Download Stroke Volume

Use of Ultrasound to Measure Left Ventricular
Stroke Volume
By HARVEY FEIGENBAUM, M.D., ADIB ZAKY, M.B.,
AND
WILLIAM K. NASSER, M.D.
With the technical assistatuce of Charles L. Hainie
Downloaded from http://circ.ahajournals.org/ by guest on April 28, 2017
SUMMARY
Several empirical observations made during ultrasound examinations for pericardial
effusion have led to the possibility that this diagnostic technique might be used to
measure left ventricular stroke volume in man. The hypothesis that left ventricular stroke
volume was proportional to the amplitude of an echo originating from a portion of
the left ventricle near the mitral ring (MREa) times the distance between the echoes
from the anterior and posterior ventricular walls (LVD) lhas been validated.
Ultrasoutnd examiniation.s were performed simultaneously with cardiac output determinations using the direct Fick method on 16 patients proven to have competent
mitral and aortic valves. Correlation between the two methods of measuring left ventricular stroke volume was excellent (r = 0.973; P < 0.001). WVhen ultrasound measurements were used in the regression equation to predict stroke volume, the calculated
values were within 11 ml or 15% of those determined by the Fick method.
The fact that an ultrasound examination is simple, can be done as a bedside determination in a matter of minutes, and is totally harmless to the patient make it a promising diagnostic procedure. The measurement of left ventricular stroke volume represents
another addition to the growing list of medical uses for this intriguing technique.
Additional Indexing Words:
Cardiac outptut
Ultrasouin(d myocardial function
Echographv
when we were attempting to diagnose pericardial effusion.2 On correlating the location
and pattern of this echo with those of echoes
originating from the cage of Starr-Edwards
mitral valves,4 this echo appeared to be originating from a portion of the left ventricle
near the mitral ring or annulus and, thus for
convenience, was designated as the mitral ring
echo (MRE)." A similar echo has been recognized by other investigators, and they too
believed that it originated from an area near
the mitral ring or atrioventricular wvall.. Upon
further investigation of the MRE, it was noted
that the curve which it inscribed was surprisingly similar to known left ventricular
volume curves with the amplitude varying
directly with left ventricular stroke volume
and inversely with left ventricular volume."
Edler5 likewise noted that the amplitude of
the atrioventricular wall echo varied with
different disease states. It thus seemed pos-
IN THE COURSE of earlier studies using
diagnostic ultrasound to detect the presence of pericardial effusion,1 2 several observations were made which suggested the
possibility of using this technique to measure
left ventricular stroke volume. The first of
these observations was the finding of a consistent and strong ultrasound signal or echo
near the echo from the anterior leaflet
of the mitral valve' :1 (fig. 1D). This new
echo was initially of interest because it was
confused with the true mitral valve echo.Secondly, this echo offered some difficulty
From the Departmient of Medicine, Indiana University School of Medicine, and the Krannert Heart
Research Institute, Marion County General Hospital,
Indianapolis, Indiana.
This investigation was supported in part by the
Hermian C. Krannert Fund, U. S. Public Health Service Grants HE-09815-02, HE-6308, HTS-5363, anid
HE-5749, and the Indiana Heart Association.
1092
Circulation. Volune XXXV, June 1967
LEFT VENTRICULAR STROKE V
OOLUME
10)93
there is some functional relationship of MREa
to left ventricular stroke volume and left ventricular volume, this ventricular dimension
possibly could be substituted for left ventricular volume, and thereby MREa could be used
to calculate left ventricular stroke volume.
This study attempts to test the validity of
these observations and assumptions by using
the amplitude of the mitral ring echo and
the distance between the anterior and posterior heart wall echoes to calculate left ventricuilar stroke volume in the intact hulman
sull)ject.
