the transmitral pressure-flow velocity relationcirc.ahajournals.org/content/78/6/1459.full.pdf ·...

1459

The TransmitralPressure-Flow Velocity Relation

Effect of Abrupt Preload Reduction

Michael Courtois, MA, Zvi Vered, MD, Benico Barzilai, MD, Nancy A. Ricciotti, MSN,Julio E. Perez, MD, and Philip A. Ludbrook, MB, BS, FRACP

Although recent animal and clinical studies suggest that Doppler-derived indexes may be usefulfor the characterization of ventricular diastolic behavior, the hemodynamic basis for the preloaddependency of these indexes has not previously been fully elucidated. Accordingly, effects ofreduction of left atrial load on the pressure-flow velocity relation were characterized in 10anesthetized, closed-chest dogs during transient inferior vena caval occlusion by means ofsimultaneously recorded left atrial and left ventricular micromanometric pressure measurementand transesophageal Doppler echocardiograms. Within four or five beats after inferior vena cavalballoon occlusion, left atrial loading was reduced as evidenced by a decrease in the slope of the leftatrial v wave from 21±4 to 13±4 mm Hg/sec (p<0.001) and by a decrease in the first crossoverpoint of left atrial and left ventricular pressures from 5.6± 1.1 to 2.9±+1.5 mm Hg (p<0.001). Thisdecrease in left atrial loading resulted in reductions during early diastole of minimum leftventricular pressure (from 1.0±0.8 to -0.4±+1.2 mm Hg, p<0.001), the maximum early forward(i.e., left atrial pressure>left ventricular pressure) transmitral pressure gradient (from 2.8±0.8to 2.4±0.5 mm Hg, p<0.01); the slope of the rapid filling pressure wave (from 44±11 to 38±10mm Hg/sec, p<0.025); and the area of the reversed (i.e., left ventricular pressure>left atrialpressure) transmitral pressure gradient (from 79±42 to 53+33 mm Hg. msec, p<0.05). Duringlate diastole, both the heights and slopes of the left atrial and left ventricular a waves fell, resultingin a decrease in the maximum late transmitral pressure gradient (from 1.2±0.7 to 0.9+0.5 mmHg, p<0.05). Vena caval occlusion also altered Doppler transmitral velocity profiles during boththe early and late phases of diastole. Peak velocity of the E wave decreased (from 50±11 to 41±7cm/sec, p<0.01) as did acceleration (from 880±222 to 757±+258 cm/sec2, p<0.025) and deceler-ation (from 597±260 to 429±t197 cm/sec2, p<0.025). Peak velocity of the A wave also fell (from29+9 to 22±5 cm/sec, p<0.005). Abrupt inferior vena caval occlusion did not significantly changeheart rate or mean aortic pressure. Significant correlations were obtained between peak flowvelocity during early diastole and the peak forward pressure gradient during early diastole(r=0.67, p<0.05), between peak flow velocity during late diastole and the peak pressure gradientduring late diastole (r=0.80, p<0.01), and between deceleration rate of the Doppler E wave andthe magnitude of the reversed transmitral pressure gradient (r=0.60, p<0.01). These resultsindicate that an acute reduction in left atrial loading conditions significantly alters the diastolictransmitral pressure relation and thus profoundly affects the Doppler flow velocity profile.(Circulation 1988;78:1459-1468)

From the Cardiovascular Division, Washington University ulsed Doppler-derived indexes of left ven-School of Medicine, St. Louis, Missouri. tricular diastolic function have been used

Presented in part at the 60th Scientific Sessions of the American 1 to characterize diastolic behavior in van-Heart Association, November 17, 1987, Anaheim, Califonia. oniosSupported in part by National Heart, Lung, and Blood Insti- ous cardiac diseases and age-related conditions.1-5tute Grant ROI-HL-25430 and by SCOR in Ischemic Heart However, recent evidence has indicated that theDisease Grant HL-17646, National Institutes of Health, Bethesda, Doppler time-velocity profile is affected by reduc-Maryland. tions in left ventricular preload.6-9 Because Dopp-Address for correspondence: Michael Courtois, MA, Washing- e e i a w

ton University School of Medicine, Cardiovascular Division, 660a clncalol, it isc al thatwthe alteptanceS. Euclid, Box 8086, St. Louis, MO 63110. as a clinical tool, it iS crucial that the alterations in

Received February 1, 1988; revision accepted June 23, 1988. cardiac hemodynamics that underlie these preload-

by guest on May 7, 2018

http://circ.ahajournals.org/D

ownloaded from

http://circ.ahajournals.org/

1460 Circulation Vol 78, No 6, December 1988

ECG

4-

8r

6

to

EE2-

FIGURE 1. Record ofmeasurements designed to characterize the transmitral pressure relation. Measurements includedslope ofthe left atrial v wave,first transmitral pressure crossover point (XI), minimum left ventricular pressure, maximumearly forward pressure gradient where left atrial pressure exceeds left ventricular pressure, the slope of the leftventricular rapidfilling pressure wave, the area ofthe reversed gradient between the second (X2) and third (X3) points oftransmitral pressure crossover where left ventricular pressure exceeds left atrial pressure, slopes and heights of both theleft atrial and left ventricular a waves, and maximum pressure gradient during atrial contraction.

related alterations in the Doppler velocity profilebe fully elucidated and understood. Accordingly,this study was designed to define the effects of asudden reduction in preload on the transmitralpressure-flow velocity relation.

