pressure-diameter relations during early diastole...

357

Pressure-Diameter Relations during EarlyDiastole in Dogs

Incompatibility with the Concept of Passive LeftVentricular Filling

HANI N. SABBAH AND PAUL D. STEIN

SUMMARY We studied left ventricular (LV) pressure-diameter relations in 13 open-chest mongreldogs to explore whether ventricular filling is passive, active, or a combination of both. Left ventricularinternal diameter along the minor axis and thickness of the LV free wall were measured with ultrasonicdimension gauges. Pressures in the LV, left atrium (LA), and aorta were measured with catheter-tipmicromanometers. Pressure in the LV during the rapid filling phase of diastole decreased from 13 ± 1mm Hg to 3 ± 0.4 nun Hg (mean ± SE). During this period, LV internal diameter increased from 32.6 ±1.5 to 36.7 ± 1.5 nun, and this represented 49% of the total change of LV diameter that occurred duringdiastole. The pressure-diameter relation that we observed during the rapid phase of filling suggeststhat active enlargement rather than passive distension of the left ventricle occurred. Reduction ofpressure within the left ventricle during rapid filling, while the ventricular diameter increased, appearsto be due to forces within the LV wall that act to restore the LV to its diastolic dimensions. This allowsthe LV to draw blood from the atrium by creating a pressure in the LV that is lower than in the LA. Onthe other hand, during the latter part of diastole, following atrial contraction, the pressure-diameterrelation suggests that the left ventricle undergoes passive distention. Circ Res 45: 357-365, 1981

WHETHER the left ventricle fills passively or ac-tively assists in its filling has been a subject ofinvestigation for many years. In general, it isthought that the ventricle is filled by a head ofpressure in the atrium during early and mid diastoleand by atrial contraction in the later part of dias-tole. In this respect, filling would be considered apassive process. This concept forms the basis forthe assessment of ventricular compliance by dia-stolic pressure-volume relations (Mirsky, 1969; Dia-mond et al, 1971; Mirsky and Parmley, 1973; andMirsky, 1976). Another concept is that of activefilling, in which the ventricle fills itself by drawingblood from the atrial reservoir. This form of ven-tricular filling has been characterized as "diastolicsuction" (Katz, 1930; Brecher, 1956; Bloom andFerris, 1956a; Fowler et al., 1958). If ventriculardiastolic suction exists, then ventricular filling inearly diastole would result from an active processof changing ventricular diastolic dimensions eitherby elastic recoil (Bloom and Ferris, 1956b) or mus-cular contraction (Gausp, 1954). Under certain cir-cumstances, the ventricle can develop a negativepressure during early diastole, and this has beenthe major evidence in favor of non-passive ventric-

From the Department of Medicine (Division of Cardiovascular Med-icine), and Department of Surgery, Henry Ford Hospital, Detroit, Mich-igan.

Supported in part by the U.S. Public Health Service National Heart,Lung, and Blood Institute Grant HL 23699-01.

Address for reprints: Paul D. Stein, M.D., Henry Ford Hospital, 2799West Grand Boulevard, Detroit, Michigan 48202.

Received April 25, 1980; accepted for publication October 1, 1980.

ular filling (Katz, 1930; Brecher, 1956; Bloom andFerris, 1956a; Sabbah et al., 1980). An inability todemonstrate negative intraventricular pressure dur-ing diastole under normal physiological conditionshas caused the concept of active filling to be largelydisregarded.

The purpose of this investigation was to explorethe relation between pressure and diameter duringdiastole in the normal canine left ventricle. Thepressure-diameter relation during diastole mighthelp determine whether filling of the ventricle ispassive, active, or a combination of both.

MethodsSimultaneous measurements of left ventricular,

left atrial, and aortic pressure, as well as measure-ments of left ventricular internal diameter and wallthickness, were made in 13 mongrel dogs. The dogsweighed between 21 and 26 kg, and were anesthe-tized with droperidol, 1.2 mg/kg, and fentanyl cit-rate, 0.24 mg/kg (Innovar-Vet 0.6 ml/kg, iv). Thedepth of the anesthesia was maintained throughoutthe procedure with a continuous intravenous infu-sion of droperidol, 0.7 mg/min; fentanyl citrate,0.01/min; and pentobarbital, 0.17 mg/min. In eachdog, a left thoracotomy was performed and thepericardium was opened.

Left ventricular internal diameter, along its mi-nor axis was measured using ultrasonic crystal di-mension gauges (Schuessler) (Fig. 1). The methodfor insertion of the endocardial ultrasonic crystalswas similar to that described by Horwitz et al.

by guest on May 19, 2018

http://circres.ahajournals.org/D

ownloaded from

http://circres.ahajournals.org/

358 CIRCULATION RESEARCH VOL. 48, No. 3, MARCH 1981

FIGURE 1 Diagrammatic representation of the loca-tion of the ultrasonic crystals used for the measurementof instantaneous changes of left ventricular internaldiameter and wall thickness.

