132

Evaluation of Left Ventricular Function in Patients with Aortic Regurgitation Using Afterload Stress JAMES L. BOLEN, M.D., EARL L. HOLLOWAY, M.D., JULIAN C. ZENER, M.D., DONALD C. HARRISON, M.D., AND EDWIN L. ALDERMAN, M.D.

SUMMARY Left ventricular function was assessed in 14 patients with chronic aortic regurgitation by afterload elevation with angiotensin. Seven of 14 patients maintained their resting ejection fraction with angiotensin (group A), while the remaining seven experienced a decline of greater than 0.10 in ejection fraction (group B). Six of seven group A patients showed an appropriate rise in left ventricular stroke work index in response to the angiotensin-induced rise in left ventricular end-diastolic pressure. In contrast, six of seven group B patients showed abnormal, flat, or declining stroke work indices. Included in the seven group B patients were two patients with left ventricular dysfunction secondary to coronary artery disease. The five other group B patients, who did not have coronary disease, exhibited similar stressinduced ventricular dysfunction, despite the absence of any significant resting hemodynamic differences from patients in

A. These five stress-induced dysfunction patients were distinctive from patients who maintained their ejection fraction level in that the former all had regurgitant fractions of greater than 0.50, whereas all group A patients had regurgitant fractions of less than 0.50. Similarly, these five stress-induced dysfunction patients had significantly larger left ventricular end-diastolic volumes than did the group A patients. These data suggest that patients with pronounced aortic regurgitation measured in terms of regurgitant fraction greater than 0.50 and left ventricular end-diastolic volume of greater than 160 cm3/m2 exhibit impaired ventricular function if appropriately stressed. As most of the patients with stress-induced dysfunction had a normal ejection fraction at rest, it may be that stressinduced dysfunction represents a stage before overt resting dysfunction and cardiac failure.

O EVALUATION OF LEFT VENTRICULAR (LV) FUNCTION in patients with aortic regurgitation remains an important clinical problem. There exist no firm physiologic or hemodynamic criteria to indicate at what point in time myocardial dysfunction develops or at what time valve replacement should be performed in order to avert or reverse myocardial dysfunction. Spagnuolo et al. found that the triad of marked radiographic evidence of LV enlargement, electrocardiographic diagnosis of ventricular hypertrophy, and wide pulse pressure identified patients at high risk for the development of congestive heart failure, angina, or death.' However, these data are of limited assistance in assessing indications for aortic valve replacement in asymptomatic or mildly symptomatic patients who exhibit some of the features of this triad. The severity of aortic regurgitation as measured by large regurgitant fractions2 and markedly elevated end-diastolic volumes2 3 correlates with the severity of clinical symptoms, but has not been correlated with quantitative measures of myocardial dysfunction. Left ventricular end-diastolic pressure has not been found to be a good index of LV dysfunction in that it correlates poorly with end-diastolic volume and ejection fraction in patients with aortic regurgitation,3 and may be normal in patients with advanced disease. Depression of forward cardiac output in patients with

regurgitation reflects very advanced disease with regurgitation and overt congestive heart failure.4 Gault et al..5 in a small number of patients studied pre- and postoperatively, found that a lowered velocity of circumferential fiber shortening (VCF), indicative of abnormal myocardial function, was not reversed by aortic valve replacement, suggesting that resting abnormalities might not be reversible by surgery. We have therefore studied left ventricular function during stress induced by augmenting afterload with angiotensin in a group of patients who at rest had relatively normal indices of myocardial function. We have examined the resultant hemodynamic and quantitative volumetric responses and found that stress abnormalities of left ventricular function can be detected in some patients, particularly those with the most severe degrees of regurgitation. Methods Fourteen consecutive patients fulfilling the clinical criteria of chronic isolated aortic regurgitation and who agreed to participate, with informed consent, comprise the study (table 1). Patients with concurrent mitral valve disease or aortic stenosis (> 10 mm Hg mean left ventricular to aortic gradient) were excluded. Resting cardiac catheterization was performed after premedication with 100 mg secobarbital intramuscularly. Cournand or Birdseye catheters with Statham P23Db transducers were employed for right-sided pressures, and 6.7 French polyethylene end-hole catheters with Micron MP- 15 transducers or Millar transducer-tipped catheters were used for the left-sided pressures. Forward cardiac outputs (COF) were determined either by the direct Fick method or duplicate indocyanine green indicator dilution techniques. Pressures were computer determined6 and checked manually. Resting single plane angiography was performed using a 6.7 French angiographic catheter or #8