Materials and MethodsMongrel dogs weighing 27-41 kg were sedated

with morphine sulfate (1 mg/kg s.c.) 30 minutesbefore induction of general anesthesia with sodiumpentothal (12.5 mg/kg i.v.) and a-chloralose (70mg/kg i.v.). Each dog was intubated and ventilatedwith room air with a Harvard respirator. The rightjugular vein, left common carotid artery, right fem-oral artery and vein, and left femoral vein wereisolated and a valved sheath (Hemoquet 8F, USCI,Billerica, Massachusetts) placed in each. A trans-esophageal two-dimensional phased-array echocar-diographic 5-MHz transducer with pulsed Dopplercapabilities (Model 77020A Ultrasound System,Hewlett-Packard, Andover, Massachusetts) waspositioned so that the interatrial septum was c learlyvisualized. A Mullins transseptal catheter intro-ducer set (8F, USCI) and Brockenbrough needle (18gauge, USCI) were directed under fluoroscopy intothe right atrium, and aided by two-dimensionalechocardiographic targeting, the entire assemblywas pressed against the interatrial septum andadvanced through it into the left atrium. Aftersuccessful puncture was verified by injection andvisualization of contrast in the left atrium and left

ventricle with two-dimensional echocardiography,the dilator and needle were removed and a bolus ofheparin sodium (4,000 USP units) administered.Under fluoroscopy, micromanometric-angiographiccatheters (8F; Model 484A, Millar Instruments,Houston, Texas) were directed into both the leftatrium via the transseptal sheath and the left ven-tricle retrogradely across the aortic valve via theleft carotid artery. A Swan-Ganz thermodilutioncatheter (Model 93-131-7F, American Edwards,Santa Ana, California) was directed from the jugu-lar vein to the superior vena cava; a fluid-filledpigtail catheter (7F, Cordis) was directed from thefemoral artery to the arch of the aorta, and anatrioseptostomy balloon catheter (Model 83-051-5F, American Edwards) was directed from theleft femoral vein to the inferior vena cava. Beforeintroduction into the animal, the micromanometerswere calibrated relative to atmospheric pressure.Before each experimental recording, the microma-nometric signal obtained from the left atrial catheterwas aligned with the pressure measured by itsfluid-filled lumen connected to a transducer posi-tioned at the midthoracic level. Left ventricularmicromanometric pressure was then aligned withleft atrial micromanometric pressure during longdiastatic periods associated with slow heart rates orcompensatory pauses after premature ventricularcontractions (Figure 1). A low-gain signal (100 mmHg= 12.5 cm) from the aortic fluid-filled catheterand a high-gain signal (20 mm Hg= 10 cm) from the

o0-



ownloaded from


Courtois et al Preload Reduction During Diastolic Filling

left atrial micromanometric catheter were transmit-ted to a photographic recorder (visicorder Model1508B, Honeywell, Denver, Colorado), and a high-gain signal from the left ventricular micromanomet-ric catheter was transmitted to both the photo-graphic recorder and a heat-sensitive recorder(Model 77500B, Hewlett-Packard) interfaced withthe ultrasound imaging system. At the start of eachexperimental recording, a square wave was deliv-ered to both recorders via the left ventricular trans-ducer control unit (Model TCB 100, Millar Instru-ments, Houston, Texas), facilitating precisealignment of the transmitral pressures and the Dopp-ler time-velocity profiles.Because regional pressure gradients have been

shown to exist in the left ventricle during both theearly and the late phases of diastole,'0 the positionof the micromanometer in the left ventricle wasstandardized at 4 cm from the apex. This wasaccomplished by advancing the left ventricularmicromanometric-angiographic catheter under fluo-roscopy to the cardiac apex. Once contact with theapex was made, the catheter was pulled back 2.5cm. Because the measured distance of the micro-manometer from the leading edge of the pigtail was1.5 cm, this maneuver positioned the transducerapproximately 4 cm from the ventricular apex.Because previous experience in our laboratory with

. . I . * . . I . . . . I . . *

.ECG

l ll l

Or-,

0

E(U

dogs of the size used in this study indicates that themajor chord is in the range of 7-8 cm, this positioncorresponds to a midventricular level.

All hemodynamic and echocardiographic measure-ments were made with the animal in the supineposition. Transmitral time-velocity profiles wereobtained by adjusting the transesophageal probe toobtain a modified four-chamber view, with thesample volume positioned at the level of the mitralanulus.'1 Electrocardiographic activity was moni-tored via a precordial (V4) electrode. Core temper-ature was monitored throughout the study andmaintained with a circulating water (38° C) heatingpad. At no time did any of the animals' coretemperature fall below 36.60 C.12 Arterial bloodgases were measured periodically, and respiratoryrate and volume adjusted accordingly.

