ILAR Journal Online, Volume 43(3) 2002: Advanced Physiological Monitoring in Rodents

Online Issues

<< All Back-issues

<< This Issue's Table of Contents

ILAR Journal V43(3) 2002
Advanced Physiological Monitoring in Rodents

View/Download article (PDF):
Noninvasive

Noninvasive Cardiovascular Phenotyping in Mice
Craig J. Hartley, George E. Taffet, Anilkumar K. Reddy, Mark L. Entman, and Lloyd H. Michael

The authors are from the Section of Cardiovascular Sciences (CVS) in the Department of Medicine and the DeBakey Heart Center (DHC), Baylor College of Medicine (BCM), and The Methodist Hospital, Houston, Texas. Craig J. Hartley, Ph.D., is a Professor and Director of the DHC Instrumentation Core; George E. Taffet, M.D., is an Associate Professor; Anilkumar K. Reddy, Ph.D., is an Instructor; Mark L. Entman, M.D., is a Professor, CVS Section Chief, and DHC Scientific Director; and Lloyd H. Michael, Ph.D., is a Professor, Director of the DHC Animal Core, and BCM Associate Dean.

Abstract

With the growth of genetic engineering, mice have become common as models of human diseases, which in turn has stimulated the development of techniques to monitor and image the murine cardiovascular system. Invasive methods are often more quantitative, but noninvasive methods are preferred when measurements must be repeated serially on living animals during development or in response to pharmacological or surgical interventions. Because of the small size and high heart rates in mice, high spatial and temporal resolutions are required to preserve signal fidelity. Monitoring of body temperature and the electrocardiogram is essential when animals must be anesthetized for a measurement or other procedure. Several other groups have developed cardiovascular imaging modalities suitable for murine applications, and ultrasound is the most widely used. Our group has developed and applied high-resolution Doppler probes and signal processing for measuring blood velocity in the heart and peripheral vessels of anesthetized mice noninvasively. We can measure cardiac filling and ejection velocities as indices of systolic and diastolic ventricular function and for timing of cardiac events; velocity pulse arrival times for determining pulse-wave velocity and arterial stiffness; peripheral velocity waveforms as indices of arterial resistance, compliance, and wave reflections; stenotic velocities for estimation of pressure drop and detection of vorticity; and tail artery velocity for determining systolic and diastolic blood pressure using a pressure cuff. These noninvasive methods are convenient and easy to apply and have been used to detect and evaluate numerous cardiovascular phenotypes in mutant mice.

Key Words: cardiovascular physiology; Doppler ultrasound; hemodynamics; mice; peripheral vascular flow; phenotyping; pulse-wave velocity; ventricular function

Introduction

The ability to alter the genotype of the mouse has created a need to evaluate the phenotypic changes that occur in the animal during development and maturation (Cowley 1997). However, the estimation and validation of "normal" values for many parameters in mice has been recent. A literature review reveals few publications prior to 1990 reporting mouse cardiovascular physiology; however, beginning in the mid-1990s, the number of publications involving mice genetically engineered to model human physiology and pathophysiology has increased exponentially. The mutations, whether targeted or random, can alter the structure, anatomy, pathology, and physiology of cells, organs, or systems in ways that are often subtle and unpredictable and that may change with time (Chien 1996; James et al. 1998; Kass et al. 1998). Available methods for evaluating the cardiovascular systems of mice include dissection and microscopy for evaluating pathology, imaging with various modalities for displaying anatomy (Hoit and Walsh 1998), and invasive techniques for measuring physiological variables (Doevendans et al. 1998). In response to the need for serial assessment of function, several groups have developed and applied implantable pressure and flow sensors that incorporate telemetry (Mitchell et al. 1998). The major problems in adapting methods for use in mice relate to the small size and high heart rates, which place extreme demands on both spatial and temporal resolution (Dawson 1991). We focus here on noninvasive methods that can be used both to evaluate cardiovascular physiology and function in anesthetized mice and serially to follow development, aging, or the effects of surgical or pharmacological interventions.