(1968). Two no. 00 control stitches were placed justinferior and perpendicular to the first diagonalbranch of the left anterior descending coronaryartery. The wires of one of the ultrasonic crystalswere threaded through a 13-gauge, 4-inch-longstainless steel needle from which the hub had beenremoved. A stab incision was made through theanterior left ventricular wall between the stitches.The needle, through which the wires of the ultra-sonic gauge were threaded, was inserted into theincision and passed across the ventricle completelythrough the posterior left ventricular wall. Theultrasonic crystal then was pushed through theincision and the wires were pulled until the crystalwas positioned against the endocardial surface ofthe posterior wall. The control stitches, previouslyinserted, then were pulled toward each other toprevent bleeding. The wires of the second crystalwere passed through the lumen of a 4-cm-long rigidplastic tube with an internal diameter of 1.4 mm.The plastic tube was used to push the crystalthrough the incision in the anterior wall. The tubewas then removed. A gentle pull on the wires pos-titioned the ultrasonic crystal against the anteriorendocardial surface of the left ventricle. The crys-tals were anchored in position by purse-string su-tures. We did not find it necessary to occlude theinferior or superior vena cava during the procedureas had been done by others to reduce bleedingduring the procedure (Horwitz et al., 1968). Thecrystals that were used for the measurement of theinternal diameter were hemispheric with a diameterof 5 mm and a maximal thickness of 1.5 mm. Bothcrystals were placed on the endocardial surface withthe convex surfaces of the crystals pointing towardeach other. The location of the crystals on theendocardial surface was confirmed at autopsy.

Simultaneous changes of thickness of the leftventricular anterior wall were also measured with

a pair of ultrasonic crystals (Fig. 1). One crystal wassutured to the epicardial surface; the other wasinserted as near as possible to the endocardial sur-face without perforating into the left ventricularcavity. The subendocardial crystal was introducedat an angle of 45° relative to the epicardial surfaceto avoid injury of the myocardium between the twocrystals. In implanting the deep crystal, a tunnelwas created with an 18-gauge needle. The epicardialcrystal was sutured in direct alignment with thesubendocardial crystal. The epicardial crystal was1.5 mm thick and 4 mm in diameter; whereas thedeep crystal was 1 mm thick and 2 mm in diameter.When in place, the convex surfaces of both crystalspointed toward each other. The location of thesubendocardial crystal was confirmed at autopsy.

Aortic, left ventricular, and left atrial pressureswere measured with catheter-tip micromanometers(Millar Instruments). The micromanometer used tomeasure left atrial pressure was introduced directlyinto the left atrium through an incision in the leftatrial appendage. Bleeding was controlled by tyingthe tip of the atrial appendage around the catheter.Atrial fibrillation was not induced in any of thedogs. The hematocrits in these dogs, after completeinstrumentation, ranged from 35 to 46 ml/100 ml.

The pressure sensors had a frequency responsethat was flat within ±2% at 5 kHz and within ±5%at 10 kHz. The phase lag of this type of sensor is90° at 35 kHz, which is equivalent to a time delayof approximately 7 j^sec. Tests in our laboratoryshowed a baseline drift of less than 1 mm Hg in 3hours. The zero baseline of each transducer wasadjusted to atmospheric pressure at the beginningof each procedure. All transducers were made equi-sensitive. Because of baseline shifts that may occurwith temperature differences between the body androom air, the level of zero pressure was reestab-lished after withdrawal of the catheter-tip micro-manometers to room air. This level of zero was usedin all cases as the true zero level. The equisensitivityof the transducers was checked at the end of eachstudy.

Pressures, wall thickness, and ventricular diam-eter were recorded on a VR-12 photographic re-corder (Electronics for Medicine) at paper speedsof 25 to 250 mm/sec. The frequency response of therecorder was flat to 2500 Hz. Pressures, however,were filtered with a 250 Hz low pass filter. Thefrequency response of the ultrasonic dimensiongauge system was flat from 0 to 60 Hz.

Diastole was defined as the period from thatpoint at which left ventricular pressure fell belowleft atrial pressure to the point at which left ven-tricular pressure increased above left atrial pres-sure. The phase of rapid filling was defined as theportion of diastole during which the diameter in-creased rapidly. Diastasis was defined as the periodof slowly increasing or unchanging diameter duringthe mid-portion of diastole. Atrial contraction was



ownloaded from


VENTRICULAR FILUNG/Sabbah and Stein 359

defined as the portion following the onset of the leftatrial A wave.

Calculations of ventricular diameter and wallthickness were made when left ventricular pressurefell below left atrial pressure (the point of cross-over), at the time of minimal left ventricular pres-sure, at the onset of the left atrial A wave, and atthe end of diastole when left ventricular pressurecrossed over left atrial pressure. The difference ofpressure between the left ventricle and left atriumduring diastole was calculated and plotted using anelectronic digitizer (Numonics) on line with a 21MX computer (Hewlett-Packard). The pressure-di-ameter curves of single beats were constructed fromdata obtained throughout the entire diastolic pe-riod. Left ventricular pressure and left ventricularinternal diameter along the minor axis were digi-tized at 1-mm intervals at paper speeds of 250 mm/sec, which corresponded to intervals of 4 msec.

