The Echocardiogram of the Anterior Leaflet of the Mitral Valve Correlation with Hemodynamic and Cineroentgenographic Studies in Dogs By GERALD M. POHOST, M.D., ROBERT E. DINSMORE, M.D., JOEL J. RUBENSTEIN, M.D., DENNIS D. O'KEEFE, M.D., R. NATHAN GRANTHAM, M.D., HUGH E. SCULLY, M.D., EDWARD A. BEIERHOLM, M.D., JAMES W. FREDERIKSEN, M.D., MYRON L. WEISFELDT, M.D., AND WILLARD M. DAGGETT, M.D. SUMMARY
The echocardiogram of the anterior leaflet of the mitral valve (ECHO) was compared to hemodynamic and cineroentgenographic data to evaluate its accuracy in timing mitral valve (MV) opening and closure, and to validate it as an indicator of MV motion. The ECHO, high speed cineroentgenography at 250 frames/sec, and/or measurement of intracardiac pressures allowed accurate timing of the events of MV motion in dogs on right heart bypass. The intersection of left ventricular and left atrial pressures in early diastole preceded the onset of rapid anterior motion of the ECHO (D' point) by 17 to 33 ± 7.6 msec; r = 0.98. The onset of left ventricular systole occurred before the termination of final rapid posterior motion of the ECHO in end diastole (C. point) by 25 i 10 msec; r = 0.96. Radiopaque clips were attached to the free edges of both leaflets of the MV. Cineroentgenographically determined plots of clip distance from the ultrasound transducer were morphologically similar to the simultaneously recorded ECHO. A delay of 23 ± 3 (0 to 40) msec was observed in the ECHO peaks of diastolic anterior excursion compared to clip motion. Contrast medium advances beyond the free edges of MV leaflets mixing with left ventricular blood 43 ± 3 msec after initial separation. These cineroentgenographic studies elucidate nonuniformity of leaflet motion responsible for ECHO delays. Thus, ECHO D' and CO correlate well with hemodynamic indicators of MV opening and closure. However, ECHO motion, although qualitatively similar, is unpredictably delayed compared to cineroentgenography of clips on the MV free edge. Since the ECHO correlates well with hemodynamic indices of MV opening and closure, this noninvasive technique can be used as a reference in the timing of intracardiac events and in the determination of systolic and diastolic time intervals.
Additional Indexing Words: Mitral valve opening Angiocardiography
Mitral valve closure Echocardiography
ECHOCARDIOGRAPHY HAS PERMITTED noninvasive observation of a variety of intracardiac structures." 2 One of the most distinctive echocardiographic patterns is that of the anterior leaflet of the mitral valve." 3 4 In 1961 Edler con-
Mitral valve motion
eluded after advancing a long needle into a beef heart parallel to the ultrasound transducer4 that this characteristic echocardiogram originated from the anterior leaflet. Further documentation of the origin of this echo has been provided by injection of indocyanine green dye into the left ventricle during ultrasound recording.5 While studies relating echocardiographic mitral valve motion to hemodynamic data have been accomplished,6 10 most were performed in the presence of mitral valve disease or left ventricular dysfunction. Few studies are available relating normal mitral valve motion to the echocardiographic pattern.4 Furthermore, it is not known whether the mitral valve echocardiogram is capable of accurately defining opening and closure, although this has been suggested in a number of clinical studies.'0' " The purpose of the present study was to evaluate the echocardiogram of the anterior leaflet of the mitral valve in timing mitral valve opening and closure, and to validate it as an in-
From the Cardiac and Surgical Cardiovascular Units of the Medical, Surgical and Radiology Services, Massachusetts General Hospital, and the Departments of Medicine, Surgery and Radiology, Harvard Medical School, Boston, Massachusetts. Supported in part by USPHS grants HL-12777, HL-12322, HL14209, and Program Project Grant HL-06664. This work was done during the tenure of a research fellowship for Dr. O'Keefe from the American Heart Association, Cape & Islands Massachusetts Chapter - #1144-F. Dr. Daggett is an Established Investigator of the American Heart
Association.
Address for reprints: Gerald M. Pohost, M.D., Cardiac Unit, Massachusetts General Hospital, Boston, Massachusetts 02114. Received June 11, 1974; revision accepted for publication September 4, 1974.
88
Circulation, Volume 51. January 1975
Downloaded from http://circ.ahajournals.org/ by guest on June 14, 2016
MV ANTERIOR LEAFLET ECHO
89
AP (mmHg)
A
A TRIALL CL CCTROGRAM
VENT RICUL AR
ELECTROGRAM L
AP
(mmHgo 30
30r
L P
l,,
~~ 150
0,
~ ~ ~ Bt. . . . .
(rnmt/g/Sec)500
L VP
100F
tmmHfg
B
0
1t
1 00 msec
Figure 1 Representative recording of simultaneons aortic pressure (AXP), echocardiograrn of atnterior leaflet of the mitral valce (E CI1O), atrial elect rogramn, vent ricular electrogran, left venr trieclar diastolic p)ressure (L VDP), left atrial press ure (LAP),
left uenitricutlar tiPIdt (IV dPIdt), and left ventricular P= posterior;
dicator of mitral valve motion. It was anticipated that this study would define the role of the echocardiogram of the mitral valve anterior leaflet as a reference in the timing of intracardiac events and in the determination of systolic and diastolic time intervals. Recent advances in the understanding and clinical application of these time intervals further emphasize the potential clinical value of this technique.'2-1i
Methods llemodynamic
Echocardiographic
Correlation
of
Valve Opening and Closure Ten mongrel dogs anesthetized intravenously with
warmed chloralose (60 mg/kg) and urethane (600 mg/kg) were placed on right heart bypass as described in detail previously.9 Aortic pressure controlled by reservoir height, cardiac output determined by the input from the bypass pump, and heart rate controlled by atrioventricular pacing were varied over a wide range. Aortic, left ventricular, and left atrial pressures were measured through large bore short metal cannulae with Statham P23Db transducers and recorded simultaneously with epicardial electrograms, elec(ircullation, Volumne
.31,
pressure
(LVP). Note that
the
echocardiogram is inverted.
anterior.
A
Jatntuary
tronically differentiated left ventricular pressure, and the echocardiogram of the anterior leaflet of the mitral valve (ECtHO) on a Beckman S-IL oscillograph (fig. 1). Echocardiograms were obtained with an echograph* having a fre(luency output of 2.25 megacycles/sec and a repetition rate of 1000 pulses/sec equipped with an 0.5 inch diameter 10 focused transducer.I A time analog gate preamplifier used to isolate the ECHO. The time delay in this ultrasound system is less than one millisecond, and the inherent frequency response limitation occurs above 500 Hz.t The ultrasound transducer was held in contact with the epicardium of the anterior cardiac wall adjacent to the left anterior descending coronary artery. The transducer was oriented so as to maximize the amplitude of anterior mitral valve leaflet excursion. Anterior leaflet motion was observed at a depth between 3 and 5 cm in all preparations. The left atrial and left ventricular pressure crossover point in early diastole was selected as the hemodynamic indicator of mitral valve opening and was obtained by manually tracing the left atrial pressure superimposed onto the left ventricular pressure. The point of onset of left ventricular systole was cm
was
IEcholine 20, Smith Kline Instruments Corporation. f Aerotech (:12. Data provided bv Smith Kline Instruments Corporation.
1.97.3
Downloaded from http://circ.ahajournals.org/ by guest on June 14, 2016
90
POHOST ET AL.
the hemodynamic indicator of mitral valve closure the left atrial and left ventricular pressures often equilibrated in mid and end diastole at lower diastolic
Table 1 Range of Hemoldynarnic Variables
pressures.
Heart rate Mean aortic pressure Peak left- ventricular dP/dt Peak negative left venitricutlar rIP/dt Cardiac inpu1t
used as
because
order to determine the timing error induced by the wide bore short metal cannula-Statham P23Db pressure measLurement system used in the present study, recordings of left ventricular diastolic pressure measured with this system were compared with those measured through an SF-1 Statham catheter tip manometer. The pressure crossover point and the point of onset of left ventricular systole on the SF-1 preceded that of the P23Db curve by 4.5 msec (-2 to 10 msec) and 5.1 msec (0 to 8 msec), respectively. Echocardiograms were analyzed with respect to the timing of these points. The echocardiogram of the normal mitral valve anterior leaflet has an "M- shaped configuration anid has been divided, by convention, into several segnments by a series of points (fig. 2). Opening of the valve has been considered to occur between D and E where the E point represents maximal anterior excursion. The E point is followed by posterior motion to point F during ventricular filling. Anterior motion again occurs after point F and reaches a second peak at point A during atrial contraction. Valve closure is thought to occur between points A and C. For the purpose of this analysis, additional points D' and C, wvere selected, representing the onset of most rapid anterior motion in early diastole and the end of final rapid posterior motion in late diastole respectively (fig. 2). Points D' and CO were easily defined echocardiographic events which could be determined most consistently and approximated mitral valve opening and closure, respectively. Three intervals which relate to mitral valve opening and/or closure were compared echocardiographically and hemodynamically: 1) the isovolumic relaxation period (aortic dicrotic notch to echocardiographic D' point or left atrial In
Figure 2 Echocardiogram of the anterior leaflet of the normal mitral ualve. Diastole occurs between points D and C. After D the leaflet moves atnteriorly through D' to peak at E. This is followed by posterior mvotion between E and F and second anterior peak associated with atrial contraction at point A. Valve closure occurs between points A a
anid C. During systole (C
to
D) the
calve
slowly
moves
anteriorly.
