Use of echocardiography for the phenotypic assessment of genetically altered mice1
Keith A. Collins1,
Claudia E. Korcarz2 and
Roberto M. Lang1
1 NONINVASIVE CARDIAC IMAGING LABORATORY, UNIVERSITY OF CHICAGO, CHICAGO, ILLINOIS 60637
2 Department of Cardiology, University of Wisconsin-Madison, Madison, Wisconsin 53792
 |
ABSTRACT
|
---|
Transgenic mice displaying abnormalities in cardiac development and function represent a powerful new tool for understanding molecular mechanisms underlying normal cardiovascular function and the pathophysiological bases of human cardiovascular disease. Complete cardiac evaluation of phenotypic changes in mice requires the ability to noninvasively assess cardiovascular structure and function in a serial manner. However, the small mouse heart beating at rates in excess of 500 beats/min presents unique methodological challenges. Two-dimensional and Doppler echocardiography have been recently used as effective, noninvasive tools for murine imaging, because quality images of cardiac structures and valvular flows can be obtained with newer high-frequency transthoracic transducers. We will discuss the use of echocardiography for the assessment of 1) left ventricular (LV) chamber dimensions and wall thicknesses, 2) LV mass, 3) improved endocardial border delineation using contrast echocardiography, 4) LV contractility using ejection phase indices and load-independent indices, 5) vascular properties, and 6) LV diastolic performance. Evaluation of cardiovascular performance in closed chest mice is feasible in a variety of murine models using Doppler echocardiographic imaging.
Doppler echocardiography; left ventricular performance
 |
INTRODUCTION
|
---|
GENETIC MODIFICATIONS are frequently used to produce mice models that are used to investigate the molecular basis of cardiac growth and development (79, 18, 24, 25) These technologies have led to a proliferation of transgenic and knockout mice models displaying a variety of cardiovascular phenotypes. To evaluate these phenotypes, it is necessary to develop accurate, reproducible, and noninvasive methods to assess cardiac morphology and function in a serial manner.
Although it is currently possible to assess cardiac hemodynamics and function using ventricular catheterization, radiolabeled microspheres, and thermodilution techniques in mice, these methods require invasive instrumentation, which restrict the ability to measure physiological changes in a serial manner. Ultrasound imaging has been increasingly applied to identify and characterize structural and functional features of different cardiac phenotypes and pathophysiological responses to surgical and pharmacological interventions in large animal models of human disease (79, 16, 17, 27, 43). The small size of the mouse heart, which beats at heart rates in excess of 500 beats/min, presents unique methodological challenges for cardiac ultrasound. Recently developed broadband, phased-array, and linear transducers have small footprints, which are capable of both high frame-rate imaging and improved near-field imaging, thereby generating high-quality images of the mouse heart.
Imaging of the small, fast-beating murine hearts also requires special technical attention to the selection of proper anesthesia. In addition to anesthetic effects, in this review, we will discuss the preparation for image acquisition using currently available echocardiographic equipment. With these considerations in mind, we will discuss how two-dimensional (2D) and Doppler echocardiographic methods may be reliably used in murine models of cardiac disease for the evaluation of left ventricular (LV) cardiovascular structure and function.
 |
Technical Considerations
|
---|
Anesthesia
Genetically and surgically altered mice models offer a great potential to identify molecular and genetic factors contributing to the pathophysiological progression of disease. Consequently, a reliable and accurate assessment of cardiac function is critical to screen or characterize the altered or impaired cardiac state compared with control mice. Although anesthetic agents are frequently employed to immobilize and sedate mice for better image acquisition during echocardiographic imaging, these drugs are known to have significant effects on cardiovascular function. General anesthetics may directly affect organ function as well as blunt both the respiratory drive and respiratory muscle function, potentially inducing cardiac dysfunction by means of hypoxia and acidosis. To assess and minimize this confounding variable, investigators have routinely measured end-tidal carbon dioxide tension or mechanical ventilation parameters while simultaneously attempting to minimize the duration of anesthesia. Heart rates in conscious normal animals have been reported to vary from 600 to 650 beats/min by telemetry methods (20, 39, 44). The reduction in heart rate caused by anesthetics in different research models may result in better temporal resolution for improved image visualization and measurement, but may simultaneously confound the physiological issue in question.
Recently, increased attention has been placed on the variable effects of anesthetics on murine cardiac function during echocardiographic recordings (1, 3, 13, 31). Many different regimens of general anesthesia have been utilized in mice, including intraperitoneal injection of ketamine:xylazine mixture, tribromoethanol (Avertin), chloral hydrate, or barbiturates, as well as inhalation of isoflurane or halothane. It is important to consider when sedating mice to perform an echocardiographic study that most anesthetics including barbiturates and inhalants cause cardiovascular and respiratory depression. Although ketamine reportedly has less of a cardiodepressor effect, when administered with xylazine or diazepam, the combination results in hypothermia and negative inotropic and chronotropic effects (1, 13, 31). Roth et al. (31) recently reported that the intraperitoneal agents, tribromoethanol, ketamine-midazolam, or ketamine:xylazine all cause early cardiodepression and decreased shortening fraction over the course of a 20-min echocardiographic study, although tribromoethanol resulted in lesser hemodynamic compromise. Comparatively, isoflurane anesthesia resulted in the most stable fractional shortening and end-diastolic dimension values during the study and the most reproducible measurements in repeated studies.
To avoid the confounding effects of anesthesia, recordings of echocardiographic data in conscious mice have been attempted. Several groups (38, 39, 44) have reported the feasibility of echocardiographic data acquisition in conscious mice to assess LV systolic and diastolic function in normal mice and murine models of hypertension or myocardial infarction. However, acquisition of echocardiographic studies in conscious mice requires multiple training sessions in handling the animals, increases the operators demand to focus on the animal and environmental conditions, and is not possible in catheterized or instrumented animals (38, 39, 44). The consensus among investigators performing echocardiographic studies in murine models is that the choice of anesthetic, dosing regimen, and timing of data acquisition must be carefully adapted not only to the particular experiments, but also to the mouse strain, sex, age, and mutation.
