Effects of left ventricular volume overload produced by mitral regurgitation on diastolic

function

MICHAEL R. ZILE, MASAAKI TOMITA, KIYOHARU NAKANO, ISRAEL BRUCE USHER, JOHN LINDROTH, AND BLASE A. CARABELLO

MIRSKY,

Department of Medicine (Cardiology Division), Medical University of South Carolina, Veterans Affairs Medical Center, and Gazes Cardiac Research Institute, Charleston, South Carolina 29425

ZILE, MICHAEL ISRAEL MIRSKY, A. CARABELLO.

duced by mitral

R., BRUCE

MASAAKI USHER,

TOMITA, KIYOHARU JOHN LINDROTH,

NAKANO, AND BLASE

Effects of left ventricular volume overload proregurgitation on diastolic function. Am. J. Phys-

iol. 261 (Heart Circ. Physiol. 30): H1471-H1480, 1991.-We hypothesized that the left ventricle’s ability to compensatefor the volume overload produced by mitral regurgitation (MR) depends,at least in part, on associatedchangesin left ventricular (LV) diastolic function. Indexes of the rate of LV pressure decline, the rate and extent of early diastolic filling, and LV diastolic stiffness were measuredwith simultaneousechocardiography and catheterization in the baseline state (baseline), immediately after creation of MR (acute MR), and 3 mo after creation of MR (chronic MR). Data are means t SD. MR causedLV dilation; end-diastolic dimensionincreasedfrom 4.3 t 0.4 in baselineto 4.7 & 0.5 in acute MR and 5.8 t 0.1 cm in chronic MR (P < 0.05 vs. baseline for both). Chronic MR caused eccentric LV hypertrophy; LV-to-body weight ratio increasedfrom 3.6 t 0.3 in baselineto 4.5 * 0.2 g/kg in chronic MR (P < 0.05 vs. baseline). Acute MR increased LV enddiastolic pressurefrom 8 t 4 in baselineto 15 k 3 mmHg (P < 0.05vs. baseline);chronic MR did not further increaseLV enddiastolic pressure(14 t 4 mmHg). MR increasedthe transmitral pressuregradient from 5 & 1 in baselineto 14 * 3 in acute MR and 20 t 6 mmHg in chronic MR (P < 0.05 vs. baseline for both). MR increasedLV early diastolic filling rate; peak rate of increasein minor axis dimension increasedfrom 11 t 2 baselineto 18 t 2 in acute MR and 19 & 2 cm/s in chronic MR (P < 0.05 vs. baselinefor both). Acute MR did not change LV stiffness constants. Chronic MR decreasedLV stiffness; the modulus of chamber stiffness decreasedfrom 7.1 t 2.8 in baselineto 2.9 t 1.6 in chronic MR (P < 0.05 vs. baseline). Thus MR causedcompensatory changesin LV diastolic function. These changes resulted from an increased transmitral pressuregradient and increasedLV distensibility. hypertrophy; left ventricular function; relaxation; stiffness

WITH MITRAL regurgitation (MR) may remain compensated, without symptoms of congestive heart failure for many years. However, to remain compensated, the left ventricle must dilate, increase stroke volume, and maintain acceptable left ventricular (LV) diastolic pressures. How the left ventricle is able to accomplish these compensatory changes is not completely understood. It is likely, however, that these compensatory changes are dependent, at least in part, on alterations in LV diastolic function. This hypothesis is supported by noninvasive clinical studies (1, 25) and a recent experiPATIENTS

mental study (with a canine model of MR) (l7), which showed that MR increased the rate and extent of LV early diastolic filling, one index of LV diastolic function. A number of experimental studies of LV volume overload have examined other adaptive mechanisms, including increased sarcomere stretch (Frank-Starling preload reserve) (28, 29), the development of LV hypertrophy (24, 34), and variable changes in LV contractile state (3, 5, 19,31). However, neither these experimental studies nor previous noninvasive clinical studies have completely defined the effects of acute or chronic MR on LV diastolic function; nor have these studies defined the mechanisms by which alterations in LV diastolic function occur. Defining these mechanisms may be essential to understanding how the left ventricle adapts to MR. Therefore, the purposes of this study were to define the effects of MR on LV diastolic function during the development of volume overload and to define the mechanisms that cause these changes in diastolic function. METHODS

In 10 dogs, simultaneous echocardiography and catheterization were used to obtain indexes of ventricular volume, mass, and diastolic function. Ten dogs were studied in the baseline state (baseline). MR was then created with the techniques discussed below. Five dogs were studied immediately after the creation of MR (acute MR); and 10 dogs were studied 3 mo after the creation of MR (chronic MR). Creation of MR. MR was produced with a technique previously described in detail (3). Briefly, a urological calculus-retrieving forceps was introduced into the left ventricle through a sheath and was used to grasp chordae tendineae or the mitral valve leaflets. Forcible retraction of the grasping forceps disrupted the chordae tendineae producing MR. After each attempt at producing MR, measurement of LV pressure, pulmonary capillary wedge pressure, cardiac output, and auscultation of the heart were performed. A fall in cardiac output by 5076, a fall in systemic systolic pressure to 85 mmHg, and a rise in pulmonary capillary wedge pressure to 20 mmHg suggested that severe MR had been produced. Repeat thermodilution cardiac output determination and ventriculography were performed to angiographically assess the amount of MR that had been created and to calculate regurgitant fraction. The amount of MR produced in the H1471

Downloaded from www.physiology.org/journal/ajpheart by ${individualUser.givenNames} ${individualUser.surname} (192.236.036.029) on November 8, 2018. Copyright © 1991 American Physiological Society. All rights reserved.