Analysis of DataAll pressure and time-velocity measurements were

made during the end-expiratory portion of the res-piratory cycle within four or five beats after ballooninflation, before any increase in heart rate. Hemo-dynamic measurements included (Figure 1) the slopeof the left atrial v wave, the point of first crossoverof the left atrial and ventricular pressures (XI),minimum left ventricular pressure, the maximumearly forward (left atrial pressure>left ventricular

. * **

FIGURE 2. Record of measurementsLVP designed to characterize the Doppler

time-velocity profile. Recordings wereobtainedfrom a transesophageal view.In this position, the flow of blood dur-ing diastolic filling is away from thetransducer; hence, the time-velocitywaveforms are negative. Measure-

> ments included acceleration and decel-eration rate of the early (E) diastolicfilling wave and peak velocity of boththe early and late (A) diastolic fillingwaves.

20 -

j.

1461

40i



ownloaded from



=_X7i~~~~~~~~~~~~X \.mmuvi1.-a

LAP

*- 4LVP _. M

. .i.

FIGURE 3. Top panel: Record of transmitral pres-sures (mm Hg) at baseline, before inferior venacaval occlusion. Bottom panel: Record of trans-mitral pressures (mm Hg) during inferior venacaval occlusion. Within four or five beats afterballoon occlusion of the inferior vena cava, fillingofthe left atrium is greatly reduced as evidenced bya decrease in the slope and height of the left atrialv wave. This decrease in left atrial loading leads toa reduction in minimum left ventricular pressure, adecrease in the early forward maximum pressuregradient during early diastolic filling, and a reduc-tion in the slope of the left ventricular rapid fillingpressure wave and subsequent overshoot of leftatrial pressures in the form of a reversed pressuregradient. Decreases also occur in both the heightsand slopes of the left atrial and left ventricularpressure a waves and in the maximum pressuregradient generated during atrial contraction.

A0 LAP0~~~~~~v^*

pressure) transmitral pressure gradient, the slope ofthe left ventricular rapid filling pressure wave, thearea of the reversed (left ventricular pressure>leftatrial pressure) transmitral pressure gradient circum-scribed by the atrial and ventricular pressure signalsbetween the second (X2) and third (X3) points ofcrossover, the heights and slopes of both the leftventricular and left atrial a waves, and the maxi-mum pressure gradient during atrial contraction. Tofacilitate precise measurement of these gradients

and areas, hemodynamic recordings were magnifiedwith a high-quality copier.

Transesophageal Doppler time-velocity profilemeasurements included (Figure 2) acceleration anddeceleration rate of the early (E) diastolic fillingwave and peak velocity of both the early and late(A) diastolic filling waves.

Five to eight occlusions of the inferior vena cavawere peformed on each animal. Subsequent analy-ses of the pressure and Doppler waveforms were

4

0

A

1

LVP....1

- 1. / A----1 if 1-. 1 j &.

lo. .00

11+



ownloaded from



TABLE 1. Changes in Hemodynamics During Inferior Vena CavalOcclusion

Baseline IVCO(n=10) (n=10) p

HR (beats/min) 84±16 85±18 NSAoM (mm Hg) 99±14 96±15 NSXI (mm Hg) 5.6±1.1 2.9±1.5 0.001v wavesiope (mm Hg/sec) 21±4 13±4 0.001LVPmin (mm Hg) 1.0±0.8 -0.4±1.2 0.001ETPmax (mm Hg) 2.8±0.8 2.4±0.5 0.008RFWslope (mm Hg/sec) 44±11 38±10 0.023ARP (mm Hg* msec) 79±42 53±33 0.033LA a wave,],p, (mm Hg/see) 60±24 37±14 0.002LA a wavemax (mm Hg) 6.6±1.1 4.3±1.2 0.001LV a wave,,op, (mm Hg/sec) 27±12 18±8 0.001LV a wavemax (mm Hg) 6.2±1.3 3.5±1.6 0.001LTPmax (mm Hg) 1.2±0.7 0.9±0.5 0.049LVEDP (mm Hg) 6.5±1.0 3.8±1.5 0.001

IVCO, inferior vena caval occlusion; HR, heart rate; AoM,mean aortic pressure; X1, first crossover of transmitral pres-sures; v wave.1p0, upslope of the left atrial v wave; LVPmjn,minimum left ventricular pressure; ETPm,x, maximum earlydiastolic transmitral pressure gradient; RFW,10pe, slope of therapid filling pressure wave; ARP, area of reversed pressuregradient where left ventricular pressure exceeds left atrial pres-sure; LA, left atrial; a wave,10p,, upslope of the a wave; awavemax, peak of a wave; LV, left ventricular; LTPmax, maximumlate diastolic transmitral pressure gradient; LVEDP, left ventric-ular end-diastolic pressure.

carried out on recordings that were free of prema-ture ventricular contractions, demonstrated well-defined Doppler profiles, showed minimal (<5%)changes in heart rate and aortic pressure, andexhibited an end-expiratory respiratory phase in thebeats immediately after inferior vena caval occlu-sion. Data from recordings meeting these criteriawere averaged; this was generally two recordingsfor each animal.

Differences between baseline and interventionconditions were detected with the paired t test. Alldata are reported as mean±SD.