Anesthesia

The use of anesthesia, which is required for most studies in living mice, is problematic in physiological and pharmacological studies because all agents alter cardiovascular control in some way. Larger domesticated animals can often be studied in the conscious unanesthetized state using implantable sensors either with or without telemetry because the animals are accustomed to humans and can be trained either to lie quietly or to exercise (Hanley et al. 1975; Hartley et al. 1978). Mice have also been studied in the awake state (Desai et al. 1997; Yang et al. 1999b); however, they are usually under stress in the laboratory situation, and it is more difficult to perform controlled studies in awake mice unless telemetry is used to minimize handling and interaction with humans (Gehrmann et al. 2000; Johansson et al. 1999; Mitchell et al. 1998). Therefore, anesthesia is usually used to provide a consistent and controlled setting in which to study mice, recognizing that the type of agent must be chosen carefully to minimize the effect on the system under study. We have used a variety of agents including inhaled isoflurane gas (Hartley et al. 2000), intraperitoneal sodium pentobarbital (Hartley et al. 1999), etomidate (Kass et al. 1998), and a rodent "cocktail" mixture consisting of ketamine, xylazine, and acepromazine (Hartley et al. 1995, 1999). Induction and recovery are faster with gas anesthesia, and we use the mixture for studies of relatively short duration. Pentobarbital, etomidate, and isoflurane gas maintain heart rate at physiological levels, but the rodent cocktail slows heart rate significantly in mice while maintaining blood pressure. As with any form of anesthesia, it is important to monitor the electrocardiogram (ECG¹) and respiration and to maintain body temperature during experimental procedures in mice (Hartley et al. 1995).

Interventions

Many phenotypes are subtle, and mice can often accommodate the mutations by using compensatory mechanisms to maintain blood pressure and cardiac output (Kass et al. 1998). Thus, resting values for these and other parameters may be nearly normal, and interventions must be performed to reveal phenotypic differences in the response to stress. Interventions can include inotropic (Georgakopoulos et al. 1998; Hartley et al. 1995, 1999), chronotropic, and vasoactive (Banda et al. 1997; Hartley et al. 1997) pharmaceuticals, exercise (Desai et al. 1997; Spencer et al. 2000), and/or surgical manipulations such as coronary occlusion (Briaud et al. 2001; Jones et al. 1999; Kurrelmeyer et al. 2000; Michael et al. 1995, 1999; Patten et al. 1998; Verdouw et al. 1998) or aortic constriction (Hongo et al. 1997; Rockman et al. 1993; Zhang et al. 2000). The ability to perform serial noninvasive studies and to measure the responses to interventions is powerful because the animal can be used as its own control, and it is not necessary to assess or assume so-called normal values.

Imaging Methods

Many of the methods used to evaluate the murine cardiovascular system noninvasively currently use some form of imaging. These methods include magnetic resonance imaging (Chacko et al. 2000; James et al. 1998; Siri et al. 1997; Wiesmann et al. 2000), x-ray angiography (Rockman et al. 1994), two-dimensional (2-D¹) (Gardin et al. 1996; Scherrer-Crosbie et al. 1998; Youn et al. 1999) and M-mode (Fentzke et al. 2001; Hoit et al. 1995; Milner et al. 1999) ultrasound, and nuclear angiography (Hartley et al. 1999). All of these techniques have been adapted from human systems or those designed for larger animals and have marginal spatial and temporal resolutions (frame rates) when used with mice. At a typical mouse heart rate of 600/min or 10/sec, the standard video frame rate of 30/sec would produce only three images/cardiac cycle. Magnetic resonance imaging requires ECG and respiratory gating (Siri et al. 1997) averaged over several minutes, and the best 2-D ultrasound systems generate at most 150 frames/sec or 15 images/cardiac cycle (Youn et al. 1999).

M-mode echocardiography, which displays the position and motion of reflecting structures along one line of sight as echo-depth versus time, is the most common noninvasive method for evaluating cardiac function in mice (Hoit and Walsh 1998). It has the highest temporal resolution at 1 msec, but the limited sweep speeds available with clinical instruments are not optimal for mice. Nevertheless, it is possible to generate quantitative signals revealing left ventricular (LV¹) diameter throughout the cardiac cycle (Hoit and Walsh 1998; Kurrelmeyer et al. 2000). From these signals, one can obtain several useful indices of LV function, including LV diameter and diameter shortening fraction and wall thickness and thickening fraction if the heart is known to be homogeneous. However, none of the imaging methods can estimate loading conditions or pressures that necessitate the use of more invasive methods.