H1472

DIASTOLE

IN MITRAL

dogs discussed in this study was comparable to our previous studies (3); regurgitant fraction for dogs included in this study was 59 t 10% immediately after creating MR and 59 t 12% 3 mo after creating MR. Simultaneous LV e&cardiography and catheterization. Animals were sedated with Innovar-Vet (droperidol

fentanyl; 0.075 ml/kg im). With the use of a sterile technique and 2% Xylocaine anesthesia, an incision was made over the cervical vessels to expose and isolate the carotid artery and jugular vein. A thermodilution SwanGanz catheter was advanced to the pulmonary capillary wedge position with use of fluoroscopic and hemodynamic guidance. A 7-Fr double micromanometer-tipped catheter (PC-780 Millar Instruments, Houston, TX) was externally calibrated to mercury at 37°C and advanced to the left ventricle under fluoroscopic guidance. The distal micromanometer was positioned in the left ventricle; the proximal micromanometer was placed in the proximal aorta. The calibration for both micromanometers was confirmed and matched to a 5-Fr fluid-filled pigtail catheter connected to a transducer (Statham P23 Db, Oxnard, CA) placed at the midchest level. LV, aortic, and pulmonary capillary wedge pressures were recorded simultaneously with LV echocardiography at a paper speed of 100 mm/s. Two dimensional, M-mode and Doppler echocardiographic studies (ATL Ultramark VI, 2.25- and 3.5-mHz transducers, Bothell, WA) were performed from the right parasternal and apical areas. The short axis and long axis views were obtained in all dogs. These methods have been previously described in detail (33). Echocardiographic data were measured by use of the American Society of Echocardiography criteria (30), including the leading edge convention. End diastole was defined at the Q wave of the electrocardiogram. End systole was defined at the peak downward notion of the interventricular septum (end ejection). In normal subjects this corresponds to the aortic dicrotic notch and the second heart sound (aortic valve closure). We recognize that this relationship may not be constant in subjects with MR. However, in the current study [as in our previous study (3)], end-ejection measurements were used because they were easily defined and reproducible. Mitral valve inflow velocities (m/s) were measured with pulsed-wave Doppler studies recorded from the apical view. The Doppler sample volume was positioned just within the inflow portion of the left ventricle, midway between the annular margins of the mitral valve. The transducer position was then finely adjusted to maximize the peak early diastolic filling velocity (E wave velocity). A minimum of ten cycles was measured in each study. Maximum E wave and A (transmitral filling during atria1 contraction) wave velocities and the ratio of E to A wave velocities were measured. LV pressure and echocardiograms were processed with a semiautomated technique similar to that used in previous studies (33). Pressure and M-mode echo records were placed on a digitizing tablet (Summasketch, Summagraphics, Fairfield, CT) and manually traced with a cursor. The position of the cursor was detected and converted to digital coordinates for processing by a microcomputer system (National Cash Register PCS, Ak-

REGURGITATION

ron, OH). LV pressure, minor axis dimension, and wall thickness were digitized with a sampling interval of 5 ms. Data were smoothed with a seven point third-order least-squares orthogonal polynomial fit. The derivatives of pressure, dimension, and thickness with respect to time were obtained from the smoothed data. Typical examples of pressure, dimension, and thickness vs. time plots are shown in Figs. 3-5 for baseline, acute MR and chronic MR. Calculations. LV mass was calculated with the recently validated formula of Feneley et al. (8) LV mass = echo cross-sectional

area (CSA)

X long axis dimension

Muscle CSA was calculated

(1)

as

CSA(cm2) = 74Dm/2 + hd2

- ~DED/~)~

(2)

where DED is the end-diastolic minor axis dimension and hEDis the end-diastolic wall thickness. In our laboratory, there has been a good correlation between echo-derived LV mass and angiography-derived LV mass and mass measured at autopsy in normal dogs and dogs with MR (Fig. 1). Fractional shortening was calculated as FS(%) = (DED - DES)/DED x 100

(3)

where DES is the end-systolic dimension. Stroke dimension was calculated as DED - DES and was used as an index of total stroke volume. Circumferential, global average wall stress (a) was calculated assuming a cylindrical (a~) or spherical (a,) geometry a,(g/cm2)

= (PD/2h)

a,(g/cm2)

PD = 4h(l + hlD)

X

1.36 X 1.36

(4)

(5)

where P is LV pressure, D is short axis dimension, and h is wall thickness. Instantaneous values of P, D, and h were used to calculate stress through the cardiac cycle. Values of stress measured at end diastole, peak systole, end systole, and mitral value opening are presented in Table 2. Indexes of peak early diastolic filling rate were obtained from echo dimension and thickness transients and Doppler velocity measurements. Peak rate of increase in dimension (peak +dD/dt) and peak wall thinning rate (peak -dh/dt) were measured in centimeters per second and normalized by dividing their maximum values by instantaneous dimension and thickness; they were then expressed as +dD/dt/D and -dh/dt/h in secends-‘. The amount of filling that occurred during the first 40% of diastole (filling extent 40%) was calculated as the dimension reached after the first 40% of cliastole (D40%) minus the end-systolic dimension. In these measurements, 40% of diastole refers to a fraction of the duration of diastole in milliseconds. The fraction of filling that occurred during the first 40% of diastole (FF40) was calculated as FF40(%) = (DED - D40%)/(DED

- DES) x 100

(6)

Downloaded from www.physiology.org/journal/ajpheart by ${individualUser.givenNames} ${individualUser.surname} (192.236.036.029) on November 8, 2018. Copyright © 1991 American Physiological Society. All rights reserved.

DIASTOLE

IN MITRAL

H1473

REGURGITATION

FIG. 1. Correlation between echocardiographitally (echo) derived left ventricular (LV) mass and mass measured angiographically (angio) (A) and mass measured at autopsy (B). There was good correlation between LV mass-echo and LV massangio or autopsy. I

50

1

75 100 LEFT VENTRICULAR

I

I

125 150 MASS-ANGIO

1

175 (g)

The left atrial-to-LV transmitral pressure gradient was not measured directly. Rather, the difference between the pulmonary capillary wedge pressure at the peak of the V wave (PCWP,) and the LV minimum pressure was used as an index of the early diastolic transmitral pressure gradient. The PCWPv was chosen because it occurs at or near the time of mitral valve opening (i.e., the time at which LV pressure falls to a value equal to the peak of the V wave in the PCWP). In so doing, we took into account the delay between pressure transients measured by the pulmonary capillary wedge catheter and those measured in the left atrium. This method therefore did not allow measurement of instantaneous transmitral gradients throughout filling but did provide an index of the maximum early diastolic transmitral pressure gradient at mitral valve opening. The time constant of isovolumic pressure decline (7) was calculated from time-expanded recordings of LV pressure digitized at 5-ms intervals beginning at peak pressure decline with respect to time (-dP/dt) and ending at mitral valve opening. rw was calculated using the original method described by Weiss et al. (36) with the assumption that the baseline pressure toward which the monoexponential decays (PB) equals zero p = p(@w

(7)

where e is the base of the natural logarithm, t is the time in milliseconds after peak -dP/dt, PO is the pressure at peak -dP/dt. 7w was defined as the time required for LV pressure to fall to the value PO/e and was obtained from the linear regression analysis of the natural log of pressure vs. time. It is recognized that PB may not be zero in this preparation (39); therefore we also used a three constant, nonlinear regression method described in Mirsky and Pasipoularides (22) to calculate TV, using the following model P = Poemat + Pg When t = 7111this equation 7m = (l/a)

