Left ventricular end-systolic stress-volume index ratio in aortic and mitral regurgitation with normal ejection fraction To evaluate the left ventricular contractile state in regurgitant valvular disease with normal ejection fraction, we analyzed the end-systolic stress-volume index relationship (ESSVR) by means of cineangiography in 15 normal subjects, 11 patients with aortic regurgitation (AR), and 10 patients with mitral regurgitation (MR) whose ejection fraction (EF) was 60% or more. The end-systolic stress-volume index ratio in normal subjects was 5.57 + 0.60 kdyne/cm5/m2 (mean + standard deviation), and we defined the range including f 2 standard deviations of the ratio as the normal ESSVR range. Six patients with AR and five patients with MR placed inside the normal ESSVR range, termed AR IN and MR IN, but the remaining five patients with AR and MR placed to the right of the normal range, termed AR OUT and MR OUT. EF did not differ between patients with AR IN and AR OUT (69.4 + 5.4 versus 70.7 ? 6.1%) and between MR IN and MR OUT (71.6 + 3.6 versus 71.1 f 7.9%). The EF of the subdivided groups with AR and MR also did not differ from that of normal subjects (70.7 +- 7.3%). This finding showed that the left ventricular contractile state was depressed in patients with AR OUT and MR OUT despite a normal EF. In AR and MR the end-systolic stress and end-systolic volume index of OUT did not differ from those of IN, but the end-diastolic volume index of OUT was larger than that of IN (AR OUT 156.6 + 27.9 versus AR IN 110.6 + 24.1 ml/m2, MR OUT 160.5 f 44.7 versus MR IN 101.0 f 16.6 ml/m2; both p < 0.05), and the regurgitant fraction of OUT was higher than that of IN (AR OUT 52.6 + 13.6 versus AR IN 29.7 f 13.3%, MR OUT 52.9 t- 10.2 versus MR IN 30.2 -t 11.4%; both p < 0.05). In addition, there was a linear inverse correlation between the end-systolic stress-volume index ratio and the end-diastolic volume index in all subjects (r = -0.62, n = 36). In normal subjects there was a linear inverse correlation between end-systolic stress and the EF (r = -0.91, n = 15), but this relationship failed to separate patients with OUT from those with IN. Results of the present study suggest that some patients with AR and MR whose EF was normal had a depressed contractile state, and these patients had a large end-diastolic volume index and a high regurgitant fraction. Therefore it may be worthwhile to investigate further the proposition that the end-systolic stress-volume index ratio can provide a more reliable means than the EF of evaluating the preoperative left ventricular contractile state in patients with AR and MR. (AM HEART J 1990; 120:692.)

Makoto Nakagawa, MD, Kunio Shirato, MD, Tadasu Ohyama, MD, Masahito Sakuma, MD, and Tamotsu Takishima, MD. Send& Japan

The timing of valve replacement in valvular heart disease is very important in the clinical course, because it is too late after irreversible left ventricular dysfunction has occurred. To assess the preoperative left ventricular contractile state in patients with valvular disease, numerous studies have been performed.1-8 The ejection fraction (EF) is now

From the First of Medicine. Received

Department

for publication

Reprint requests: Medicine, Tohoku dai, Japan 980. 4/l/22835

892

Tamotsu University

of Internal Dec.

Medicine,

21, 1989;

Takishima, School

accepted

MD, First of Medicine,

Tohoku May

University

School

4, 1990.

Department 1-1, Seiryo

of Internal machi, Sen-

accepted as a useful index for evaluating preoperative left ventricular systolic function even in patients with regurgitant valvular disease. However, some patients with aortic regurgitation (AR) and mitral regurgitation (MR) whose EF is normal have symptoms of heart failure and poor prognosis in spite of medical treatment and sometimes require valve replacement. This means that these patients have a depressed contractile state despite a normal EF. Recently the left ventricular end-systolic pressurevolume relationship has been used for evaluating myocardial contractility in the isolated canine heartg-l2 because this index is sensitive to the contractile state but insensitive to preload and afterload. Some investigators have applied this index in clini-

Volume Number

120 4

End-systolic

cal studies.13-I5 However, heart size and wall thickness vary among patients with valvular disease, and these may affect afterload in the left ventricle. For this reason it has been recognized that wall stress is preferable to pressure for precise assessment of left ventricular afterload.1-5, l6 Therefore to clarify the discrepancy between the left ventricular contractile state and the EF in regurgitant valvular disease, we calculated the circumferential stress of the left ventricle by means of the thick wall formula of Mirsky,lg and we analyzed the left ventricular end-systolic stress-volume index ratio by means of cineangiography in patients with AR and MR whose EF was 60% or more. METHODS Patients.