Instantaneous distensibility of the left ventriclein early diastole was calculated as the ratio of theinstantaneous value of left ventricular diameter toinstantaneous left ventricular pressure. The ratio ofdiameter to pressure was plotted as a function oftime at intervals of 4 msec, starting at the time ofcrossover of left ventricular and left atrial pressure,and ending at the time of minimal left ventricularpressure.

ResultsThe crossover of left ventricular and left atrial

pressure occurred at a pressure of 13 ± 1 mm Hg(mean ± SE). Left ventricular pressure continued todiminish and reached a minimal value of 3 ± 0.4mm Hg an average of 66 msec (range, 50-90 msec)after the point of crossover (Fig. 2). This timeperiod represented 29% of diastole (Figs. 2, 3). Pastthe point of crossover, left atrial pressure decreasedmore slowly than left ventricular pressure (Figs. 3,4). The pressure difference developed because left

ventricular pressure diminished more rapidly thanleft atrial pressure.

The internal diameter of the left ventricle in-creased as pressure within the left ventricle dimin-ished (Figs. 3, 4). The left ventricular internal di-ameter, measured at the moment of left ventricularand left atrial pressure crossover, was 32.6 ± 1.5mm. It increased to 36.7 ± 1.5 mm at the time ofminimal left ventricular diastolic pressure. This rep-resented an average of 49% (range, 24-82%) of thechange of diameter that occurred during diastole.

Left ventricular wall thickness, measured at themoment of pressure crossover, was 13.0 ± 0.5 mm.The wall thinned to 11.4 ± 0.5 mm at the momentof minimal left ventricular pressure (Figs. 2-4). Thisreduction of wall thickness represented an averageof 54% (range, 24-89%) of the thinning that occurredduring diastole. The left ventricular wall becamethinner, therefore, as pressure within the left ven-tricle diminished.

During the rapid filling phase of diastole, leftventricular pressure-diameter relations showed adiminishing pressure in the left ventricle associatedwith a prominently increasing ventricular diameter(Fig. 5).

Left ventricular pressure began to increaseshortly after it reached a minimal value, and itincreased gradually or was nearly constant duringdiastasis. During diastasis, left atrial pressure wasnearly constant or diminished (Figs. 3, 4).

Atrial contraction resulted in an increase of bothleft atrial and left ventricular pressure. At this time,left atrial pressure increased more than left ventric-ular pressure. The pressure difference between thechambers increased. During atrial contraction, leftventricular diameter increased from 38.8 ± 1.4 to41.2 ± 1.4 mm. This accounted for 28% of the totalchange of diameter during diastole. Wall thicknessdiminished from 10.5 ± 0.5 to 10.0 ± 0.5 mm duringatrial contraction, and this accounted for 17% of the

LV WALLTHICKNESS

(MM) 10

34LV

DIAMETER(MM) 31

100

PRESSURE(MM HG)

FIGURE 2 Typicalrecording of left ven-tricular (L V) freewall thickness, LVinternal diameter,and pressure in theaorta (Ao), LV, leftatrium (LA), andlead II of the electro-cardiogram (ECG).Portions of the leftventricular pressuretracing were re-touched for clarity ofreproduction.



ownloaded from



RAPID LVFILLING DIASTASIS

ATRIALCONTRAC-1 TION

4 3

LVDIAMETER

(MM)

35

20

PRESSURE(MM HG)

I.05

I. 10 . 1 5

1.20 .25

I.30

LV WALLTHICKNESS

(MM)

FIGURE 3 Computer output of left ventricular(L V) internal diameter, wall thickness, and L Vand left atrial (LA) pressures during diastoleof a single beat of the same dog shown in Figure1. The broken line represents the differencebetween LA and LV pressure (LA-LV). Theheart rate was HI beats/min.

TIME (SEC)

r 17

LV WALL

THICKNESS(MM)

0 .04 .08 .12 .16 .20 .24

TIME (SEC)

FIGURE 4 Computer output of pressures, in-ternal diameter, and wall thickness duringdiastole in a single beat of a dog with a heartrate of 145 beats/min. Abbreviations are as inFigure 2.

total reduction of wall thickness (Figs. 3, 4). Duringatrial contraction, therefore, the further increase ofdiameter and thinning of the wall of the left ventri-cle were accompanied by an increased pressurewithin the left ventricle.

Distensibility of the left ventricle calculated asthe ratio of left ventricular internal diameter to leftventricular intracavitary pressure increased in earlydiastole in all the dogs (Fig. 6). During this period,starting at the time of crossover of left ventricular



ownloaded from


VENTRICULAR FILLING/Sa66a/i and Stein 361

oX

CO

30

20

10

1 1 1 1 1

LV-LA PRESSURE/ CROSSOVER

\\

\

1 1 1 1 1

r 1 1 1

LV-LA PRESSURECROSSOVER \

1 1 1 1

FIGURE 5 Pressure-diameter relationconstructed during diastole for a singlebeat in the same dog as shown in Figure1. The curve starts at the point of leftventricular and left atrial (LV-LA) pres-sure crossover and ends at the LV-LApressure crossover at the end of diastole.

