasto'.eis encroached upon. Acute, spontaneous heartfailure was accompanied by an augmentedleft ventricular eud-diastolic circumferencefor any given end-diastolic pressure, an effectwhich may be considered to reflect an increasein myocardial extensibility. The relationshipbetween left ventricular end-diastolic pres-sure and end-diastolic circumference was notmodified either by changes of aortic pressureor of cardiac output. There was no constantrelationship between the left ventricular end-diastolic circumference and the tension-timeindex. Indeed, it was possible to manipulateaortic pressure and cardiac output so that.these 2 parameters moved in opposite direc-tions. These observations are not consonantwith the view that myocardial oxygen con-sumption is primarily dependent on end-dia-stolic fiber length.

Summario in InterlinguaLe relation inter le tension termino-diastolic sinis-

tro-vcntricula r e le circumferentin esseva studiate in27 canes a thornce aperte, utilisante le calibrator aresistentia de merenrio pro mesurar le circumferentiaterniino-diastolic sinistro-ventrioular. Tn 6 experimen-tos, le tension aortic, le rendimento cardiac, e lefrequcntia del corde esseva variate independentemen-te. Tachycardia in ultra de mi limite critic proonine corde individual elevava le tension tennino-diastolic sinistro-ventricular pro onine circumferentintermino-diastolic particular. Hypothermia, con eon-stantia del frequentia cardiac habeva le niesme effecto.Es opiimte que le .literate relationes inter le tensiontermino-diastolic sinistro-ventricular e le circumfer-entia termino-diastolic que resulta de tachycardia ede hypothermia es relationate al incomplete relaxationventricular occurrente in tanto que le duration deldiastole es compromittite. Acute, spontnnee disfalli-mento cardiac esseva accompaniate de un augmento

del circumferentia termino-diastolic sinistro-ventricu-lar pro omne tension termino-diastolie particular. IIpare justificate considerar iste effecto como reflectenteun augmento del extensibilitate myocardial. Le rela-tion inter le tension termino-diastolic sinistro-ven-tricular e le circumferentia termino-diastolic nonesseva modificate per alterationes del tension aortice non per alterationes del rendimento cardiac. Essevatrovate nulle constante relation inter le circumferentiatermino-diastolic sinistro-vontricular e le indice detension e tempore. De facto, il esseva. possibile ma-nipular le tension aortic e le rendimento cardiac domaniera quo iste duo parametros so moveva in di-rection contrari. Iste constatationes non es in con-gruentia con le these que le consumption myocardialde oxygeno depende primn.riiiii':'.tc del longor termino-diastolic del fibras.

References1. STARLING, E. H.: Law of the Heart Beat. New

York, Longmans, Green & Co., 1918.2. PATTERSON, S. W., PIPER, H., AND STARLING, E.:

Regulation of the heart beat. J. Physiol. 48:465, 1914.

3. KOZATVA, S.: Mechanical regulation of the heartbeat in the tortoise. J. Physiol. 49: 233,1914-35.

4. FRANK, 0.: Zur Dynamik des Hermuskels.Ztsehr. Biol. 32: 370, 1895.

;•>. STRAUU, H.: Dynamik des siiugctierherzens. I.Deutsche Arch. f. Win. Med. 115: 531, 1914.

6. DIPALJIA, J. 1?., AND REISS, R. A.: Myographicstudy of the cat's heart: Effect of changesin venous return and in peripheral resistanceon ventricular contraction. Am. J. Physiol.155: 327, 1948.

7. GESELL, K. A.: Cardiodynamics in heart blockas affected by auricular systole, auricularfibrillation and stimulation of the vagus nerve.Am. J. Physiol. 40: 2G7, 1916.

5. WIGGERS, C. J., AND KATZ, L. X.: Contour ofthe ventricular volume curves under differentconditions. Am. J. Physiol. 58: 439, 1932.

9. —: Present status of cardiodynamic studies onnormal and pathologic hearts. Arch. Int. Med.27: 475, 1921.

10. KATZ, L. N.: Symposium on regulation of per-formance of heart: Analysis of several factorsregulating performance of heart. Physiol. Rev.35: 91, 1955.