Points D' and CO are reference points related to opening and closure of the nitral valve respectively.
Perfusioni temlper ature
65- 200 beats mill 70-130 mm lig 11003-600 mm IIg/sec 700 2300 mm Hlg/sec 1000-2500 mI/mill 31 `-400 C
and ventricular pressure crossover), 2) the electrocardiographic Q wave to valve closure interval (echocardiographic C0U point or point of onset of left ventricular systole), and 3) the diastolic filling period (echocardiographic D)' to C0 or pressure crossover to point of onset of left ventricular systole). In addition to aortic pressure, cardiac input and heart rate, blood perfusion temperature was varied (table 1) to obtain a broad range of intervals. The intervals determined hemodynamically were compared to those obtained echocardiographically by calculation of a correlation coefficient, standard error, and plot of the best fit regression line on a Xerox Sigma 3 computer and Calcomp plotter. Cineroentgenographic Studies of Mitral Valve Motion
Mitral valve motion in eleven additional dogs was studied by cineroentgenographic techniques. All experimental preparations underwent multiple injections of radiopaque contrast medium (meglumine diatrizoate and sodium diatrizoate)* into the left atrium (10 cc) and/or the pulmonary artery (30 cc) with left-sided follow through with the dog in approximately 800 left anterior oblique position to maximize visualization of the mitral valve. High speed 16 mm cineangiograms (250 frames/sec) were obtained. t An oscilloscopic image of the time-analog echocardiographic signal of the anterior leaflet of the mitral valve was simultaneously recorded on the cineangiogram. The delay in the oscilloscope was measured to be less than 0. 1 msec, and the total delay in the ultrasound system was less than 1 msec. Thus mitral valve motion as shown by cineangiography could be accurately compared to simultaneously recorded echocardiographic pattern. In four preparations on right heart bypass, radiopaque tantalum clips were attached to the mitral valve anterior leaflet free edge in all preparations, the posterior leaflet free edge in three preparations, and an anterior chorda tendinea adjacent to the anterior leaflet in one preparation. In one of these preparations two separate clips were placed on the free edge of the anterior leaflet (fig. 3). In the three preparations in which clips were attached to both leaflets, care was taken to insure that apposing clips would approximate one another when the mitral valve was closed. This methodology allowed visualization of mitral valve free edge motion and leaflet separation during high speed cineroentgenography with simultaneous echocardiographic recording. Heart rate ranged between 100 and 160 beats/min and cardiac output, aortic pressure and temperature were maintained within a similar range in all preparations. Single frame analysis of appropriate segments of the cineroentgenographic studies were used to plot the
*Renografini-76, Squibb and Sorus. tSiemens Corporation. ( ire tulatlio
Downloaded from http://circ.ahajournals.org/ by guest on June 14, 2016
Vu/olane 51, Janutary 197.5
91
MV ANTERIOR LEAFLET ECHO
chorda tendinea from the ultrasound transducer, and 4) simuiltaneously obtained echocardiogram. These data were scaled to match maximum and minimum amplitudes and plotted against time (4 msec for each data point or cine frame) on a Calcomp plotter driven by a Xerox Sigma 3 computer. The scaled echocardiogram of mitral valve motionr was compared to clip motion on at least two plots representing a different transducer position from each of fouir experimental preparations. The time of maximal early diastolic anterior excursion and of maximal anterior excursion during atrial contraction were compared cineroentgenographically and echocardiographically. The clip which pivoted on the valvular end of the arnterior chorda tendinea, and thus was flow-directed similar to a weather vanie, provided ari additional meanis for dletermination of the onset of diastolic filling Results Mitral Valve Opening and Closure: Comparison of Echocardiographic and Ilernodynamic Data
Figure 3 Open h2eart (lisplayi of position of tantalumn steel clips oni anterior (2 clips o0 the right sidce of t/ie ptotogrtipl) arid posterior (1 clip) oni the left sideI of the phlotographi) leaflets of it/ir mitral value. This specirnill is front the canin e prepa ration used to obtain figures 7 and S. following: 1) distance of clips orn the anterior leaflet from the ultrasound transducer, 2) points of separation and apposition of anterior and posterior mitral valve leaflet clips, 3) distance of the tip of the mobile clip on the anterior leaflet
The isovolumic relaxation period (IRP) measured from the time of the aortic dicrotic notch to the left ventricular and left atrial pressure crossover point (IRPH) and to D' on the echocardiogram (IRPE) (fig. 4A) were closely related. This relationship is described by the regression equation: IRPE = (1.1) IRPH + 8.6 msec; r - 0.98 (se\i ± 7.6 msec) The duration of the isovolumic relaxation period determined hemodynamically ranged between 30 and 190 msec. Substituting for IRPH, the range for IRPE would be 42-218 msec. Thus, there is a 12-28 msec delay between the pressure crossover and echocardiographic D' points. Addition of the 5 msec delay in =
C
B
A
S~~~
Ate it
x
cs.
L
lit
to
(3)0Z
IRP HEMODYNAM/C
x 10 1
O TO CLOSURE HEMODYNAMIC x /of
DFf/HEMODNYhAMIC
x JO
Figure 4 A) Regression lite relating isovolumiic relaxation period (lRP) determiined echocardiographically to that determined htemodynta ntically. B) Regressiolio n e relating the interval between the first t ransient of the electr cardiographic Q RS and valve closutre or 9 to closure interval determined echocardiographically and hemodynamnically. C) Regression line relating the diastolic filling period (DFP) determined echocardiographically and hemodynamiically. ('ircitlationt, Votliirt7e 5), Januttary 1.975 Downloaded from http://circ.ahajournals.org/ by guest on June 14, 2016
POHOST ET AL.
92
the
pressure 17-33 msec.
measuring system increases this delay to
The interval between the onset of electrical activity and valve closure event (electrocardiographic Q wave to closure interval or QCI) was measured from the first transient of the ventricular electrogram to the point of onset of left ventricular systole (QCIH) and to the C0 point on the echocardiogram (QCIE). These intervals illustrated in figure 4B are related by the following equation:
QCIE = QCIH + 20 msec; 0.96 (SEM = ± 10 msec) Thus, there is a delay of 20 ± 10 msec between the hemodynamic and the echocardiographic valve closing event. The total delay including that incurred in the pressure measuring system is 25 10 msec. Finally, the diastolic filling period (DFP) was measured hemodynamically from pressure crossover to the point of onset of left ventricular systole (DFPH) and echocardiographically from D' to C0 (DFPE). These intervals illustrated in figure 4C are related by the regression equation: DFPE = DFPH msec; r = 0.99 (SEM = ± 20 msec) r
=
.q
E
VENTr
C
"
V %,
Tm
j MITRAL VALVE OPENING
MIRALVALVE CLURE
Figure 5 Composite diagram relating the time of hemodynamically and cineroentgenographically determined events to the echocardiogram of the mitral valve anterior leaflet (ECHO). The left ventricular (LV) and left atrial (LA) pressures and the ECHO are illustrated above. Below, events are indicated adjacent to their timing relative to the earliest event for mitral valve opening (clip separation) and for mitral valve closure (onset of LV systole).