ResultsEarly Diastolic Transmitral Pressures andDoppler Flow Velocity During Inferior VenaCaval Occlusion

Within four to five beats after balloon occlusionof the inferior vena cava, pressure in the left atriumwas greatly reduced (Figure 3) as evidenced by aprominent reduction in the slope and height of theleft atrial v wave. This reduction in left atrialloading was associated with a decline in minimumleft ventricular pressure, a decrease in the earlymaximum forward pressure gradient during earlyfilling, and a reduction in the slope of the leftventricular rapid filling pressure wave and subse-quent overshoot of left atrial pressures in the formof a reversed pressure gradient. Group data derivedfrom transmitral pressures for the baseline and

occlusion conditions are presented in Table 1. Allmeasured parameters except heart rate and meanaortic pressure changed significantly after vena cavalocclusion. These changes in early transmitral pres-sures greatly affected the configuration of the earlyDoppler time-velocity profile (Figure 4), manifestedby significant reductions in acceleration, maximumvelocity, and deceleration. Group data derived fromthe Doppler E wave is presented in Table 2. Indi-vidual data points for the maximum early pressuregradient and peak E wave velocity are presented inFigures 5 and 6, respectively.

Late Diastolic Transmitral Pressures andDoppler Flow Velocity During Inferior VenaCaval Occlusion

After reduction of left atrial loading, significantchanges in transmitral pressure and flow velocityalso occurred during the late phase of diastolic fillingassociated with atrial contraction. Significantdecreases were noted in the heights and slopes ofboth the left atrial and left ventricular pressure awaves, resulting in a decline in the maximum pres-sure gradient generated during atrial contraction(Figure 3). This reduction in the pressure gradientacross the mitral valve during late diastole wasassociated with a reduction in the maximum velocityof the Doppler A wave (Figure 4). Group dataderived from left atrial and left ventricular pressuresand the Doppler time-velocity profiles during atrialcontraction are presented in Tables 1 and 2, respec-tively. Significant changes were observed in all var-iables after vena caval occlusion. Individual datapoints for the maximum pressure gradient during latediastole and peak velocity of the Doppler A wave arepresented in Figures 7 and 8, respectively.

Correlation Between Doppler FlowVelocity Measurements and TransmitralPressure GradientsTo normalize for differences in mitral valve resis-

tance, the ratios of the paired data (baseline dividedby occlusion) were used.13 With this technique,peak velocity of the Doppler E wave and themaximum forward pressure gradient during earlydiastolic filling exhibited a moderate correlation(r=0.67, p<0.05) (Figure 9), whereas peak velocityof the Doppler A wave and the maximum pressuregradient during atrial contraction correlated some-what more strongly (r=0.80, p<0.01) (Figure 10).In addition, a significant, albeit weak, correlationwas found between deceleration of the early fillingwave and the magnitude of the reversed transmitralpressure gradient (r=0.60, p<0.01) (Figure 11).

DiscussionThe present study confirms recent reports that

reductions in left ventricular preload significantlyalter the Doppler transmitral inflow time-velocityprofile during early diastolic filling. We also noted asignificant change in the late diastolic filling wave.

1463



ownloaded from



8

EEI

ECG

(EQ)

8

EO0

C;E:)

F- 0.4sec, I a 11 11 a a ..a A a ...a t |L

FIGURE 4. Left panel: Record ofthe Doppler time-velocity profile at baseline before inferior vena caval occlusion. Rightpanel: Record of the Doppler time-velocity profile during balloon occlusion of the inferior vena cava. Compared withbaseline, decreases are noted in the acceleration and deceleration rates of the E wave and in the heights of both the Eand A Doppler time-velocity waveforms.

More important, the present study describes indetail the underlying transmitral pressure changesthat accompany and are responsible for these alter-ations in flow velocity.

Other investigators have reported that a decreasein preload is accompanied by significant changes inthe early filling wave, but only small, generallystatistically nonsignificant changes in the late fillingwave. Choong et a16 effected a reduction in preloadwith nitroglycerin infusion in humans undergoingcardiac catheterization and noted reductions in accel-eration, deceleration, and maximum velocity of theearly filling wave but no change in maximum Awave velocity. Using lower body negative andpositive pressure to alter preload by altering venousreturn in humans with coronary artery disease, Taka-hashi et a18 showed the maximum velocity of theearly diastolic filling wave to be very sensitive tochanges in left atrial pressure, whereas significantdifferences in A wave maximum velocity were seenonly at the extreme of applied negative pressure.Bhatia et a19 decreased left ventricular preload innormal humans with nitroglycerin and reported asignificant difference in maximum E wave velocitybut no change in A wave velocity.The present study confirms that the early dia-

stolic Doppler flow velocity waveform is substan-tially altered by changes in left ventricular preload.Significant changes were found in acceleration, decel-eration, and peak velocity of flow as measured from

the Doppler E wave. However, contrary to theresults of other studies, we also found a significantdifference in the maximum velocity of the late fillingwave associated with atrial contraction. This dis-crepancy may be due in part to modest increases inheart rate that accompanied preload reduction inthese other studies. Choong et a16 reported nochange in peak A wave velocity and noted a statis-tically significant increase in heart rate (from 60 to65 beats/min). Likewise, Bhatia et a19 reported asignificant increase in heart rate (from 65 to 73beats/min) but noted no changes in the Doppler Awave. Increases in heart rate are well known toaffect the late phase of diastolic filling. 14 Two recentpreliminary reports of the effects of heart rate ontransmitral Doppler profiles concluded that latemitral inflow patterns are highly sensitive to heartrate changes. In humans, Parker et al'5 noted a

TABLE 2. Changes in Doppler Time-Velocity Profile DuringInferior Vena Caval Occlusion

Baseline IVCO(n= 10) (n=10) p

E acceleration (cm/sec2) 880+222 757+258 0.011Maximum E velocity (cm/sec) 50+11 41+7 0.002E deceleration (cm/sec2) 597±260 429+197 0.024Maximum A velocity (cm/sec) 29±9 22±5 0.005

IVCO, inferior vena caval occlusion; E, early Doppler time-velocity filling wave; A, late Doppler time-velocity filling wave.