Preparation for Imaging
Following sedation, the animal should be placed in the left lateral decubitus position with electrocardiographic electrodes (adhesive, needle, or metal conductors) applied to the paws. To increase probe contact and reduce air bubbles, the imaging area must be closely shaved and wetted with alcohol or water. Since the effects of anesthesia-induced temperature changes on cardiac function and heart rate are well documented (1, 16, 19), some method of thermoregulation should be used to avoid hypothermia (e.g., circulating warming pad, heating lamps, autoregulated heating blankets, or warming of acoustic gel). Importantly, the sonographer acquiring the images needs to avoid placing excessive pressure on the chest cavity with the transducer, since even the weight of this imaging instrument alone may cause bradycardia and hypotension. Accordingly, a slight upward lifting of the transducer while continuing contact with the chest wall is recommended (43).
Echocardiographic Equipment
Echocardiographic assessment of murine hearts has historically been limited by 1) the low frame-rate image acquisition relative to the high heart rate of the mouse and 2) inappropriate transducer frequencies for near-field imaging. Recently developed broadband annular phased-array and linear transducers have smaller footprints and are capable of high frame-rate imaging, which are appropriate for the murine chest and heart rate, respectively. Because of the improved near-field imaging of these new transducers, it is now possible to avoid the use of acoustic standoffs to obtain adequate images. It is important to conduct phantom tests monthly to ensure reliable performance of the equipment, especially in long-term or serial studies.
 |
Structural Descriptors Of The Left Ventricle
|
---|
Chamber Dimensions and Wall Thicknesses
Cardiac ultrasound is a relatively quick and versatile method that allows serial, noninvasive quantitative assessment of chamber dimensions, thickness, and valvular structures, as well as cardiac function. The parasternal long-axis view is usually acquired initially because this view allows a general assessment of overall LV size and function. When the probe is positioned to simultaneously visualize the mitral valve and the LV apex in its maximum length, this view, while difficult to achieve in a single plane, provides a measure of LV length, left atrial (LA) size, mitral valve function, and aortic root dimensions. From this transducer position, a clockwise 90° rotation at the papillary muscle level will depict the short-axis view. From this view, 2D- or targeted M-mode-derived measures of LV area or dimensions can be obtained, using the leading edge-to-leading edge convention adopted by the American Society of Echocardiography (36). 2D imaging also allows visualization and measurement of the aortic root and main pulmonary artery diameters. M-mode imaging allows better temporal resolution and the estimation of aortic diameter, LV chamber dimensions, and thickness.
Temporal changes between LV end-systolic dimension (LVESD) and LV end-diastolic dimension (LVEDD) are used for the calculation of shortening fraction (FS), as follows
Diastolic measurements are made at the time of apparent maximal LV diastolic dimension, whereas LV end-systolic dimension are obtained at the time of the incisura of the aortic pressure tracing or at the time of minimal LV dimension.
Intra- and inter-observer variability between measurements of LVEDD, LVESD, FS, velocity of circumferential shortening (Vcf), and posterior wall thickness (PWTh) are
10% either by 2D or M-mode method (17), which is acceptable for accurate measurements. Measurements of individual parameters such as LVEDD and PWTh may vary in normal mice up to 25% (4, 17), requiring a comparable control group to accurately estimate and assess minor dimensional changes. This variability may be explained not only by differences between strains but also due to differences in anesthetic regimens and loading conditions.
Transthoracic 2D echocardiography has been employed extensively to discern structural and functional differences among genetically engineered mice. In our laboratory, transgenic mice expressing a dominant-negative form of the CREB transcription factor (CREBA133) were produced to develop a dilated cardiomyopathy (DCM) that closely resembles the anatomical, physiological, and clinical features of the human disease (79). Serial echocardiography was employed to evaluate the progression of the phenotype. Like humans with DCM, these animals displayed progressive four-chamber dilatation, significantly depressed LV systolic function, abnormal diastolic relaxation, and attenuated contractile responses to the ß-adrenergic agonist, isoproterenol (Fig. 1). This noninvasive assessment serves as a powerful complement to genetic and pathological interrogation when studying an animal model of human disease.

View larger version (68K):
[in this window]
[in a new window]
|
Fig. 1. M-mode echocardiographic analysis of CREBA133 mice (Tg), a transgenic model of dilated cardiomyopathy, compared with nontransgenic control littermates (NTg). Two-dimensionally (2D) targeted M-mode echocardiograms were obtained at the level of the papillary muscles at baseline (left) and after a continuous intravenous infusion of 1 ng/min isoproterenol (right). Note the increased end-diastolic (EDD) and end-systolic (ESD) dimensions and the decreased response to isoproterenol in the Tg mice, compared with the NTg controls. Phenotypic differences between the adult Tg mice compared with its littermate control may be observed by the larger heart size, body weight, and presence of generalized edema (far right). Reprinted with permission from Fentzke et al. (7).
|
|
Left Ventricular Mass
LV mass is a commonly used descriptor of cardiac status. Previous studies have examined the accuracy of M-mode and 2D LV mass measurement methods in small cohorts of predominantly normal mice (11, 23). Although M-mode echocardiography has yielded LVM estimates with relatively good correlation compared with necropsy values in mice of uniform geometry (11, 23, 40), this method is limited in mice with irregularly shaped hearts. With this method, images are obtained in a single plane, and thus LVM calculation may be subject to greater error than when using formulas derived from multi-planar images
In humans, 2D area-length-based estimates of LV mass have been shown to be more accurate than M-mode-based estimates (5, 36). In a recent study, we evaluated the accuracy of known echocardiographic formulas for estimating LV mass in a large number of mice with a wide range of LV sizes, cardiac geometries, and weights and determined the sources of error inherent to each formula (4). Specifically, we compared LV masses calculated using the cubed formula (M-mode) as well as area-length and truncated ellipsoid formulas (2D) with necropsy LV weights. Measurements were obtained as displayed in Fig. 2. The following formulas were used.

View larger version (153K):
[in this window]
[in a new window]
|
Fig. 2. Representative echocardiograms and measurements from the left ventricle (LV) of a normal CD-1 mouse. The 2D parasternal short-axis image (A) and long-axis image (B) are shown at end diastole. The 2D guided M-mode image (C) was obtained at the level of the papillary muscles. t, myocardial wall thickness; L, LV length; a, full major radius; b, minor axis radius; d, truncated major radius; IVS, interventricular septal thickness; LVID, LV internal diameter; LVPW, LV posterior wall thickness. Reprinted with permission from Collins et al. (4).
|
|
The M-mode (cubed) method.