(8)

becomes

In[ePo/(Po

+ PB - ePB)]

(9)

where PO, Pg, and a are obtained from a nonlinear regression analysis of LV pressure (P) vs. time (t ). It is also recognized that in the presence of MR, LV pressure decline from peak -dP/dt to mitral valve opening may not be totally isovolumic. The amount of LV volume that is ejected into the left atrium after aortic valve closure is quite small and should not significantly

effect this analysis. However, in addition to examining 7w and 7m (indexes of ventricular relaxation), 76 (an index of myocardial relaxation) was also examined. 7, was calculated as CT= aoe-t/q7 (10) where 76 is’ obtained from the linear regression analysis of the natural log of stress vs. time. This index of myocardial relaxation was first used by Pouleur et al. (27) in patients with coronary artery disease. Its use has been reviewed by Mirsky and Pasipoularides (22) and proposed for use in mitral and aortic regurgitation. Like Eq. 7, one limitation of this method is that it assumes a monoexponential decay to zero stress. Typical examples of pressure vs. time plots for each mathematic model used to calculate 7 (7,, TV, and 7J are shown for the baseline state in Fig. 2. The average correlation coefficient for the linear regression analysis exceeded 0.99 for each of these three methods. Regional chamber stiffness constant (Kc) and regional myocardial stiffness constant (Km) were calculated from LV echocardiography-catherization data. The calculation of chamber and myocardial stiffness constants was based on the analysis of the curvilinear diastolic pressure-volume and stress-strain relationship. Methods used to apply these concepts to the pressure-dimensionthickness data used in this study were developed by Mirsky and Pasipoularides (22). Assuming that the ventricle can be modeled as a cylindrical annulus, regional chamber stiffness constant normalized for LV mass was calculated as p = AeK”(?rD2/4CSA)

where the linear and normalized Regional as

(10

relation between dP/d(rD2/4CSA) and P is K,, (the slope of this linear relation) is the modulus of regional chamber stiffness. muscle stiffness constant (Km) was calculated

K m = %i(Dda*)/(dD) (1% where the stress difference (CT*), assuming a cylindrical geometry, was calculated as (J* = [P(D + 2h)2]/[2h(D + h)] (13) Typical examples of the pressure vs. ?rD2/4CSA analysis used to calculate Kc and the stress difference vs. dimension analysis used to calculate Km are shown in Figs. 6 and 7 for baseline, acute MR, and chronic MR. The average correlation coefficient for each analysis exceeded 0.95.

Downloaded from www.physiology.org/journal/ajpheart by ${individualUser.givenNames} ${individualUser.surname} (192.236.036.029) on November 8, 2018. Copyright © 1991 American Physiological Society. All rights reserved.

H1474

DIASTOLE

IN

MITRAL

-0

REGURGITATION

E

FIG. 2. Example of 3 methods used to calculate time constant of LV isovolumic pressure decline; these analyses were performed on same animal in baseline state. A: rw, method of Weiss et al. (36). of Mirsky and PasipouB: 7111, method larides (22). C: r,, method of Pouleur et al. (27). Each mathematical model fits actual pressure closely and had large correlation coefficient (r > 0.99). Heavy dashed line represents mathematical fit produced by each method. Dotted line represents rate of change of pressure with respect to time (dP/dt).

5

20 - -2000

200

400

TIME

L

~38

(ma)

ms

r = 0.988 - 2000

4000

-i 3 'a

'a

A r)

200

I

-2 z

-0

-

%

-2000

1 -4000

0

6

s

5

d

.g

100 -4000

s

u

t

0 0

200

400

TIME TIME

0

'a Y

i

(ms)

(ma)

Statistical analysis. Data are presented as means t SD. Differences among baseline, acute MR, and chronic MR were examined for significance using a repeated measures analysis of variance and a Newman-Keuls multiple sample comparison test. P < 0.05 was considered to be significant. All animals received humane care in compliance with the “Principles of Laboratory Animal Care,” formulated by the National Society for Medical Research, and the “Guide for the Care and Use of Laboratory Animals” [DHEW(DHHS) Publication No. (NIH) 85-23, revised 1985, Office of Science and Health Reports, DRR/NIH, Bethesda, MD].

TABLE

1. Hemodynamic changes produced by MR

Heart rate, beats/min LV mass, g WBW g/k

Acute MR caused immediate LV dilation (end-diastolic dimension increased from 4.3 t 0.4 in baseline to 4.7 t 0.5 cm in acute MR, P C 0.05). End-systolic dimension fell, fractional shortening increased, and total stroke volume increased (stroke dimension increased from 1.5 t 0.4 in baseline to 2.1 t 0.2 cm in acute MR, P c 0.05). LV systolic pressure and wall stress fell; peak-systolic stress was not significantly changed, end-systolic stress fell significantly from 152 t 30 in baseline to 105 t 31 g/cm2 in acute MR (P c 0.05). Chronic MR caused further LV dilation (end-diastolic dimension increased to 5.8 t 0.1 cm in chronic MR, P C 0.05 vs. baseline and acute MR) and the development of modest LV hypertrophy (LV-to-body wt ratio increased’ from 3.6 t 0.3 in baseline to 4.5 t 0.2 g/kg in chronic MR, P c 0.05). End-systolic dimension increased compared with baseline and acute MR. Fractional shortening fell compared with acute MR but remained greater than baseline. Total stroke volume remained elevated (stroke dimension was 2.3 t 0.1 cm in chronic MR, P < 0.05 vs. baseline). Peak-systolic stress and end-systolic stress

Acute MR

Chronic MR

86t15 loot9 3.6t0.3

lllkl2* 105t13 3.6iO.3

85+10t 125+9”t 4.5+0.2*t

108+7* 93t5” 20t4” 4t2 15*3* 15t3* 14t3*

91+10*t 77+10*t 23t8* 2k3 14t4* 15*5* 20&6*

Catheterization LV pressure, mmHg Peak systole End systole Mitral valve opening Minimum End diastole Pulmonary wedge pressure, PCWP, - LVP,i,, mmHg

RESULTS

LV volume, mass, and wall stress (Tables 1 and 2).