37 39 4 1 43

LEFT VENTRICULAR DIAMETER (MM)

45

and left atrial pressure and ending at the time ofminimal left ventricular pressure, the change ofdistensibility with time was relatively constant in 8dogs, and it showed an increasing change with timein five dogs (Fig. 6).

DiscussionThe relation between left ventricular pressure

and diameter that we observed during the rapidphase of ventricular filling suggests that left ven-tricular filling may not be a passive process. Duringearly diastole, we observed a diminishing left ven-tricular pressure as left ventricular diameter in-creased. Such a behavior is compatible with selfenlargement of the left ventricle, possibly secondaryto an active process of myocardial relaxation.

A diminishing left ventricular pressure associatedwith an increasing ventricular volume also conceiv-ably could occur if the distensibility of the leftventricular wall, during early diastole, markedlyincreased during that time. This would require alower stress (ventricular pressure) to achieve a pas-sive enlargement of the ventricle. To explore thispossibility we examined the distensibility of the leftventricle during early diastole. The minor axis ofthe left ventricle has been shown to relate directlyto left ventricular volume (Rankin et al., 1980). Ifone defines instantaneous distensibility of the ven-tricle as the ratio of ventricular diameter (andtherefore ventricular volume) to ventricular pres-sure, then

= D-P (1)

where V is instantaneous ventricular volume (ml),P is instantaneous ventricular pressure (mm Hg),and D is instantaneous distensibility (ml/mm Hg).The use of instantaneous pressure-volume relationsis compatible with the work of Suga and Sagawa(1974). If all three parameters are time dependent,then by differentiating Equation 1, one would ob-tain the following relation:

AV = AP-D +AD-P (2)

where AV is the change of volume (ml) for a given

period of time; AP is the change of pressure (mmHg) during this time and AD is the change ofdistensibility (ml/mm Hg) during this time.

It could happen that AD is sufficiently large topermit AP to be negative, yet result in a positiveincrement of volume. However, when the data fromour dogs were substituted in Equation 2, starting atthe time of the left ventricular and left atrial pres-sure crossover and ending at the time of minimalleft ventricular pressure, AP • D + AD • P was alwaysnegative (Fig. 6). This suggests that a reduction ofpressure would be accompanied by reduction ofvolume if there were passive distention. However,during the rapid filling phase, ventricular diameterincreased while pressure within the ventricle de-creased. Therefore, ventricular enlargement doesnot appear to result from passive filling. A reductionof left ventricular pressure during filling suggeststhat the ventricle causes its diameter to increaseand therefore contributes to its filling. Elastic recoilor muscular contraction within the left ventricularwall during diastole may produce such restoringforces and explain these observations. Such a pro-cess could produce thinning of the ventricular wallduring diastole. Prominent thinning of the free wallof the left ventricle was observed as the ventricleincreased in diameter during rapid filling.

Although there was a pressure difference be-tween the left atrium and left ventricle during rapidfilling, it should be considered in perspective withthe self-filling action of the left ventricle. The pres-sure difference occurred due to a more rapid fall ofleft ventricular pressure than left atrial pressure.This can occur only as a result of the self-enlargingaction of the left ventricle. Blood flowed into theleft ventricle because of the pressure difference.However, the pressure difference was caused by thereduction of left ventricular pressure below leftatrial pressure.

A large number of studies of ventricular compli-ance based on pressure-volume relations duringdiastole indicate passive filling of the left ventricle.How are these studies compatible with our obser-vations? The answer seems to relate to the portion



ownloaded from



39 r-

38

37

< 36

35

340 4 8 12 16 20 24 28 32 36 40

TIME ( msec)

o>

crCOCOLU

£

20i-

16

12

j

8 12 16 20 24 28 32 36 40

TIME (msec)

o _v- o>cr pLU pi- ^LU g

< .§Q

LU

* »

a.Icr

2 0 r

16

12

t ,

0 4 8 12 16 20 24 28 32 36 40

TIME (msec)

FIGURE 6 Top: Left ventricular (LV) internal diameteris shown as it changes with time during early diastolein a single beat. t\ is the time of crossover of left ventric-ular and left atrial pressure, and t2 is the time of minimalleft ventricular pressure. Middle: Left ventricular (L V)

of diastole included in such pressure-volume rela-tions. The portion of diastole that suggests an activeparticipation of the left ventricle in the filling pro-cess is excluded from pressure-volume studies thatrelate to ventricular compliance (Noble et al., 1969;Gaasch et al., 1975; Rankin et al., 1977). Construc-tion of the pressure-volume curve is usually initi-ated at the time of minimal left ventricular pressure(Noble et al., 1969; Gaasch et al., 1975; Rankin etal., 1977). Exclusion of the portion of diastole duringwhich left ventricular pressure falls as filling occursis based upon the presumption that the left ventri-cle is not fully relaxed during this time (Rankin etal., 1980). The increment of mural force above thestatic relation during dynamic filling was inter-preted to be a manifestation of parallel viscousproperties of the myocardium (Rankin et al., 1980).