From the Cardiology Division, Stanford University School of Medicine, Stanford, California. Presented in part at the American College of Cardiology meetings in Houston, Texas, February, 1975. Supported in part by NIH Grant No. HL-5866, Program Project Grant No. I-POI-15833, and a grant from the Bay Area Heart Research Committee. Address for reprints: Edwin L. Alderman, M.D., Assistant Professor of Medicine, Cardiology Division, Stanford University School of Medicine, Stanford, California 94305. Received March 28, 1975; revision accepted for publication July 23, 1975.

group

severe severe

Downloaded from http://circ.ahajournals.org/ at CONS CALIFORNIA DIG LIB on June 25, 2015

VENTRICULAR FUNCTION IN AORTIC REGURGITATION/Bolen et al.

133

TABLEx 1. Clinical Data NYHA Class

Pt.

Age/Sex

1 2 3 4 5 6 7 8 9 10 11 12 13

31/F 58/M 46/M 36/M 35/M 49/F 63/M 49/M 27/F 39/M 49/M 43/M 47/M

III I III I I I III II III II II II III

14

63/M

III

Valvular etiology

RHD, ?SBE Bicuspid RHD U U RHD, SBE Leutic RHD RHD, SBE U U RHD Ankylosing spondylitis Aortic sclerosis*

ECG LVH

Chest X-ray LVH

+ + + + + + + + + + + +

+ + + + + + + + + + + +

+

+

Coronary arteriograms

ND 50% distal RCA 50% mid-RCA ND ND ND Normal Normal ND ND Normal ND 75% proximal RCA 50% LAD Diffuse 3-vessel disease*

*Found at postmortem examination. Abbreviations: NYHA = New York Heart, Association classification; ECG = electrocardiographic; LVH - left ventricular hypertrophy; RHD = rheumatic heart disease; SBE = subacute bacterial endocarditis; RCA = right coronary artery; LAD = left anterior descending coronary artery; U = unknown or uncertain; M = male; F = female; ND = not done.

French Millar angiographic catheter passed retrogradely across the aortic valve with the patient in the 300 right anterior oblique position. End-diastolic and end-systolic LV volumes and ejection fraction were determined by light-pen computer processing of video disc recorded LV images as previously described.7 Three sinus systolic-diastolic cycles, excluding postextrasystolic beats, were analyzed and averaged. Left ventricular cardiac output (COLV) was determined by multiplying angiographically-determined stroke volume by heart rate. Heart rate at the time of the angiogram and at the measurement of forward cardiac output were within eight beats of one another. Regurgitant fraction (RF) was then calculated according to the formula COLV

-

COF

RF =

COLV Left ventricular stroke work index (SWILv) was calculated using the formula: SWILV = (LV stroke index) X (mean LV systolic pressure during ejection) X (0.0136). After resting hemodynamics and LV angiography were performed, LV pressure was monitored until returning to baseline levels. Angiotensin was then infused through a peripheral vein, beginning at a rate of 0.4 ,ug/min and increasing 0.4 ,ug/min every three minutes until LV systolic pressure was raised by 20-50 mm Hg. Doses of angiotensin required were 0.4-2.0 ,g/min. Hemodynamics were then allowed to stabilize for ten minutes, at which time LV pressure was recorded. In nine of the 14 patients, simultaneous aortic pressure was also measured, and in six, forward cardiac output was determined during angiotensin infusion (table 2). While angiotensin infusion was maintained, the LV angiogram was repeated and quantitative volumetric data, ejection fraction, and SWILV were calculated for angiotensin (table 2). At least 30 minutes separated this angiogram from the first. There were no complications in performing this second angiogram. Data analysis was performed by computation of mean and standard error of the

mean (± SEM). Analysis of paired data and statistical significance was evaluated by means of the paired Student's

t-test.