Downloaded from http://circ.ahajournals.org/ by guest on April 28, 2017
Method
Stuidies wer e donie oni 16 patienits uniildergoinig
diagnostic heart catheterizationi. The patients inchlidled botlh nmales awl females, aniid their ages
ranged firom 15 to 61 years. No patienit lhad any
eviden-ice of mitr-al or aortic inisuifficiency. In 15
of the patients valvuilair insuifficienicy was excluded
by left hear-t catheterization aind selective cineFigure I
Fourt cardiac
echoes in a nonimal indiividuial. Aniterior
is at the top of eaclh tiracing A and B illu-strate echloes
froml the poster ior- hleart wall, the echo in C is frontI
the
ainterior leaflet of
the
mitral valve, andl the strong
in1ovin1g ech1o in D i.s front a portion of the lef tventr-icle near th7e mnitydl ring (AIRE). (FrJ omtI Feigen1batmttn,
II., Zaky, A., anid Waldhatusen, J. A.: Anni IniternlMeci
65: 443, 1966, bl) periniission of the Ametican College
of Ph ysicia its.)
sible that then
'-
miglht
be
some
fiunctional
relationiship amonig the MRE amplitude
(MREa ), left venitricuilar voluime, anid left
ventricuilar stroke volume.
The seconid observation which made this
sttidy feasible was that the distance between
the anterior chest wvall anid the echo originating from the posterior xvall of the heart wa.s
rouighly proportional to the size of the left
ventricle
(fig. 2). The correlation betwveen
this tiltrasouind measurement and the clinical
estimation of left ventricular volume improved
eveIn fuirther when a more accurcate left ventricular dimenision was obtained by learning
how to record the faint aniterior heart wall
echo in most patieiIts wvith or without pericardial effusion" (fig. 3). Thuis, asstuming that
2
Circulation,
Volume
XXXV, June 1967
angiocardiography. The one patient who lhad
only riglt lheart cathleterization had nio heart muirmul.urs and was beinig stuidiedI to evaluate a recent
ptulmoar-y einbolus.
Thie principles of diagnostic tultrasotunid lIave
leen reviewed in previouis papers.7-10 The uiltrasouniid exatminiation-s were perfofrmed with a comm-tercially available uiltrasonioscope utilizing a 2.25mc, 0.75-inichl transducer withl a repetitioni rate
of 1,000 imptulses per secon-id. For recordinig the
echoes, the "slow sweep" or "time-motion" presenitatioIn was used whlereby dlistanice is plotted
against time and mllovinig echloes are plotted as
wavvy lines. The patients wveie examined in the
recumbent positionl and AqLuasonic gel was used
as a} coupling medliuim to enisuir-e airless conitact
between the trainsduceI and the skin. To record
the anterior anidl posterior lheart wall eclhoes, the
transducer was placed alonig the left sternal bordler at the four-th in-tercostal space anid pointed
(dir-ectlv posteriorly (fig. 4). Jtust as is done when
excimininig for per icardial effusion, the gain,
clamping, and reject modalities, as well as the
exact positioining of the tran-sducer-, were adjusted
to record the sin-gle most dominan-t echo originatinig from the viciniity of the posterior wall of
the heart. In addition, the gaini conitrolling the
iiear field was adjusted to record the nonimoving
aniterior clhest wall eclhoes plus the fuizzy tlhin
echo juist immediately posterior to the chlest wall
echoes. In maiiy patienits, despite its somewhat
inidistinct chiaracter, this anter-ior heart wall echo
couild be seeni to move in a directioni opposite
to the posterior lieaut wall (fig. 3).
1094
FEIGENBAUM ET AL.
T
Downloaded from http://circ.ahajournals.org/ by guest on April 28, 2017
P
Figure 2
ch (F) in a 110a1001 subject (A) and in a patient twith a dilated left
Posteil hteai watll
f
clif
centr-iclc (13). T t pi-esents tlhe ccho originating fr-onti the tt-ansdtcer. (A) The posterior hieart
naill echo (FP) is 10 cot frorn the transAdcer. (B) The ce/Io P is nicarly 14 crtt fronti ecllo T.
(I"tI out Jcigenbanitn II., Zakoy, A., and Waldhatsicn,' J. A.: Ann Itntern Mecd 65: 443, 1966,
b1)1 pO tnitsusiott of the Antet-io-an Co/lage of Physicians.)
The echo originiatinig from the mitral rinlg
(NIRE) was obtain-ed by placinig the transducer
_s_s_
~~AW_
Pw.
