- Email: [email protected]
Echocardiographic and cardiac doppler assessment of mice
Echocardiographic and Cardiac Doppler Assessment of Mice Charles Pollick, MBChB, Sharon L. Hale, BS, and Robert A. Kloner, MD, PhD, Los Angeles, California
Recent studies have suggested that intermediatefrequency M - m o d e transthoracic echocardiographic imaging is a promising method for evaluating the left ventricle in transgenic mice. However, there is a paucity o f data regarding two-dimensional (2-D) echocardiography and cardiac Doppler echocardiography in this model. Therefore we studied 15 mice (body weights 38 to 65 gm) with an ultrasound system equipped with a 9 M H z transducer. M-mode, 2-D, pulsed, and color-flow Doppler studies were performed. Mean _+SD for septal, posterior wall, and left ventricular cavitary dimensions at end diastole were the following: M-mode: 1.1 + 0.2, 1.0 _+0.2, and 3.7 -+ 0.7 mm; 2-D: 1.0 -+ 0.2, 1.1 _+0.3, and 3.0 + 0.6 nun. Left ventricular fractional shortening was assessed from the M-mode echocardiogram: mean 53.7% _+10.7% (range 42% to 77%). 2-D assessment o f left ventricular mass correlated better with left ventricular mass identified at necropsy than left ventricular mass identified by M - m o d e echocardiography (r = 0.70; p = 0.007 versus r = 0.07; p not significant). 2-D visualization o f left ventricle, proximal aorta, and aortic and mitral valves
was excellent and was obtained mainly from a "parasternal" window. Apical views were more difficult to obtain. Mean + SD for aortic peak and mean velocities and velocity-time integral were 0.53 + 0.13, 0.32 + 0.08, and 0.025 + 0.008 m/see. Estimated stroke volume was 0.0506 + 0.018 ml/beat. Cardiac o u t p u t was 12.64 + 7.87 m l / m i n . Mean + SD for mitral peak E, peak A, and E / A ratio were 0.45 + 0.09 m/sec, 0.19 + 0.06 m/sec, and 2.4 + 0.66 m/see, respectively. I n all mice the E / A ratio was greater than 1 (range 1.76 to 3.6). Color-flow imaging dearly displayed normal mitral inflow and left ventricular outflow. In one mouse, aortic regurgitation was recorded by pulsed Doppler echocardiography. Echocardiographic, pulsed, and color-flow Doppler assessment o f mice is feasible. I n this study left ventricular mass was assessed better by 2-D measurement o f left ventricular dimensions. Assessment o f left ventricular performance is feasible. Color Doppler-guided evaluation o f aortic flow and aortic root measurement permits assessment o f stroke volume and cardiac output. (] AM S o c ECHOCARDIOGR1995 ;8:602-10.)
]g, chocardiography is an established procedure for the investigation o f cardiac performance in research studies that use large animals (e.g., cats or dogs). T h e limited resolution for standard echocardiographic diagnostic transducers has prevented the application o f echocardiography for smaller animals (e.g., rats and mice). W i t h the increasing use o f transgcnic mice in cardiovascular research, the need for a reliable n o n invasivc test o f cardiac performance has b e c o m e apparent. Recent studies 1 3 have suggested that M - m o d e transthoracic echocardiography permits assessment o f left ventricular mass. However, there is a paucity o f information on two-dimensional (2-D) cardiac structure and function in the mouse. Re-
search studies o n cardiac performance in the mouse are n o t rcstricted to evaluation o f left ventricular mass and, for this m o d e l to be m o r e useful, it should be s h o w n to be able to assess left ventricular function and cardiac performance fully. I n this study we r e p o r t the findings o f c o m b i n e d echocardiography and cardiac D o p p l e r with b o t h color-flow imaging and pulsed-wave Doppler echocardiography to evaluate cardiac performance in 15 healthy mice.
From the Heart Institute, the Hospital of the Good Samaritan. Reprint requests: Charles Pollick, MBChB, the Heart Institute, the Hospital of the Good Samaritan, 616 S. Wirmer St., Los Angeles, CA 90017. Copyright 9 1995 by the American Society of Echocardiography. 0894-7317/95 $5.00 + 0 27/1/61704 602
METHODS
Fifteen mice weighing from 38 to 65 gm (mean 48 + 7.6 gm) were assessed. Anesthetic administered was ketamine, 3.3 rag/mouse, and xylazine,0.3 mg/mouse, and was injected into the peritoneum; the mouse was positioned on a bed of gauze to permit echocardiographic assessment (Figure 1). M-mode and 2-D echocardiographic studies were performed with an ATL Interspec Apogee CX200 ultrasound
Journal of the AmericanSocietyof Echocardiography Volume 8 Number 5, Part 1
Figure 1
Pollick, Hale, mad Kloner
603
Position o f transducer during echocardiographic study.