2 . . . 5 . . . . 5 . . . .a. a . .. a . . . . a 1 _I-



ownloaded from


Courtois et al Preload Reduction During Diastolic Filling 1465

-0-

.-W-0--D--a-

-U-

2345678910

IVC OCCLUSION

3-

2-

1 -

m

-

z-

4

FM2

2

BASELINE IVC OCCLUSION

FIGURE 5. Plot of individual data points for 10 dogs atbaseline and during inferior vena caval occlusion for thepeak pressure gradient recorded during the early dia-stolic filling phase.

uniform rise in peak A wave velocity with increas-ing heart rate, with an excellent correlation(r=-0.97) between heart rate and the ratio ofmaximum E to maximum A wave velocity. Like-wise, Gillam et al'6 measured mitral inflow profileswith Doppler in patients with programmable pace-makers and reported that changes in heart rategreatly influenced late inflow patterns. An increase inheart rate from 60 to 70 beats/min was associatedwith a statistically significant 10% increase in peakvelocity of the A wave. Thus, small increases inheart rate may negate any decrease in A wave peakvelocity that might otherwise accompany a decreasein preload. In the present study in which hemody-namics and Doppler data were recorded within fouror five beats of the onset of vena caval occlusion, noincrease in heart rate was noted (from 84 to 85 beats/min). Hence, the results of the preceding reportsmay represent changes associated with preload reduc-tion in a clinical setting where a number of uncon-trollable factors, including heart rate, may be expectedto vary, whereas the results of the present study mayrepresent more nearly pure preload reduction.A second possible explanation for the discrep-

ancy between our report and those of others mightbe the degree of preload reduction produced byinferior vena caval occlusion compared to that, for

70 -

60 -

50 -

40 -

30 -

FIGURE 7. Plot of individual data points for 10 dogs atbaseline and during inferior vena caval occlusion for thepeak pressure gradient recorded during the late diastolicfilling phase.

example, of nitroglycerin. However, Choong et al,6who found only a small, statistically nonsignificantdecrease in peak A wave velocity (from 61 to 59 cm/sec), reported a 58% decrease in the peak of thepulmonary wedge v wave (from 12 to 5 mm Hg)with nitroglycerin, which would seem to comparereasonably with the XI data reported in this study (a48% decrease from 5.6 to 2.9 mm Hg). Still, the ideathat changes in the A wave are related to the degreeof preload reduction is reinforced by the report ofTakahashi et al,8 who noted a statistically signifi-cant 17% decrease in peak A wave velocity at theextreme of applied negative lower-body pressure(-40 mm Hg) but found statistically nonsignificantchanges at higher pressures.A third possible explanation for this discrepancy

may be found on close examination of the individualdata points presented in Figure 8. As can be seen, thegreatest changes in A wave peak velocity are foundin animals exhibiting the largest A waves at baseline.Mean change for the four animals with the largest Awaves at baseline is -35% (from 38 to 25 cm/sec;peak E: peak A ratio at baseline, 1.20). Mean changefor the remaining six animals is - 17% (from 24 to 20cm/sec; peak E: peak A ratio at baseline, 2.32). Thus,the substantial (-25%) decrease in the A wavereported in this study may have been influenced by a

50 -

-0-14- 2= 3.4- 4-4 5-0- 6-fr 7-h- 8- 9.-- 10


FIGURE 6. Plot of individual data points for 10 dogs atbaseline and during inferior vena caval occlusion forpeak E wave velocity.

40i)9 40

c) 30 -

0

_U

ct20

10-

-G-

.4--0--U--a--A--6-

.-m-4-


FIGURE 8. Plot of individual data points for 10 dogs atbaseline and during inferior vena caval occlusion forpeak A wave velocity.

5 -

4-

3-

2-

E-

zW

0C_

cc

-J9a2

.2 1 l4 *

p -p.- v

BASELINE

-- 1-*- 2-- 3-0- 4-U 5-0- 6-kr 7- 8_- 9

+ 10

0)E

0-J5

0

2345678910

0 a

20 1. 1 a

.---10



ownloaded from



1.5 -

X:

Ui-2<LUX

1.4 -

y = 0.19 + 0.89x r = 0.67 pc.05

1.3-

1.1-

1.0E Y 1.D1RA DEARLY GRAD PRE / EARLY GRAD OCCL

FIGURE 9. Plot of relation of the early peak forwardtransmitral pressure gradient to peak Doppler E wave

velocity after normalizing for differences in mitral valveresistance.

heightened sensitivity of the late diastolic fillingphase to abrupt preload reduction in animals withdepressed E:A ratios. However, the control meanE:A ratio for the patients in the Choong et a16 studywas 0.97, suggesting the presence of abnormal dia-stolic function in their sample group.