Figure 3
Anttcrito (AW) t(d po)sterni(r (PW1) lIcart trall eclcoe.s
in a tItorttIal stbiject. The ciho ftom tite aniterior heart
a
trall
is
faittt arndl
jitst
posterior to the ec/Ioes origirnalt-
frottl te
ah tteti ioI chtest waill. (Fr1omti Feigernbantn,
H., Zakil, A., at(l (Girabhottn, L. L.: Circttdationt 34:
6i11, 1966, bhy' peitmission of Atterican Heart A.sns.,
itig
1Inc.)
over the apex of the heart an2d pointing the
transducer posteriorly, mediallv, anid a little
superiorly (fig. 5 ) The exact direction was adjitsted to recor-d the strong echo wlich was just
posterior to the imnitral valve eeho and which hadl
a characteristic patterin of motion. The MRE
inove(l toward the tr-anisduticer witl-h ventricular
svstole an(d away from the t -anisduicer- with diastole. The diastolic motioni was divided inito
three plases: rapid iiovement eairly in diastole,
eitlher less rapid or no motioni in mid-diastole,
and theni a-pidI movemeent following atrial systole
(fig. 5). A simuiltaneous electrocardiogram aided
in identificationi of all eclhoes. B3y placinig the transduicer over the apex of the heart, the maximal
aimiplitudle of the MRE was recorded." This appr-oach also lhelped to standardize the examination
by allowinig for dillerences in. heart size fr-om
one patienit to) another.
The uiltrasouniid recor-dings were obtained simutltaneously with the determiniation of cardiac
output ais ml-easuri-ed by the direct Fick method.
A simultatneouls electrocardiogranm oni a standard(
strip-chart recorlder was use(l to ineasuire the
Circultion, Volumie XXXV, June 1967
LEFT VENTRICULAR STROKE VOLUME
01()95
Downloaded from http://circ.ahajournals.org/ by guest on April 28, 2017
):
Figure 4
Diagramit oar! actu1al u-iltrasountdftracing illu,strfatiiig the mitetIodi byl whlich the left ventricular
C
and posterior wealls
diareter (LVD) is obtainiedl. CWV = ch5est twall; AWV antd PW1 = anterior
of left venitricle; TI- echo froni transducer.
heart rate. In the thlree patients witlh aitriIa]
fibrillatioIn, ain effor-t waVis miale to utilize those
ultrasouind measurements wlhich followed anm B-B
interval corresponding to the averalge hear t rate
during the determination- of cardliac outpuit.
Results
The ultrasound measurements which vere
uised to determine left ventricullar stroke volume (LVSV) vere left ventricular diameter
(LVD), which wvas the distance betveen the
echoes from the anterior and posterior lheart
xvalls duiring diastole (fig. 4), and the mitral
ring echo amplitude (MREa), which wits the
total amplitude of motion inscribed by thc
mitral ring echo from the beginning of systole
to the beginninig of diastole (fig. 5).
The formula used to calculate LVSV was
derived primarily from empirical findings.
As stated before, MRlEa seemed to vary with
stroke volume for a given left ventricuilar
Circulalion, Volumue XXXV, June 1967
end-diastolic volume (LVV). If LVV were
increased an1d LVSV were niot chaiiged, then
MREa decrease-J. Thus, MLIEa appeared to
vary directly with LVSV and iniversely with
oV r LVSV-MIEa
LVV, that is, M IEa- L'NV
x LVV. By using LVD as a function of LVV,
the relatively simple formula IVSV-MREa x
LVD xxwas employed. Whien M I
x x D xvas
plotted against the stroke voltume, as determine(l by the Fick method, tlere was a
straight line relationship betxveen the txvo
valuies (fig. 6). The correlation coefficient
xvas r 0.973 (P <0.091) xvith a regression
ecqtiatiunvi = 87.5 + 10.40 (x 8.9). Table
1 shoxvs the results of uisinag this regression
equiation to predict stroke volume from the
ultrasound measurements (SV,,). The greatest disc repancy between the stroke volu-mes
Downloaded from http://circ.ahajournals.org/ by guest on April 28, 2017
sS_^;~
FEIGENBAUM ET AL.