system (Ambler, Pa.). The transducer frequency was 9 M H z for echocardiographic studies and 6.5 M H z for Doppler studies. The transducer was positioned on the left anterior side o f the chest after the animal's chest had been shaved. Parasternal long-axis views were obtained and recorded on videotape. 2-D echocardiography-directed M-mode of the left ventricle at the chordal level (from either the long- or short-axis 2-D view, depending on which view gave the best M-mode endocardial resolution) was obtained and recorded at a paper speed of 100 mm/sec. A 2-D short-axis view of the left ventricle was then obtained at the chordal level. From the M-mode tracings, measurements (inner edge to inner edge 4) were made o f left ventricular cavitary dimension, septal and posterior wall thickness, at end diastole and end systole, and fractional shortening. From the 2-D still frames, septal and posterior wall thickness and left ventricular cavitary dimension at end diastole were determined (inner edge to inner edge4). M-mode and 2-D measurements were made off-line from the video with a Freeland Systems (Colorado) Cardiology Workstation or Nova Microsonics ImageVue (Indianapolis, Ind.). Two to five beats were measured and the mean was recorded. Measurements were made by one experienced observer (C.R) with previously reported minimal intraobserver and interobserver variaiblity for left ventricular cavitary and wall dimensionsr Left ventricular mass (LVM) was determined according to a previously determined formula ~ for assessing left ventricular mass in rats: LVM = 1.055[(LVd + IVS + PW) 3 - (LVd3)] where LVd = left ventricular cavity in diastole, IVS =
inrerventricular septal thickness, and PW = posterior wall thickness. For Doppler studies, a midprecordial long-axis view of the heart was acquired to obtain flow as parallel to the transducer as possible (usually within 15 degrees); purely vertical apical four-chamber views were difficult to obtain. Color-flow imaging was performed and pulsed-Doppler assessment of left ventricular inflow and left ventricular outflow was recorded with the color-flow imaging a a guide to place the pulsed-Doppler sample volume. Peak mitral and aortic velocities were assessed in 10 and nine mice and measured off-line from the video. Necropsy assessment o f left ventriuclar mass was obtained by excluding the left ventricle from the rest of the heart and fixing it in formalin.
RESULTS Feasibility Excellent visualization o f the lcft vcntriclc was seen in all mice, w i t h clear M - m o d e images o f t h e left side o f the s e p t u m and the c n d o c a r d i u m a n d c p i c a r d i u m o f the p o s t e r i o r wall ( F i g u r e 2). T h e r i g h t side o f the i n t e r v e n t r i c u l a r s e p t u m was less well visualized, possibly because t h e r i g h t ventricular cavity was so small. 2 - D parasternal l o n g - a n d short-axis views were also o b t a i n a b l e in all the mice ( F i g u r e 3). A l t h o u g h true apical views were difficult to o b t a i n , it was nevertheless possible to acquirc m i d p r c c o r d i a l views with the
Journal of the American Society of Echocardiography
604
Pollick, Hale, and Kloner
September-October 1995
Figure 2 M-mode echocardiography of left ventricle. Depth markers are 2 mm.
Figure 3 Parasternal 2-D long-axis view of left ventricle. Depth markers are 2 ram. LV, Left ventricle; LA, left atrium; AO, aorta.
long axis o f the heart approximately 60 degrees from the horizontal view (apical view was at 90 degrees). In most mice the aortic root, aortic valve, left atrium, and mitral valve and apparatus could be identified.
Measurements
The normal M - m o d e values and 2-D cross-sectional short-axis values are shown in Tablc 1. Lcft vcm tricular mass corrclations with actual nccropsy-
Journal of the American Society of Echocardiography Volume 8 Number 5, Part 1
Pollick, Hale, a n d Kloner
0.20
605
2-D Analysis Left VentricularMass
0,18
~"
0.16
>
0.14
"5 _? t~ 0
0.12 _
~
r=0.71
0.10
~"
0.08 0.08
9
I
I
I
I
I
I
I
0.09
0.10
0.11
0.12
0.13
0.14
0.15
LV Weight (g) Figure 4
Correlation o f necropsy w i t h 2-D-derived left ventricular
(LV) mass.
M-Mode Analysis Left VenlricularMass
028 0.26 0,24 "~ 0.22 =- 0.20 > ~
0.18 0,16
---------1I
ot~ 0.14 o 0.12
9
y=.26 (x) -~ 12 r=O.07
0,10 0.08 0.01
I
I
I
[
I
I
I
0.09
0.10
0.11
0.12
O13
0.14
0.15
LV Weight (g) Figure 5
Correlation of necropsy with M-mode-derived left ventricular
d e t e r m i n e d left ventricular mass in 13 mice w e r e as follows: 2 - D - d e r i v e d left ventricular mass: m e a n 0.11 + 0.03 g m , r a n g e 0.05 to 0.18 g m , r = 0.70, a n d p = 0 . 0 0 7 0 ; M - m o d e - d e r i v e d left ventricular mass: m e a n 0.15 -+ 0.05 gin, r a n g e 0.09 t o 0.25 gin,
(LV) mass.
r = 0.07, a n d p -- 0.82; actual left ventricular mass: 0.12 + 0.01 g m a n d range 0.09 to 0 . 1 4 g m (Figures 4 a n d 5). Left ventricular f u n c t i o n was assessed from the M - m o d e e c h o c a r d i o g r a m . H e a r t rate varied f r o m
606
Journal of the American Society of Echocardiography September-October 1995
Pollick, Hale, and Kloner
Figure 6 Color-flow image of mitral inflow. Arrows point to pulmonary veins (inflow depicted in red). Abbreviations are given in Figure 3.