Baseline

mmHg

123t12 110t9 822 2t2 824 6kl 5*1

Echocardiography End-diastolic dimension, cm End-systolic dimension, cm End-diastolic thickness, cm Fractional shortening, % Stroke dimension, cm End-diastolic radius-to-thickness ratio LA/SA

4.3t0.4 2.9t0.3 l.Ot0.04 35t7 1.5kO.4 2.120.2

4.7*0.5* 2.5t0.5* 0.9&0.01* 46_t6* 2.lt0.2* 2.6t0.3*

5.8+0.1*t 3.4*0.1*? 0.85t0.08” 40*1*-t 2.3*0.1* 3.7+0.3*-f

1.6t0.2

1.4tO.l*

1.1*0.1*t

Values PCWPv,

are means t SD. LV/BW, left ventricle-to-body wt ratio; pulmonary capillary wedge pressure at peak of V wave; minimum early diastolic LV pressure; LA/SA, end-diastolic LVPrni*, long to short axis ratio. * P < 0.05 vs. control. t P < 0.05 vs. acute MR.

were not significantly

different

from baseline values.

LV diastolic function (Table 3, Figs. 3-9). Acute MR

caused impairment of both myocardial and LV relaxation rates (Fig. 3): 7w increased from 29 t 3 in baseline to 40 t 8 ms in acute MR, and 7, increased from 31 t 4 in baseline to 49 t 8 ms in acute MR, both P < 0.05. The rate of LV early diastolic filling increased (Figs. 4 and 5); the peak rate of increase in minor axis dimension (peak +dD/dt) increased from 11 t 2 in baseline to 18 t 2 cm/s in acute MR, the peak thinning rate (peak -dh/dt) increased from 3.6 t 0.6 in baseline to 6.3 t 1.3

Downloaded from www.physiology.org/journal/ajpheart by ${individualUser.givenNames} ${individualUser.surname} (192.236.036.029) on November 8, 2018. Copyright © 1991 American Physiological Society. All rights reserved.

DIASTOLE

IN MITRAL

2. Effect of MR on LV wall stress

TABLE

Wall

Stress,

g/cm’

Baseline

Acute

MR

Chronic

MR

Cylinder Peak systole 318t54 312t38 367*58 End systole 152k30 105t31’ 131*19 Mitral valve opening 10t3 24k7* 39t12’ End diastole 26k6 52&11* 66*20* Sphere Peak systole 132*21 127*18 148t23 End systole 52t12 34t12’ 42kll Mitral valve opening 3*1 9t3’ 15t6* End diastole 10*3 2124’ 29*9* Values are means t SD. LV, left ventricular. * P < 0.05 vs. control. TABLE

3. Effect of MR on diastolic function

L V pressure

Peak -dP/dt, 7w9ms 7rn9

mmHg/s

decline

1,920+300 29t3 31*4 31k4

ms

7 “, ms

LV diastolic

+dD/dt, cm/s +dD/dt/D, s-l -dh/dt, cm/s -dh/dt/h, s-’ Filling extent 4O%, mm Filling fraction 40%, % Doppler E velocity, m/s Doppler A velocity, m/s Doppler E/A ratio

Km

40t8* 43*9* 49k8’

1,470+500 29&5-t

11t2

18&2*

0.7t0.2 0.4&O. 1

5.3&1.2* 6.3t1.3’ 4.9kO.8’ 10*2 51t11 1.3kO.2’ 0.6&O. 1*

33+11t

1.9kO.4

2.1kO.5

52tlO

19t2” 4.5kO.4’ 6.0tl.O’ 4.9t0.8’ 18*2*t 79+7*-f 1.4t0.3’ 0.6tO. 1’ 2.6t0.4*

t

constant

7.1k2.8 8.1k3.5

6.1k3.2 7.8t4.2

H1475

peak -dh/dt was 6.0 t 1.0 cm/s, and Doppler E velocity was 1.4 m/s t 0.3 in chronic MR). The extent of early filling was increased: the filling fraction at 40% increased from 52 t 10% in baseline to 79 t 7% in chronic MR (P < 0.05); and the filling extent at 40% increased to 1.8 t 0.2 cm in chronic MR, P < 0.05 vs. baseline. The pulmonary capillary wedge to LV pressure gradient increased in chronic MR (20 t 6 mmHg, P < 0.05 vs. baseline). In chronic MR mean pulmonary wedge pressure, and LV end-diastolic pressure were increased compared with baseline but were unchanged compared with acute MR. The pressure-dimension and pressure-mass normalized dimension curves were shifted to the right. There was a significant fall in the chamber stiffness constant (Kc fell from 7.1 t 2.8 in baseline to 2.9 t 1.6 in chronic MR, P < 0.05). There were no significant changes in the myocardial stiffness constant (K, was 8.1 t 3.5 in baseline and 10.3 t 5.7 in chronic MR). DISCUSSION

32t5t

filling

3.4t0.6 3.6t0.6 2.8t0.5 8k3

Stiffness

KC

1,320+350*

REGURGITATION

2.9t1.6’t 10.3t5.7

Values are means k SD. TV, time constant using Weiss et al. (36) method; 7311,time constant using Mirsky and Pasipoularides (22) method; T,, myocardial relaxation rate; +dD/dt, peak rate of increase in minor axis dimension; -dh/dt, peak wall thinning rate; KC, chamber stiffness constant; Km, myocardial stiffness constant. * P < 0.05 vs. control. t P < 0.05 vs. acute MR.

cm/s in acute MR, and the Doppler E wave velocity increased from 0.7 t 0.2 in baseline to 1.3 t 0.2 m/s in acute MR, P < 0.05. There was a significant increase in heart rate and a shortening of the diastolic filling time. The early diastolic filling fraction (FF40%) was not changed, and the filling extent after 40% of diastole increased slightly from 0.8 t 0.3 in baseline to 1.0 t 0.2 cm in acute MR. The pulmonary capillary wedge to LV pressure gradient during early diastole rose from 5 t 1 in baseline to 14 t 3 mmHg in acute MR, P < 0.05. The mean pulmonary wedge pressure increased from 6 t 1 in baseline to 15 t 3 mmHg in acute MR, P < 0.05. LV end-diastolic pressure increased from 8 t 4 in baseline to 15 t 3 mmHg in acute MR, P < 0.05. The pressuredimension and pressure-mass normalized dimension (i.e., LV pressure vs. 7rD2/4CSA) curves were shifted upward but neither the chamber nor muscle stiffness constants were significantly affected by acute MR (Figs. 6-9). Chronic MR did not significantly alter 7W or 7, compared with the baseline state. The rate of early diastolic filling was increased compared with baseline and similar to that caused by acute MR (peak +dD/dt was 19 t 2,