A total of 36 patients underwent diagnostic cardiac catheterization in our laboratory from April 1983to July 1987. None of them had significant coronary artery stenosisgreater than 50% luminal diameter, and none had regional hypokinesis or akinesis.Fifteen normal subjects werecatheterized to evaluate chestpain syndromeand past history of ventricular tachycardia, but they had no abnormal hemodynamic parameters. Eleven patients with AR and 10patients with MR were alsocatheterized but had no other valvular lesionssuch as aortic or mitral stenosis.All subjectswere in sinusrhythm, and QRS duration did not exceed0.11second.Long-term medicationsreceived before cardiac catheterization included diuretics, digitalis, calcium antagonists,/3 blockers, and nitrates. Procedures. Cardiovascular medicationswere withheld on the day before catheterization. Catheterization of the right and left sidesof the heart wasperformed in all subjects, and biplane left ventricular cineangiography was performed with 35 mm tine film at a rate of 50 frames/set in the 30-degreeright anterior oblique and the 60-degree left anterior oblique projections with patients in the supine position. Left ventricular pressurewasdetermined simultaneously with cineangiography by meansof an 8F Millar micromanometer-angiographiccatheter retrogradely via the femoral artery in 15 subjects. In the remaining 21 subjectsleft ventricular pressurewasrecordedwith a fluid-filled system just before cineangiography. Forward cardiac output was measuredby the dye-dilution method (DCR 702, Waters Instruments, Inc, Rochester,Minn.). Left ventricular pressure, ECG, and tine film markers were recorded during cineangiographyat a paper speedof 250mm/set to achieve precise synchronization with pressureand tine film. Data analysis. Left ventricular volume was measured frame by frame from the right anterior oblique view by means of the area-length method,i7gl* becauseregional wall hypokinesis or akinesis was absent. The calculated ventricular volume wasindexed by body surface area. Left ventricular wall thicknesswasmeasuredat the midportion of the anterior wall in the right anterior oblique view. Correction factors for each view were derived from the grids. EF (% ) = TSViEDV and regurgitant fraction (% ) =

stress-uolume

ratio in valvular

regurgitation

893

(TSV - FSV)/TSV, where TSV is angiographic (total) stroke volume, EDV is end-diastolic volume, and FSV is forward stroke volume (all in milliliters) determined by forward cardiac output. Systemic vascular resistance(SVR in dynes . sec/cm5) was calculated by the formula: SVR = 80(MAP - RAP)/CO, where MAP is mean systemic arterial pressure(mm Hg), RAP is mean right atria1 pressure(mm Hg), CO is forward cardiac output (L/min), and 80 is a conversion factor. Circumferential wall stress(kdyne/cm2) was calculated by means of the thick wail formula of Mirskyig: Stress = 1.332(Pb/h) (1 - h/2b - b2/2a2+ h2/8a2), where P is left ventricular pressure(mm Hg), a = midwall semimajor axis (L + h)/2, b = midwall semiminor axis (D + h)/2, L is the long axis, h is wall thickness, D is diameter (all in centimeters) ascomputed from the area-length method of the left ventricle, and 1.332 is the conversion factor. Dynamic wall thickness was calculated from the end-diastolic wall thickness with the assumptionthat left ventricular massis constant throughout systole.20 The left ventricular wall stress(y axis) - volume index (x axis) loop wasplotted frame by frame in one cardiac cycle, and end systole was defined asthe point of maximum stress-volumeindex ratio. In subjectswhosepressurewas recorded with the fluid-filled system,end-systolic pressure wasdefined as left ventricular pressureat the time of the aortic dicrotic notch and end-systolic volume was defined asthe smallestone.lv2*13*l4 End diastole wasconsideredto be the midpoint of the QRS complex. Statistics. Correlations betweenthe end-systolic stressvolume index ratio and the end-diastolic volume index relationship and betweenend-systolic wall stressand the EF relationship werefitted by simplelinear regressionwith the least-squaresmethod. Differences in measurementsbetween the groupswere determined with the unpaired t test. Variables were consideredto be significantly different at the p < 0.05 level. All data reported here are given as mean + standard deviation (SD). RESULTS

Fig. 1 shows the left ventricular end-systolic stressvolume index relationship (ESSVR) data from 15 normal subjects. The value of the end-systolic stressvolume index ratio in normal subjects was 5.57 f 0.60 kdyne/cm5/m2 (mean +- standard deviation [SD]), and we defined the range including & 2 SD of this ratio as the normal ESSVR range. Fig. 2 shows the ESSVR data from patients with AR and MR. Six of 11 patients with AR and 5 of 10 patients with MR placed inside the normal ESSVR range, termed AR IN and MR IN. However, the remaining five patients with AR and five patients with MR placed to the right of the normal range, termed AR OUT and MR OUT. The left ventricular contractile state was depressed in these patients whose ESSVR values lay to the right of the normal range. Therefore we subdivided the patients with AR and

894

Nakagawa

et al.