Table 1 M-mode and 2-D measurements (n = 15) Echocardiography
IVS (ram)
PW (mm)
LVd (mm)
LVs (ram)
FS (%)
LV mass (gm; n ~ 13)
1.13 0.20
0.99 0.17
3.67 0.7I
1.68 0.45
53.7 10.7
0.16 0.05
1.03 0.15
1.11 0.289
2.98* 0.58
M-mode
Mean SD 2-D Mean SD
m
0.11 0.03
IVS, Interventricular septum; PW, posterior wall; LV, left ventficular cavity (d, diastole; s, systole); FS, fractional shortening. *p = 0.003, 2-D left ventricular cavity in diastole versus M mode.
129 to 353 beats/min (mean 236 + 66 beats/min). Fractional shortening was (mean) 53.7% + 10.7% and (range) 42% to 77%. Maximum left ventricular shortening rate was (mean) 0.88 + 0.30 cm/sec; maximum left ventricular filling rate was (mean) 0.91 +_ 0.28 cm/sec. Doppler flow studies were obtainable in each of the 10 mice in which it was attempted. Color-flow imaging clearly demonstrated laminar flow coming into the left atrium and left ventricle, and in at least half o f the animals flow coming from the pulmonary veins could be seen (Figure 6). Left ventricular and aortic outflow could also be seen on color-flow imaging (Figure 7). N o valvular regurgitation was detected on color-flow imaging in any o f the mice. Laminar flow was seen in all studies. Pulsed Doppler evaluation ofmitral inflow and left
ventricular or aortic outflow was performed in 10 mice. Mitral inflow pattern showed a typical double peak with early diastolic (E) and later peak atrial (A) contraction (Figure 8). The values are shown in Table 2. The aortic velocity was also a typical single peak (Figure 9), and the velocities are shown in Table 3. Stroke volume was determined in eight mice by the following equationT: cross-sectional area ( d / 2 2 ) o f the aortic root x aortic velocity-time integral, where d = aortic root diameter. Aortic root measured 1.66 -+ 0.2 mm (range 1.4 to 2.0 ram). Mean _+ SD for stroke volume and cardiac output was 0.0505 + 0.0185 m l / b e a t and 12.64 _+ 7.87 m l / m i n , respectively. In one mouse there was a turbulent signal in diastole consistent with aortic regurgitation (Figure 10).
Journal of the American Societ3r of Echocardiography Volume 8 Number 5, Part i
Pollick, Hale, and Kloner 607
Figure 7 Color-flowimage of aortic outflow (AO) (blue).
Figure 8 Pulsed Doppler velocityprofile of mitral inflow.
DISCUSSION
This study demonstrates the feasibility of performing M-mode and 2-D transthoracic echocardiography and cardiac Doppler echocardiography with both color-flow imaging and pulsed-wave Doppler echo-
cardiography in mice. For relativity to clinical echocardiography it seems worthwhile to point out that excellent visualization was obtained of cardiac structure (down to the fine structure of chordae and papillary muscles [Figure 3] and pulmonary veins [Figure 6]) and Doppler function in the mouse heart
Journal of the American Society of Echocardiography
608
Pollick, Hale, and Kloner
September-October 199S
Table 2 Mitral inflow velocity (n = 10) Velocity (cm/sec)
Mean SD
E
A
E/A ratio
44.6 8.9
19.3 5.7
2.4 0.7
Table 3 Aortic veloctiy (n = 9)
EF slope (cm/sec z)
Deceleration time (msec)
743 341
S0 IS
Pressure half-time (msec) 16 6
that in its total transverse dimension o f approximately S mm is half the thickness o f the normal human scptum. The lack o f previous studies demonstrating the ability to portray 2-D echocardiographic and Doppler information in mice may be related to the type o f echocardiographic equipment used. To obtain adequate echocardiographic images of small structures (e.g., a mouse's heart), it is necessary to use a relatively high ultrasound frequency that has a high resolution (i.e., can distinguish small measurements down to 0.1 or 0.2 mm). The limitation o f highfrequency transducers is that they usually have reduced penetration capabilities. In this study, however, the transducer uses electronic phased-array technology that creates a narrow, tight focused beam that extends throughout the entire field o f view.8 This technology permits improved penetration for the same-frequency transducer compared with more conventional linear phased array systems. 8'9 Another advantage o f the transducer used in this study is the small footprint inherent in the spheric design o f the surface o f the probe, which may enable better probechest contact than the larger flat-head transducers on other echocardiography machines. Echocardiographic measurements obtained from 2-D short-axis views showed good correlation with necropsy-derived left ventricular mass, suggesting that wall thickness and measurements would be useful in mice undergoing research studies in which left ventricular wall thickness might be affected. M-mode-derived left ventricular mass measurements correlated poorly with left ventricular mass. Two possible reasons exist for this discrepancy. First, the right ventricle is not well seen and therefore definition o f the right side o f the interventricular septum is unclear. Because there was no significant difference between septal thickness derived from M-mode or 2-D data (p = 0.10), however, this reason seems unlikely. Second, the M-mode-derived left ventricular cavitary dimension may have been inaccurate if the beam was not perpendicular to the left ventricular long axis. Indeed, the left ventricular cavitary dimension was significantly different from M-modederived left ventricular dimension (p = 0.003). The M-mode was usually taken from the parasternal long-axis view, and this may explain the different left
Mean SD
Peak velocity (m/sec)
Peak gradient (ram Hg)
Systolic VTI (m)
El" (sec)
AT (see)
0.$3 0.13
0.52 0.23
0.02S 0.008
0.063 0.007
0.019 0.007
VTI, Velocity-time integral; ET, ejection time; AT, acceleration time.