Our observations do not contradict the impres-sion that filling of the ventricle during diastasis andatrial contraction may be passive. Changes of pres-sure and diameter of the left ventricle during dias-tasis and, particularly, during atrial contraction,were compatible with the dynamics of filling of apassive body. During atrial contraction, both leftventricular pressure and diameter increased andwere associated with a reduction of left ventricularwall thickness. The difference of pressure betweenthe left atrium and left ventricle during these pe-riods was caused by an elevation of left atrial pres-sure above left ventricular pressure.

Analyses of the pressure-volume or pressure-di-ameter relations of individual beats during diastolehave shown, as we have shown, that filling of theventricle occurs as pressure diminishes during thephase of rapid filling (Dodge et al., 1962; Horwitzand Bishop, 1972; Kennish et al., 1975; Rankin etal., 1977; Ling et al., 1979; Grossman and Barry,1980). The interpretation of the events during thisperiod has been variable. Some have suggested thatthere is a residual influence of systolic relaxation ondiastolic filling (Weiss et al., 1976). This wouldcause significant increments in mural force above

pressure is shown as it changes with time during thesame cardiac cycle and during the same period of timeas above. Bottom: Distensibility is shown as it varieswith time during early diastole. Distensibility was cal-culated as the ratio of diameter (top panel) to pressure(middlepanel). From these graphs, data can be obtainedto substitute in the equation AV = AP-D + AD-P. Fromthe bottom panel, the change of distensibility over thetime t\ to t% AD =11 mm/mm Hg, and the averagedistensibility, D = 8 mm/mm Hg. From the middle panel,the change of left ventricular pressure, AP = —9 mm Hgand the average LV pressure, P = 6 mm Hg. Therefore,by substitution in the equation it is clear that the changeof volume, AV, would be negative. Assuming passivedistention, therefore, for a given reduction of pressure areduction of volume would be predicted. This did notoccur.



ownloaded from


VENTRICULAR FILLING/Sa66a/i and Stein 363

the static relationship present during dynamic fill-ing (Rankin et al., 1980). Some have calculated anegative stiffness during the early phase of ventric-ular filling (Kennish et al., 1975). Inertia of thecardiohemic system was excluded as a likely mech-anism for this apparent negative stiffness (Kennishet al., 1975). It was felt that the negative stiffnesswas a manifestation of incomplete relaxation (Ken-nish et al., 1975). The observations were thought tobe explainable on the basis of a model containingan element whose deformation is rate dependent,that is, a parallel viscous element.

A continuing active decay of stress within the leftventricle may be an important factor underlyingthe continuing decline of pressure within the leftventricular wall during rapid filling (Kennish et al.,1975; Weiss et al., 1976). The likelihood of this issupported by the observation that pressure withinthe wall of the left ventricle continues to fall wellafter the onset of diastole (Stein et al., 1980). Toexplain the increasing left ventricular volume thataccompanies the diminishing left ventricular pres-sure, one must postulate the existence of a time-varying reduction of stiffness that is sufficient toallow distention of the left ventricle by a progres-sively diminishing pressure.

The observed variation of the difference betweenleft atrial and left ventricular pressure during rapidfilling also is incompatible with the concept of pas-sive filling during this period. If passive filling werepresent, pressure would not diminish in the leftventricle at a greater rate than in the left atrium.

The issue of whether the ventricle is filled pas-sively or whether it draws blood from the atriumhas rested primarily on whether ventricular dia-stolic suction can be demonstrated under normalphysiological conditions. In 1956, Brecher recordednegative diastolic pressures, -1.3 to —10.8 cm H2O,in the left ventricle of dogs when the mitral orificewas temporarily occluded. Others subsequentlyshowed similar results either following obstructionof mitral inflow (Bloom and Ferris, 1956a; Fowleret al., 1958; Tyberg et al., 1970), or following rapidbleeding (Fowler et al., 1958). An autopsy report ofa man who died after a gunshot wound of the rightventricle seems to reflect this mechanism (Robertset al., 1979). The left ventricle following hemorrhageappeared to create a sucking action sufficientlystrong to cause the left atrial appendage to invagin-ate through the mitral orifice into the left ventricle(Roberts et al., 1979).