Therefore, the echocardiographically and hemodynamically determined diastolic filling periods are essentially equivalent. Mitral Valve Opening and Closure: Cineroentgenographic and Echocardiographic Studies
The time relationships of the events described below are illustrated in figure 5. Mitral Valve Opening: Three of the events constituting mitral valve opening as shown by cineroentgenography were compared with the echocardiogram: 1) the time of separation of apposing clips on the anterior and posterior leaflets, 2) the time of onset of rapid anterior motion of the anterior leaflet clip in early diastole (similar to ECHO D'), and 3) the time at which the advancing front of contrast medium within the left atrium mixes with left ventricular blood. The interval between the onset of mitral valve clip separation and the D' point on the ECHO was determined on 20 occasions in three canine preparations. Clip separation occurred 26 ± 3 msec before the echocardiographic D' point. The D' point occurs approximately 19-24 msec after pressure crossover within a physiologic range of IRP (50-100 msec).*20 Thus, clip separation occurs within 2 to 7 msec of pressure crossover. This delay is not statistically significant. The time of onset of rapid anterior motion in early diastole of the anterior leaflet clip occurs an average of 6 msec after separation of anterior and posterior leaflet clips. Contrast medium was observed to enter the left ventricle from the left atrium as a convex front behind the mitral valve until the valve opened widely in all cineangiographic studies. Disruption of this contrast medium front and mixture with left ventricular blood occurred 17 ± 2 msec after the D' point of the ECHO. Thus, an interval of approximately 43 msec between initial separation of mitral valve anterior and posterior leaflet and the flow of blood beyond the free edges of the mitral valve describes mitral valve opening. In one preparation, a mobile clip was attached to an anterior chorda tendinea near its insertion onto the anterior leaflet free edge. The distance of this "weather vane" clip and a clip on the anterior leaflet free edge from the ultrasound transducer and the simultaneous ECHO were plotted on the same scale (fig. 6). The flow directed "weather vane" clip changed direction 9 msec after the echo D' point, about 8 msec prior to the disruption of the contrast medium front. Mitral Valve Closure: The CO point on the ECHO *Substituting 50 to 100 msec for IRPH in the regression equation IRPE = 1.1 IRPH + 8.6 reveals a range of 64 to 119 msec for IRPE. Thus ECHO D' occurs 14 to 19 msec after the measured crossover point or 19 to 24 msec after the actual crossover point. Circulation, Volume 51, January 1975
Downloaded from http://circ.ahajournals.org/ by guest on June 14, 2016
93
MV ANTERIOR LEAFLET ECHO
1
Figure 6 C"07ompuiter )lot of moiotion of radiopaqute clip on the anterior leaflet of the miitral valve (,XLMV) and the free enid of mobile clip with the other end attached to an anterior chorda tendinea (CT) the ultrasound transducer arid the iweather vanie' relative echocardiogram (ECHO) of the anterior leaflet, Note the abrupt change in direction of CT at the time of peak anterior excursion of a
a
to
ALMV.
occurred 18 2 msee after coaptation of free edge clips (fig. 5). We have shown that the onset of left ventricular systole occurred 25 msec before the echocardiographic C0 point. Thus, clip coaptation occurs shortly after the onset of left ventricular systole (7 msec), an interval which is not statistically significant. The termination of rapid posterior motion of the anterior leaflet clip (similar to echocardiographic C0) occurred within 2 msec of anterior and posterior leaflet clip coaptation. Mitral Valve Motion: Cineroentgenographic and Echocardiographic Studies
The time relationship of mitral valve anterior leaflet motion determined cineroentgenographically and
echocardiographically was analyzed by two additional methods: 1) comparison of the two points of maximal anterior leaflet excursion during early diastolic filling and during atrial systole, and 2) comparison of computer plots of distance of anterior leaflet clips from the ultrasound transducer with the simultaneous echocardiogram. The peak anterior excursion of the anterior leaflet clips in early diastole occurred 24 ± 5 msee (0 to 40 msee) prior to the corresponding echocardiographic event (E point), while the peak anterior excursion during atrial contraction occurred 15 ± 4 msee (0 to 38 msec) before the A point of the echo. Considering both of these delays, a mean delay of 20 ± 3 msee has been demonstrated between clip motion and the ECHO.
The position of the ultrasound transducer and the clips on the free edge of the anterior and posterior leaflets is illustrated in eight frames of a representative cineroentgenographic sequence in figure 7. The (t
a
tiot,
simultaneous ECHO appears on the right-hand side of each cine frame. The configuration of the valve as determined by contrast medium injection during previous sequences is sketched into each frame. The computer plot of clip motion and echocardiographic motion for this sequence is illustrated in figure 8. The plots of clip motion are qualitatively similar to the echocardiogram. Note that peak anterior motion in early diastole occurred 20 msec before the echocardiographic E point and that peak anterior motion during atrial systole occurred 16 msee before the echocardiographic A point. Figure 8 demonstrates the relationship of clip separation (point X) and the onset of rapid anterior motion in early diastole (clip D'), and of clip coaptation (point Y) and the termination of final rapid posterior motion in late diastole (clip C). Echocardiographic D' and C. are thus analogous to clip events which closely approximate separation and coaptation of the mitral valve free edges. Discussion Despite a large number of clinical studies in the use of diagnostic cardiac ultrasound,' 2, 21 only a few studies of validation exist.4' ' 'o Although numerous studies delineate abnormal mitral valve echocardiographic morphology, normal morphology has not been rigorously defined. Although the echocardiogram provides a convenient method of observing mitral valve motion, it has not been widely used as a means of timing opening or closure of the mitral valve, nor has the relationship of mitral valve opening and closure to the echocardiogram been accurately correlated. The present study demonstrates an excellent correlation between hemodynamic and echocardiographic indexes of mitral valve opening and closure. However, a variable echocardiographic delay of up to 40 msec is evident in comparison to motion of anterior leaflet free edge radiopaque clips as shown by cineroentgenography. Plots of the simultaneously recorded echocardiogram of the anterior leaflet of the mitral valve and cineroentgenographic motion of radiopaque clips on the leaflet free edge are morphologically similar. Clips on two separate points of the free edge of the anterior leaflet display similar morphology, although a discrepancy in timing is evident. Thus, in figure 8, clip ALMV2 leads clip ALMVI by up to 20 msee. The echocardiographic events occur 15-35 msec after comparable events of clip ALMV2 and 15-25 msec after comparable events of ALMVI. These delays are compatible with nonuniformity of motion of the free edge of the leaflet. Nonuniformity of mitral valve motion is evident upon inspection of the cinematographic studies by Edler.4
Jarutai ]uar l 197 5
Downloaded from http://circ.ahajournals.org/ by guest on June 14, 2016
94
POHOST ET AL.
Figure 7 Representative cineroentgenographic frames illustrating two radiopaque clips on the free edge of the anterior leaflet of the mitral valve (ALMVI, ALMV2), and a clip on the posterior leaflet of the mitral valve (PLMV), the position of the ultrasound transducer (T) and the cinetrace of the echocardiogram of the anterior leaflet of the mitral valve (ECHO). The anatomic structures have been drawn in and labeledfor illtustrative purposes. (Ao aorta, LV left ventricle, LA - left atrium.) 1 mid systole, 2 8-12 msec prior to mitrail valve clip separation, 3 = 8-12 mnsec after mitral valve clip separation, 4 - maximal anterior leaflet anterior excursion, 5 - maximal anterior leaflet excursion during atrial contraction, (6 less than 4 msec prior to mitral valve clip apposition, 7 less than 4 msec after mitral valve clip contact, 8 early =
=
=
=
=
=
sy,stole. Cirulatifol, Vlolutne 51, JajuaItry 197.5 Downloaded from http://circ.ahajournals.org/ by guest on June 14, 2016
MV ANTERIOR LEAFLET ECHO
95
Ec,h.ho
Mot ion
!¾
,/ ALMV 2
ALMV 1
Clip D' \i C co 2Omsec Clip \.\' - 1 / ALMV 2 / y ALMV, 1
ALMV 1
1
ANT
ALMV2
2
3
4
5
6,7
POST
8
Figure 8
Conmputer plot of the rnotion of the two radiopaquie clips on the anterior leaflet of the mitral valve (ALMV] and ALMXN2) relative to the ulttrasound transducer an d the simultaneouts cchocardiogram (Echo). Anterior (AAN T) and pos terior (POS T) nmotioni are identified orn the right. Points X and Y indicate the time of separation and apposition, respectively, of clips on1 an te rior aiil pos terior leaflets. Clip DT is the point Of onset of rapid anterior motion associa ted with valve opening acnd clip1 C>, the point of termination of final rapid posterior nmotion associated with valve closuire. These points are analogotus to echocardiographiic 1)' and C, The nu nibers along the base of this figure correspond to te events illustrrated in fig ure 7
Thus, the edges at the midportion of the valve appear to separate before the edges closer to the commissures. The selective echocardiographic sampling of the free edge may account for a portion of the timing discrepancy between echocardiogram and clip cineroentgenography of anterior leaflet motion. The ultrasound transducer samples motion within a cylindrically shaped column with a diameter approximately equal to that of the transducer (fig. 9). Observation of the mitral valve during contrast medium injection demonstrated that following leaflet separation, the anterior leaflet free edge leads the remainder of the leaflet with a whipping motion progressing from free edge to base. The lag in motion of the remainder of the leaflet behind free edge could account for a portion of the time delay between cineroentgenographic and echocardiographic motion. The possibility that tantalum clips would alter mitral valve motion is recognized. However, mitral valve
Canine Range for MV in Present Study
05" r-
motion was similar in the studies performed with radiopaque contrast medium injection alone and in the studies with clips attached to the free edge(s) of the mitral valve. Another factor in the observed delay is a parallax phenomenon. The transducer receives reflections only when the ultrasound beam interacts perpendicularly to an interface. As the mitral valve opens, the anterior leaflet free edge changes from a relatively perpendicular to a nearly parallel orientation with respect to the ultrasound transducer (fig. 6). Thus, as the mitral valve opens, a larger area of anterior leaflet is sampled by the ultrasound beam and relatively less of its echocardiogram is composed of leading leaflet free edge. Finally, echocardiographic time resolution is related to the number of bursts of ultrasound energy emitted over a given time or the pulse repetition rate. The "echo dropout" phenomenon referred to by
-1
l
Clinical Range for MV
1
1
051 -LI Transducer
/
Full AmplituSde Beam Profile
7'
Hal/f Amplituxde Beam Profile
Figure 9 Ultrasound beam width profile of the 10 cm focused or collimated transducer (approximately full power = diagonal stripes; half power = dots). (Constructed from data courtesy of Aerotech Laboratories.) The center of the beam is divided into half-inch intervals. Note that the beam dimensions are similarfor the mitral valve (MV) range in the present and for the usual clinical range.
stuidy
(irculation, volume 51, Januiary f975 Downloaded from http://circ.ahajournals.org/ by guest on June 14, 2016
POHOST ET AL.