With the M-mode (cubed) method
where IVS and LVPW are the interventricular septal and posterior wall thicknesses, respectively, and LVID is the LV internal diameter (5).
The 2D area-length method.
With the 2D area-length method
where 1.05 is the specific gravity of muscle, A1 and A2 are the epicardial and endocardial parasternal short-axis areas, respectively, L is the parasternal long-axis length, and t is the wall thickness calculated from A1 and A2 (36).
Truncated ellipsoid method.
With the truncated ellipsoid method
where b is the minor axis radius of the LV measured at the level of the papillary muscle tip. Its placement determines the division of the measured LV length (L) into a full major radius (a) and a truncated major radius (d). The average wall thickness, t, is calculated from A1 and A2 (5, 36).
We found that 1) the M-mode method systematically overestimated gravimetric LV mass; 2) LV mass estimates by 2D area-length method are more accurate than those obtained by the M-mode method (significantly reduced bias, and error and had higher correlation with necropsy values) (Fig. 3); and 3) although the 2D area-length method yielded mass estimates that correlated highly with necropsy weights (r = 0.90) and had insignificant bias (<2%), the error of individual estimates were relatively large (
25%, from Bland-Altman analysis). Surprisingly, the use of a truncated ellipsoid geometric model, which better accounts for LV shape variability (i.e., aortic-banded models, transgenic models of DCM), did not result in improved correlation with necropsy LV weights.

View larger version (14K):
[in this window]
[in a new window]
|
Fig. 3. Bland-Altman analyses showing the agreement as a percent difference between necropsy LV weight and LV mass (LVM) calculated by the 2D guided M-mode method (A) or by the area-length (AL) method (B) at end systole in 89 normal CD-1, transgenic CREBA133, and aortic-banded mice. Horizontal reference lines are zero percent difference (bold), the bias (center thin line), and 2 standard deviations (SD) above and below the bias (outlying thin lines). Bias values (±2 SD) for the AL method, compared with the M-mode method, were 1.7 ± 28.6% vs. 23.0 ± 42.1%, respectively. Reprinted with permission from Collins et al. (4).
|
|
Sample size calculations indicate that the lower inter- and intra-observer variabilities in LV mass measurements calculated with the area-length method provide an additional advantage over the M-mode method, because with the use of this formula, future studies would require smaller groups of mice to detect significant differences.
A recent study by Kanno et al. (21) applied the Simpson rule for reconstructing infarct area size, total LVM, and derived LV ejection fraction. The Simpson rule is advantageous to evaluate LV mass in experiments involving regional remodeling because this method represents the LV cavity as a stack of disks wherein the volumes of all disks are summed to obtain a close approximation of LV volume. LV mass is calculated by subtracting the volume of the LV cavity at end systole from the LV volume measured at end diastole and multiplying by the density of myocardium (1.055). In this manner, Kanno et al. (21) accurately measured infarct size induced by left anterior descending (LAD) occlusion in mice using 2D echocardiography (Fig. 4). In summary, we believe that future improvements in accuracy of LVM measurements are likely to result from instrumentation modifications rather than refinement of geometric models.

View larger version (97K):
[in this window]
[in a new window]
|
Fig. 4. Echocardiographic analysis of LV volumes and infarct size. A: sequential short-axis images of LV were acquired from base to apex. B: endocardial border was traced at end diastole (Endo-ed) and end systole (Endo-es) and epicardial border at end diastole (Epi-ed). Borders of akinetic myocardial segments were also traced. C: distances between short-axis slices and LV length were measured from long-axis image. D: LV volumes, LV mass, and infarct mass were calculated based on 3D reconstruction of LV geometry by method of disks. Reprinted with permission from Kanno et al. (21).
|
|
Advances in clinical echocardiography to further delineate endocardial borders in humans, such as the use of intravenous contrast agents, have recently been extended to murine echocardiographic studies. In our laboratory, a bolus of Optison, an albumin-coated microsphere containing a high-molecular-weight, perfluorocarbon gas, was injected intravenously, resulting in enhanced endocardial border definition (25). This allowed quantitative assessment of LV size and function using frame-by-frame endocardial border tracing to generate LV area-time curves (Fig. 5).

View larger version (81K):
[in this window]
[in a new window]
|
Fig. 5. To improve endocardial visualization and thus facilitate endocardial tracings, contrast enhancement of the LV cavity can be achieved with intravenous injections of echocardiographic contrast. The contrast-enhanced end-diastolic (A) and end-systolic (B) frames in a normal CD-1 mouse are displayed in short-axis view of the LV with LV endocardial border traced and corresponding areas calculated. LV area-over-time waveforms were obtained by manually tracing LV endocardial boundary frame by frame. Example of LV area vs. time waveforms obtained in a normal nontransgenic (NTg) mouse (C) and a transgenic CREBA133 (Tg) mouse (D) are displayed under baseline conditions (BL, solid circles) and with dobutamine infusion (DOB, open circles). Reprinted with permission from Mor-Avi et al. (25).
|
|
Scherrer-Crosbie et al. (35) also employed an ultrasound contrast agent (Definity) to study myocardial perfusion. This was done by injecting the contrast agent to quantify the ischemic risk area after coronary occlusion using time-videointensity curves. The manually traced perfusion defect areas correlated highly with postmortem die exclusion techniques (r = 0.93, slope = 0.77). The use of contrast in murine echocardiographic studies may be useful in future experiments to both improve the delineation of endocardial border and assess myocardial perfusion. However, this technique is limited by the need for intravenous access. To date, there is no standardized dosing regimen for contrast infusion in mice for LV opacification or perfusion studies.
Techniques for automated border detection, such as acoustic quantification (28), theoretically could be used with or without contrast to eliminate the need for manual tracings. Although this technique is feasible in mice with slow heart rates due to anesthesia or genetic manipulation, acoustic quantification is currently limited by its rather slow temporal resolution (Fig. 6).