Forward Transmitral Pressure Gradients DuringInferior Vena Caval Occlusion

During balloon occlusion of the inferior venacava, left atrial pressure was greatly reduced, asevidenced by a significant decrease in left atrial vwave upslope and a reduction of the first transmitralpressure crossover point XI (pressure at mitralvalve opening). In response to this reduction in leftatrial load, minimum left ventricular pressure dur-ing early diastolic filling fell immediately, maintain-ing the early transmitral pressure gradient. Thisdrop in minimum left ventricular pressure withmaintenance of the transmitral gradient probablyreflects a shift in the mechanism of early diastolicfilling to primarily one of ventricular suction. 17 Thisreduction ofminimum pressure was, however, insuf-ficient to compensate for the loss in left atrialpressure, and the overall effect was a reduction inthe gradient across the mitral valve.

Thus, the present study demonstrates that in thepresence of reduced preload, the left side of the

y = 0.81 + 0.38x r = 0.80 p<.01

0

0

2<

0.

Z.

0 1 2 3LATE GRAD PRE / LATE GRAD OCCL

FIGURE 10. Plot of relation of the late forward peakpressure gradient to peak Doppler A wave velocity afternormalizing for differences in mitral valve resistance.

W

cm

z0F

0-0

1200 -

800

600 -

400 -

200 -

y = 268.82 + 3.72x r = 0.60 p<.01

0 50 1 00 1 50AREA OF REVERSED PRESSURE (mmHg-msec)

FIGURE 11. Plot of relation between the area of thereversed transmitral pressure gradient and decelerationrate of the Doppler E wave.

heart is less capable of accelerating blood as mea-sured by Doppler echocardiography. The reductionin acceleration and maximum flow velocity duringearly diastole appears to be directly due to thereduction in the pressure gradient across the mitralvalve. This loss of pressure gradient due to thesudden reduction in left atrial loading cannot becompensated for by elastic recoil alone despite apossible reduction in end-systolic volume that isknown to accompany reductions in preload'8 andleads to augmented suction.'9 In this study, themaximum transmitral pressure gradient during earlydiastole was found to correlate significantly with thepeak velocity of the Doppler E wave. This is inkeeping with a recent report by Ishida et al,13 whodescribed a significant linear correlation betweenearly peak transmitral flow measured by electromag-netic flow probe and the peak transmitral pressuredifference during early diastolic filling.The results of the present study are consistent

with a model of the heart that views early diastolicfilling as resulting from two processes: elastic poten-tial energy developed in the left atrial wall duringthe previous right ventricular contraction20 and elas-tic potential energy stored in the left ventricularwall during the previous left ventricular con-traction.21 If, as in this study, atrial loading isreduced dramatically, forces developed in the atriumthat normally contribute to diastolic filling duringearly diastole are attenuated'7 and subsequent rapidventricular filling may not be completed as effec-tively. Although the right side of the heart was notevaluated in this experiment, we assume that dia-stolic filling of the right ventricle might be similarlyinfluenced by abrupt preload reduction.During late diastolic filling, a significant decrease

is seen in the height and upslope of the left atrial awave during inferior vena caval occlusion. Thisprobably relates to the biphasic nature of pulmo-nary blood flow and is the result of reducedrefilling of the atrium after the ventricular rapidfilling phase, resulting in unloading of the atriumbefore atrial contraction.22 Thus, the velocity ofthe blood subsequently injected into the ventricleduring left atrial contraction is reduced. This isalso reflected in a reduction in the upslope and

0



ownloaded from



height of the left ventricular a wave as less impactis produced in the ventricle. As expected, a sig-nificant correlation was also found between themaximum pressure gradient and Doppler-derivedpeak flow velocity during this phase of diastole.

Deceleration of Early Flow Velocity and theReversed Transmitral Pressure Gradient

Because blood accelerated during early diastolereturns to near zero velocity before systolic ejec-tion, a force comparable in magnitude and oppositein direction to the accelerative force should bepresent during deceleration of flow.23 The down-slope of the early time-velocity transmitral fillingwave has previously been shown to be temporallyrelated to the occurrence of a reversed transmitralpressure gradient.241 0 The weak, though significant,correlation between E wave deceleration and degreeof transmitral pressure reversal obtained in thepresent study also supports the idea that thesephenomena are related. The low correlation is mostprobably attributable to variation in factors notspecifically measured in this study, such as relativeintraventricular catheter position10 and ventricularcompliance. 13The mechanism for this observed reduction in the