96
Kingsley'0 at a slow repetition rate (200/sec) would not be expected to be an important factor in delays greater than one millisecond at a repetition rate of 1000 as was employed in the present study. The results of the present study are consistent with data of other investigators who demonstrate in mitral stenosis that valve closure as determined hemodynamically occurs immediately before the most posterior motion of the anterior leaflet'0 and that the opening snap on external phonocardiography occurs approximately 20 msec before the echocardiographic E point.22 Considering the delay demonstrated in the present study the opening snap would occur coincident with maximal anterior excursion in early diastole. Since mitral valve leaflets separate nonuniformly, the time at which free edge clips separate depends upon their location. This nonuniform motion does not influence hemodynamic indicators of opening and closure, and is minimized by the echocardiogram due to the integrating properties of the ultrasound beam. Thus, echocardiographic indicators of mitral valve opening and closure correlate well with hemodynamic indicators of these events. In the clinical setting, the mitral anterior leaflet is further from the ultrasound transducer (7-10 cm) than in the canine preparation used in the present study (3-5 cm). The area of anterior leaflet represented by the echocardiographic tracing in the clinical range with a focused or collimated transducer is approximately equivalent to the leaflet area represented in the present study (fig. 9). A good correlation with hemodynamic indicators of opening and closure similarly would be expected in a clinical study. Inspection of cineroentgenography of mitral valves with clips on anterior and posterior leaflets and cineangiography elucidates the mechanism of mitral valve opening and left ventricular filling. Separation of the free edge clips begins 26 msec before D' on the echocardiogram (the point of onset of most rapid anterior mitral valve motion in early diastole) as the valve bulges into the left ventricle. An advancing front of contrast medium retains the convex configuration of the bulging mitral valve, while opening continues. The contrast medium does not begin to mix with left ventricular blood until the valve is almost fully opened. Hence, the onset of flow beyond the mitral valve free edges occurs considerably after the initial separation of anterior and posterior leaflets. Thus mitral valve opening is a complex sequence of events rather than a single event in time. Mitral valve closure determined hemodynamically and cineroentgenographically correlated well with the echocardiographic C0 point, although there was a delay of the C0 point of approximately 20 msec (fig. 8). Hemodynamic and cineroentgenographic data
demonstrate that the onset of left ventricular systole and coaptation of free edge clips occur at approximately the same time. The interrelationship of events of mitral valve motion determined echocardiographically, hemodynamically, and cineroentgenographically has been demonstrated in the present study. Hemodynamic indicators of opening and closure correlate well with echocardiographic indicators, whereas a greater uncertainty (0 to 40 msec) is evident when comparing motion of anterior leaflet mitral valve free edge markers to the echocardiogram of the anterior leaflet. The echocardiogram provides a reliable noninvasive means of determining intervals which relate to hemodynamic indices of mitral valve opening and closure. The echocardiogram also provides a means of observing mitral valve motion qualitatively or quantitatively within the limits presented herein. Acknowledgment The authors gratefully acknowledge the valuable assistance of Mr. John B. Newell and Mr. Peter E. Smith of the Myocardial Infarction Research Unit in the statistical analyses and the computerized plotting techniques presented herein. The authors also acknowledge the technical assistance of Messrs. Theodore Kelly, James S. Titus, Bruce F. Robertson, Alvin Yadgood, Gilbert L'Italien, Tony Zingerelli and Tony Wielkiewicz in the conduct of the experiments. We appreciate the very capable secretarial assistance of Misses Margaret Miller, Lorraine Facchina and Stephanie Murray.
References 1. FEIGENBAL-i H: Echocardiography. Philadelphia, Lea & Febiger, 1972 2. GIAnIIAK R, SHAH PM, KRAi-IEi DH: Ultrasound cardiography: Contrast studies in anatomy and function. Radiology 92: 939, 1969 3. JOYNER CR, REID JM: Applications of ultrasound in cardiology and cardiovascular physiology. Prog Cardiovasc Dis 5: 482, 1963 4. EDLER I, Gu-STAFSON A, KARLEFORS T, CHRISTENSSON B: Mitral and aortic valve movements recorded by an ultra-sonic echo method. An experimental study. In Ultrasound Cardiology. Acta Med Scand 370 (suppl): 68, 1961 5. FEIGENBAU.NI H, STONE JM, LEE DA, NASSER WK, CHANG S:
Identification of ultrasound echoes from the left ventricle using intracardiac injections of indocyanine green. Circulation 41: 615, 1970 6. ZAKY A, NASSER WK, FEIGENBAUM H: Study of mitral valve action recorded by reflected ultrasound and its application in the diagnosis of mitral stenosis. Circulation 37: 789, 1968 7. KONECKE LL, FEIGENBALM H, CHANG S, CORYA BC, FISCHER
JC: Abnormal mitral valve motion in patients with elevated left ventricular diastolic pressures. Circulation 47: 989, 1973 8. GLSTAFSON A: Correlation between ultrasoundcardiography, hemodynamics and surgical findings in mitral stenosis. Am J Cardiol 19: 32, 1967 9. ZAKY A, STEINMETZ EF, FEIGENBAIM H: Role of the atrium in closure of the mitral valve in man. Am J Physiol 217: 1952, 1969 10. KINGSLEY B, FLINT GB JR, RABER GT, SEGAL B: Another look at Circutlation, Volume 51, January 1975
Downloaded from http://circ.ahajournals.org/ by guest on June 14, 2016
MV ANTERIOR LEAFLET ECHO echocardiography. Am J Cardiol 19: 108, 1967 11. RUBENSTEIN JJ, POHOST GM, DINSMOIRE RE, SA.NDERS CA: The noninvasive determination of isovolumetric relaxation. (abstr) Circulation 46 (suppl IV): IV-63, 1972 12. BENCHIMIOL A, ELLIS JG: A study of the period of isovolumic relaxation in normal subjects and in patients with heart disease. Am J Cardiol 19: 196, 1967 13. LISA C, HOOD G, TAVIEL M: The jugular venous pulse in pericardial constriction: Its differentiation from that of cardiomyopathy. Am Heart J 84: 409, 1972 14. RUBENSTEIN JJ, POHOST GM, FOSTER JR, DINSNIORE RE: The echocardiographic determination of the isovolumic relaxation period in patients with normal and diseased coronary arteries. (abstr) Clin Res 21: 446, 1973 15. WAYNE H: Noninvasive Technics in Cardiology. Chicago, Year Book Medical Publishers, Inc., 1973, pp 155-214 16. WEISFELDT ML, SCLLLY HE, FREDERIKSE.N J, RUBENSTEIN JJ, POHOST GM, BEIERHOLM E, BELLO AG, DAGGETT WM: Hemodynamic determinants of maximum negative dP/dt
Circulation, Volume 51, January
97 and the periods of diastole. Am J Physiol, in press 17. LEwvIS RP, LEIGHTON RF, FORESTER WF, WEISSLER AM: Systolic time intervals. In Noninvasive Cardiology, edited by WEISSLER AM. New York, Grune & Stratton, 1974, pp 301-368 18. WEISSLER AM, HARRIS WS, SCHOENFELD CD: Systolic time intervals in heart failure in man. Circulation 37: 149, 1968 19. DAGGETT WM, BIANCO JA, POWELL WJ, AUSTEN WG: Relative contributions of the atrial systole-ventricular systole interval and of patterns of ventricular activation to ventricular function during electrical pacing of the dog heart. Circ Res 27: 69, 1970 20. AREVALO F, SAKANIOTO T: On the duration of the isovolumetric relaxation period (IVRP) in dog and man. Am Heart J 67: 5, 1964 21. FEIGENBALNI H: Clinical applications of echocardiography. Prog Cardiovasc Dis 14: 531, 1972 22. SECGAL BL, LIKOFF W, KINCSLEY B: Echocardiography. JAMA 195: 99, 1966
1975
Downloaded from http://circ.ahajournals.org/ by guest on June 14, 2016
The echocardiogram of the anterior leaflet of the mitral valve. Correlation with hemodynamic and cineroentgenographic studies in dogs. G M Pohost, R E Dinsmore, J J Rubenstein, D D O'Keefe, R N Grantham, H E Scully, E A Beierholm, J W Frederiksen, M L Weisfeldt and W M Daggett Circulation. 1975;51:88-97 doi: 10.1161/01.CIR.51.1.88 Circulation is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231 Copyright © 1975 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/51/1/88
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/
Downloaded from http://circ.ahajournals.org/ by guest on June 14, 2016
The echocardiogram of the anterior leaflet of the mitral valve (ECHO) was compared to hemodynamic and cineroentgenographic data to evaluate its accuracy in timing mitral valve (MV) opening and closure, and to validate it as an indicator of MV motion. The ECHO, high speed cineroentgenography at 250 frames/sec, and/or measurement of intracardiac pressures allowed accurate timing of the events of MV motion in dogs on right heart bypass. The intersection of left ventricular and left atrial pressures in early diastole preceded the onset of rapid anterior motion of the ECHO (D' point) by 17 to 33 ± 7.6 msec; r = 0.98. The onset of left ventricular systole occurred before the termination of final rapid posterior motion of the ECHO in end diastole (C. point) by 25 i 10 msec; r = 0.96. Radiopaque clips were attached to the free edges of both leaflets of the MV. Cineroentgenographically determined plots of clip distance from the ultrasound transducer were morphologically similar to the simultaneously recorded ECHO. A delay of 23 ± 3 (0 to 40) msec was observed in the ECHO peaks of diastolic anterior excursion compared to clip motion. Contrast medium advances beyond the free edges of MV leaflets mixing with left ventricular blood 43 ± 3 msec after initial separation. These cineroentgenographic studies elucidate nonuniformity of leaflet motion responsible for ECHO delays. Thus, ECHO D' and CO correlate well with hemodynamic indicators of MV opening and closure. However, ECHO motion, although qualitatively similar, is unpredictably delayed compared to cineroentgenography of clips on the MV free edge. Since the ECHO correlates well with hemodynamic indices of MV opening and closure, this noninvasive technique can be used as a reference in the timing of intracardiac events and in the determination of systolic and diastolic time intervals.