The purposes of this study were to 1) define the effects of mitral regurgitation on left ventricular diastolic function during the development of volume overload and 2) define the mechanisms that cause these compensatory changes in diastolic function. Data from the current study indicate that the volume overload produced by mitral regurgitation caused significant changes in both early and late left ventricular diastolic function. During early diastole, both acute and chronic mitral regurgitation caused an increase in the extent and rate of left ventricular early diastolic filling. In acute mitral regurgitation, the determinants that played a role in augmenting left ventricular early diastolic filling were an increase in the transmitral pressure gradient, a decrease in left ventricular end-systolic stress, and an increase in left ventricular systolic shortening. After 3 mo of mitral regurgitation (chronic mitral regurgitation), the rate and extent of left ventricular early diastolic filling remained increased, at levels comparable with acute mitral regurgitation. In chronic mitral regurgitation, the determinants that played a role in augmenting left ventricular early diastolic filling were a persistent increase in transmitral pressure gradient and fractional shortening and a return to normal in the rate of left ventricular pressure decline. Although many determinants may have played a role in augmenting early filling, an increased transmitral gradient was the one consistent factor promoting left ventricular early filling in both acute and chronic mitral regurgitation. Therefore, we interpreted these data to indicate that early filling was augmented in mitral regurgitation principally by an increase in the left atria1 driving force. During mid-late diastole, both acute and chronic mitral regurgitation caused an increase in left ventricular diastolic pressures. In acute mitral regurgitation, the nearly twofold increase in left ventricular end-diastolic pressure was associated with a 10% increase in left ventricular end-diastolic dimension, an upward shift in the pressuredimension relationship, and no change in chamber or myocardial stiffness constants. Thus, in acute mitral regurgitation, increases in left ventricular diastolic pres-

Downloaded from www.physiology.org/journal/ajpheart by ${individualUser.givenNames} ${individualUser.surname} (192.236.036.029) on November 8, 2018. Copyright © 1991 American Physiological Society. All rights reserved.

H1476

DIASTOLE

IN MITRAL

REGURGITATION

80

0

200

TIME

B

L

b

200

343

FIG. 3. Example of effects of mitral regurgitation (MR) on time constant of LV relaxation using Weiss et al. (36) method (TV). A: baseline. B: acute MR prolonged time constant. C: chronic MR caused no significant change in time constant. Results were similar with 7111and

1 600

400

(ms)

ms

400

600

0

200

TIME (ms)

400

TIME

sures were caused by acute left ventricular dilation and the consequent shift upward to a steeper portion of the normal pressure-volume relationship. In contrast, chronic mitral regurgitation did not cause a further increase in left ventricular diastolic pressures above that seen in acute mitral regurgitation. After 3 mo of mitral regurgitation, left ventricular diastolic pressures remained unchanged compared with acute mitral regurgitation despite a 25% increase in left ventricular endA

600

(ms)

diastolic dimension. This lack of increase in diastolic pressures was associated with an increase in the volumemass ratio, a rightward shift in the pressure-dimension relationship, and a decrease in the chamber stiffness. Thus, in chronic mitral regurgitation, ventricular remodeling and a decrease in left ventricular stiffness allowed the left ventricle to become more distensible. In chronic mitral regurgitation, these changes in diastolic function allowed the ventricle to dilate and increase stroke volume 20

10 s

0

5 s %

-10

-20 200

400

TIME (ms)

60

20

1

s

FIG. 4. Example of effects of MR on LV dimension and rate of change in dimension (dD/dt). A: baseline. B: acute MR. C: chronic MR.

-20 0

200

400

TIME (ms)

Downloaded from www.physiology.org/journal/ajpheart by ${individualUser.givenNames} ${individualUser.surname} (192.236.036.029) on November 8, 2018. Copyright © 1991 American Physiological Society. All rights reserved.

DIASTOLE

IN MITRAL

H1477

REGURGITATION

10

S S z 0 ii z 0

0

-S

-10 0

200

400 TIME

(ms)

10

0

0

1

1

200

400 TIME

-l@ -10

0

200

400 TINE

(mr)

without causing excessive increases in left ventricular diastolic pressure. Determinants of left ventricular

early diastolic filling.

The current study was concordant with previous clinical studies that demonstrated that indexes of left ventricular early diastolic filling rate were enhanced in chronic mitral regurgitation (1, 25). Clinical studies using echocardiographic techniques showed that the peak rate of increase in minor axis dimension and the peak rate of left ventricular wall thinning were increased in patients with mitral regurgitation (25). Studies with echo-Doppler techniques demonstrated that the maximum early mitral flow velocity (E velocity), the E/A flow velocity ratio, and the deceleration half time of early diastolic flow were -

(mm)

increased in patients with mitral regurgitation (1). None of these studies, however, examined the mechanisms causing these changes in left ventricular early diastolic filling rates. Possible mechanisms include a decrease in left ventricular afterload, an increase in systolic shortening, an increase in the left atrial-to-left ventricular transmitral pressure gradient, or an increase in the rate of myocardial deactivation (40, 42). In their recent study, Katayama et al. (17) were the first to attempt to define the mechanisms that cause change in early diastolic function in mitral regurgitation. They concluded that mitral regurgitation of a-day to 4wk duration increased early diastolic filling because of “increased systolic shortening, increased elastic recoil, and increased early diastolic chamber elastance.” How-

BASEUNE ----k-

0.0

FIG. 5. Example of effects of MR on LV wall thickness and rate of change of wall thickness (dh/dt). A: baseline. B, acute MR. C: chronic MR.

I

I

I

I

J

0.4

0.8

1.2

1.6

2.0

It 02/4

2.0

CSA

FIG. 6. Example of effects of MR on chamber stiffness. Acute MR caused LV diastolic pressure vs. mass normalized dimension (,rrD*/ 4CSA) relationship to move upward along curve similar to baseline state. Chronic MR, however, caused this relationship to shift to right and become less steep, indicating decrease in LV chamber stiffness. CSA, muscle cross-sectional area.

BASELINE ACUTEMR cxRoNcMR

3.0

LV DIASTOLIC

4.0

DIMENSION

5.0

6.0

(cm)

7. Example of effects of MR on myocardial stiffness. Acute MR caused LV diastolic stress vs. dimension relationship to move upward along curve similar to baseline state. Chronic MR caused this relationship to shift to right but did not change slope of this relationship, indicating no change in myocardial stiffness. FIG.