View larger version (58K):
[in this window]
[in a new window]
|
Fig. 6. Automated border detection at end diastole (A) and at end systole (B) acquired with acoustic quantification in a normal CD-1 mouse. C: LV area vs. time plot acquired with acoustic quantification and its derivative. The morphology of the waveform, similar to that observed in humans, clearly demonstrates the different phases of the cardiac cycle, including ejection, rapid filling, diastasis, and atrial filling.
|
|
 |
Descriptors Of Lv Function
|
---|
Ejection Phase Indices Including Cardiac Output
LV fractional area shortening and ejection fraction are the most commonly used ejection phase indices to measure LV systolic performance. In addition, traditional high-fidelity LV pressure tracings and their derived maximum rate of pressure generation (dP/dtmax) have also been commonly used. These indices are load dependent and as such are incapable of separating changes in ventricular contractility from those caused by altered loading conditions (18, 24, 29).
Cardiac output has been previously measured in mice with the use of radioactive soluble indicators and microspheres (42), while other investigators have employed implantable transit time flow probes or electromagnetic flow meters in the ascending aorta (20). These methods are technically challenging and are not suitable for serial assessment. To circumvent this limitation, investigators have estimated cardiac output as the difference between 2D-determined LV end-systolic and end-diastolic volumes (20). More recently, Doppler ultrasound techniques, which allow the noninvasive assessment of cardiac output multiple times per experiment and the ability to follow changes serially, have been used.
Stroke volume and ascending aortic blood velocity can be calculated from continuous wave aortic Doppler velocity recordings, acquired from the parasternal long-axis view using 12- to 15-MHz transducers and 2D-targeted M-mode echocardiographic measurements of the diameter of the proximal ascending aorta (9). Other investigators (3, 43) have used different approaches, recording peak aortic velocities from the suprasternal approach or the apical window. More recently, with the advent of linear probes, high parasternal long-axis views have also been used to record maximal aortic velocities (8, 26, 27).
Stroke volume can be calculated by multiplying the aortic flow velocity integral by the aortic cross-sectional area. Cardiac output can then be calculated as the product of stroke volume and heart rate (Fig. 7). The major sources of error in estimating cardiac output using Doppler echocardiography in mice include 1) inadequate alignment between the sound beam and the ascending aorta and 2) the inability to measure the ascending aortic diameter accurately. Peak aortic velocities obtained in our laboratory are comparable to those obtained by Hoit et al. (17) and Pollick et al. (29) using similar equipment. Hartley et al. (15), using a 0.5-mm diameter, 20-MHz pencil-like transducer developed in their laboratory, was able to record higher peak aortic velocities, due to both improved alignment between the ultrasound beam and the aortic flow and smaller sampling volume (Fig. 8). With the development of new linear probes, flow alignment with the ascending thoracic aorta will be further optimized, allowing for less angle correction. Because of these constraints, as well as the limitation of frame rate, Doppler measurements of cardiac output are probably better suited for comparative, serial measurements rather than absolute determinations.

View larger version (54K):
[in this window]
[in a new window]
|
Fig. 7. Simultaneous recording of continuous wave aortic Doppler (CW Ao Doppler) echocardiographic measurements of instantaneous ascending aortic velocities with high-fidelity central aortic pressures (A) and M-mode echocardiographic measurements of ascending aortic diameter (B) were obtained in a normal CD-1 mouse. Cardiac output (CO) can be calculated as the aortic flow velocity integral (FVIAo) times aortic cross-sectional area (CSAAo). ECG, electrocardiogram; PAo, aortic pressure; Aod, aortic diameter. Reprinted with permission from Fentzke et al. (9).
|
|

View larger version (37K):
[in this window]
[in a new window]
|
Fig. 8. Cardiac Doppler signals from a mouse using a 2-mm diameter 10-MHz probe positioned at the apex with the sample volume set at the LV outflow track to measure aortic velocity or at the LV inflow track to measure mitral velocity. Reprinted with permission from Hartley et al. (Am J Physiol Heart Circ Physiol 279: H2326H2334, 2000).
|
|
Color Doppler imaging can be also be used to assess valvular disease. In different murine models, this technique has been used to verify the absence or presence of aortic insufficiency when a catheter is introduced retrogradely across the aortic valve (8, 18). This technique can also be used to improve alignment of the ultrasound beam with blood flow but is limited by frame rate. Pulsed-wave Doppler modality provides transvalvular flow velocity waveforms that can be effectively used for the calculation of 1) aortic and pulmonic peak velocities, 2) isovolumic relaxation and contraction times (IVRT/IVCT), 3) LV ejection times, and 4) mitral inflow velocity patterns. These data can be used to calculate the Tei index, a myocardial performance index that combines diastolic and systolic cardiac function (41). Color Doppler imaging has also been used to evaluate the severity of transverse aortic constriction, which is used to induce LV hypertrophy, in a variety of transgenic mice models. An example of color Doppler flow patterns across the thoracic aortic arch in a normal mouse and the corresponding pulsed-wave Doppler waveforms of ascending and descending aortic velocities may be seen in Fig. 9.

View larger version (80K):
[in this window]
[in a new window]
|
Fig. 9. An example of color Doppler recording in a normal CD-1 mouse across the transverse aortic arch, acquired from a high parasternal long-axis view, using a 12-MHz phased-array transducer. The probe was angled to align the maximal velocity flow across both the ascending and descending thoracic aorta. Color flow Doppler images were obtained by centering the sampling area in a narrow region of interest. Note red-colored velocity jet denotes flow away from the direction of the transducer, and blue denotes flow toward the probe.
|
|
Vascular Properties
The arterial system hydraulic load is the opposition to movement of blood out of the ventricle. Arterial input impedance, defined as the harmonic ratio of aortic pressure to flow, is a comprehensive characterization of this hydraulic load. Arterial input impedance can be separated into two components: a steady term commonly known as systemic vascular resistance and a pulsatile component. By simultaneously acquiring cardiac output and mean aortic pressures, it is possible to calculate total vascular resistance (TVR), a measure of arteriolar tone. Total vascular resistance can be calculated as
where TVR is in dyn · s · cm-5, MAP is mean arterial pressure in millimeters of mercury, CO is cardiac output in milliliters per minute, and 0.080 is a conversion factor.