Additional Indexing Words: Mitral valve opening Angiocardiography
Mitral valve closure Echocardiography
ECHOCARDIOGRAPHY HAS PERMITTED noninvasive observation of a variety of intracardiac structures." 2 One of the most distinctive echocardiographic patterns is that of the anterior leaflet of the mitral valve." 3 4 In 1961 Edler con-
Mitral valve motion
eluded after advancing a long needle into a beef heart parallel to the ultrasound transducer4 that this characteristic echocardiogram originated from the anterior leaflet. Further documentation of the origin of this echo has been provided by injection of indocyanine green dye into the left ventricle during ultrasound recording.5 While studies relating echocardiographic mitral valve motion to hemodynamic data have been accomplished,6 10 most were performed in the presence of mitral valve disease or left ventricular dysfunction. Few studies are available relating normal mitral valve motion to the echocardiographic pattern.4 Furthermore, it is not known whether the mitral valve echocardiogram is capable of accurately defining opening and closure, although this has been suggested in a number of clinical studies.'0' " The purpose of the present study was to evaluate the echocardiogram of the anterior leaflet of the mitral valve in timing mitral valve opening and closure, and to validate it as an in-
From the Cardiac and Surgical Cardiovascular Units of the Medical, Surgical and Radiology Services, Massachusetts General Hospital, and the Departments of Medicine, Surgery and Radiology, Harvard Medical School, Boston, Massachusetts. Supported in part by USPHS grants HL-12777, HL-12322, HL14209, and Program Project Grant HL-06664. This work was done during the tenure of a research fellowship for Dr. O'Keefe from the American Heart Association, Cape & Islands Massachusetts Chapter - #1144-F. Dr. Daggett is an Established Investigator of the American Heart
Association.
Address for reprints: Gerald M. Pohost, M.D., Cardiac Unit, Massachusetts General Hospital, Boston, Massachusetts 02114. Received June 11, 1974; revision accepted for publication September 4, 1974.
88
Circulation, Volume 51. January 1975
Downloaded from http://circ.ahajournals.org/ by guest on June 14, 2016
MV ANTERIOR LEAFLET ECHO
89
AP (mmHg)
A
A TRIALL CL CCTROGRAM
VENT RICUL AR
ELECTROGRAM L
AP
(mmHgo 30
30r
L P
l,,
~~ 150
0,
~ ~ ~ Bt. . . . .
(rnmt/g/Sec)500
L VP
100F
tmmHfg
B
0
1t
1 00 msec
Figure 1 Representative recording of simultaneons aortic pressure (AXP), echocardiograrn of atnterior leaflet of the mitral valce (E CI1O), atrial elect rogramn, vent ricular electrogran, left venr trieclar diastolic p)ressure (L VDP), left atrial press ure (LAP),
left uenitricutlar tiPIdt (IV dPIdt), and left ventricular P= posterior;
dicator of mitral valve motion. It was anticipated that this study would define the role of the echocardiogram of the mitral valve anterior leaflet as a reference in the timing of intracardiac events and in the determination of systolic and diastolic time intervals. Recent advances in the understanding and clinical application of these time intervals further emphasize the potential clinical value of this technique.'2-1i
Methods llemodynamic
Echocardiographic
Correlation
of
Valve Opening and Closure Ten mongrel dogs anesthetized intravenously with
warmed chloralose (60 mg/kg) and urethane (600 mg/kg) were placed on right heart bypass as described in detail previously.9 Aortic pressure controlled by reservoir height, cardiac output determined by the input from the bypass pump, and heart rate controlled by atrioventricular pacing were varied over a wide range. Aortic, left ventricular, and left atrial pressures were measured through large bore short metal cannulae with Statham P23Db transducers and recorded simultaneously with epicardial electrograms, elec(ircullation, Volumne
.31,
pressure
(LVP). Note that
the
echocardiogram is inverted.
anterior.
A
Jatntuary
tronically differentiated left ventricular pressure, and the echocardiogram of the anterior leaflet of the mitral valve (ECtHO) on a Beckman S-IL oscillograph (fig. 1). Echocardiograms were obtained with an echograph* having a fre(luency output of 2.25 megacycles/sec and a repetition rate of 1000 pulses/sec equipped with an 0.5 inch diameter 10 focused transducer.I A time analog gate preamplifier used to isolate the ECHO. The time delay in this ultrasound system is less than one millisecond, and the inherent frequency response limitation occurs above 500 Hz.t The ultrasound transducer was held in contact with the epicardium of the anterior cardiac wall adjacent to the left anterior descending coronary artery. The transducer was oriented so as to maximize the amplitude of anterior mitral valve leaflet excursion. Anterior leaflet motion was observed at a depth between 3 and 5 cm in all preparations. The left atrial and left ventricular pressure crossover point in early diastole was selected as the hemodynamic indicator of mitral valve opening and was obtained by manually tracing the left atrial pressure superimposed onto the left ventricular pressure. The point of onset of left ventricular systole was cm
was
IEcholine 20, Smith Kline Instruments Corporation. f Aerotech (:12. Data provided bv Smith Kline Instruments Corporation.
1.97.3
Downloaded from http://circ.ahajournals.org/ by guest on June 14, 2016
90
POHOST ET AL.
the hemodynamic indicator of mitral valve closure the left atrial and left ventricular pressures often equilibrated in mid and end diastole at lower diastolic
Table 1 Range of Hemoldynarnic Variables
pressures.
Heart rate Mean aortic pressure Peak left- ventricular dP/dt Peak negative left venitricutlar rIP/dt Cardiac inpu1t
used as
because
order to determine the timing error induced by the wide bore short metal cannula-Statham P23Db pressure measLurement system used in the present study, recordings of left ventricular diastolic pressure measured with this system were compared with those measured through an SF-1 Statham catheter tip manometer. The pressure crossover point and the point of onset of left ventricular systole on the SF-1 preceded that of the P23Db curve by 4.5 msec (-2 to 10 msec) and 5.1 msec (0 to 8 msec), respectively. Echocardiograms were analyzed with respect to the timing of these points. The echocardiogram of the normal mitral valve anterior leaflet has an "M- shaped configuration anid has been divided, by convention, into several segnments by a series of points (fig. 2). Opening of the valve has been considered to occur between D and E where the E point represents maximal anterior excursion. The E point is followed by posterior motion to point F during ventricular filling. Anterior motion again occurs after point F and reaches a second peak at point A during atrial contraction. Valve closure is thought to occur between points A and C. For the purpose of this analysis, additional points D' and C, wvere selected, representing the onset of most rapid anterior motion in early diastole and the end of final rapid posterior motion in late diastole respectively (fig. 2). Points D' and CO were easily defined echocardiographic events which could be determined most consistently and approximated mitral valve opening and closure, respectively. Three intervals which relate to mitral valve opening and/or closure were compared echocardiographically and hemodynamically: 1) the isovolumic relaxation period (aortic dicrotic notch to echocardiographic D' point or left atrial In
Figure 2 Echocardiogram of the anterior leaflet of the normal mitral ualve. Diastole occurs between points D and C. After D the leaflet moves atnteriorly through D' to peak at E. This is followed by posterior mvotion between E and F and second anterior peak associated with atrial contraction at point A. Valve closure occurs between points A a
anid C. During systole (C
to
D) the
calve
slowly
moves
anteriorly.