Results Clinical Data

Clinical data are presented in table 1, and hemodynamic data in table 2. Patient ages ranged from 27 to 63 years (mean 45 years). Clinical disease severity, as assessed by New York Heart Association (NYHA) classifications, was Class I or II for eight patients, and Class III for six patients. Five of the patients had had rheumatic fever in the past, and two of these had had subacute bacterial endocarditis at some time remote from the present study. One patient had ankylosing spondylitis, and one syphillis, as probable etiologies for their valvular disease. Thirteen of the 14 patients had left ventricular hypertrophy by radiographic and electrocardiographic criteria (table 1). Patient 14 died suddenly six months after study and was found to have aortic valve scarring and severe arteriosclerotic coronary disease with diffuse myocardial fibrosis. Five patients had coronary arteriography because of either chest pain or age. Three showed coronary narrowings, and only one had significant multivessel disease (patient 13). All patients except 4, 6, and 14 had aortic valve replacement; the coronary arteries were inspected and palpated at this time and, excluding patient 13, were judged to be free of major disease. Response to Afterload Elevation

Left ventricular systolic pressure was raised for all patients by an average of 24% (148 to 183 mm Hg) and LV end-diastolic pressure by 65% (24.1 to 39.6 mm Hg) (table 2). There was, however, no direct correlation between peak LV systolic pressure elevation and the rise in LV enddiastolic pressure. Forward cardiac index was depressed by angiotensin in each of the six patients in whom this was determined. Regurgitant volume and regurgitant fraction in-

Downloaded from http://circ.ahajournals.org/ at CONS CALIFORNIA DIG LIB on June 25, 2015

Vol. 53, No. 1, January 1976

CIRCULATION

134

66

6666

I

I

66

I

.)

0D 0_

Lo e ko 6 666

LO

6

tz

= "IC 6~~~ 66

co VD IZ 67

110

coo cO 00

o0 o

o.-to

to to co to o 0o o0

N

00

C 01I 0 to o~ cvto NE:0 Lo oo 00 C< to0'n

N 0 t c 00000s0 O-OLOLOO0N ioc 0 00000 00 O04 C1 CO

1'0 t-

L1O tO 01i

0 to 00 '~4 N ~0 tO 01 00 ---q C-A -

0 o

0

.t

r + m

N m

m ":t llq m CeD

r-

?--

?-

- 0 0 C. QC 0 00 to t tLo "to to -o N to to N 0 0t l X

0000

.0

di o

'O

00 N

C9 00000 c0 O0

lt

1

I It Cq

t- c

.ce 0)0 s -0)S 0..-

C

C) D

t

In0 z 0 oIn c00 1 00 oc c0 t4 m Cq N. Cl

N 0 ro 000 0

c t 1C0 Nt0 01 0

to

0101

to to

00 c-ic C'iq -: C6 00

_n-

r N

O~ 0

II

0

0 co C0 0 ' o N-~

-4H

00 :j 00 '-4r--4t

00 Cl Cr) 010 00 CC

T-

C T-

.0

C0'i 1

N0 to '-

to 0 00 00

v

0 n-

0)O

M C1.C --qci

-0C t--

O *_

r

0)

-

s

CS --p

0

010b- 0_0 _ 1_ N0 COCOC'00 N 00000000100 ~0~ -

a

O0

010e0

0

b to'

01-

4C)

N 0 '~0 00000 ~0 0 t~- tLO O 0-'

000 c

O0

ON

0C0

0

0 D0

to

000

00000

'-

000C ^

4

CnZ

C1

-4

-4

-4

-4 14

000 o 01 NV 0b c0

00 to00000c - C4 -4 ,-4 ,-4 ,-4 ,-4 ,-4

-

'

tNN'-^

,-.4 ,-- ,-q 01

c30

00 t0

x 0_ o 4 C" t-_*t 0 * 30 CSC CC e 00COO -0N !0 to

rXc)c ~ 0 - to:0 N~ ~ _I s _zz 0

00 _ oc o to

c

s-I eD _

00

00

1

0

0

0

0

-

-

01

01

V¢

Ve

t-

1 4

00 _ _I

- tCQ 010 OC N '-40 '

00N0

N

-

¢