ventricular diastolic dimensions from the 2-D shortaxis left ventricular dimensions, although care was tal~en to ensure the M-mode beam to be perpendicular. Manning et al.3 used 2-D-guided M-mode echocardiography in the short-axis views and found a good correlation between M-mode measurements and left ventricular mass. Better correlation coefficients would likely have been obtained (in our study) if a wider range o f hearts with significantly varying masses had been examined and possibily if we had used only short-axis-derived M-mode measurements. Despite the differences between M-mode- and 2-D-derived estimates o f left ventricular mass, digitization o f M-mode left ventricular tracings was feasible because the left side o f the interventricular septum and the endocardium o f the posterior wall were well seen. Fractional shortening and other M-mode-derived changes in left ventricular dimension and 2-D area are easily measured, permitting future studies on systolic performance. Pulsed Doppler flow measurements were obtainable with color-flow imaging as a way of ensuring velocities that are maximal and parallel to the beam. Mitral inflow patterns were similar to normal human mitral inflow Doppler velocity profiles. In particular, the E / A ratio was always greater than 1. Peak mice mitral inflow velocities (0.4 m / s e e ; range 0.3 to 0.6 m / s e e ) were less than human values (normal range ~~ o f adult human mitral inflow velocity 0.6 to 1.3 m/see). Studies on diastolic function in mice could therefore be studied with pulsed Doppler echocardiography o f mitral inflow. Aortic velocity profiles were similar in contour to human profiles. Peak mice aortic velocities (mean 0.S m/see; range 0.3 to 0.7 m/see) were less than human values (normal range ~~ o f adult human aortic velocity 1.0 to 1.7 m/see). The estimated cardiac output was 12.64 m l / m i n , which is similar to the estimated cardiac output from a previously derived formula ~ based on body weight (cardiac output = 0.762 [body weight gm] ~ and, for this group o f mice, =14.67 m l / m i n . The Doppler values may be a slight underestimation of the true value because flow was approximately 15 degrees to the Doppler beam. Doppler color flow was well seen, and normal laminar flow o f mitral inflow and left ventricular
Journal of the American Society of Echocardiography Volume 8 Number 5, Part 1
Figure 9
Pollick, Hale, and Kloner
609
Pulsed Doppler velocity profile of left ventricular outflow. Arrows point to systole.
Figure 10 Pulsed Doppler velocity profile of left ventricular outflow shows aortic regurgitation (arrow).
outflow were observed consistently. In several mice, pulmonary venous flow could also be demonstrated. In one mouse aortic regurgitation was detected by pulsed Doppler echocardiography.
In conclusion, echocardiographic, pulsed, and color-flow Doppler assessment o f mice is feasible. In this study left ventricular mass was assessed better by 2-D measurement o f left ventricular dimensions. M-
610
Pollick, Hale, and Kloner
m o d e a n d 2 - D a s s e s s m e n t o f left v e n t r i c u l a r p e r f o r m a n c e is feasible. C o l o r D o p p l e r - g u i d e d e v a l u a t i o n of aortic flow and aortic root measurement permits a s s e s s m e n t o f s t r o k e v o l u m e a n d cardiac o u t p u t .
REFERENCES
1. Gardin JM, Siri F, Kitsis RN, Leinwand LA. Intermediatefrequency transthoracic echocardiographic imaging: a promising method for evaluating the left ventricle in transgenic mice [Abstract]. J AM Soc ECHOCARDIOGR1993;6:$31. 2. Harfley CJ, Michael LH, Johnston AL. High resolution echocardiography in mice [Abstract]. Circulation 1993;88:1276. 3. Manning WJ, Wei IY, Katz SE, Litwin SE, Douglas PS. In vivo assessment of LV mass in mice using high-frequency cardiac ultrasound: necropsy validation. Am J Physiol 1994;266:H1672-5. 4. Picard MH. M-mode echocardiography: principles and examination techniques. In: Weyman AE, ed. Principles and practice of ecbocardiography. 2nd ed. Philadelphia: Lea & Febiger, 1994:294.
Journal of the AmericanSocietyof Echocardiography September-October 1995
5. Pollick C, Fitzgerald PJ, Popp RL. Variability of digitized echocardiography: size, source, and means of reduction. Am J Cardiol 1983;51:576-82. 6. De Simone G, Wallerson DC, Volpe M, Devereux RB. Echocardiographic measurement of left ventricular mass and volume in normotensive and hypertensive rats: necropsy validation. Am J Hypertens i990;3:688-96. 7. Marshall SA, Weyman AE. Doppler estimation of volumetric flow. In: Weyman AE, ed. Principles and practice of echocardiography. 2nd ed. Philadelphia: Lea & Febiger, 1994:95578. 8. Ryan T, Armstrong WF, Feigenbaum H. Annular array technology: application to cardiac imaging. Echocardiography I987;4:203-14. 9. Faletra F, Cipriani M, Corno R, et al. Transthoracic highfrequency echocardiographic detection of atherosclerotic lesions in the descending portion of the left coronary artery. J AM Soc ECHOCARDIOGR1993;6:290-8. 10. Feigenbaum H. Echocardiography. 5th ed. Philadelphia: Lea & Febiger, 1994:675. 11. Foster HL, Small JD, Fox JG. The mouse in biomedical research, rot 3. Boston: Academic Press, 1983:251.