Points D' and CO are reference points related to opening and closure of the nitral valve respectively.
Perfusioni temlper ature
65- 200 beats mill 70-130 mm lig 11003-600 mm IIg/sec 700 2300 mm Hlg/sec 1000-2500 mI/mill 31 `-400 C
and ventricular pressure crossover), 2) the electrocardiographic Q wave to valve closure interval (echocardiographic C0U point or point of onset of left ventricular systole), and 3) the diastolic filling period (echocardiographic D)' to C0 or pressure crossover to point of onset of left ventricular systole). In addition to aortic pressure, cardiac input and heart rate, blood perfusion temperature was varied (table 1) to obtain a broad range of intervals. The intervals determined hemodynamically were compared to those obtained echocardiographically by calculation of a correlation coefficient, standard error, and plot of the best fit regression line on a Xerox Sigma 3 computer and Calcomp plotter. Cineroentgenographic Studies of Mitral Valve Motion
Mitral valve motion in eleven additional dogs was studied by cineroentgenographic techniques. All experimental preparations underwent multiple injections of radiopaque contrast medium (meglumine diatrizoate and sodium diatrizoate)* into the left atrium (10 cc) and/or the pulmonary artery (30 cc) with left-sided follow through with the dog in approximately 800 left anterior oblique position to maximize visualization of the mitral valve. High speed 16 mm cineangiograms (250 frames/sec) were obtained. t An oscilloscopic image of the time-analog echocardiographic signal of the anterior leaflet of the mitral valve was simultaneously recorded on the cineangiogram. The delay in the oscilloscope was measured to be less than 0. 1 msec, and the total delay in the ultrasound system was less than 1 msec. Thus mitral valve motion as shown by cineangiography could be accurately compared to simultaneously recorded echocardiographic pattern. In four preparations on right heart bypass, radiopaque tantalum clips were attached to the mitral valve anterior leaflet free edge in all preparations, the posterior leaflet free edge in three preparations, and an anterior chorda tendinea adjacent to the anterior leaflet in one preparation. In one of these preparations two separate clips were placed on the free edge of the anterior leaflet (fig. 3). In the three preparations in which clips were attached to both leaflets, care was taken to insure that apposing clips would approximate one another when the mitral valve was closed. This methodology allowed visualization of mitral valve free edge motion and leaflet separation during high speed cineroentgenography with simultaneous echocardiographic recording. Heart rate ranged between 100 and 160 beats/min and cardiac output, aortic pressure and temperature were maintained within a similar range in all preparations. Single frame analysis of appropriate segments of the cineroentgenographic studies were used to plot the
*Renografini-76, Squibb and Sorus. tSiemens Corporation. ( ire tulatlio
Downloaded from http://circ.ahajournals.org/ by guest on June 14, 2016
Vu/olane 51, Janutary 197.5
91
MV ANTERIOR LEAFLET ECHO
chorda tendinea from the ultrasound transducer, and 4) simuiltaneously obtained echocardiogram. These data were scaled to match maximum and minimum amplitudes and plotted against time (4 msec for each data point or cine frame) on a Calcomp plotter driven by a Xerox Sigma 3 computer. The scaled echocardiogram of mitral valve motionr was compared to clip motion on at least two plots representing a different transducer position from each of fouir experimental preparations. The time of maximal early diastolic anterior excursion and of maximal anterior excursion during atrial contraction were compared cineroentgenographically and echocardiographically. The clip which pivoted on the valvular end of the arnterior chorda tendinea, and thus was flow-directed similar to a weather vanie, provided ari additional meanis for dletermination of the onset of diastolic filling Results Mitral Valve Opening and Closure: Comparison of Echocardiographic and Ilernodynamic Data
Figure 3 Open h2eart (lisplayi of position of tantalumn steel clips oni anterior (2 clips o0 the right sidce of t/ie ptotogrtipl) arid posterior (1 clip) oni the left sideI of the phlotographi) leaflets of it/ir mitral value. This specirnill is front the canin e prepa ration used to obtain figures 7 and S. following: 1) distance of clips orn the anterior leaflet from the ultrasound transducer, 2) points of separation and apposition of anterior and posterior mitral valve leaflet clips, 3) distance of the tip of the mobile clip on the anterior leaflet
The isovolumic relaxation period (IRP) measured from the time of the aortic dicrotic notch to the left ventricular and left atrial pressure crossover point (IRPH) and to D' on the echocardiogram (IRPE) (fig. 4A) were closely related. This relationship is described by the regression equation: IRPE = (1.1) IRPH + 8.6 msec; r - 0.98 (se\i ± 7.6 msec) The duration of the isovolumic relaxation period determined hemodynamically ranged between 30 and 190 msec. Substituting for IRPH, the range for IRPE would be 42-218 msec. Thus, there is a 12-28 msec delay between the pressure crossover and echocardiographic D' points. Addition of the 5 msec delay in =
C
B
A
S~~~
Ate it
x
cs.
L
lit
to
(3)0Z
IRP HEMODYNAM/C
x 10 1
O TO CLOSURE HEMODYNAMIC x /of
DFf/HEMODNYhAMIC
x JO
Figure 4 A) Regression lite relating isovolumiic relaxation period (lRP) determiined echocardiographically to that determined htemodynta ntically. B) Regressiolio n e relating the interval between the first t ransient of the electr cardiographic Q RS and valve closutre or 9 to closure interval determined echocardiographically and hemodynamnically. C) Regression line relating the diastolic filling period (DFP) determined echocardiographically and hemodynamiically. ('ircitlationt, Votliirt7e 5), Januttary 1.975 Downloaded from http://circ.ahajournals.org/ by guest on June 14, 2016
POHOST ET AL.
92
the
pressure 17-33 msec.
measuring system increases this delay to
The interval between the onset of electrical activity and valve closure event (electrocardiographic Q wave to closure interval or QCI) was measured from the first transient of the ventricular electrogram to the point of onset of left ventricular systole (QCIH) and to the C0 point on the echocardiogram (QCIE). These intervals illustrated in figure 4B are related by the following equation:
QCIE = QCIH + 20 msec; 0.96 (SEM = ± 10 msec) Thus, there is a delay of 20 ± 10 msec between the hemodynamic and the echocardiographic valve closing event. The total delay including that incurred in the pressure measuring system is 25 10 msec. Finally, the diastolic filling period (DFP) was measured hemodynamically from pressure crossover to the point of onset of left ventricular systole (DFPH) and echocardiographically from D' to C0 (DFPE). These intervals illustrated in figure 4C are related by the regression equation: DFPE = DFPH msec; r = 0.99 (SEM = ± 20 msec) r
=
.q
E
VENTr
C
"
V %,
Tm
j MITRAL VALVE OPENING
MIRALVALVE CLURE
Figure 5 Composite diagram relating the time of hemodynamically and cineroentgenographically determined events to the echocardiogram of the mitral valve anterior leaflet (ECHO). The left ventricular (LV) and left atrial (LA) pressures and the ECHO are illustrated above. Below, events are indicated adjacent to their timing relative to the earliest event for mitral valve opening (clip separation) and for mitral valve closure (onset of LV systole).
Therefore, the echocardiographically and hemodynamically determined diastolic filling periods are essentially equivalent. Mitral Valve Opening and Closure: Cineroentgenographic and Echocardiographic Studies
The time relationships of the events described below are illustrated in figure 5. Mitral Valve Opening: Three of the events constituting mitral valve opening as shown by cineroentgenography were compared with the echocardiogram: 1) the time of separation of apposing clips on the anterior and posterior leaflets, 2) the time of onset of rapid anterior motion of the anterior leaflet clip in early diastole (similar to ECHO D'), and 3) the time at which the advancing front of contrast medium within the left atrium mixes with left ventricular blood. The interval between the onset of mitral valve clip separation and the D' point on the ECHO was determined on 20 occasions in three canine preparations. Clip separation occurred 26 ± 3 msec before the echocardiographic D' point. The D' point occurs approximately 19-24 msec after pressure crossover within a physiologic range of IRP (50-100 msec).*20 Thus, clip separation occurs within 2 to 7 msec of pressure crossover. This delay is not statistically significant. The time of onset of rapid anterior motion in early diastole of the anterior leaflet clip occurs an average of 6 msec after separation of anterior and posterior leaflet clips. Contrast medium was observed to enter the left ventricle from the left atrium as a convex front behind the mitral valve until the valve opened widely in all cineangiographic studies. Disruption of this contrast medium front and mixture with left ventricular blood occurred 17 ± 2 msec after the D' point of the ECHO. Thus, an interval of approximately 43 msec between initial separation of mitral valve anterior and posterior leaflet and the flow of blood beyond the free edges of the mitral valve describes mitral valve opening. In one preparation, a mobile clip was attached to an anterior chorda tendinea near its insertion onto the anterior leaflet free edge. The distance of this "weather vane" clip and a clip on the anterior leaflet free edge from the ultrasound transducer and the simultaneous ECHO were plotted on the same scale (fig. 6). The flow directed "weather vane" clip changed direction 9 msec after the echo D' point, about 8 msec prior to the disruption of the contrast medium front. Mitral Valve Closure: The CO point on the ECHO *Substituting 50 to 100 msec for IRPH in the regression equation IRPE = 1.1 IRPH + 8.6 reveals a range of 64 to 119 msec for IRPE. Thus ECHO D' occurs 14 to 19 msec after the measured crossover point or 19 to 24 msec after the actual crossover point. Circulation, Volume 51, January 1975
Downloaded from http://circ.ahajournals.org/ by guest on June 14, 2016
93
MV ANTERIOR LEAFLET ECHO
1
Figure 6 C"07ompuiter )lot of moiotion of radiopaqute clip on the anterior leaflet of the miitral valve (,XLMV) and the free enid of mobile clip with the other end attached to an anterior chorda tendinea (CT) the ultrasound transducer arid the iweather vanie' relative echocardiogram (ECHO) of the anterior leaflet, Note the abrupt change in direction of CT at the time of peak anterior excursion of a
a
to
ALMV.