Invasive Physiological Measurements

In larger animals, cardiovascular physiology and mechanics are usually assessed using pressure, flow, and/or dimension sensors either attached or inserted into the vessels or organs of interest (Hartley et al. 1978). Several manufacturers are producing miniaturized versions of pressure (Lorenz and Robbins 1996; Wang et al. 2000; Williams et al. 1998) (Millar, Houston, Texas), flow (Feldman et al. 2000a; Yang et al. 1999a) (Transonic, Ithaca, New York), dimension (Feldman et al. 2000a; Kubota et al. 1998) (Sonometric, Ontario, Canada), volume (Feldman et al. 2000b; Georgakopoulos et al. 1998) (Millar), and telemetry (Mitchell et al. 1998) sensors and instrumentation (Data Sciences, St. Paul, Minnesota) specifically designed for use in mice. These methods are quantitative and allow good control of loading conditions, although size, resolution, fidelity, accuracy, and calibration are major concerns (Kass et al. 1998). These methods all require precise surgical manipulations, chronic implantation of the sensors is problematic, and most of the studies are terminal and cannot be repeated in the same mouse (Kass et al. 1998). Several excellent reviews on evaluating cardiovascular physiology in mice are available (Hoit and Walsh 1998; James et al. 1998; Kass et al. 1998).

Noninvasive Physiological Methods

Among the noninvasive methods available for physiological studies in mice are ECG (Berul et al. 1996; Mitchell et al. 1998; Richards et al. 1953; Wehrens et al. 2000) (adapted from large animal and human systems), tail-cuff blood pressure (Krege et al. 1995) (adapted from rat systems), and M-mode and Doppler echocardiography (Gui et al. 1996; Youn et al. 1999) (using unmodified clinical systems). Several years ago, we began a program to design and optimize these and other techniques further so that we could make rapid, noninvasive serial measurements of cardiovascular function in mice during growth and development (Hartley et al. 1995, 1997, 1999). The goal was to facilitate quantitative assessment of cardiac and vascular phenotypes in mice (Hartley et al. 2000).

Murine ECG

The first recording of an ECG in a mouse was reported more than 70 yr ago (Agduhr and Stenstrom 1929), and Richards and colleagues (1953) performed seminal studies on location of the T-wave in the mid 1950s. A recent review summarizes the progress over the last 30 yr (Wehrens et al. 2000). Early studies attempted to elucidate the electrical markers of repolarization and the Q-T segment in mice (Goldbarg et al. 1968). At the time of this writing, 6- and/or 12-lead ECG recordings have been analyzed in transgenic and wild type mice to determine a variety of time intervals and conduction velocities (Berul et al. 1996, 1997; Sah et al. 1999; Vaidya et al. 1999). An ECG measurement in a mouse requires consideration of animal positioning, lead attachment, number of leads, noise suppression, signal fidelity, amplifier bandwidth, and status of the animal, among other requirements. Importantly, amplifiers with filters optimized for human ECG recording seriously compromise signal fidelity and distort timing when used in mice. A conventional amplifier with active-lead noise suppression and a bandwidth of 0.1 to 2 kHz is suitable for use in mice (Hartley et al. 2000).

A related concern with mice is maintenance of body temperature during surgical and monitoring procedures. Significant alterations occur in heart rate and cardiac function with changes in temperature (Hartley et al. 1995; Richards et al. 1953). Heat lamps or heating pads are therefore used commonly, but they are inconvenient and are often a source of 60-Hz electrical noise. By incorporating a low-voltage heater into an ECG/restraint board, automatic thermal control can be achieved with no electrical interference (Hartley et al. 2000). The board is shown in Figure 1 (left) and contains four ECG electrodes, an array of surface-mount resistors, and a temperature sensor. The anesthetized mouse is positioned over the resistor array, and each limb is taped to an electrode to which a small drop of electrode cream has been applied. A controller applies current to the resistor array to maintain the board (or the mouse) at a selected temperature. This general approach for ECG monitoring and thermal support is used during surgical procedures, during myocardial ischemia and reperfusion, and for invasive and noninvasive physiological measurements and interventions.