Recent studies have suggested that intermediatefrequency M - m o d e transthoracic echocardiographic imaging is a promising method for evaluating the left ventricle in transgenic mice. However, there is a paucity o f data regarding two-dimensional (2-D) echocardiography and cardiac Doppler echocardiography in this model. Therefore we studied 15 mice (body weights 38 to 65 gm) with an ultrasound system equipped with a 9 M H z transducer. M-mode, 2-D, pulsed, and color-flow Doppler studies were performed. Mean _+SD for septal, posterior wall, and left ventricular cavitary dimensions at end diastole were the following: M-mode: 1.1 + 0.2, 1.0 _+0.2, and 3.7 -+ 0.7 mm; 2-D: 1.0 -+ 0.2, 1.1 _+0.3, and 3.0 + 0.6 nun. Left ventricular fractional shortening was assessed from the M-mode echocardiogram: mean 53.7% _+10.7% (range 42% to 77%). 2-D assessment o f left ventricular mass correlated better with left ventricular mass identified at necropsy than left ventricular mass identified by M - m o d e echocardiography (r = 0.70; p = 0.007 versus r = 0.07; p not significant). 2-D visualization o f left ventricle, proximal aorta, and aortic and mitral valves
was excellent and was obtained mainly from a "parasternal" window. Apical views were more difficult to obtain. Mean + SD for aortic peak and mean velocities and velocity-time integral were 0.53 + 0.13, 0.32 + 0.08, and 0.025 + 0.008 m/see. Estimated stroke volume was 0.0506 + 0.018 ml/beat. Cardiac o u t p u t was 12.64 + 7.87 m l / m i n . Mean + SD for mitral peak E, peak A, and E / A ratio were 0.45 + 0.09 m/sec, 0.19 + 0.06 m/sec, and 2.4 + 0.66 m/see, respectively. I n all mice the E / A ratio was greater than 1 (range 1.76 to 3.6). Color-flow imaging dearly displayed normal mitral inflow and left ventricular outflow. In one mouse, aortic regurgitation was recorded by pulsed Doppler echocardiography. Echocardiographic, pulsed, and color-flow Doppler assessment o f mice is feasible. I n this study left ventricular mass was assessed better by 2-D measurement o f left ventricular dimensions. Assessment o f left ventricular performance is feasible. Color Doppler-guided evaluation o f aortic flow and aortic root measurement permits assessment o f stroke volume and cardiac output. (] AM S o c ECHOCARDIOGR1995 ;8:602-10.)
]g, chocardiography is an established procedure for the investigation o f cardiac performance in research studies that use large animals (e.g., cats or dogs). T h e limited resolution for standard echocardiographic diagnostic transducers has prevented the application o f echocardiography for smaller animals (e.g., rats and mice). W i t h the increasing use o f transgcnic mice in cardiovascular research, the need for a reliable n o n invasivc test o f cardiac performance has b e c o m e apparent. Recent studies 1 3 have suggested that M - m o d e transthoracic echocardiography permits assessment o f left ventricular mass. However, there is a paucity o f information on two-dimensional (2-D) cardiac structure and function in the mouse. Re-
search studies o n cardiac performance in the mouse are n o t rcstricted to evaluation o f left ventricular mass and, for this m o d e l to be m o r e useful, it should be s h o w n to be able to assess left ventricular function and cardiac performance fully. I n this study we r e p o r t the findings o f c o m b i n e d echocardiography and cardiac D o p p l e r with b o t h color-flow imaging and pulsed-wave Doppler echocardiography to evaluate cardiac performance in 15 healthy mice.
From the Heart Institute, the Hospital of the Good Samaritan. Reprint requests: Charles Pollick, MBChB, the Heart Institute, the Hospital of the Good Samaritan, 616 S. Wirmer St., Los Angeles, CA 90017. Copyright 9 1995 by the American Society of Echocardiography. 0894-7317/95 $5.00 + 0 27/1/61704 602
METHODS
Fifteen mice weighing from 38 to 65 gm (mean 48 + 7.6 gm) were assessed. Anesthetic administered was ketamine, 3.3 rag/mouse, and xylazine,0.3 mg/mouse, and was injected into the peritoneum; the mouse was positioned on a bed of gauze to permit echocardiographic assessment (Figure 1). M-mode and 2-D echocardiographic studies were performed with an ATL Interspec Apogee CX200 ultrasound
Journal of the AmericanSocietyof Echocardiography Volume 8 Number 5, Part 1
Figure 1
Pollick, Hale, mad Kloner
603
Position o f transducer during echocardiographic study.