1096
t- J~~~~~~~~~g
A~~~~~~~~~A,
Figure 5
ho
I)iagran and tiltrasountld recorlin, demonstatinlgio
aimiplittide (AIREa) is obtainiedl.
V9.
M.0
,E
s.ol
LVD x MREa
Figure 6
Stroke volulme as determined byl the Fick mnethod
plotted against the prodtuct of tire tultrasouind mleastrem?wents, LVD X MREa.
calculated by the Fick and the tiltrasotind
methods was 11 ml or 15%.
Discussion
Many attempts have been made to measure
left ventricular stroke voluime by utilizing
theinitral ring e((ho (AIR1 ) and( its
Various (liagniostic procedures. Probably the
most widely accepted methodl is measturing
the cardiac otutput by either tlhe Fick6X or the indicator-diluition- technique and theni dividing
the cardiac outpuit by the heart rate. This
method has mainy obvious disadvantages.
First, right heart catheterization is niecessary.
Secondly, this technique caininot l)e used to
measure truie stroke volume in the Presence
of mitral in.suifficiency or aortic insufficiency.
Thirdly, this teclhniqtue (loes niot allow for
heat to beat measurements of stroke volume.
Direct measuirements of left ventricular volume by either the angiographic' I 1 2 or inr
dicator-dilution techniquies1; are more recent
procedures uised to measiure left ventricular
stroke volume. These techniques again require cardiac catheteriizatioD. The anigiographic proceduire possesses severail inherent objections. It introduces a foreign substance
which may alter the hemodynamic situation;
the technique offer.s som-ne risk to the patient;
(rculhaion, Vo/lume XXXV. June 1967
LEFT VENTRICULAR STROKE VOLUME
I'l
--o
1097
co
't
o-
c
X
i In CS -o
oo
++++++++I
10
I
10 10
co r- I- c
00
*li'
00
'I"
t-
oo OO N
N
+++++++I
+±
-f
c
++I
iD_co
I>t
>
10 't 0 1
r-Kt cscs
ao
+tl
CZ
00CD
-
cs
UC
Cl)
>
:rA
O)10 M
00
-- t
m
5
ta-
_-
--
*0
Downloaded from http://circ.ahajournals.org/ by guest on April 28, 2017
N
Iin
00 CO 0
Co 00 m
-_
._
10
° 'It )n
in C
CS
oo t :0oob b
o
cr
t-
aQ
I'l
0
O
E
cCC
:6
10 00
+1
CO
~6
1
CZ
1--
"-.
.
x
.
1)'t
c
VO
.
.
c
r-
1-
+1
.
"t1
)1
cor
00
0~ U
co
a
11CZ o6
CO
r--l
CZ
P-10010)0)1~cq-00
0.
cu
CZ
00
-
s>E
04
C)
't O
CO
.
1
.
.
C)
"t C0
.
-0t-N -t
--
+1-0 -,t
o6 6
t-
Ct
t0
CZ
coC 000 00 0001
CS O o6 o6 o6
c6
Ci)
CA
CO
ozQ._ E
m CZ E.
t ¢
*
C
,.O
.
S
X
c
C
o 0
n
t
cs
cs _cs
CZ
b
C
cO
;Z 0
_
K
t
CZ00
C")CIDin
0O=
0C,
-0 0
c:
CA
CZ
10
'C
01C)-"T
Clo
co:)
H-
CIO
Circulation,
Volume
XXXV,
June
1967
'
01
oot_nm
m
.0
1CbO
C
_t
r
rn cO 0)
Cc
r-
c
/
Uw=P4P
;4
;
00
O0)0~= -C01Nco "t10CC0
-,uw-~~-;
,-
--
r-q
r-q
r--
r-
O
FEIGENBAUM ET AL.
1098
Downloaded from http://circ.ahajournals.org/ by guest on April 28, 2017
it does not lend itself to many repeated examinations, and the actual calculations are
tedious and time consuming.4 In addition,
both the angiographic and indicator-dilution
techniques are still being questioned as to
their accuracy in truly measuring left ventricular volume.-'
The results of the present study indicate
that diagnostic ultrasound with its many inherent advantages may provide another way
of measuring left ventricular stroke volume.