downslope of the E wave in association with pre-load reduction is unknown, but it may be speculatedthat it might be related to changes in the process ofelastic recoil. During normal early diastolic filling,the ventricular wall, augmented by elastic potentialenergy stored in and then released by the left atrialwall, presumably recoils promptly to some "fequi-librium" conformation.25 Consequently, blood decel-erating by collision with the ventricular wall will beimpacting with an essentially stationary structureand will therefore decelerate rapidly. If left atrialloading is substantially reduced, forces ordinarilyprovided by normal atrial filling are reduced, andthe rate of ventricular recoil might tend to beattenuated. Thus, blood decelerating by impactionagainst the ventricular wall will collide with a struc-ture that is still in motion, moving in the samedirection as the flow of blood. Hence, the impact ofcollision is reduced, and deceleration, as evidencedby the downslope of the Doppler E wave, is lessabrupt. Clinical pathological examples of theextremes of this scenario are represented byrestrictive-constrictive physiology and mitral ste-nosis. In the case of restrictive ventricular physiol-ogy, wall expansion is halted abruptly, and short-ened deceleration times for the E wave are notedacross both the mitral and tricuspid valves.26 At theother extreme, mitral stenosis represents a situationin which wall expansion occurs over a prolongedperiod because of restricted mitral inflow, and Ewave deceleration times are markedly prolonged.27If it is conceded that the upstroke of the left ven-tricular rapid filling pressure wave results from thecollision of blood with the ventricular wall duringearly diastolic filing,24,28 the decrease in the slope of

the rapid filling pressure wave and extent of subse-quent transmitral pressure reversal after preloadreduction may also be due in part to this phenom-enon and in part to the reduction in peak velocityreached by the blood during the rapid filling phase.Another possible explanation for the slower E

wave deceleration rate recorded after preload reduc-tion relates to factors that contribute to the func-tional compliance of the left ventricle. Because reduc-tions in left ventricular volume accompany reductionsin preload, the ventricle may be operating on a lowerportion of the pressure-volume curve,29 and it isconceivable that the functionally more compliantventricle may be less effective at rapidly deceleratingblood. In addition, substantial decreases in rightventricular volume that accompany vena caval occlu-sion may also contribute to shifting the left ventricleto a lower portion of the pressure-volume curve ordisplacing the entire curve downward.30,31

Implications and ConclusionsThe present study demonstrates that in this prep-

aration, a sudden reduction of left atrial pressure byinferior vena caval occlusion significantly alters thetransmitral pressure relation and, thus, profoundlyaffects the transmitral Doppler flow velocity profileduring both the early and late phases of diastole.Hence, experimental or clinical studies that useparameters derived from Doppler time-velocity plotsto characterize left ventricular diastolic functionmust rigorously control or make allowances for leftatrial loading conditions.

AcknowledgmentsWe thank Burton E. Sobel, MD, and Sandor J.

Kovacs Jr., PhD, MD, for their critical review ofthe manuscript and Donna Marquart, RT, for hertechnical assistance.

References1. Takenaka K, Dabestani A, Gardin JM, Russell D, Clark S,

Allfie A, Henry WL: Pulsed Doppler echocardiographicstudy of left ventricular filling in dilated cardiomyopathy.Am J Cardiol 1986;58:143-147

2. Takenaka K, Dabestani A, Gardin JM, Russell D, Clark S,Ailfie A, Henry WL: Left ventricular filling in hypertrophiccardiomyopathy: A pulsed Doppler echocardiographic study.J Am Coll Cardiol 1986;7:1263-1271

3. Fujii J, Yoshizumi Y, Sawada H, Aizawa T, Watanabe H,Kato K: Noninvasive assessment of left and right ventricularfilling in myocardial infarction with a two-dimensional Dopp-ler echocardiographic method. J Am Coll Cardiol 1985;5:1 155-1 160

4. Snider AR, Gidding SS, Rocchini AP, Rosenthal A, Dick MII, Crowley DC, Peters J: Doppler evaluation of left ventric-ular diastolic filling in children with systemic hypertension.Am J Cardiol 1985;56:921-926

5. Miyatake K, Okamoto M, Kinoshita N, Owa M, Nakasone1, Sakakibara H, Nimura Y: Augmentation of atrial contri-bution to left ventricular inflow with aging as assessed byintracardiac Doppler flowmetry. Am J Cardiol 1984;53:586-589

6. Choong CY, Herrmann HC, Weyman AE, Fifer MA: Pre-load dependence of Doppler-derived indexes of left ventric-

1467



ownloaded from



ular diastolic function in humans. J Am Coll Cardiol1987;10:800-808

7. Leeman DE, Feldman MD, Diver DJ, Santinga JT, ComePC: Effects of decreases in preload on pulsed Dopplerindices of left ventricular filling (abstract). Circulation 1986;74(suppl lI):I1-46

8. Takahashi T, Sato H, lizuka M, Serizawa T, Ohya T,Takatoshi M, Sakamoto T: Left ventricular early filling flowis sensitive to changes in left atrial pressure: A study usinglower body negative and positive pressure (abstract). Circu-lation 1986;74(suppl Il):II-477

9. Bhatia SJS, Theard MA, Plappert T, St. John Sutton MG:Effects of altered loading conditions on transmitral Dopplerblood flow velocity (abstract). Circulation 1987;76(supplIV):IV-124

10. Courtois M, Kovacs SJ Jr, Ludbrook PA: The transmitralpressure-flow velocity relationship: The importance ofregional pressure gradients in the left ventricle. Circulation1988;78:661-671

11. Gardin JM, Dabestani A, Takenaka K, Rohan MK, Knoll M,Russell D, Henry WL: Effect of imaging view and samplevolume location on evaluation of mitral flow velocity bypulsed Doppler echocardiography. Am J Cardiol 1986;57:1335-1339