system (Ambler, Pa.). The transducer frequency was 9 M H z for echocardiographic studies and 6.5 M H z for Doppler studies. The transducer was positioned on the left anterior side o f the chest after the animal's chest had been shaved. Parasternal long-axis views were obtained and recorded on videotape. 2-D echocardiography-directed M-mode of the left ventricle at the chordal level (from either the long- or short-axis 2-D view, depending on which view gave the best M-mode endocardial resolution) was obtained and recorded at a paper speed of 100 mm/sec. A 2-D short-axis view of the left ventricle was then obtained at the chordal level. From the M-mode tracings, measurements (inner edge to inner edge 4) were made o f left ventricular cavitary dimension, septal and posterior wall thickness, at end diastole and end systole, and fractional shortening. From the 2-D still frames, septal and posterior wall thickness and left ventricular cavitary dimension at end diastole were determined (inner edge to inner edge4). M-mode and 2-D measurements were made off-line from the video with a Freeland Systems (Colorado) Cardiology Workstation or Nova Microsonics ImageVue (Indianapolis, Ind.). Two to five beats were measured and the mean was recorded. Measurements were made by one experienced observer (C.R) with previously reported minimal intraobserver and interobserver variaiblity for left ventricular cavitary and wall dimensionsr Left ventricular mass (LVM) was determined according to a previously determined formula ~ for assessing left ventricular mass in rats: LVM = 1.055[(LVd + IVS + PW) 3 - (LVd3)] where LVd = left ventricular cavity in diastole, IVS =
inrerventricular septal thickness, and PW = posterior wall thickness. For Doppler studies, a midprecordial long-axis view of the heart was acquired to obtain flow as parallel to the transducer as possible (usually within 15 degrees); purely vertical apical four-chamber views were difficult to obtain. Color-flow imaging was performed and pulsed-Doppler assessment of left ventricular inflow and left ventricular outflow was recorded with the color-flow imaging a a guide to place the pulsed-Doppler sample volume. Peak mitral and aortic velocities were assessed in 10 and nine mice and measured off-line from the video. Necropsy assessment o f left ventriuclar mass was obtained by excluding the left ventricle from the rest of the heart and fixing it in formalin.
RESULTS Feasibility Excellent visualization o f the lcft vcntriclc was seen in all mice, w i t h clear M - m o d e images o f t h e left side o f the s e p t u m and the c n d o c a r d i u m a n d c p i c a r d i u m o f the p o s t e r i o r wall ( F i g u r e 2). T h e r i g h t side o f the i n t e r v e n t r i c u l a r s e p t u m was less well visualized, possibly because t h e r i g h t ventricular cavity was so small. 2 - D parasternal l o n g - a n d short-axis views were also o b t a i n a b l e in all the mice ( F i g u r e 3). A l t h o u g h true apical views were difficult to o b t a i n , it was nevertheless possible to acquirc m i d p r c c o r d i a l views with the
Journal of the American Society of Echocardiography
604
Pollick, Hale, and Kloner
September-October 1995
Figure 2 M-mode echocardiography of left ventricle. Depth markers are 2 mm.
Figure 3 Parasternal 2-D long-axis view of left ventricle. Depth markers are 2 ram. LV, Left ventricle; LA, left atrium; AO, aorta.
long axis o f the heart approximately 60 degrees from the horizontal view (apical view was at 90 degrees). In most mice the aortic root, aortic valve, left atrium, and mitral valve and apparatus could be identified.
Measurements
The normal M - m o d e values and 2-D cross-sectional short-axis values are shown in Tablc 1. Lcft vcm tricular mass corrclations with actual nccropsy-
Journal of the American Society of Echocardiography Volume 8 Number 5, Part 1
Pollick, Hale, a n d Kloner
0.20
605
2-D Analysis Left VentricularMass
0,18
~"
0.16
>
0.14
"5 _? t~ 0
0.12 _
~
r=0.71
0.10
~"
0.08 0.08
9
I
I
I
I
I
I
I
0.09
0.10
0.11
0.12
0.13
0.14
0.15
LV Weight (g) Figure 4
Correlation o f necropsy w i t h 2-D-derived left ventricular
(LV) mass.
M-Mode Analysis Left VenlricularMass
028 0.26 0,24 "~ 0.22 =- 0.20 > ~
0.18 0,16
---------1I
ot~ 0.14 o 0.12
9
y=.26 (x) -~ 12 r=O.07
0,10 0.08 0.01
I
I
I
[
I
I
I
0.09
0.10
0.11
0.12
O13
0.14
0.15
LV Weight (g) Figure 5
Correlation of necropsy with M-mode-derived left ventricular
d e t e r m i n e d left ventricular mass in 13 mice w e r e as follows: 2 - D - d e r i v e d left ventricular mass: m e a n 0.11 + 0.03 g m , r a n g e 0.05 to 0.18 g m , r = 0.70, a n d p = 0 . 0 0 7 0 ; M - m o d e - d e r i v e d left ventricular mass: m e a n 0.15 -+ 0.05 gin, r a n g e 0.09 t o 0.25 gin,
(LV) mass.
r = 0.07, a n d p -- 0.82; actual left ventricular mass: 0.12 + 0.01 g m a n d range 0.09 to 0 . 1 4 g m (Figures 4 a n d 5). Left ventricular f u n c t i o n was assessed from the M - m o d e e c h o c a r d i o g r a m . H e a r t rate varied f r o m
606
Journal of the American Society of Echocardiography September-October 1995
Pollick, Hale, and Kloner
Figure 6 Color-flow image of mitral inflow. Arrows point to pulmonary veins (inflow depicted in red). Abbreviations are given in Figure 3.