As stated in many previous publications,7 10
ultrasound in the power ranges used for diagnostic purposes is entirely innocuous to the
patient. The technique used in this study is
simple and the equipment is portable. Thus,
the examination can be performed as a bedside procedure in a short time. The ease of
the examination is borne out by the fact th-at
for each patient in this study the ultrasound
recordings were obtained within the 4 minuites it took to measure the cardiac output
by the Fick method. The close correlation
between the calculated left ventricular stroke
volume and the expected left ventricular
stroke volume as measured by the Fick method indicates that the accuracy of this ultrasound technique is probably as good as, if not
better than, any of the other currently available methods.
The advantages of a simple, accurate
method for measuring left ventricular stroke
volume are numerous. First of all, when such
a technique is combined with a Fick cardiac
output, theoretically a quantitative measure
of mitral or aortic insufficiency can be ob-
does represent some function of left ventricular volume and might prove to be of value as
a qualitative or semiquantitative measure of
left ventricular volume.
It is somewhat difficult to explain entirely
why these two relatively simple ultrasound
measurements should be directly related to
left ventricular stroke volume. The logical tendency is to try to fit the measurements into
some mathematical formula used to measure
left ventricular volume. Obviously there are
too few data to justify such an attempt. In
fact, one cannot entirely justify placing cm"
or ml after the ultrasound calculations. Left
ventricular diameter is actually measured in
cm, as is MREa, thus the answer should really
be in cm.2 The only justification one might
have to use ml is that LVD supposedly represents a function of left ventricular volume
in ml, and MREa is a function of how much
left ventricular volume changes with ventricular systole. In any case, for the time
being the findings must be considered entirely empirical. It is possible that future
attempts to explain these findings may yield
valuable information concerning the shape of
the left ventricle and the manner in which
this chamber ejects blood.
tained. In patients without mitral or aortic
valvular incompetence, this technique provides a bedside method for measuring cardiac
output. In addition, since the formula LVSVMREa x LVD seems to be valid, the fact that
the curve inscribed by the MRE is similar to
left ventricular volume curves is probably
more than coincidental. Thus, the recording
of this echo may lend itself to hemodynamic
studies involving changes in left ventricular
volume. The validation of this formula also
indicates that the ultrasound measurement
of left ventricular diameter (LVD) probably
3. ZAKY, A., GRABHORN, L., ANDI) FEIGENBAUM,
H.: Movenment of the mitral ring: Study in
References
1. FEIGENBAUM, H., WALDHAUSEN, J. A., AND HYDE,
L. P.: Ultrasouind diagnosis of pericardial effu-
s,ion. JAMA 191: 107, 1965.
2. FEIGENBAUm, H., ZAKY, A., AND WVALDHAUSEN,
J. A.: Use of ultrasound in the diagnosis of
pericardial effusion. Ann Intern Med 65:
443, 1966.
ultrasoundcardiograpby. Cardiov Res Cen Bull.
In press.
4. GIMENEZ, J. L., WINTERS,
V. L., JR., DAVILA,
J. C., CONNELL, J., AND KLEIN, K. S.: Dynamiies
of the Starr-Edwards ball valve prosthesis:
A cinefluorographic and ultrasonic study in
humans. Amer J Med Sci 250: 652, 1965.
5. EDLER, I.: Examination of the Cardiac Patient,
edited by A. A. Luisada. New York, McCrawHill Book Co., 1965, p. 4.
6. FEIGENBAUM, H., ZAKY, A., AND GRABHOIR,
L.: Cardiac motion in patients with pericardial
effusion: A study using reflected ultrasound.
Circulation 34: 611, 1966.
(irculation, VZolime XXXV. !nne I96-
LEFT VENTRICULAR STROKE VOLUME
1099
7. EDLER, I., GUSTOFSoN, A., KARLEFORS, T., AND
12. CHAPMAN, C. B., BAKER, O., MITCHEL, J. A.,
AND COLLIER, R. G.: Experiences with a cinefluorographic method for measuring ventricular volume. Amer J Cardiol 18: 25, 1966.