12. Courtois MR, Kurnik PB, Ludbrook PA: Sensitivity ofisovolumic relaxation to hypothermia during myocardialinfarction. JAm Coil Cardiol 1988;11:201-206

13. Ishida Y, Meisner JS, Tsujioka K, Gallo JI, Yoran C, FraterRWM, Yellin EL: Left ventricular filling dynamics: Influ-ence of left ventricular relaxation and left atrial pressure.Circulation 1986;74:187-196

14. Nolan SP, Dixon SH, Fisher RD, Morrow AG: The influenceof atrial contraction and mitral valve mechanics on ventric-ular filling. Am Heart J 1969;77:784-791

15. Parker TG, Cameron D, Serra J, Morgan CD, Sasson Z: Theeffect of heart rate and a-v interval on Doppler ultrasoundindices of left ventricular diastolic function (abstract). Cir-culation 1987;76(suppl IV):IV-124

16. Gillam LD, Homma S, Novick SS, Rediker DE, Eagle KA:The influence of heart rate on Doppler inflow patterns(abstract). Circulation 1987;76(suppl IV):IV-123

17. Tyberg JV, Keon WJ, Sonnenblick EH, Urschel CW:Mechanics of ventricular diastole. Cardiovasc Res 1970;4:423-428

18. Mahler F, Ross J Jr, O'Rourke RA, Covell JW: Effects ofchanges in preload, afterload and inotropic state on ejection

and isovolumic phase measures of contractility in the con-scious dog. Am J Cardiol 1975;35:626-634

19. Hori M, Yellin EL, Sonnenblick EH: Left ventricular dia-stolic suction as a mechanism of ventricular filling. Jpn CircJ 1982;46:124-129

20. Yellin EL, Sonnenblick EH, Frater RWM: Dynamic deter-minants of left ventricular filling: An overview, in Baan J,Arntzenius AC, Yellin EL (eds): Cardiac Dynamics. TheHague, Martinus Nijhoff, 1980, pp 145-158

21. Robinson TF, Factor SM, Sonnenblick EH: The heart as asuction pump. Sci Am 1986;254:84-91

22. Keren G, Meisner JS, Sherez J, Yellin EL, Laniado S:Interrelationship of mid-diastolic mitral valve motion, pul-monary venous flow, and transmitral flow. Circulation1986;74:36-44

23. Ling D, Rankin JS, Edwards CH, McHale PA, AndersonRW: Regional diastolic mechanics of the left ventricle in theconscious dog. Am J Physiol 1979;236 (Heart Circ Physiol5):H323-H330

24. Van de Werf F, Minten J, Carmeliet P, De Geest H,Kesteloot H: The genesis of the third and fourth heartsounds: A pressure-flow study in dogs. J Clin Invest 1984;73:1400-1407

25. Brecher GA, Kolder H, Horres AD: Ventricular volume ofnonbeating excised dog hearts in the state of elastic equilib-rium. Circ Res 1966;19:1080-1085

26. Appleton CP, Hatle LK, Popp RL: Demonstration of restric-tive ventricular physiology by Doppler echocardiography. JAm Coll Cardiol 1988;1:757-768

27. Hatle L, Brubakk A, Tromsdal A, Angelsen B: Noninvasiveassessment of pressure drop in mitral stenosis. Br Heart J1978;40: 131-140

28. Van de Werf F, Boel A, Geboers J, Minten J, Willems J, DeGeest H, Kesteloot H: Diastolic properties of the left ven-tricle in normal adults and in patients with third heartsounds. Circulation 1984;69:1070-1078

29. Gaasch WH, Cole JS, Quinones MA, Alexander JK: Dynamicdeterminants of left ventricular diastolic pressure-volumerelations in man. Circulation 1975;51:317-323

30. Taylor RR, Covell JW, Sonnenblick EH, Ross J Jr: Depen-dence of ventricular distensibility on filling of the oppositeventricle. Am J Physiol 1967;213:711-718

31. Ludbrook PA, Byrne JD, McKnight RC: Influence of rightventricular hemodynamics on left ventricular diastolicpressure-volume relations in man. Circulation 1979;59:21-31

KEY WORDS * Doppler echocardiography transmitralpressure-flow relations . preload reduction * diastolehemodynamics



ownloaded from


M Courtois, Z Vered, B Barzilai, N A Ricciotti, J E Pérez and P A LudbrookThe transmitral pressure-flow velocity relation. Effect of abrupt preload reduction.

Print ISSN: 0009-7322. Online ISSN: 1524-4539 Copyright © 1988 American Heart Association, Inc. All rights reserved.

is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231Circulation doi: 10.1161/01.CIR.78.6.1459

1988;78:1459-1468Circulation.

http://circ.ahajournals.org/content/78/6/1459the World Wide Web at:

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

http://circ.ahajournals.org//subscriptions/

is online at: Circulation 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 information about this process is available in the

located, click Request Permissions in the middle column of the Web page under Services. FurtherEditorial 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 theCirculationpublished in Requests for permissions to reproduce figures, tables, or portions of articles originallyPermissions:



ownloaded from

http://circ.ahajournals.org/content/78/6/1459

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

http://www.lww.com/reprints

http://circ.ahajournals.org//subscriptions/


the transmitral pressure-flow velocity relationcirc.ahajournals.org/content/78/6/1459.full.pdf ·...

Documents