Table 1 M-mode and 2-D measurements (n = 15) Echocardiography
IVS (ram)
PW (mm)
LVd (mm)
LVs (ram)
FS (%)
LV mass (gm; n ~ 13)
1.13 0.20
0.99 0.17
3.67 0.7I
1.68 0.45
53.7 10.7
0.16 0.05
1.03 0.15
1.11 0.289
2.98* 0.58
M-mode
Mean SD 2-D Mean SD
m
0.11 0.03
IVS, Interventricular septum; PW, posterior wall; LV, left ventficular cavity (d, diastole; s, systole); FS, fractional shortening. *p = 0.003, 2-D left ventricular cavity in diastole versus M mode.
129 to 353 beats/min (mean 236 + 66 beats/min). Fractional shortening was (mean) 53.7% + 10.7% and (range) 42% to 77%. Maximum left ventricular shortening rate was (mean) 0.88 + 0.30 cm/sec; maximum left ventricular filling rate was (mean) 0.91 +_ 0.28 cm/sec. Doppler flow studies were obtainable in each of the 10 mice in which it was attempted. Color-flow imaging clearly demonstrated laminar flow coming into the left atrium and left ventricle, and in at least half o f the animals flow coming from the pulmonary veins could be seen (Figure 6). Left ventricular and aortic outflow could also be seen on color-flow imaging (Figure 7). N o valvular regurgitation was detected on color-flow imaging in any o f the mice. Laminar flow was seen in all studies. Pulsed Doppler evaluation ofmitral inflow and left
ventricular or aortic outflow was performed in 10 mice. Mitral inflow pattern showed a typical double peak with early diastolic (E) and later peak atrial (A) contraction (Figure 8). The values are shown in Table 2. The aortic velocity was also a typical single peak (Figure 9), and the velocities are shown in Table 3. Stroke volume was determined in eight mice by the following equationT: cross-sectional area ( d / 2 2 ) o f the aortic root x aortic velocity-time integral, where d = aortic root diameter. Aortic root measured 1.66 -+ 0.2 mm (range 1.4 to 2.0 ram). Mean _+ SD for stroke volume and cardiac output was 0.0505 + 0.0185 m l / b e a t and 12.64 _+ 7.87 m l / m i n , respectively. In one mouse there was a turbulent signal in diastole consistent with aortic regurgitation (Figure 10).
Journal of the American Societ3r of Echocardiography Volume 8 Number 5, Part i
Pollick, Hale, and Kloner 607
Figure 7 Color-flowimage of aortic outflow (AO) (blue).
Figure 8 Pulsed Doppler velocityprofile of mitral inflow.
DISCUSSION
This study demonstrates the feasibility of performing M-mode and 2-D transthoracic echocardiography and cardiac Doppler echocardiography with both color-flow imaging and pulsed-wave Doppler echo-
cardiography in mice. For relativity to clinical echocardiography it seems worthwhile to point out that excellent visualization was obtained of cardiac structure (down to the fine structure of chordae and papillary muscles [Figure 3] and pulmonary veins [Figure 6]) and Doppler function in the mouse heart
Journal of the American Society of Echocardiography
608
Pollick, Hale, and Kloner
September-October 199S
Table 2 Mitral inflow velocity (n = 10) Velocity (cm/sec)
Mean SD
E
A
E/A ratio
44.6 8.9
19.3 5.7
2.4 0.7
Table 3 Aortic veloctiy (n = 9)
EF slope (cm/sec z)
Deceleration time (msec)
743 341
S0 IS
Pressure half-time (msec) 16 6
that in its total transverse dimension o f approximately S mm is half the thickness o f the normal human scptum. The lack o f previous studies demonstrating the ability to portray 2-D echocardiographic and Doppler information in mice may be related to the type o f echocardiographic equipment used. To obtain adequate echocardiographic images of small structures (e.g., a mouse's heart), it is necessary to use a relatively high ultrasound frequency that has a high resolution (i.e., can distinguish small measurements down to 0.1 or 0.2 mm). The limitation o f highfrequency transducers is that they usually have reduced penetration capabilities. In this study, however, the transducer uses electronic phased-array technology that creates a narrow, tight focused beam that extends throughout the entire field o f view.8 This technology permits improved penetration for the same-frequency transducer compared with more conventional linear phased array systems. 8'9 Another advantage o f the transducer used in this study is the small footprint inherent in the spheric design o f the surface o f the probe, which may enable better probechest contact than the larger flat-head transducers on other echocardiography machines. Echocardiographic measurements obtained from 2-D short-axis views showed good correlation with necropsy-derived left ventricular mass, suggesting that wall thickness and measurements would be useful in mice undergoing research studies in which left ventricular wall thickness might be affected. M-mode-derived left ventricular mass measurements correlated poorly with left ventricular mass. Two possible reasons exist for this discrepancy. First, the right ventricle is not well seen and therefore definition o f the right side o f the interventricular septum is unclear. Because there was no significant difference between septal thickness derived from M-mode or 2-D data (p = 0.10), however, this reason seems unlikely. Second, the M-mode-derived left ventricular cavitary dimension may have been inaccurate if the beam was not perpendicular to the left ventricular long axis. Indeed, the left ventricular cavitary dimension was significantly different from M-modederived left ventricular dimension (p = 0.003). The M-mode was usually taken from the parasternal long-axis view, and this may explain the different left
Mean SD
Peak velocity (m/sec)
Peak gradient (ram Hg)
Systolic VTI (m)
El" (sec)
AT (see)
0.$3 0.13
0.52 0.23
0.02S 0.008
0.063 0.007
0.019 0.007
VTI, Velocity-time integral; ET, ejection time; AT, acceleration time.