Observations of the capability of the mammalianleft ventricle to create a negative diastolic pressureusually have been limited to nonphysiological con-ditions, such as inflow occlusion, severe hemor-rhage, or stimulation by isoproterenol (Brecher,1956; Bloom and Ferris, 1956a; Fowler et al., 1958).Whether diastolic suction can occur under physio-logical conditions has been unsettled. These obser-vations, however, did not reflect the effects of neg-ative intrathoracic pressure, because the chest was

open in the dogs in these studies. Several investi-gators have suggested that a negative diastolic pres-sure occurs in the left ventricle only when the end-systolic volume is small (Bloom, 1956; Fowler et al.,1958; Brecher and Kissen, 1967; Roberts et al.,1979). We have demonstrated, however, that nega-tive left ventricular pressure can occur in patientswith mitral stenosis in the presence of a normalend-systolic volume (Sabbah et al., 1980). The neg-ative diastolic pressure in these patients appears tobe related to the maintenance of a vigorous con-traction in the presence of mitral stenosis. Theventricle appears to have exerted a sucking actionto draw blood through the reduced mitral orifice.

Any force that lowers pressure in a region towardwhich flow occurs can be termed suction, whetheror not the pressure developed in that region dropsbelow atmospheric zero (Brecher and Galletti,1966). The term "suction" has been defined as areduction of pressure at some point in a system bythe application of a force which results from anenergy conversion process, such as muscular con-traction or elastic recoil (Brecher and Galletti,1966). The definition of suction in these terms wasreputed by Rushmer et al. (1953), who stated thatthe term must be used only when a negative pres-sure is present. Our observations of pressure-diam-eter relations during rapid filling suggest that theventricle draws blood from the atrium. This oc-curred in the absence of a negative left ventricularpressure. The observation is consistent with suc-tion, as defined by Brecher and Galletti (1966).

Tyberg et al. (1970) suggested that the compli-ance of the atrium allows large changes of volumewithout changes of pressure. Consequently, onewould be less likely to observe negative ventriculardiastolic pressure in the normal situation than in aninflow limiting disease (Tyberg et al., 1970). Theyadded that the absence of negative diastolic pres-sure does not necessarily imply that the ventricle isnot contributing to the filling process, especially ifthe function of the inflow channel is normal. Theyalso suggested that, in the presence of normal ven-tricular volume and afterload (aortic pressure), re-storing forces in diastole are probably very small,but may still be evident.

One postulated form of active process by whichthe ventricle may draw blood from the atrium is thecontraction of muscle fibers (Gausp, 1954). It wassuggested that, owing to their anatomic arrange-ment, such contraction of the muscle fibers couldenlarge the ventricular cavity during diastole(Gausp, 1954). Another suggested active process isthe development of a force during diastole that actsto lengthen the muscle fibers upon completion oftheir contraction (Brecher and Galletti, 1966). An-other process that may permit the ventricle to drawblood from the atrium is elastic recoil (Bloom andFerris, 1956b; Brecher and Kissen, 1967). Such anelastic recoil would tend to restore ventricular dia-stolic dimensions. Perhaps myocardial fibers of dif-



ownloaded from



ferent layers pull against each other and applystretch to the connections between them (Rushmeret al., 1953). This action may represent potentialenergy which is wasted as far as systolic ejection isconcerned. However, as the ventricles begin to re-lax, this potential energy would be released andcould act to return the ventricular chambers totheir original diastolic dimensions. Under these con-ditions, the inflowing blood would encounter a smallresistance to distention. Therefore, in early diastole,rapid filling would be facilitated (Rushmer et al.,1953). The restoring force also may result fromdistortion of the intramural structure of the walland in the heart may derive in part from compres-sion of the ultrastructural unit of contraction, thesarcomere (Tyberg et al., 1970). If individual musclefibers are made to contract, their sarcomeres willshorten and with relaxation will elongate them-selves (Parsons and Porter, 1966). With relaxation,sarcomeres will spring back to the longer length,demonstrating a restoring force. In this sense, theshortening sarcomere overcomes a resistance of un-known magnitude and retains this energy as elasticrecoil on relaxation (Tyberg et al., 1970).

A change of diameter of the left ventricle oc-curred in some of the dogs before the crossover ofleft atrial and left ventricular pressure (Fig. 1). Thismay have related to the role of the papillary musclein opening the mitral valve (Marzilli et al., 1980).Opening of the mitral valve has been observedbefore the crossover of left atrial and left ventricularpressure (Tsakaris et al., 1978). Action of the pap-illary muscle may explain why opening of the mitralvalve is present at that time (Marzilli et al., 1980).On the other hand, it is also possible that changesof shape of the left ventricle may occur duringisovolumic relaxation. It has been shown, neverthe-less, that the minor axis of the left ventricle relatesdirectly to left ventricular volume (Rankin et al.,1980).

In conclusion, left ventricular filling during earlydiastole is accompanied by pressure-diameterchanges that are incompatible with passive fillingof the ventricle. The decreasing left ventricularpressure during rapid ventricular filling appears toreflect an active restoration of left ventricular di-mensions. The nature of such restorative forces isundetermined.

ReferencesBloom WL (1956) Diastolic filling of the beating excised heart.