Downloaded from www.physiology.org/journal/ajpheart by ${individualUser.givenNames} ${individualUser.surname} (192.236.036.029) on November 8, 2018. Copyright © 1991 American Physiological Society. All rights reserved.

DIASTOLE +

BASELINE ACLE

IN MITRAL

1 MR

cliKNc

Ml-3

/I +-!+

+‘. l

:

BASELINE ACUTE

MR

CtimcMR

0

2

0

3.0

LV DIASTOLIC

4.0

5.0

DIMENSION

6.

(cm)

FIG. 9. Effects of MR on myocardial stiffness; data are means t SD for each group. Acute MR caused LV diastolic stress vs. dimension relationship to shift upward along curve similar to baseline state. Chronic MR caused this relationship to shift to right but did not change slope, indicating no change in myocardial stiffness.

ever, as the authors themselves point out, many of the potential determinants of early diastolic filling rate (such as “altered systolic and relaxation loading, left atriumto-left ventricular pressure gradient, and accelerated myocardial inactivation”) were not examined (17). Therefore, the current study attempted to extend the work of Katayama et al. (17) and others (1, 25) by 1) studying animals with mitral regurgitation of a longer duration (12 wk), 2) characterizing the effects of mitral regurgitation on both early and late diastolic function, and 3) attempting to more completely define the mechanisms causing these changes in diastolic function. Ventricular load. Experimental studies have shown that acute mitral regurgitation caused a significant decrease in afterload and a significant increase in systolic shortening (13, 17, 35). Clinical studies of patients with chronic compensated mitral regurgitation have shown that afterload (peak, end, and mean systolic stress) and systolic shortening were normal in patients with well-

REGURGITATION

preserved ejection performance (6,37,41). None of these studies, however, examined these changes in afterload and shortening in light of their potential effects on left ventricular early diastolic function. Previous studies have made it clear that both acute and chronic mitral regurgitation cause an increase in left atria1 pressure; however, there are no previous studies that have measured the effect of acute or chronic mitral regurgitation on minimum left ventricular pressure, the left atrial-toleft ventricular transmitral pressure gradient, or the left ventricular stress during early filling. The current study demonstrated that mitral regurgitation caused an increase in shortening, a decrease in left ventricular endsystolic stress, and an increase in the left atrial-to-left ventricular pressure gradient compared with baseline. Myocardial deactivation. In addition to left ventricular load, the rate of myocardial deactivation (as indexed by the rate of left ventricular pressure decline) is believed to be an important determinant of peak early diastolic filling rate (4, 40, 42). The rate of myocardial deactivation is also an important descriptor of left ventricular early diastolic function. Quantitating the rate of isovolumic pressure decline is problematic in volume overload lesions such as mitral regurgitation and aortic regurgitation; left ventricular pressure decline may not be totally isovolumic in the presence of these lesions. To date, however, there are no studies in chronic mitral regurgitation that have carefully quantitated the changes in left ventricular volume that occur between peak -dP/dt and mitral valve opening (the time over which left ventricular pressure measurements are made to calculate the relaxation time constant). Two recent studies (2, 18), suggested that there was little or no mitral valve flow and little change in left ventricular volume after peak -dP/ dt. None the less, to compensate for this limitation in examining left ventricular relaxation, an index quantifying myocardial relaxation rate (the time course of left ventricular wall stress decline) was developed by Pouleur et al. (27) and Mirsky and Pasipoularides (22). In the current study both left ventricular pressure and stress decline were well fit by an exponential function. Changes in the time constant using the Weiss et al. (36) method or Mirsky and Pasipoularides (22) method closely parallel the change in myocardial relaxation rate. These data indicated that acute mitral regurgitation prolonged the rate of ventricular and myocardial relaxation, whereas chronic mitral regurgitation produced no significant changes in these values. Data from the current study in dogs with chronic mitral regurgitation differ from that of Hirota (16) who measured the time constant in patients with mitral regurgitation and Eichhorn et al. (7) who measured the time constant in patients with aortic insufficiency. In both of these clinical studies, chronic volume overload hypertrophy caused myocardial relaxation to be prolonged. Possible reasons for the differences between the current experimental study and these previous clinical studies may be the chronicity of the volume overload hypertrophy and the presence or absence of clinical symptoms. Whether afterload, systolic shortening, and myocardial deactivation act as primary determinants of left

Downloaded from www.physiology.org/journal/ajpheart by ${individualUser.givenNames} ${individualUser.surname} (192.236.036.029) on November 8, 2018. Copyright © 1991 American Physiological Society. All rights reserved.

DIASTOLE

IN MITRAL

ventricular early filling, independent of their actions on the transmitral pressure gradient, is unclear. Current data support the studies of Yellin and colleagues (39), which suggested that flow across the mitral valve during diastole is primarily determined by the atrioventricular pressure gradient. Other factors (such as afterload or systolic shortening) probably influence left ventricular filling by altering the magnitude of the transmitral pressure gradient either by affecting left atria1 pressure or left ventricular pressure during diastole. In addition, left atria1 loading and left atria1 contractile conditions may also affect the transmitral gradient. In acute mitral regurgitation, left atria1 volume is increased; consequently left atria1 pressure must also be increased. The pressurevolume relationships of the left atrium were not examined in this study, but defining the effects of mitral regurgitation on these relationships may provide important information about compensatory mechanisms in volume overload. Determinants of left ventricular end-diastolic pressure.

In addition to causing changes in early diastolic function, data from the current study indicate that mitral regurgitation caused alterations in mid-late diastolic function. Left ventricular mean diastolic and end-diastolic pressures are determined by left ventricular chamber stiffness. Chamber stiffness, in turn, is influenced by changes in left ventricular volume, mass, geometry, the extent of myocardial relaxation, myocardial stiffness, pericardial constraint, and other factors (12). In the current study, we were able to consider many but not all of these determinants. Chamber stiffness. Studies describing the effect of volume overload hypertrophy on chamber stiffness have been variable; some studies demonstrated no change in chamber stiffness (14-16, 32), some showed a reduction in chamber stiffness (9,20,23,26,38), and others showed an increase in chamber stiffness (11,21). This variability may have been based on the lesion itself (mitral regurgitation vs. aortic insufficiency vs. aorta-caval fistula), the method used to measure chamber stiffness (overall chamber stiffness vs. end-diastolic operative stiffness), the duration of the lesion (acute vs. chronic), or the presence or absence of clinical symptoms. In the current study, acute mitral regurgitation caused the pressure-dimension curve to shift upward along the normal baseline curve but did not change the chamber stiffness constant. This represents an increase in left ventricular end-diastolic operating stiffness with no change in overall stiffness. In contrast, chronic mitral regurgitation caused the pressure-dimension curve to shift to the right, indicating a fall in chamber stiffness. These changes were associated with an increase in the volume-mass ratio but no significant changes in the myocardial stiffness constant (discussed below). The influence of the pericardium on the diastolic pressure-volume relationship in acute and chronic mitral regurgitation was probably minimal. As suggested by our pressure-dimension data, acute mitral regurgitation may not have increased overall cardiac volume enough to elicit a pericardial constraining effect. Freeman and LeWinter (10) have shown that the pericardium adapts to chronic increases in heart size by itself increasing in size (and