Characterization of the pulsatile component can be approximated by parameters derived over different frequency ranges [e.g., low (global arterial compliance) and high (aortic characteristic impedance)]. Quantification of high-frequency parameters requires determination of instantaneous aortic pressure and flow with high temporal resolution (1 ms for the mouse). The fast Fourier transform processor used in commercially available Doppler ultrasound equipment does not have the temporal resolution required to accurately determine instantaneous aortic flow, particularly during peak flow acceleration or deceleration. Consequently, in our murine studies, global arterial compliance was calculated, for which our temporal Doppler blood flow resolution is adequate (6, 7, 9).
Arterial compliance for the two-element Windkessel model can be calculated using the area method of Liu et al. (22)
where Carea is global arterial compliance in centimeters cubed per millimeters of mercury, Ad is the area under the diastolic portion of the arterial pressure wave, Pes is the end-ejection pressure, and Pd is the diastolic pressure.
Load-Independent Assessment of Myocardial Contractility
The majority of studies that have assessed cardiac performance in mice using cardiac ultrasound have used ejection phase indices, such as ejection fraction, shortening fraction, cardiac output, and maximum dP/dt, all of which are limited by their load (18, 24, 29). The ability to separate alterations in myocardial contractility from simultaneously occurring changes in loading conditions would enable a more physiological understanding of the different cardiovascular phenotypes of genetically altered mice. Several relatively load-independent indices of LV performance, such as the end-systolic wall stress (
es) rate-corrected velocity of fiber shortening (Vcfc) relationship, have been used for the load-independent assessment of contractile state in humans. In a recent study, we demonstrated the feasibility of adapting this relationship for use in mice (8). To this effect, we performed simultaneous recordings of 2D-targeted M-mode echocardiography together with high-fidelity central aortic pressures at baseline and during infusion of the
-agonist, methoxamine, and isoproterenol (Fig. 10).

View larger version (47K):
[in this window]
[in a new window]
|
Fig. 10. Simultaneous recordings of ECG, high-fidelity aortic pressures, and continuous wave aortic Doppler velocities at baseline and during infusions methoxamine and isoproterenol. The data hereby depicted are required for the calculation of arterial compliance.
|
|
The rate-corrected velocity of LV fiber shortening (Vcfc) is calculated using the following formula
where ETc is the rate-corrected LV ejection time divided by the square root of the preceding R-R interval. Compared with shortening fraction, Vcfc has the advantage of being relatively preload and heart rate independent over the physiological range.
LV meridional systolic wall stress (
), a measure of true LV afterload, was calculated as a function of time using the following angiographically validated formula (2)
where
is in g/cm2, [P] is aortic pressure, [D] is LV internal dimension, [h] is systolic LV wall thickness, and 1.35 is a unit conversion factor. LV pressures during LV ejection were assumed to be equal to systolic aortic pressures.
The Vcfc -
es data points were obtained over a wide range of afterloads induced by jugular vein infusion of methoxamine. The Vcfc -
es relationships were inverse and fairly linear (r2 = 0.94) (Fig. 11), similar to the previously reported LV stress-shortening data in mice by Hoit et al. (17). During inotropic stimulation with intraperitoneal isoproterenol, Vcfc was found to be higher at any given level of end-systolic wall stress in all experiments. It is possible that in the future, with the use of minimally invasive or noninvasive (tail-cuff) determinations of aortic pressures, this index could be noninvasively determined.

View larger version (8K):
[in this window]
[in a new window]
|
Fig. 11. Rate-corrected velocity of fiber shortening (Vcfc) vs. end-systolic LV wall stress ( es) relationship in normal (left) and transgenic CREBA133 mice (right). A minimum of four data points (solid symbols) was obtained in each animal to determine baseline contractile state over wide range of afterloads. The correlation coefficient for this animal is given. Administration of isoproterenol caused increase in contractility in the control mouse, as indicated by upward deviation from the methoxamine-generated baseline contractility relationship (open symbol, dashed line). Contractile reserve was reduced in the CREBA133 mouse (open triangle, dashed line). Reprinted with permission from Fentzke et al. (9).
|
|
LV afterload (total load) can be thought of as LV systolic wall stress. According to Laplaces principle, LV wall stress is directly related to chamber dimension and inversely related to wall thickness. By digitizing the M-mode echocardiogram with the simultaneously recorded aortic pressures, we demonstrated that LV wall stress in mice reaches its peak within the first one-third of ejection and then declines throughout the remainder of systole as the ventricle becomes smaller and thicker, an observation similar to that in humans (9). This occurs despite rising pressures throughout most of the ejection period and emphasizes the importance of the decline in LV size and wall thickness as determinants of LV systolic wall stress. In normal mice examined in a recent study, LV end-systolic wall stress was on average less than 50% of the peak value. Stress vs. time plots were used to calculate the integral of instantaneous wall stress, which is one of the major determinants of myocardial oxygen consumption. The magnitudes of LV wall stress at the start, peak, and end of ejection and the integral of wall stress over ejection each have unique physiological significance (Fig. 12).

View larger version (18K):
[in this window]
[in a new window]
|
Fig. 12. LV afterload (i.e., wall stress, ) calculated over the course of ejection from simultaneous measurements of pressure (PAo), dimension (D), and wall thickness (h) in a normal nontransgenic mouse (NTg) and a transgenic CREBA133 mouse (Tg). Systolic wall stress peaked soon after onset of ejection and then declined throughout remainder of systole in the normal mouse. In comparison, the transgenic mouse displayed increased wall stress and LV dimension, as well as decreased aortic pressure and LV thickness. Reprinted with permission from Fentzke et al. (9).
|
|
By simultaneously digitizing instantaneous LV dimensions and pressures, it is also possible to plot LV pressure-dimension loops (Fig. 13). These may be generated by digitizing LV pressure waveforms and superimposing manually digitized septum and posterior wall dimensions to obtain continuous changes in LV dimension (43). Load may then be manipulated by either pharmacological or mechanical interventions. Preload reduction may be achieved by the infusion of nitroprusside or by decreasing venous return (balloon or ligature around vena cava, or digital compression). Afterload may be increased by using methoxamine or by physical means (balloon or ligature around the aorta) (12). These load manipulations are used to generate the end-systolic pressure-dimension relationship. This method may be used to determine contractile reserve, applying pharmacological inotropic challenges or mechanical challenges (e.g., atrial pacing) (26). Positive inotropic agents commonly used have been dobutamine and isoproterenol. Negative inotropic agents such as esmolol and propanolol have also been employed to decrease contractility. An advantage of this method over pressure:volume loops obtained with impedance catheters mounted on pressure transducers is that the smaller size of the high-fidelity pressure catheter better preserves the integrity of the aortic valve (6, 12).