occurred 18 2 msee after coaptation of free edge clips (fig. 5). We have shown that the onset of left ventricular systole occurred 25 msec before the echocardiographic C0 point. Thus, clip coaptation occurs shortly after the onset of left ventricular systole (7 msec), an interval which is not statistically significant. The termination of rapid posterior motion of the anterior leaflet clip (similar to echocardiographic C0) occurred within 2 msec of anterior and posterior leaflet clip coaptation. Mitral Valve Motion: Cineroentgenographic and Echocardiographic Studies
The time relationship of mitral valve anterior leaflet motion determined cineroentgenographically and
echocardiographically was analyzed by two additional methods: 1) comparison of the two points of maximal anterior leaflet excursion during early diastolic filling and during atrial systole, and 2) comparison of computer plots of distance of anterior leaflet clips from the ultrasound transducer with the simultaneous echocardiogram. The peak anterior excursion of the anterior leaflet clips in early diastole occurred 24 ± 5 msee (0 to 40 msee) prior to the corresponding echocardiographic event (E point), while the peak anterior excursion during atrial contraction occurred 15 ± 4 msee (0 to 38 msec) before the A point of the echo. Considering both of these delays, a mean delay of 20 ± 3 msee has been demonstrated between clip motion and the ECHO.
The position of the ultrasound transducer and the clips on the free edge of the anterior and posterior leaflets is illustrated in eight frames of a representative cineroentgenographic sequence in figure 7. The (t
a
tiot,
simultaneous ECHO appears on the right-hand side of each cine frame. The configuration of the valve as determined by contrast medium injection during previous sequences is sketched into each frame. The computer plot of clip motion and echocardiographic motion for this sequence is illustrated in figure 8. The plots of clip motion are qualitatively similar to the echocardiogram. Note that peak anterior motion in early diastole occurred 20 msec before the echocardiographic E point and that peak anterior motion during atrial systole occurred 16 msee before the echocardiographic A point. Figure 8 demonstrates the relationship of clip separation (point X) and the onset of rapid anterior motion in early diastole (clip D'), and of clip coaptation (point Y) and the termination of final rapid posterior motion in late diastole (clip C). Echocardiographic D' and C. are thus analogous to clip events which closely approximate separation and coaptation of the mitral valve free edges. Discussion Despite a large number of clinical studies in the use of diagnostic cardiac ultrasound,' 2, 21 only a few studies of validation exist.4' ' 'o Although numerous studies delineate abnormal mitral valve echocardiographic morphology, normal morphology has not been rigorously defined. Although the echocardiogram provides a convenient method of observing mitral valve motion, it has not been widely used as a means of timing opening or closure of the mitral valve, nor has the relationship of mitral valve opening and closure to the echocardiogram been accurately correlated. The present study demonstrates an excellent correlation between hemodynamic and echocardiographic indexes of mitral valve opening and closure. However, a variable echocardiographic delay of up to 40 msec is evident in comparison to motion of anterior leaflet free edge radiopaque clips as shown by cineroentgenography. Plots of the simultaneously recorded echocardiogram of the anterior leaflet of the mitral valve and cineroentgenographic motion of radiopaque clips on the leaflet free edge are morphologically similar. Clips on two separate points of the free edge of the anterior leaflet display similar morphology, although a discrepancy in timing is evident. Thus, in figure 8, clip ALMV2 leads clip ALMVI by up to 20 msee. The echocardiographic events occur 15-35 msec after comparable events of clip ALMV2 and 15-25 msec after comparable events of ALMVI. These delays are compatible with nonuniformity of motion of the free edge of the leaflet. Nonuniformity of mitral valve motion is evident upon inspection of the cinematographic studies by Edler.4
Jarutai ]uar l 197 5
Downloaded from http://circ.ahajournals.org/ by guest on June 14, 2016
94
POHOST ET AL.
Figure 7 Representative cineroentgenographic frames illustrating two radiopaque clips on the free edge of the anterior leaflet of the mitral valve (ALMVI, ALMV2), and a clip on the posterior leaflet of the mitral valve (PLMV), the position of the ultrasound transducer (T) and the cinetrace of the echocardiogram of the anterior leaflet of the mitral valve (ECHO). The anatomic structures have been drawn in and labeledfor illtustrative purposes. (Ao aorta, LV left ventricle, LA - left atrium.) 1 mid systole, 2 8-12 msec prior to mitrail valve clip separation, 3 = 8-12 mnsec after mitral valve clip separation, 4 - maximal anterior leaflet anterior excursion, 5 - maximal anterior leaflet excursion during atrial contraction, (6 less than 4 msec prior to mitral valve clip apposition, 7 less than 4 msec after mitral valve clip contact, 8 early =
=
=
=
=
=
sy,stole. Cirulatifol, Vlolutne 51, JajuaItry 197.5 Downloaded from http://circ.ahajournals.org/ by guest on June 14, 2016
MV ANTERIOR LEAFLET ECHO
95
Ec,h.ho
Mot ion
!¾
,/ ALMV 2
ALMV 1
Clip D' \i C co 2Omsec Clip \.\' - 1 / ALMV 2 / y ALMV, 1
ALMV 1
1
ANT
ALMV2
2
3
4
5
6,7
POST
8
Figure 8
Conmputer plot of the rnotion of the two radiopaquie clips on the anterior leaflet of the mitral valve (ALMV] and ALMXN2) relative to the ulttrasound transducer an d the simultaneouts cchocardiogram (Echo). Anterior (AAN T) and pos terior (POS T) nmotioni are identified orn the right. Points X and Y indicate the time of separation and apposition, respectively, of clips on1 an te rior aiil pos terior leaflets. Clip DT is the point Of onset of rapid anterior motion associa ted with valve opening acnd clip1 C>, the point of termination of final rapid posterior nmotion associated with valve closuire. These points are analogotus to echocardiographiic 1)' and C, The nu nibers along the base of this figure correspond to te events illustrrated in fig ure 7
Thus, the edges at the midportion of the valve appear to separate before the edges closer to the commissures. The selective echocardiographic sampling of the free edge may account for a portion of the timing discrepancy between echocardiogram and clip cineroentgenography of anterior leaflet motion. The ultrasound transducer samples motion within a cylindrically shaped column with a diameter approximately equal to that of the transducer (fig. 9). Observation of the mitral valve during contrast medium injection demonstrated that following leaflet separation, the anterior leaflet free edge leads the remainder of the leaflet with a whipping motion progressing from free edge to base. The lag in motion of the remainder of the leaflet behind free edge could account for a portion of the time delay between cineroentgenographic and echocardiographic motion. The possibility that tantalum clips would alter mitral valve motion is recognized. However, mitral valve
Canine Range for MV in Present Study
05" r-
motion was similar in the studies performed with radiopaque contrast medium injection alone and in the studies with clips attached to the free edge(s) of the mitral valve. Another factor in the observed delay is a parallax phenomenon. The transducer receives reflections only when the ultrasound beam interacts perpendicularly to an interface. As the mitral valve opens, the anterior leaflet free edge changes from a relatively perpendicular to a nearly parallel orientation with respect to the ultrasound transducer (fig. 6). Thus, as the mitral valve opens, a larger area of anterior leaflet is sampled by the ultrasound beam and relatively less of its echocardiogram is composed of leading leaflet free edge. Finally, echocardiographic time resolution is related to the number of bursts of ultrasound energy emitted over a given time or the pulse repetition rate. The "echo dropout" phenomenon referred to by
-1
l
Clinical Range for MV
1
1
051 -LI Transducer
/
Full AmplituSde Beam Profile
7'
Hal/f Amplituxde Beam Profile
Figure 9 Ultrasound beam width profile of the 10 cm focused or collimated transducer (approximately full power = diagonal stripes; half power = dots). (Constructed from data courtesy of Aerotech Laboratories.) The center of the beam is divided into half-inch intervals. Note that the beam dimensions are similarfor the mitral valve (MV) range in the present and for the usual clinical range.
stuidy
(irculation, volume 51, Januiary f975 Downloaded from http://circ.ahajournals.org/ by guest on June 14, 2016
POHOST ET AL.