Figure 1 Illustration of the setup used to make noninvasive electrocardiogram (ECG) and Doppler measurements showing an anesthetized mouse taped to electrodes on a temperature-controlled circuit board, Doppler and ECG signal processors, and a display of ECG and Doppler signals from the left ventricular inflow and outflow tracks. The 23 × 30-cm printed circuit board contains four stainless steel ECG electrodes, an array of 50 surface-mount resistors, a temperature sensor, a large ground plane, and electrical connections to an ECG amplifier and a temperature controller. Labeled on the Doppler tracing are the opening (o) and closing (c) of the mitral (m) and aortic (a) valves, peak ejection velocity (P) and acceleration (accel), and peak early (E) and late (A) filling velocities. From these signals, it is possible to obtain accurate timing of cardiac events such as pre-ejection time, filling and ejection times, and isovolumic contraction and relaxation times as indices of systolic and diastolic ventricular function. The rodent "cocktail" (which slows heart rate) was used for anesthesia.

Murine Doppler System

Doppler instruments measure blood velocity by detecting the difference in frequency between an emitted burst of ultrasound (10 or 20 MHz) and the returning echoes from moving blood. In a pulsed Doppler system, the sample volume can be adjusted in depth by varying the range-gate or delay between transmission of the ultrasonic burst and sampling of the returning echoes. The dimensions of the sample volume are determined by the burst length, range-gate length, and diameter of the sound beam. The resulting Doppler signal (which is in the audible range) is a summation of the Doppler shifted echoes from many blood cells moving at many velocities, and the spectrum of frequencies represents the distribution of red cell velocities within the sample volume. The Doppler equation (Δf = 2f_o(V/c)cosθ) relates the Doppler frequency (Δf) to the velocity (V) of each red cell, where f_o is the ultrasonic frequency (10 or 20 MHz), c is the speed of sound in blood (1540 M/s), and θ is the angle between the sound beam and the direction of flow. When θ is close to zero, the conversion factors become 7.5 (cm/sec)/kHz at 10 MHz and 3.75 (cm/sec)/kHz at 20 MHz (Hartley et al. 1995).

The murine pulsed Doppler system was adapted from a modular instrument originally designed for use with implantable probes to measure blood flow in small vessels of dogs and rats (Gardiner et al. 1990; Hartley and Cole 1974). For noninvasive applications in mice, small hand-held probes (Hartley et al. 1995) and a high-fidelity signal processor were designed (Hartley et al. 2000). The probes consist of a 1-mm-diameter 10- or 20-MHz ultrasonic crystal mounted at the end of a 2-mm-diameter 10-cm-long stainless steel tube. An epoxy lens is molded to the front face of the crystal to focus the sound beam at a depth of 4 to 6 mm. The resulting sample volume is less than 0.02 µl (0.3 mm diameter × 0.3 mm long) at the focus. The Doppler signal processor is a computer-based system, which Indus Instruments (Houston, Texas) designed to our specifications. The computer digitizes the audio Doppler signals at 125 kHz, generates a fast Fourier transform display in real-time, and also captures and displays up to four other signals such as ECG and pressure. After the signals are acquired, the number of the points used to calculate the fast Fourier transform and the update rate can be adjusted to optimize either temporal or frequency (velocity) resolution, depending on the application. The best velocity resolution is 5 mm/sec at 20 MHz; the maximum measurable velocity is 9.3 m/sec at 10MHz; and the best temporal resolution is 0.1 msec.

Nuclear Angiography

First-pass nuclear angiography using ¹⁷⁸Ta (T½ = 9.3 min) is a minimally invasive method to assess right and left ventricular ejection fractions in mice (Hartley et al. 1999). The human-sized nuclear camera is fitted with a pin-hole lens to achieve a 2-mm resolution with a frame rate of 160/sec. The resolution is marginal, but the method generates quantitative estimates of cardiac volumes in mice and is adequate to calculate ejection fractions. A jugular venous catheter is required for injection of 10 µl (20 mCi) of radionuclide, and images can be repeated at 30-min intervals or after several days by subcutaneous placement of the heparin-locked catheter.