13. HOLT, J. P.: Indicator-dilution methods: Indicators, injection, sampling and mixing problems
in measurement of ventricular volume. Amer
J Cardiol 18: 208, 1966.
14. DODGE, H. T., SANDLER, H., BAXLEY, W. A.,
AND HAWLEY, R. R.: Usefulness and limitations of radiographic methods for determining
left ventricular volume. Amer J Cardiol 18:
10, 1966.
15. SOLOFF, L. A.: On measuring left ventricular
volume. Amer J Cardiol 18: 2, 1966.
CHRISTENSON, B.: Ultrasound cardiography.
Acta Med Scand 170 (suppl 370): 67, 1961.
8. JOYNER, C. R., AND REID, J. M.: Application of
ultrasound in cardiology and cardiovascular
physiology. Progr Cardiov Dis 5: 482, 1963.
9. FEIGENBAUM, H.: Diagnostic ultrasound. Ann
Intern Med 65: 185, 1966.
10. SEGAL, B. L., LixOFF, W.,
AND
KINGSLEY, B.:
Echocardiography: Clinical application in mitral stenosis. JAMA 195: 161, 1966.
1 1. DODGE, H. T., HAY, R. E., AND SANDLER, H.:
Angiocardiographic method for directly determining left ventricular stroke volume in man.
Circulation Research 11: 739, 1962.
Downloaded from http://circ.ahajournals.org/ by guest on April 28, 2017
Elementary Approaches to Thrombosis
Various medical scientists look at thrombosis from different points of view. Only one,
the clinician, studies the whole patient in whom a symptomatic clot resides....
The pathologist notes that when a vessel is injured, a clumping of platelets quickly
develops at the site of injury. Under the electron microscope, he can see that the
clumped platelets tend to lose their "dense granules," perhaps releasing their clotpromoting contents.
The biochemist is interested in the chemical attraction between platelets and injured intima and the subjacent collagen; he tends to implicate the chemical, adenosine
diphosphate. The biochemist has made significant advances in the chemistry of prothrombin and its active enzyme, thrombin, as well as in the biochemistry of the
fibrinogen-fibrin reaction.
The physiologist is interested in clots that form in living vessels. He notes that undamaged vascular intima is electrically negatively charged, repelling platelets that are
similarly charged. When a vessel is damaged, its electronegativity disappears or may
be replaced by an electropositivity, which tends to attract platelets to the injured site.
The biophysicist is concerned with two rheologic principles as they may relate to
vascular thrombi. First, fluid moving through a tube generates a negative "streaming" potential at the surface of the tube. This exaggerates the normal negative charge
of the intima and is dependent upon the flow rate of the blood. Further, the flow rate
in very small tubes, such as capillaries, varies with the fourth power of the caliber
of the tube. If the diameter of a tube is reduced by one half, the flow rate diminishes
to one sixteenth.-CHARLEs A. OWEN, JR.: Hypercoagulability and Thrombosis. In Proc
Mayo Clin 40: 830, 1965.
Circulation, Volume XXXV, June 1967
Use of Ultrasound to Measure Left Ventricular Stroke Volume
HARVEY FEIGENBAUM, ADIB ZAKY, WILLIAM K. NASSER and Charles L.
Haine
Downloaded from http://circ.ahajournals.org/ by guest on April 28, 2017
Circulation. 1967;35:1092-1099
doi: 10.1161/01.CIR.35.6.1092
Circulation is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright © 1967 American Heart Association, Inc. All rights reserved.
Print ISSN: 0009-7322. Online ISSN: 1524-4539
The online version of this article, along with updated information and services, is
located on the World Wide Web at:
http://circ.ahajournals.org/content/35/6/1092
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles
originally published in Circulation can be obtained via RightsLink, a service of the Copyright
Clearance Center, not the Editorial Office. Once the online version of the published article for
which permission is being requested is located, click Request Permissions in the middle column of
the Web page under Services. Further information about this process is available in the Permissions
and Rights Question and Answer document.
Reprints: Information about reprints can be found online at:
http://www.lww.com/reprints
Subscriptions: Information about subscribing to Circulation is online at:
http://circ.ahajournals.org//subscriptions/