11. SARNOFF, S. J., BRAUNWALD, E., WELCH, G. H.,JR., CASE, R. B., STAINSBY, W. N., AND MACRUZ,R.: Heniodynamie determinants of oxygen con-sumption of the heart with special referenceto the tension-time index. Am. J. Physiol. 192:148, 1958.

12. BRAUNWALD, E., FRYE, R. L., AND ROSS, J., JR . :Effect of heart rate on end-diastolic ventricularpressure-volume relationships. Clin. Res. 7:228, 1959.

Circulation Research, Volume Vlll, November I960

by guest on May 17, 2018

http://circres.ahajournals.org/D

ownloaded from

LEFT VENTRICULAR DTASTOLIC PRESSURE 1263

13. —, —, AND —: Determinants of the relationship -9.between left ventricular end-diastolic pressureand circumference. Fed. Proc. 19: 107, 1960.

14. SARNOFF, S. J., CASE, E. B., WELCH, G. H., JR.,

BRATJNWALD, E., AND STAINSBY, W. N.: Per- 30.

formaiico clmracteristics and oxygen debt ina nonfailing, metabolically supported, isolatedheart preparation. Am. J. Physiol. 192: 141,1958. 33.

15. WHITNEY, R. J. : Measurement of volume changesin human limbs. J. Physiol. 121: 1, 1953.

16. RTJSHMER, R. F . : Pressure-circumference relationsof left ventricle. Am. .1. Physiol. 186: 115, 1956. 32.

17. LAAVTON, B. "W., AND COLLINS, C. C.: Calibra-

tion of an aortic circumference gauge. J.Appl. Physiol. 14: 465, 1959.

18. BRAUNWALD, E., SARNOFF, S. J., CASE, R. B., 33,

STAINSHY, W. N., AND WELCH, G. H., J R . :

Hemodynamie determinants of coronary flow:Effect of changes in aortic pressure and car- 34diac output on the relationship between myo-cardial oxygen consumption and coronary 3.5flow. Am. J. Physiol. 192: 157, 1958.

19. SHIPLEY, B. E., AND WILSON, C.: Improved

recording rotametor. Proc. Soc. Exper. Biol.& Mod. 78: 724, 1951.

20. SARNOPF, S. J., AND BERGLTJND, E.: Ventricular 3^

function. I. Starling's law of the heart studiedby means of simultaneous right and left ven-tricular function curves in the dog. Circulation9: 706, 1954. 3 7 -

21. ULLRICH, K. J., RIECKER, G., AND KRAMER, K.:

Das Druckvolumdiagramni des Warmbliiter-herzens. Isomotrische Gleichgewichtskurven. 3gArch. ges. Physiol. 259: 481, 1954.

22. BUCKLEY, N. M., OGDEN, E., AND LINTON, D. S.,

J R . : Effects of work load and heart rate on 39filling of the isolated right ventricle of the dogheart. Circulation Research 3: 434, 1955.

23. —, SIDKY, JI., AND OGDEN, E.: Factors altering

filling of isolated left ventricle of clog heart;effects of epinophrino and norepinephrine. 40Circulation Research 4: 148, 1956.

24. MITCHELL, J. H., LINDEN, R. J., AND SARNOPF,

S. J . : Influence of cardiac sympathetic andvagal nerve stimulation on the relation between 4j .left ventricular diastolic pressure and myo-cardial segment length. Circulation Research8: 1100, 1960. 49.

25. BRAUNWALD, E., SARNOFF, S. J., AND STAINSBY,

W. N.: Determinants of duration and meanrate of ventricular ejection. Circulation Re-search 6: 319, 1958. 43.

2(i. MEEK, W. J.: Question of cardiac tonus. Physiol.Rev. 7: 259, 1927.

27. WIGGERS, C. J. : Muscular reactions of the mam-malian ventricles to artificial surface stimuli. 44.Am. J. Physiol. 73: 346, 1925.

28. BERNE, K. M.: Myocardial function in severehypothermia. Circulation Research 2: 90, 1954.

TRAUTWEIN, W., AND DUDEL, J . : Aktionspo-

tontial und Mochanogramm des Katzenpapil-larmuskels als Funktion der Temperatur.Arch. ges. Physiol. 260: 104, 1954.