96
Kingsley'0 at a slow repetition rate (200/sec) would not be expected to be an important factor in delays greater than one millisecond at a repetition rate of 1000 as was employed in the present study. The results of the present study are consistent with data of other investigators who demonstrate in mitral stenosis that valve closure as determined hemodynamically occurs immediately before the most posterior motion of the anterior leaflet'0 and that the opening snap on external phonocardiography occurs approximately 20 msec before the echocardiographic E point.22 Considering the delay demonstrated in the present study the opening snap would occur coincident with maximal anterior excursion in early diastole. Since mitral valve leaflets separate nonuniformly, the time at which free edge clips separate depends upon their location. This nonuniform motion does not influence hemodynamic indicators of opening and closure, and is minimized by the echocardiogram due to the integrating properties of the ultrasound beam. Thus, echocardiographic indicators of mitral valve opening and closure correlate well with hemodynamic indicators of these events. In the clinical setting, the mitral anterior leaflet is further from the ultrasound transducer (7-10 cm) than in the canine preparation used in the present study (3-5 cm). The area of anterior leaflet represented by the echocardiographic tracing in the clinical range with a focused or collimated transducer is approximately equivalent to the leaflet area represented in the present study (fig. 9). A good correlation with hemodynamic indicators of opening and closure similarly would be expected in a clinical study. Inspection of cineroentgenography of mitral valves with clips on anterior and posterior leaflets and cineangiography elucidates the mechanism of mitral valve opening and left ventricular filling. Separation of the free edge clips begins 26 msec before D' on the echocardiogram (the point of onset of most rapid anterior mitral valve motion in early diastole) as the valve bulges into the left ventricle. An advancing front of contrast medium retains the convex configuration of the bulging mitral valve, while opening continues. The contrast medium does not begin to mix with left ventricular blood until the valve is almost fully opened. Hence, the onset of flow beyond the mitral valve free edges occurs considerably after the initial separation of anterior and posterior leaflets. Thus mitral valve opening is a complex sequence of events rather than a single event in time. Mitral valve closure determined hemodynamically and cineroentgenographically correlated well with the echocardiographic C0 point, although there was a delay of the C0 point of approximately 20 msec (fig. 8). Hemodynamic and cineroentgenographic data
demonstrate that the onset of left ventricular systole and coaptation of free edge clips occur at approximately the same time. The interrelationship of events of mitral valve motion determined echocardiographically, hemodynamically, and cineroentgenographically has been demonstrated in the present study. Hemodynamic indicators of opening and closure correlate well with echocardiographic indicators, whereas a greater uncertainty (0 to 40 msec) is evident when comparing motion of anterior leaflet mitral valve free edge markers to the echocardiogram of the anterior leaflet. The echocardiogram provides a reliable noninvasive means of determining intervals which relate to hemodynamic indices of mitral valve opening and closure. The echocardiogram also provides a means of observing mitral valve motion qualitatively or quantitatively within the limits presented herein. Acknowledgment The authors gratefully acknowledge the valuable assistance of Mr. John B. Newell and Mr. Peter E. Smith of the Myocardial Infarction Research Unit in the statistical analyses and the computerized plotting techniques presented herein. The authors also acknowledge the technical assistance of Messrs. Theodore Kelly, James S. Titus, Bruce F. Robertson, Alvin Yadgood, Gilbert L'Italien, Tony Zingerelli and Tony Wielkiewicz in the conduct of the experiments. We appreciate the very capable secretarial assistance of Misses Margaret Miller, Lorraine Facchina and Stephanie Murray.
References 1. FEIGENBAL-i H: Echocardiography. Philadelphia, Lea & Febiger, 1972 2. GIAnIIAK R, SHAH PM, KRAi-IEi DH: Ultrasound cardiography: Contrast studies in anatomy and function. Radiology 92: 939, 1969 3. JOYNER CR, REID JM: Applications of ultrasound in cardiology and cardiovascular physiology. Prog Cardiovasc Dis 5: 482, 1963 4. EDLER I, Gu-STAFSON A, KARLEFORS T, CHRISTENSSON B: Mitral and aortic valve movements recorded by an ultra-sonic echo method. An experimental study. In Ultrasound Cardiology. Acta Med Scand 370 (suppl): 68, 1961 5. FEIGENBAU.NI H, STONE JM, LEE DA, NASSER WK, CHANG S:
Identification of ultrasound echoes from the left ventricle using intracardiac injections of indocyanine green. Circulation 41: 615, 1970 6. ZAKY A, NASSER WK, FEIGENBAUM H: Study of mitral valve action recorded by reflected ultrasound and its application in the diagnosis of mitral stenosis. Circulation 37: 789, 1968 7. KONECKE LL, FEIGENBALM H, CHANG S, CORYA BC, FISCHER
JC: Abnormal mitral valve motion in patients with elevated left ventricular diastolic pressures. Circulation 47: 989, 1973 8. GLSTAFSON A: Correlation between ultrasoundcardiography, hemodynamics and surgical findings in mitral stenosis. Am J Cardiol 19: 32, 1967 9. ZAKY A, STEINMETZ EF, FEIGENBAIM H: Role of the atrium in closure of the mitral valve in man. Am J Physiol 217: 1952, 1969 10. KINGSLEY B, FLINT GB JR, RABER GT, SEGAL B: Another look at Circutlation, Volume 51, January 1975
Downloaded from http://circ.ahajournals.org/ by guest on June 14, 2016
MV ANTERIOR LEAFLET ECHO echocardiography. Am J Cardiol 19: 108, 1967 11. RUBENSTEIN JJ, POHOST GM, DINSMOIRE RE, SA.NDERS CA: The noninvasive determination of isovolumetric relaxation. (abstr) Circulation 46 (suppl IV): IV-63, 1972 12. BENCHIMIOL A, ELLIS JG: A study of the period of isovolumic relaxation in normal subjects and in patients with heart disease. Am J Cardiol 19: 196, 1967 13. LISA C, HOOD G, TAVIEL M: The jugular venous pulse in pericardial constriction: Its differentiation from that of cardiomyopathy. Am Heart J 84: 409, 1972 14. RUBENSTEIN JJ, POHOST GM, FOSTER JR, DINSNIORE RE: The echocardiographic determination of the isovolumic relaxation period in patients with normal and diseased coronary arteries. (abstr) Clin Res 21: 446, 1973 15. WAYNE H: Noninvasive Technics in Cardiology. Chicago, Year Book Medical Publishers, Inc., 1973, pp 155-214 16. WEISFELDT ML, SCLLLY HE, FREDERIKSE.N J, RUBENSTEIN JJ, POHOST GM, BEIERHOLM E, BELLO AG, DAGGETT WM: Hemodynamic determinants of maximum negative dP/dt
Circulation, Volume 51, January
97 and the periods of diastole. Am J Physiol, in press 17. LEwvIS RP, LEIGHTON RF, FORESTER WF, WEISSLER AM: Systolic time intervals. In Noninvasive Cardiology, edited by WEISSLER AM. New York, Grune & Stratton, 1974, pp 301-368 18. WEISSLER AM, HARRIS WS, SCHOENFELD CD: Systolic time intervals in heart failure in man. Circulation 37: 149, 1968 19. DAGGETT WM, BIANCO JA, POWELL WJ, AUSTEN WG: Relative contributions of the atrial systole-ventricular systole interval and of patterns of ventricular activation to ventricular function during electrical pacing of the dog heart. Circ Res 27: 69, 1970 20. AREVALO F, SAKANIOTO T: On the duration of the isovolumetric relaxation period (IVRP) in dog and man. Am Heart J 67: 5, 1964 21. FEIGENBALNI H: Clinical applications of echocardiography. Prog Cardiovasc Dis 14: 531, 1972 22. SECGAL BL, LIKOFF W, KINCSLEY B: Echocardiography. JAMA 195: 99, 1966
1975
Downloaded from http://circ.ahajournals.org/ by guest on June 14, 2016
The echocardiogram of the anterior leaflet of the mitral valve. Correlation with hemodynamic and cineroentgenographic studies in dogs. G M Pohost, R E Dinsmore, J J Rubenstein, D D O'Keefe, R N Grantham, H E Scully, E A Beierholm, J W Frederiksen, M L Weisfeldt and W M Daggett Circulation. 1975;51:88-97 doi: 10.1161/01.CIR.51.1.88 Circulation is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231 Copyright © 1975 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/51/1/88
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/
Downloaded from http://circ.ahajournals.org/ by guest on June 14, 2016