View larger version (35K):
[in this window]
[in a new window]
|
Fig. 13. Simultaneous recording of ECG, 2D-targeted M-mode echocardiography and high-fidelity LV pressure in a normal CD-1 mouse (A). By simultaneously digitizing instantaneous LV dimensions and pressures, it is possible to plot pressure-dimension loops (B). Load is manipulated with methoxamine to obtain the end-systolic pressure-dimension relation (straight line), a load-independent measure of LV contractility.
|
|
LV Diastolic Function
Genetic and surgical murine models of human disease also require the evaluation of diastolic function. In larger animals, relaxation and filling parameters have been widely assessed using transmitral Doppler echocardiography. Impaired LV relaxation has been associated with reduced early-to-late diastolic transmitral Doppler flow velocity ratios (i.e., decreased E/A ratio), prolonged isovolumic relaxation times (IVRT), and prolonged E wave deceleration times. Tanaka et al. (40) and Hoit et al. (18) have demonstrated that transmitral filling patterns may also be used to estimate LV diastolic function in normal and transgenic murine models of impaired relaxation. These investigators, however, were not able to consistently identify all functional phenotypes. The use of pulsed Doppler of the LV inflow for the assessment of diastolic function is limited by 1) poor alignment of the Doppler beam with mitral flow, which makes it difficult to reproducibly compare tracings obtained in different animals; and 2) the high heart rate of mice, which results in merging of the E and A waves and thus confounds the analysis of the mitral flow.
Recently, however, Schmidt et al. (37), reported that altered patterns of abnormal LV diastolic function could be identified in two murine models of altered calcium uptake. Phospholamban knockout mice exhibit enhanced calcium reuptake in the sarcoplasmic reticulum and augmented contractility. In contrast, the mutant form of phospholamban exhibited superinhibition of calcium reuptake and decreased contractility (18). The knockout mice were characterized by improved LV diastolic function, including shorter ejection times, increased Vcfc, increased E/A ratios, lower deceleration times, decreased IVRT corrected for heart rate, and increased mitral E wave propagation velocities, whereas the mutant phospholamban mice displayed a profile consistent with impaired LV relaxation (Fig. 14). Although IVRT is dependent on load and rate, color M-mode determined propagation velocity is a relatively load-independent index of LV filling.

View larger version (63K):
[in this window]
[in a new window]
|
Fig. 14. Diastolic color M-mode flow propagation velocity (vp) into LV, which is determined by slope of first aliasing isovelocity line during early filling. A: representative tracings of PLB/KO (left), WT (middle), and PLB/N27A mice (right). In WT animal, early filling progresses rapidly into LV toward apex. This is further enhanced in PLB/KO mice. In contrast, in PLB/N27A, slope of early filling is low and does not reach apex, and relatively high velocity is observed during A wave. B: mean values for color M-mode flow propagation (±SE). *P < 0.05 vs. WT. #P < 0.05 vs. PLB/KO. PLB/KO, phospholamban knockout; WT, wild type; PLB/N27A, mutant form of PLB. Reprinted with permission from Schmidt et al. (37).
|
|
The Tei index (41) was also obtained from the Doppler velocity profiles and discerned the phospholamban mutant mice from the knockout and control mice. These data demonstrate that color M-mode flow propagation velocities and spectral wave Doppler echocardiography can be effectively used to noninvasively assess LV diastolic function in transgenic murine models. The applicability and reliability in serial measurements of diastolic function remains to be determined in future studies.
 |
Conclusion
|
---|
Transgenic mice displaying abnormalities in cardiac development and function represent a powerful new tool for understanding the molecular mechanisms underlying both normal cardiovascular function and the pathophysiological basis of human cardiovascular disease. Magnetic resonance imaging and ultrafast CT scanning offer accurate estimation of LV mass (10, 32), but their use may be limited by expense and availability. Contrast-enhanced X-ray ventriculography and radionuclide techniques for the assessment of mouse cardiovascular function have also been adapted for studies in mice (14, 30). Although three-dimensional echocardiography (34) and transesophageal imaging (33) of the murine heart is feasible and may offer a theoretical potential for assessing hearts of widely varying sizes, these techniques are technically challenging for broad application to a variety of murine models or are limited by low frame rates.
Recent advances in cardiac ultrasound have responded to the challenges imposed by small, fast-beating murine hearts. 2D Doppler echocardiography currently provides a relatively inexpensive, accurate, and noninvasive method for the assessment of cardiovascular structural and functional changes in a variety of murine models.
 |
FOOTNOTES
|
---|
1 This review article was based on work originally presented at the "NHLBI Symposium on Phenotyping: Mouse Cardiovascular Function and Development" held at the Natcher Conference Center, NIH, Bethesda, MD, on October 1011, 2002. 
Article published online before print. See web site for date of publication (http://physiolgenomics.physiology.org).