MANSFELD, G., AND HECHTJ K.: Untersuchungen

iiber Herzmuskeltonus und die experimentelleDilatation des Warmbliitcrherzens. Arch. ges.Physiol. 232: 643, 1933

KIESE, M., AND GARAN, R. S.: Mechanische

Leistung, Grosse und Sauerstoffverbauch desWarmbliiterhcrzens. Arch, exper. Path. \i.Pharmakol. 188: 226, 1938.

GOLLWITZER-MEIF.R, K., AND KRUGER, E.: Zur

Verschiedenheit der Herzenergetik und Herz-dynamic bei Druck- und Volumleistung. Arch,ges. Physiol. 238: 279, 1937.

SULZER, R.: ITeber das Vorhalten des Skelett-muskels in Ruhe und in Kontraktur bei derDehnung. Ztsch. Biol. 87: 472, 1928.

— : Die mechaniscsen Eigenschaften des Hcrz-muskels. Ztschr. Biol. 92: 545, 1932.

GEHL, H., GRAP, K., AND KRAMER, K.: Das

Druckvolumdiagramm des Kaltbliitorherzcns.Die Bedeutung des plastischen Elementes fiirdie Herzmeehnnik. Arch. ges. Physiol. 261:270, 1955.

KABAT, H., AND VISSCHER, M. B.: The elastic

properties of the beating tortoise ventriclewith particular reference to hysteresis. Am.J. Physiol. 125: 437, 1939.

BUCKLEY, N. M., AND ZEIG, N. J . : Acute uni-

lateral ventricular failure in the isolated dogheart. Am. J. Physiol. 197: 247, 1959.

JOHNSON, V., AND KATZ, L. N.: Tone in the

mammalian ventricle. Am. J. Physiol. 99:579, 1932.

BENSON, E. S., HALLAWAY, B. E., AND TURBAK,

C. E.: Contractile properties of glycerol-extracated muscle bundles from the chron-ically failing canine heart. Circulation Re-search 6: 122, 1958.

KAKO, K., AND BING, R. J. : Contractility of

actomyosin bands prepared from normal andfailing human hearts. J. Clin. Invest. 37:465, 1958.

STARLING, E. H., AND VISSCHER, M. B.: Regu-

lation of the energy output of the heart. J .Physiol. 62: 243, 1926-27.

ANREP, G. V., AND BULATAO, E.: Observations

on the pulmonary circulation: Pulmonarycirculation in the heart-lung preparation. J.Physiol. 60: 175, 1925.

EVANS, C. L., AND MATSUOKA, Y.: Effect of

various mechanical conditions on the gaseousmetabolism and efficiency of the mammalianheart. J. Physiol. 49: 378, 1915.

ALELLA, A., WILLIAMS, F. L., BOLENE-WILLIAMS,

C, AND KATZ, L. N.: Interrelation betweencardiac oxygen consumption and coronaryblood flow. Am. J. Physiol. 183: 570. 1955.

Cirmtlution Jietfeurch, Volume Vlll, November 1960

by guest on May 17, 2018

http://circres.ahajournals.org/D

ownloaded from

EUGENE BRAUNWALD, ROBERT L. FRYE, JOHN ROSS, JR. and Robert LewisVentricular End-Diastolic Pressure and Circumference

Studies on Starling's Law of the Heart: Determinants of the Relationship Between Left

Print ISSN: 0009-7330. Online ISSN: 1524-4571 Copyright © 1960 American Heart Association, Inc. All rights reserved.is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231Circulation Research

doi: 10.1161/01.RES.8.6.12541960;8:1254-1263Circ Res. 

http://circres.ahajournals.org/content/8/6/1254World Wide Web at:

The online version of this article, along with updated information and services, is located on the

  http://circres.ahajournals.org//subscriptions/

is online at: Circulation Research Information about subscribing to Subscriptions: 

http://www.lww.com/reprints Information about reprints can be found online at: Reprints:

  document. Permissions and Rights Question and Answer about this process is available in the

located, click Request Permissions in the middle column of the Web page under Services. Further informationEditorial Office. Once the online version of the published article for which permission is being requested is

can be obtained via RightsLink, a service of the Copyright Clearance Center, not theCirculation Research Requests for permissions to reproduce figures, tables, or portions of articles originally published inPermissions:

by guest on May 17, 2018

http://circres.ahajournals.org/D

ownloaded from