Am J Physiol 187: 143-144Bloom WL, Ferris EB (1956a) Negative ventricular diastolic

pressure in beating heart studied in vitro and in vivo. ProcSoc Exp Biol Med 93: 451-454

Bloom WL, Ferris EB (1956b) Elastic recoil of the heart as afactor in diastolic filling. Trans Assoc Am Physicians 69: 200-206

Brecher GA (1956) Experimental evidence of ventricular dia-stolic suction. Circ Res 4: 513-518

Brecher GA, Galletti PM (1966) Functional anatomy of cardiacpumping. In Handbook of Physiology. Circulation, vol 2, ed-

ited by WF Hamilton, P Dow. Washington, D.C., AmericanPhysiological Society, p 759-798

Brecher GA, Kissen AT (1967) Relation of negative intraven-tricular pressure to ventricular volume. Circ Res 5: 157-162

Diamond G, Forrester JS, Hargis J, Parmley WW, Danzig R,Swan HJC (1971) Diastolic pressure-volume relationship inthe canine left ventricle. Circ Res 29: 267-275

Dodge HT, Hay RE, Sandier H (1962) Pressure-volume char-acteristics of the diastolic left ventricle of man with heartdisease. Am Heart J 64: 503-511

Fowler NO, Couves C, Bewick J (1958) Effect of inflow obstruc-tion and rapid bleeding on ventricular diastolic pressure. JThorac Surg 35: 532-537

Gaasch WH, Cole JS, Quinones MA, Alexander JK (1975) Dy-namic determinants of left ventricular diastolic pressure-vol-ume relations in man. Circulation 51: 317-323

Gausp FT (1954) El ciclo cardiaco, consideraciones criticas sobrela interpretacion clasica y nuevas ideas sobre el mismo. Pri-vately published monograph by Gausp from the Medical Fac-ulty of the University of Salamanca, Madrid, quoted byBrecher GA, (4th reference, this list)

Grossman W, Barry WH (1980) Diastolic pressure-volume rela-tions in the diseased heart. Fed Proc 39: 148-155

Horwitz LD, Bishop VS (1972) Left ventricular pressure-dimen-sion relationships in the conscious dog. Cardiovasc Res 6: 163-171

Horwitz LD, Bishop VS, Stone HL, Stegall HF (1968) Continu-ous measurement of internal left ventricular diameter. J ApplPhysiol 24: 738-740

Katz LN (1930) The role played by the ventricular relaxationprocess in filling the ventricle. Am J Physiol 95: 542-553

Kennish A, Yellin E, Frater RW (1975) Dynamic stiffness profilesin the left ventricle. J Appl Physiol 39: 665-671

Ling D, Rankin JS, Edwards CH II, McHale PA, Anderson RW(1979) Regional diastolic mechanics of the left ventricle in theconscious dog. Am J Physiol 235: H323-H330

Marzilli M, Sabbah HN, Lee T, Stein PD (1980) Role of thepapillary muscle in opening and closure of the mitral valve.Am J Physiol 238: H348-H354

Mirsky I (1969) Left ventricular stresses in the intact humanheart. Biophys J 9: 189-208

Mirsky I (1976) Assessment of passive elastic stiffness of cardiacmuscle: mathematical concepts, physiologic and clinical con-siderations, directions of future research. Prog Cardiovasc Dis18: 277-308

Mirsky I, Parmley WW (1973) Assessment of passive elasticstiffness for isolated heart muscle and the intact heart. CircRes 33: 233-243

Noble MIM, Milne ENC, Goerke RJ, Carlsson E, DomenechRJ, Saunders KB, Hoffman JIE (1969) Left ventricular fillingand diastolic pressure-volume relations in the conscious dog.Circ Res 24: 269-283

Parsons C, Porter KR (1966) Muscle relaxation: Evidence for anintrafibrillar restoring force in vertebrate striated muscle. Sci-ence 153: 426-427

Rankin JS, Arentzen CE, McHale PA, Ling D, Anderson RW(1977) Viscoelastic properties of the diastolic left ventricle inthe conscious dog. Circ Res 41: 37-45

Rankin JS, Arentzen CE, Ring WS, Edwards CH II, McHalePA, Anderson RW (1980) The diastolic mechanical propertiesof the intact left ventricle. Fed Proc 39: 141-147

Roberts WC, Brownlee WJ, Jones AA, Luke JL (1979) Suckingaction of the left ventricle. Demonstration of a physiologicprinciple by a gunshot wound penetrating only the right sideof the heart. Am J Cardiol 43: 1234-1237

Rushmer RF, Crystal DK, Wagner C (1953) The functionalanatomy of ventricular contraction. Circ Res 1: 162-170

Sabbah HN, Anbe DT, Stein PD (1980) Negative intraventric-ular diastolic pressure in patients with mitral stenosis: Evi-dence of left ventricular diastolic suction. Am J Cardiol 45:562-566