American

50

October 1990 Heart Journal

100

END-SYSTOLIC VOLUME INDEX (md/m’) Fig. 1. Left ventricular ESSVR in 15normal subjects(NORMAL). Two dotted lines show f 2 SD of endsvstolic stress-volumeindex ratio in normal subiects.Thereafter areabetween two dotted lines showsnormal ESSVR range. ”

MR into four groups: AR IN, AR OUT, MR IN, and MR OUT. The following parameters were analyzed to clarify the hemodynamic differences between each of the two groups within AR and MR. General hemodynamic data from the normal subjects and the four subgroups are shown in Table I. Age, heart rate, systemic vascular resistance, and left ventricular end-diastolic stress in the four subgroups did not differ from those of normal subjects, and there was no difference between each of the two subgroups within AR and MR. Fig. 3 shows the left ventricular EF, end-systolic stress, and end-systolic volume index. The EF of the four subgroups did not differ from that of normal subjects, and there was no difference between each of the two subgroups within AR and MR. End-systolic stress of AR IN was higher than that of normal subjects, but the end-systolic volume index of AR IN did not differ from that of normal subjects. The end-systolic volume index of AR OUT was larger than that of normal subjects. However, the end-systolic stress and the end-systolic volume index did not differ between each of the two subgroups within AR and MR. Fig. 4 shows the left ventricular end-diastolic volume index and regurgitant fraction. The end-diastolic volume index of AR IN, AR OUT, and MR OUT was larger than that of normal subjects, and the enddiastolic volume index of OUT was larger than that of IN. In addition, the regurgitant fraction of OUT was higher than that of IN (AR OUT 52.6 f 13.6

versus AR IN 29.7 f 13.3 % , MR OUT 52.9 of: 10.2 versus MR IN 30.2 -t 11.4%; both p < 0.05), although the forward stroke volume index did not differ between each of the two groups (AR OUT 50.8 -t 8.4 versus AR IN 52.5 f 5.4 ml/m2, MR OUT 52.6 f 14.9 versus MR IN 50.7 f 12.7 ml/m2). Furthermore, as shown in Fig. 5, there was a linear inverse correlation between the left ventricular endsystolic stress-volume index ratio and the end-diastolic volume index in all subjects (r = -0.82, n = 36). This finding showed that patients with AR and MR whose end-systolic stress-volume index ratio was low had a large end-diastolic volume index. In normal subjects there was a linear inverse correlation between end-systolic stress and EF (r = -0.91, n = 15), which meant the afterload-shortening relationship. However, as shown in Fig. 6, patients with AR and MR overlapped normal subjects, except for two patients with AR IN who placed to the right and upward in the afterload-shortening relationship. DISCUSSION

We analyzed the left ventricular contractile state in patients who had AR and MR with a normal EF by means of the end-systolic stress-volume index ratio and found that some patients placed to the right of the normal range of the ESSVR. These results indicate that some patients with AR and MR whose EF was normal had a depressed contractile state. This suggests that the end-systolic stress-volume index

Volume Number

120 4

End-systolic stress-volume ratio in valvular regurgitation

0 :

:;i

.' F.." ::

;'

... . .’

.b

0 :.. :

.

895

. :. : :

0

AR (n=ll)

0 0

: :

_’

0

50

100

END-SYSTOLIC VOLUME INDEX (mE/m’)

50

100

END-SYSTOLIC VOLUME INDEX (m@/m’) Fig. 2. Left ventricular ESSVR in 11 patients with aortic regurgitation (AR) in upper panel and 10 pa-

tients with mitral regurgitation (MR) in lower panel. Six patients with AR and five patients with MR placed inside normal ESSVR range, termed AR IN and MR IN. However, remaining five patients with AR and MR placed to right of normal ESSVR range, termed AR OUT and MR OUT.

ratio may be more reliable than the EF for evaluating the left ventricular contractile state in patients with AR and MR. All subjects in the present study had a normal EF, which was 60% or more. EF is now the most widely used value for clinical measurement of left ventricular systolic function, because it is easily determined by invasive and noninvasive techniques and requires

no measurement of pressure or absolute value of diameter or volume and therefore is dimensionless. However, ejection phase indices such as EF and mean velocity of circumferential fiber shortening are affected by preload and afterload.21p 22In valvular heart disease preload is altered; it is decreased in mitral stenosis and increased in AR and MR. Afterload is also altered; it is decreased in MR because of left

696

Nakagawa

et al.

mu IN

200

g

I

s: Ef 50-

t:

2

ii

it? L 00 “E

z -I OT

;s

23 100 z;o”

p 0

NORMAL

AR

G i2

0

MR

AR

MR

50

z

is w

Y

100

?E 0’

>i3 ?

0’

October 1990 Heart Journal

*P