Address for reprint requests and other correspondence: R. M. Lang, Univ. of Chicago Medical Center, M.C. 5084, 5841 S. Maryland Ave., Chicago, IL 60637 (E-mail: rlang{at}medicine.bsd.uchicago.edu).
 |
References
|
---|
- Arras M, Autenried P, Rettich A, Spaeni D, and Rulicke T. Optimization of intraperitoneal injection anesthesia in mice: drugs, dosages, adverse effects, and anesthesia depth. Comp Med 51: 443456, 2001.[ISI][Medline]
- Brodie BR, McLaurin LP, and Grossman W. Combined hemodynamic-ultrasonic method for studying left ventricular wall stress: comparison with angiography. Am J Cardiol 37: 864870, 1976.[ISI][Medline]
- Chaves AA, Weinstein DM, and Bauer JA. Non-invasive echocardiographic studies in mice: influence of anesthetic regimen. Life Sci 69: 213222, 2001.[ISI][Medline]
- Collins KA, Korcarz CE, Shroff SG, Bednarz JE, Fentzke RC, Lin H, Leiden JM, and Lang RM. Accuracy of echocardiographic estimates of left ventricular mass in mice. Am J Physiol Heart Circ Physiol 280: H1954H1962, 2001.[Abstract/Free Full Text]
- Devereux RB, Alonso DR, Lutas EM, Gottlieb GJ, Campo E, Sachs I, and Reichek N. Echocardiographic assessment of left ventricular hypertrophy: comparison to necropsy findings. Am J Cardiol 57: 450458, 1986.[ISI][Medline]
- Feldman MD, Erikson JM, Mao Y, Korcarz CE, Lang RM, and Freeman GL. Validation of a mouse conductance system to determine LV volume: comparison to echocardiography and crystals. Am J Physiol Heart Circ Physiol 279: H1698H1707, 2000.[Abstract/Free Full Text]
- Fentzke RC, Korcarz CE, Lang RM, Lin H, and Leiden JM. Dilated cardiomyopathy in transgenic mice expressing a dominant-negative CREB transcription factor in the heart. J Clin Invest 101: 24152426, 1998.[Abstract/Free Full Text]
- Fentzke RC, Korcarz CE, Shroff SG, Lin H, Leiden JM, and Lang RM. The left ventricular stress-velocity relation in transgenic mice expressing a dominant negative CREB transgene in the heart. J Am Soc Echocardiogr 14: 209218, 2001.[ISI][Medline]
- Fentzke RC, Korcarz CE, Shroff SG, Lin H, Sandelski J, Leiden JM, and Lang RM. Evaluation of ventricular and arterial hemodynamics in anesthetized closed-chest mice. J Am Soc Echocardiogr 10: 915925, 1997.[ISI][Medline]
- Franco F, Dubois SK, Peshock RM, and Shohet RV. Magnetic resonance imaging accurately estimates LV mass in a transgenic mouse model of cardiac hypertrophy. Am J Physiol Heart Circ Physiol 274: H679H683, 1998.[Abstract/Free Full Text]
- Gardin JM, Siri FM, Kitsis RN, Edwards JG, and Leinwand LA. Echocardiographic assessment of left ventricular mass and systolic function in mice. Circ Res 76: 907914, 1995.[Abstract/Free Full Text]
- Georgakopoulos D, Mitzner WA, Chen CH, Byrne BJ, Millar HD, Hare JM, and Kass DA. In vivo murine left ventricular pressure-volume relations by miniaturized conductance micromanometry. Am J Physiol Heart Circ Physiol 274: H1416H1422, 1998.[Abstract/Free Full Text]
- Hart CY, Burnett JC Jr, and Redfield MM. Effects of avertin versus xylazine-ketamine anesthesia on cardiac function in normal mice. Am J Physiol Heart Circ Physiol 281: H1938H1945, 2001.[Abstract/Free Full Text]
- Hartley CJ, Lacy JL, Dai D, Nayak N, Taffet GE, Entman ML, and Michael LH. Functional cardiac imaging in mice using Ta-178. Nat Med 5: 237239, 1999.[ISI][Medline]
- Hartley CJ, Taffet GE, Michael LH, Pham TT, and Entman ML. Noninvasive determination of pulse-wave velocity in mice. Am J Physiol Heart Circ Physiol 273: H494H500, 1997.[Abstract/Free Full Text]
- Hartley CJ, Taffet GE, Reddy AK, Entman ML, and Michael LH. Noninvasive cardiovascular phenotyping in mice. ILAR J 43: 147158, 2002.[Medline]
- Hoit BD, Ball N, and Walsh RA. Invasive hemodynamics and force-frequency relationships in open- versus closed-chest mice. Am J Physiol Heart Circ Physiol 273: H2528H2533, 1997.[Abstract/Free Full Text]
- Hoit BD, Khoury SF, Kranias EG, Ball N, and Walsh RA. In vivo echocardiographic detection of enhanced left ventricular function in gene-targeted mice with phospholamban deficiency. Circ Res 77: 632637, 1995.[Abstract/Free Full Text]
- Hoit BD, Kiatchoosakun S, Restivo J, Kirkpatrick D, Olszens K, Shao H, Pao YH, and Nadeau JH. Naturally occurring variation in cardiovascular traits among inbred mouse strains. Genomics 79: 679685, 2002.[ISI][Medline]
- Janssen B, Debets J, Leenders P, and Smits J. Chronic measurement of cardiac output in conscious mice. Am J Physiol Regul Integr Comp Physiol 282: R928R935, 2002; 10.1152/ajpregu.00406.2001.