ventricular diastolic dimensions from the 2-D shortaxis left ventricular dimensions, although care was tal~en to ensure the M-mode beam to be perpendicular. Manning et al.3 used 2-D-guided M-mode echocardiography in the short-axis views and found a good correlation between M-mode measurements and left ventricular mass. Better correlation coefficients would likely have been obtained (in our study) if a wider range o f hearts with significantly varying masses had been examined and possibily if we had used only short-axis-derived M-mode measurements. Despite the differences between M-mode- and 2-D-derived estimates o f left ventricular mass, digitization o f M-mode left ventricular tracings was feasible because the left side o f the interventricular septum and the endocardium o f the posterior wall were well seen. Fractional shortening and other M-mode-derived changes in left ventricular dimension and 2-D area are easily measured, permitting future studies on systolic performance. Pulsed Doppler flow measurements were obtainable with color-flow imaging as a way of ensuring velocities that are maximal and parallel to the beam. Mitral inflow patterns were similar to normal human mitral inflow Doppler velocity profiles. In particular, the E / A ratio was always greater than 1. Peak mice mitral inflow velocities (0.4 m / s e e ; range 0.3 to 0.6 m / s e e ) were less than human values (normal range ~~ o f adult human mitral inflow velocity 0.6 to 1.3 m/see). Studies on diastolic function in mice could therefore be studied with pulsed Doppler echocardiography o f mitral inflow. Aortic velocity profiles were similar in contour to human profiles. Peak mice aortic velocities (mean 0.S m/see; range 0.3 to 0.7 m/see) were less than human values (normal range ~~ o f adult human aortic velocity 1.0 to 1.7 m/see). The estimated cardiac output was 12.64 m l / m i n , which is similar to the estimated cardiac output from a previously derived formula ~ based on body weight (cardiac output = 0.762 [body weight gm] ~ and, for this group o f mice, =14.67 m l / m i n . The Doppler values may be a slight underestimation of the true value because flow was approximately 15 degrees to the Doppler beam. Doppler color flow was well seen, and normal laminar flow o f mitral inflow and left ventricular
Journal of the American Society of Echocardiography Volume 8 Number 5, Part 1
Figure 9
Pollick, Hale, and Kloner
609
Pulsed Doppler velocity profile of left ventricular outflow. Arrows point to systole.
Figure 10 Pulsed Doppler velocity profile of left ventricular outflow shows aortic regurgitation (arrow).
outflow were observed consistently. In several mice, pulmonary venous flow could also be demonstrated. In one mouse aortic regurgitation was detected by pulsed Doppler echocardiography.
In conclusion, echocardiographic, pulsed, and color-flow Doppler assessment o f mice is feasible. In this study left ventricular mass was assessed better by 2-D measurement o f left ventricular dimensions. M-
610
Pollick, Hale, and Kloner
m o d e a n d 2 - D a s s e s s m e n t o f left v e n t r i c u l a r p e r f o r m a n c e is feasible. C o l o r D o p p l e r - g u i d e d e v a l u a t i o n of aortic flow and aortic root measurement permits a s s e s s m e n t o f s t r o k e v o l u m e a n d cardiac o u t p u t .
REFERENCES
1. Gardin JM, Siri F, Kitsis RN, Leinwand LA. Intermediatefrequency transthoracic echocardiographic imaging: a promising method for evaluating the left ventricle in transgenic mice [Abstract]. J AM Soc ECHOCARDIOGR1993;6:$31. 2. Harfley CJ, Michael LH, Johnston AL. High resolution echocardiography in mice [Abstract]. Circulation 1993;88:1276. 3. Manning WJ, Wei IY, Katz SE, Litwin SE, Douglas PS. In vivo assessment of LV mass in mice using high-frequency cardiac ultrasound: necropsy validation. Am J Physiol 1994;266:H1672-5. 4. Picard MH. M-mode echocardiography: principles and examination techniques. In: Weyman AE, ed. Principles and practice of ecbocardiography. 2nd ed. Philadelphia: Lea & Febiger, 1994:294.
Journal of the AmericanSocietyof Echocardiography September-October 1995
5. Pollick C, Fitzgerald PJ, Popp RL. Variability of digitized echocardiography: size, source, and means of reduction. Am J Cardiol 1983;51:576-82. 6. De Simone G, Wallerson DC, Volpe M, Devereux RB. Echocardiographic measurement of left ventricular mass and volume in normotensive and hypertensive rats: necropsy validation. Am J Hypertens i990;3:688-96. 7. Marshall SA, Weyman AE. Doppler estimation of volumetric flow. In: Weyman AE, ed. Principles and practice of echocardiography. 2nd ed. Philadelphia: Lea & Febiger, 1994:95578. 8. Ryan T, Armstrong WF, Feigenbaum H. Annular array technology: application to cardiac imaging. Echocardiography I987;4:203-14. 9. Faletra F, Cipriani M, Corno R, et al. Transthoracic highfrequency echocardiographic detection of atherosclerotic lesions in the descending portion of the left coronary artery. J AM Soc ECHOCARDIOGR1993;6:290-8. 10. Feigenbaum H. Echocardiography. 5th ed. Philadelphia: Lea & Febiger, 1994:675. 11. Foster HL, Small JD, Fox JG. The mouse in biomedical research, rot 3. Boston: Academic Press, 1983:251.