Stein PD, Marzilli M, Sabbah HN, Lee T (1980) Systolic anddiastolic pressure gradients within the left ventricular wall.Am J Physiol 238: H625-H630



ownloaded from


AORTIC COARCTATION AND BARORECEPTOR RESETTING/Igler et al. 365

Suga H, Sagawa K (1974) Instantaneous pressure-volume rela-tionships and their ratio in the excised, supported canine leftventricle. Circ Res 35: 117-126

Tsakiris AG, Gordon DA, Padiyar R, Frechette D (1978) Rela-tion of mitral valve opening and closure to left atrial andventricular pressures in the intact dog. Am J Physiol 234:

H146-H151Tyberg JV, Keon WJ, Sonnenblick EH, Urschel CW (1970)

Mechanics of ventricular diastole. Cardiovasc Res 4: 423-428Weiss JL, Frederiksen JW, Weisfeldt ML (1976) Hemodynamic

determinants of the time-course of fall in canine left ventric-ular pressure. J Clin Invest 58: 751-760

Coarctation of the Aorta and BaroreceptorResetting

A Study of Carotid Baroreceptor Stimulus-ResponseCharacteristics before and after Surgical Repair in the

Dog

FRANZ O. IGLER, LAWRENCE E. BOERBOOM, PAUL H. WERNER, JUDITH H. DONEGAN,

E.J. ZUPERKU, LAWRENCE I. BONCHEK, AND J.P. KAMPINE

SUMMARY We studied baroreceptor function in dogs before and after surgical repair of coarctationof the aorta by direct recording of multifiber carotid sinus (CS) nerve activity (NA) during alteration ofpulsatile arterial pressure with systemic phenylephrine and nitroprusside, and during static pressurechanges using a CS pouch preparation. Coarctation was induced by banding the proximal thoracicaorta in ten 3- to 5-day old puppies. One and one-half years later, five of these coarctated animals werestudied before, and five were studied 3-7 months after, surgical repair. Five adult animals also werestudied 4-6 months after the proximal thoracic aorta had been banded. Controls were eight normaladult dogs. Threshold pressure at which NA began, saturation pressure at which NA reached amaximum, and slope (% Max NA/mm Hg) of the linear portion of the stimulus-response curve weredetermined. Pulsatile manipulations of pressure elicited normal sensitivity (slope) in dogs with coarc-tation but static nonpulsatile pressure changes showed depressed sensitivity compared to controls.After surgical repair, threshold and saturation returned toward normal; sensitivity determined withstatic pressure manipulations returned to control value. Coarctation reset CS baroreceptors to operateat higher pressures in both puppies and adult dogs and repair of coarctation returned function towardnormal. We conclude that resolution of hypertension after repair of coarctation may depend uponbaroreceptor readaptation. Circ Res 48: 365-371, 1981

SEALY et al. (1957) suggested that the initial risein systolic pressure following repair of aortic coarc-tation in children (paradoxical hypertension) maybe due to resetting of systemic arterial barorecep-tors to the chronically elevated pressure. Acutelyunloading baroreceptor afferents by surgical repairwould lead to an acute increase in sympatheticoutflow and thus an increase in systemic arterialpressure. Rocchini et al. (1976) found that paradox-

From the Departments of Anesthesiology and Cardiothoracic Surgery,The Medical College of Wisconsin, and the Wood Veterans Administra-tion Hospital, Milwaukee, Wisconsin.

This study was supported by National Institutes of Health GrantHL21071, National Institutes of Health Young Investigator Award HL19556, and the Medical Research Service of the Veterans Administration.

Address for reprints: Franz O. Igler, M.D., Ph.D., Research Service(151), Veterans Administration Hospital, Wood (Milwaukee), Wisconsin53193.

Received February 1, 1980; accepted for publication October 1, 1980.

ical hypertension after surgical repair of coarctationin children was biphasic in nature. An initial rise insystolic pressure within the first 24 hours afterrepair was related to increased sympathetic activ-ity. The subsequent gradual rise in diastolic pres-sure was related to increased plasma renin activity.Upward resetting of the baroreflex heart rate re-sponse in the awake, adult canine with aortic coarc-tation from birth was demonstrated recently byBonchek et al. (1976). The stimulus-response curverelating systemic arterial pressure to heart rate wasshifted to the right with no detectable change insensitivity (slope). Resetting of carotid and aorticbaroreceptors to chronically elevated pressure inrenal hypertension (Aars, 1976; McCubbin, 1958;Salgado and Krieger, 1973), and in the sponta-neously hypertensive rat (Nosaku and Wang, 1972;Sapru and Wang, 1976) has been demonstrated.



ownloaded from


H N Sabbah and P D Steinof passive left ventricular filling.

Pressure-diameter relations during early diastole in dogs. Incompatibility with the concept

Print ISSN: 0009-7330. Online ISSN: 1524-4571 Copyright © 1981 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.48.3.3571981;48:357-365Circ Res.

http://circres.ahajournals.org/content/48/3/357World 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:



ownloaded from

http://circres.ahajournals.org/content/48/3/357

http://www.ahajournals.org/site/rights/

http://www.lww.com/reprints

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


pressure-diameter relations during early diastole...

Documents