H1479

REGURGITATION

mass). As the pericardium expands, its effects on the pressure-volume relationship decrease. Chronic mitral regurgitation caused a marked increase in left ventricular volume compared with acute mitral regurgitation but no further increase in mean pulmonary wedge pressure or left ventricular end-diastolic pressure. Myocardial stiffness. Although myocardial stiffness has been well characterized in pressure overload hypertrophy, only a few studies have examined myocardial stiffness in volume overload hypertrophy, and no previous studies have examined the effect of chronic mitral regurgitation on myocardial stiffness. In studies of experimentally induced aortic insufficiency and aortocaval fistula and in clinical studies of patients with aortic insufficiency, myocardial stiffness was unchanged by the presence of volume overload hypertrophy (15, 26, 32). The current study is concordant with these previous studies. Myocardial stiffness was unchanged by acute or chronic mitral regurgitation. Summary. Acute mitral regurgitation caused left ventricular dilation, increased total stroke volume, and an increase in the rate and extent of early diastolic filling. These compensatory changes in left ventricular early diastolic filling rate were caused by an increase in the transmitral pressure gradient and therefore an increase in the left atria1 driving force. Chronic mitral regurgitation caused further left ventricular dilation, eccentric left ventricular hypertrophy, increased total stroke volume, an increase in the extent and rate of left ventricular early diastolic filling, and a limited increase in left ventricular end-diastolic and mean capillary wedge pressures. These changes were caused by an increase in the left atria1 driving force and left ventricular remodeling, which caused a decreased left ventricular chamber stiffness and an increase in left ventricular distensibility. The authors thank Beverly Ksenzak for assistance in the preparation of this manuscript. This research was supported by medical research funds from the Dept. of Veterans Affairs, Washington, DC. Address for reprint requests: M. R. Zile, Cardiology Div./Dept. of Medicine, Medical Univ. of South Carolina, 171 Ashley Ave., Charleston, SC 29425. Received 21 November 1990; accepted in final form 7 June 1991. REFERENCES C. P., L. K. HATLE, AND R. L. POPP. Relation of transmitral flow velocity patterns to left ventricular diastolic function: new insights from a combined hemodynamic and Doppler echocardiographic study. J. Am. Coil. Curdiol. 12: 426-440,1988. 2. BRICKNER, M. E., AND M. R. STARLING. Dissociation of end systole from end ejection in patients with long-term mitral regurgitation. 1. APPLETON,

Circukztion 81: 1277-1286, 1990. 3. CARABELLO, B. A., K. NAKANO, J. F. SPANN. Left ventricular

overload hypertrophy.

W. CORIN, R. BIEDERMAN, AND function in experimental volume Am. J. Physiol. 256 (Heart Circ. Physiol.

25): H974-H981, 1989. 4. CHENG, C., G. L. FREEMAN, TINESCU, AND W. C. LITTLE.

W. P. SANTAMORE, M. S. CONSTANEffect of loading conditions, contractile state, and heart rate on early diastolic left ventricular filling in conscious dogs. Circ. Res. 66: 814-823, 1990. 5. COOPER, G., F. J. PUGA, K. J. ZUJKO, C. E. HARRISON, AND H. N. COLEMAN. Normal myocardial function and energetics in volumeoverloaded hypertrophy in the cat. Circ. Res. 32: 140-148, 1973. 6. CORIN, W. J., E. S. MONRAD, T. MURAKAMI, H. NONOGI, N. M. HESS, AND H. P. KRAYENBUEHL. The relationship of afterload to

Downloaded from www.physiology.org/journal/ajpheart by ${individualUser.givenNames} ${individualUser.surname} (192.236.036.029) on November 8, 2018. Copyright © 1991 American Physiological Society. All rights reserved.