- Kanno S, Lerner DL, Schuessler RB, Betsuyaku T, Yamada KA, Saffitz JE, and Kovacs A. Echocardiographic evaluation of ventricular remodeling in a mouse model of myocardial infarction. J Am Soc Echocardiogr 15: 601609, 2002.[ISI][Medline]
- Liu Z, Brin KP, and Yin FC. Estimation of total arterial compliance: an improved method and evaluation of current methods. Am J Physiol Heart Circ Physiol 251: H588H600, 1986.[Abstract/Free Full Text]
- Manning WJ, Wei JY, Katz SE, Litwin SE, and Douglas PS. In vivo assessment of LV mass in mice using high-frequency cardiac ultrasound: necropsy validation. Am J Physiol Heart Circ Physiol 266: H1672H1675, 1994.[Abstract/Free Full Text]
- Milano CA, Allen LF, Rockman HA, Dolber PC, McMinn TR, Chien KR, Johnson TD, Bond RA, and Lefkowitz RJ. Enhanced myocardial function in transgenic mice overexpressing the ß2-adrenergic receptor. Science 264: 582586, 1994.[ISI][Medline]
- Mor-Avi V, Korcarz C, Fentzke RC, Lin H, Leiden JM, and Lang RM. Quantitative evaluation of left ventricular function in a transgenic mouse model of dilated cardiomyopathy with 2-dimensional contrast echocardiography. J Am Soc Echocardiogr 12: 209214, 1999.[ISI][Medline]
- Nemoto S, DeFreitas G, Mann DL, and Carabello BA. Effects of changes in left ventricular contractility on indexes of contractility in mice. Am J Physiol Heart Circ Physiol 283: H2504H2510, 2002. First published September 19, 2002; 10.1152/ajpheart.0765.2001.[Abstract/Free Full Text]
- Patten RD, Aronovitz MJ, Bridgman P, and Pandian NG. Use of pulse wave and color flow Doppler echocardiography in mouse models of human disease. J Am Soc Echocardiogr 15: 708714, 2002.[ISI][Medline]
- Perez JE, Davila-Roman VG, and Miller JG. Assessment of myocardial viability by ultrasonic tissue characterization. Coron Artery Dis 6: 613618, 1995.[ISI][Medline]
- Pollick C, Hale SL, and Kloner RA. Echocardiographic and cardiac Doppler assessment of mice. J Am Soc Echocardiogr 8: 602610, 1995.[Medline]
- Rockman HA, Ono S, Ross RS, Jones LR, Karimi M, Bhargava V, Ross J Jr, and Chien KR. Molecular and physiological alterations in murine ventricular dysfunction. Proc Natl Acad Sci USA 91: 26942698, 1994.[Abstract]
- Roth DM, Swaney JS, Dalton ND, Gilpin EA, and Ross J Jr. Impact of anesthesia on cardiac function during echocardiography in mice. Am J Physiol Heart Circ Physiol 282: H2134H2140, 2002; 10.1152/ajpheart.00845.2001.[Abstract/Free Full Text]
- Ruff J, Wiesmann F, Hiller KH, Voll S, von Kienlin M, Bauer WR, Rommel E, Neubauer S, and Haase A. Magnetic resonance microimaging for noninvasive quantification of myocardial function and mass in the mouse. Magn Reson Med 40: 4348, 1998.[ISI][Medline]
- Scherrer-Crosbie M, Steudel W, Hunziker PR, Foster GP, Garrido L, Liel-Cohen N, Zapol WM, and Picard MH. Determination of right ventricular structure and function in normoxic and hypoxic mice: a transesophageal echocardiographic study. Circulation 98: 10151021, 1998.[Abstract/Free Full Text]
- Scherrer-Crosbie M, Steudel W, Hunziker PR, Liel-Cohen N, Ullrich R, Zapol WM, and Picard MH. Three-dimensional echocardiographic assessment of left ventricular wall motion abnormalities in mouse myocardial infarction. J Am Soc Echocardiogr 12: 834840, 1999.[ISI][Medline]
- Scherrer-Crosbie M, Steudel W, Ullrich R, Hunziker PR, Liel-Cohen N, Newell J, Zaroff J, Zapol WM, and Picard MH. Echocardiographic determination of risk area size in a murine model of myocardial ischemia. Am J Physiol Heart Circ Physiol 277: H986H992, 1999.[Abstract/Free Full Text]
- Schiller NB, Shah PM, Crawford M, DeMaria A, Devereux R, Feigenbaum H, Gutgesell H, Reichek N, Sahn D, Schnittger I, et al. Recommendations for quantitation of the left ventricle by two-dimensional echocardiography. American Society of Echocardiography Committee on Standards, Subcommittee on Quantitation of Two-Dimensional Echocardiograms. J Am Soc Echocardiogr 2: 358367, 1989.[Medline]
- Schmidt AG, Gerst M, Zhai J, Carr AN, Pater L, Kranias EG, and Hoit BD. Evaluation of left ventricular diastolic function from spectral and color M-mode Doppler in genetically altered mice. J Am Soc Echocardiogr 15: 10651073, 2002.[ISI][Medline]
- Suehiro K, Takuma S, Cardinale C, Hozumi T, Shimizu J, Yano H, Di Tullio MR, Wang J, Smith CR, Burkhoff D, and Homma S. Assessment of segmental wall motion abnormalities using contrast two-dimensional echocardiography in awake mice. Am J Physiol Heart Circ Physiol 280: H1729H1735, 2001.[Abstract/Free Full Text]
- Takuma S, Suehiro K, Cardinale C, Hozumi T, Yano H, Shimizu J, Mullis-Jansson S, Sciacca R, Wang J, Burkhoff D, Di Tullio MR, and Homma S. Anesthetic inhibition in ischemic and nonischemic murine heart: comparison with conscious echocardiographic approach. Am J Physiol Heart Circ Physiol 280: H2364H2370, 2001.[Abstract/Free Full Text]
- Tanaka N, Dalton N, Mao L, Rockman HA, Peterson KL, Gottshall KR, Hunter JJ, Chien KR, and Ross J Jr. Transthoracic echocardiography in models of cardiac disease in the mouse. Circulation 94: 11091117, 1996.[Abstract/Free Full Text]
- Tei C, Ling LH, Hodge DO, Bailey KR, Oh JK, Rodeheffer RJ, Tajik AJ, and Seward JB. New index of combined systolic and diastolic myocardial performance: a simple and reproducible measure of cardiac function: a study in normals and dilated cardiomyopathy. J Cardiol 26: 357366, 1995.[Medline]
- Wang P, Ba ZF, Burkhardt J, and Chaudry IH. Trauma-hemorrhage and resuscitation in the mouse: effects on cardiac output and organ blood flow. Am J Physiol Heart Circ Physiol 264: H1166H1173, 1993.[Abstract/Free Full Text]
- Williams RV, Lorenz JN, Witt SA, Hellard DT, Khoury PR, and Kimball TR. End-systolic stress-velocity and pressure-dimension relationships by transthoracic echocardiography in mice. Am J Physiol Heart Circ Physiol 274: H1828H1835, 1998.[Abstract/Free Full Text]
- Yang XP, Liu YH, Rhaleb NE, Kurihara N, Kim HE, and Carretero OA. Echocardiographic assessment of cardiac function in conscious and anesthetized mice. Am J Physiol Heart Circ Physiol 277: H1967H1974, 1999.[Abstract/Free Full Text]