H1480

DIASTOLE

IN

MITRAL

ejection performance in chronic mitral regurgitation. Circulation 76: 59-67, 1987. EICHHORN, P., J. GRIMM, R. KOCH, 0. HESS, J. CARROLL, AND H. P. KRAYENBUEHL. Left ventricular relaxation in patients with left ventricular hypertrophy secondary to aortic valve disease. Circulation 65: 13951404, 1982. FENELEY, M. P., J. W. GAYNOR, G. W. MAIER, S. A. GALL, J. A. KISSLO, AND J. S. RANKIN. In vivo estimation of left ventricular wall volume in volume-overloaded canine hearts. Am. J. Physiol. 255 (Heart Circ. Physiol. 24): H1399-H1404, 1988. FLORENZANO, F., AND S. A. GLANTZ. Left ventricular mechanical adaptation to chronic aortic regurgitation in intact dogs. Am. J. Physiol. 252 (Heart Circ. Physiol. 21): H969-H984, 1987. FREEMAN, G. L., AND M. M. LEWINTER. Pericardial adaptations during chronic cardiac dilation in dogs. Circ. Res. 54: 294-300, 1984. GAULT, J. H., J. W. COVELL, E. BRAUNWALD, AND J. Ross. Left ventricular performance following correction of free aortic regurgitation. Circulation 42: 773-780, 1970. GILBERT, J. C., AND S. A. GLANTZ. Determinants of left ventricular filling and of the diastolic pressure-volume relation. Circ. Res. 64: 827-852,1989. GROSSMAN, W., D. JONES, AND L. P. MCLAURIN. Wall stress and patterns of hypertrophy in the human left ventricle. J. Clin. Invest. 56:56-64,1975. GROSSMAN, W., L. P. MCLAURIN, AND M. A. STEFADOUROS. Left ventricular stiffness associated with chronic pressure and volume overloads in man. Circ. Res. 35: 793-800, 1974. HESS, 0. M., M. RI?TER, J. SCHNEIDER, J. GRIMM, M. TURINA, AND H. P. KRAYENBUEHL. Diastolic stiffness and myocardial structure in aortic valve disease before and after valve replacement. Circulation 69: 855-865, 1984. HIROTA, Y. A clinical study of left ventricular relaxation. Circulation 62: 756-766, 1980. KATAYAMA, K., T. TAJIMI, B. D. GUTH, M. MATSUZAKI, J. LEE, R. SEITELBERGER, AND A. L. PETERSON. Early diastolic filling dynamics during experimental mitral regurgitation in the conscious dog. Circulation 78: 390-400,1988. KEREN, G., T. H. LEJEMTEL, A. A. ZELCER, J. S. MEISNER, A. BIER, AND E. L. YELLIN. Time variation of mitral regurgitant flow in patients with dilated cardiomyopathy. Circulation 74: 684-692, 1986. LEWINTER, M. M., R. L. ENGLER, AND J. S. KARLINER. Enhanced left ventricular shortening during chronic volume overload in conscious dogs. Am. J. Physiol. 238 (Heurt Circ. Physiol. 7): H126H133, 1980. LEWIS, B. S., AND M. S. GOTSMAN. Left ventricular function during systole and diastole in mitral incompetence. Am. J. Cardiol. 34:635-643, 1974. MCCULLAGH, W. H., J. W. COVELL, AND J. Ross. Left ventricular dilatation and diastolic compliance changes during chronic volume overloading. Circulation 45: 943-951, 1972. MIRSKY, I., AND A. PASIPOULARIDES. Clinical assessment of diastolic function. Prog. Cardiovusc. Dis. 32: 291-318, 1990. MIRSKY, I., J. M. PFEFFER, AND M. A. PFEFFER. Mechanical properties of normal and hypertrophied myocardium: is there a relationship between diastolic and systolic function? In: Myocardial Hypertrophy and Fuilure, edited by N. R. Alpert. New York: Raven, 1983, p. 39-45. NEWMAN, W. H. Contractile state of hypertrophied left ventricle in long-standing volume overload. Am. J. Physiol. 234 (Heart Circ. Physiol. 3): H88-H93, 1978. OSBAKKEN, M. D., AND A. A. BOVE. Use of left ventricular filling

REGURGITATION

26.

27.

29.

30.

31.

32.

33.

34.

35.

36.

37.

38.

39.

40.

41.

42.

and ejection patterns in assessing severity of chronic mitral and aortic regurgitation. Am. J. Cardiol. 53: 1054-1060, 1984. PAPADIMITRIOU, J. M., B. E. HOPKINS, AND R. R. TAYLOR. Regression of left ventricular dilation and hypertrophy after removal of volume overload. Circ. Res. 35: 127-135, 1974. POULEUR, H., M. F. ROUSSEAU, C. VANEYLL, AND A. A. CHARLIER. Assessment of regional left ventricular relaxation in patients with coronary artery disease: importance of geometric factors and changes in wall thickness. Circulation 69: 696-702, 1984. ROSS, J., JR. Adaptation of the left ventricle to chronic volume overload. Circ. Res. 35 Suppl. II: 11-64-11-70, 1974. Ross, J., JR., E. H. SONNENBLICK, R. TAYLOR, H. M. SPONITZ, AND J. W. COVELL. Diastolic geometry and sarcomere length in the chronically dilated canine left ventricle. Circ. Res. 28: 49-61, 1971. SAHN, D. J., A. DEMARIA, J. KISSLO, AND A. WEYMAN. Recommendations regarding quantitation in M-mode echocardiography: results of a survey of echocardiographic measurements. Circulation 58:1072-1083,1978. TAYLOR, R. R., J. W. COVELL, AND J. Ross, JR. Left ventricular function in experimental aorto-caval fistula with circulatory congestion and fluid retention. J. Clin. Invest. 47: 1333~1342,1968. TAYLOR, R. R., AND B. E. HOPKINS. Left ventricular response to experimentally induced chronic aortic regurgitation. Cardiovusc. Res. 6: 404-414, 1972. TOMITA, M., F. G. SPINALE, F. A. CRAWFORD, AND M. R. ZILE. Changes in left ventricular volume, mass and function during the development and regression of supraventricular tachycardia induced cardiomyopathy: disparity between recovery of systolic versus diastolic function. Circulation 83: 635-644, 1991. TURINA, M., W. D. BUSSMANN, AND H. P. KRAYENBUHL. Contractility of the hypertrophied canine heart in chronic volume overload. Cardiovasc. Res. 3: 486-495, 1969. URSCHEL, C. W., J. W. COVELL, E. H. SONNENBLICK, J. Ross, AND E. BRAUNWALD. Myocardial mechanics in aortic and mitral valvular regurgitation: the concept of instantaneous impedance as a determinant of the performance of the intact heart. J. Clin. Invest. 47: 867-883, 1968. WEISS, J. L., J. W. FREDERIKSEN, AND M. L. WEISFELDT. Hemodynamic determinants of the time course of fall in canine left ventricular pressure. J. Clin. Invest. 58: 751-760, 1976. WISENBAUGH, T., J. F. SPANN, AND B. A. CARABELLO. Differences in myocardial performance and load between patients with similar amounts of chronic aortic versus chronic mitral regurgitation. J. Am. Coll. Cardiol. 5: 811-826, 1985. WONG, C. Y. H., AND H. M. SPOTNITZ. Systolic and diastolic properties of the human left ventricle during valve replacement for chronic mitral regurgitation. Am. J. Cardiol. 47: 40-50, 1981. YELLIN, E. L., S. NIKOLIC, AND R. W. M. FRATER. Left ventricular filling dynamics and diastolic function. Prog. Cardiovusc. Dis. 32: 247-271,199O. ZILE, M. R., AND W. H. GAASCH. Mechanical loads and the isovolumic and filling indices of left ventricular relaxation. Prog. Cardiovasc. Dis. 32: 333-346, 1990. ZILE, M. R., W. H. GAASCH, AND H. J. LEVINE. Left ventricular stress-dimension-shortening relations before and after correction of chronic aortic and mitral regurgitation. Am. J. Cardiol. 56: 99105,1985. ZILE, M. R., W. H. GAASCH, A. W. WEIGNER, K. G. ROBINSON, AND 0. H. L. BING. Mechanical determinants of the rate of isotonic lengthening in rat left ventricular myocardium. Circ. Res. 60: 815823,1987.

Downloaded from www.physiology.org/journal/ajpheart by ${individualUser.givenNames} ${individualUser.surname} (192.236.036.029) on November 8, 2018. Copyright © 1991 American Physiological Society. All rights reserved.