Examples and Applications

In our facility, ECG, M-mode echocardiography, Doppler ultrasound, and ¹⁷⁸Ta nuclear imaging are used to evaluate cardiovascular function noninvasively in genetically engineered anesthetized mice with and without additional interventions. All of these measurements can be repeated within minutes and over several days or weeks so that animals can be used as their own controls and so that changes in function and responses to interventions can be followed over time. An example of cardiac Doppler signals from a mouse using a 10-MHz transducer placed at the base of the sternum and pointed toward the heart is shown in Figure 1 (right) along with the ECG. A photograph of a mouse on the ECG/heater board is also shown (left). The probe was not focused, and the relatively large sample volume included both the inflow and outflow tracks of the left ventricle. With a more focused probe, signals from the inflow track and the aortic root can be isolated. From this approach, cardiac timing and several indices of LV systolic function (peak velocity and acceleration) and diastolic function (peak early and late filling velocities and their ratio) can be measured (Taffet et al. 1996). The rodent cocktail was used to produce the slower heart rate shown in Figure 1 because at physiological rates (>500 beats/min), the early and late fillings waves are usually merged.

One of the models we have evaluated is an ischemia-reperfusion model created by occluding (and reperfusing) the left anterior descending coronary artery in a mouse and then following cardiac function for several weeks or months using Doppler and nuclear imaging methods (Hartley et al. 1999; Michael et al. 1995, 1999). The ECG is monitored for S-T changes before, during, and after occlusion of the left anterior descending coronary artery to verify creation of ischemia, as shown in Figure 2. A summary of ECG, nuclear angiography, and Doppler ultrasound data from control, sham-operated (but not occluded), permanently occluded, and occlusion/reperfusion mice taken 2 to 35 wk after operation is shown in Figure 3. Ejection fraction is markedly reduced in both the occluded and reperfused groups compared with controls, but peak aortic velocity is only slightly lower. Pre-ejection time (another index of contractility) is elevated in both ischemic groups. When followed over several months, Doppler indices of systolic (peak aortic velocity) and diastolic function (peak early filling velocity) are initially depressed in sham-operated, occluded, and reperfused groups; however, function returns to control values in the sham-operated group after 2 wk and in the reperfused group after 4 wk (Michael et al. 1999). Ejection fraction remains depressed. It appears that the mice compensate for the presence of a significant aneurysm such that stroke volume and cardiac output are preserved at nearly normal levels.

Figure 2 Lead 2 ECG signals from a mouse before, during, and after occlusion of the left anterior descending (LAD) coronary artery. Although T-waves are not often seen in mice, there is a marked S-T elevation often with arrhythmias (as shown at 15 min) after occlusion. Usually the arrhythmias resolve and the S-T elevation is reduced with time. After reperfusion, the shape of the ECG is often significantly altered from the control as shown in the bottom panel.

Figure 3 Heart rate (HR), ejection fraction (EF), peak aortic velocity (Vel), and pre-ejection time (PET) in normal (N), sham-operated (S), permanently occluded (O), and reperfused (R) mice measured 2 to 35 wk after occlusion (and reperfusion). The numbers of mice in each group are shown, and the bars are scaled as percentages of normal, with the average values shown below. Sodium pentobarbital was used for anesthesia.

Using a 20-MHz probe and a knowledge of anatomy to locate vessels and to estimate their angles, velocity signals can be obtained from many peripheral arterial sites, as shown in Figure 4. By measuring peak and mean velocity and evaluating the shapes of the waveforms from several arterial sites, subtle changes in peripheral vascular impedance have been detected in an apolipoprotein-E (ApoE¹) knockout (KO¹) mouse model of atherosclerosis (Hartley et al. 2000). In Figure 5, aortic arch velocity (V)and acceleration (dV/dt) waveforms from an ApoE-KO and an age-matched wild-type mouse are shown. The waveform from the ApoE-KO mouse has a biphasic acceleration phase in which the second peak is higher than the first peak. This phenotypic waveform may be caused by increased amplitude of wave reflections from the stenosed carotid bifurcation in ApoE-KO mice. ApoE-KO mice also have elevated aortic and mitral velocities due either to decreased cross-sectional areas or to increased cardiac output and decreased total peripheral resistance.

Figure 4 Doppler signals from several peripheral arterial sites in an anesthetized mouse using the 2-mm-diameter probe shown above. All signals were obtained with the mouse supine except for the renal signals, which were obtained with the mouse prone and the probe placed lateral to the spine.

Figure 5 Aortic arch velocity and acceleration (dV/dt) signals from a 13-mo-old apolipoprotein-E knockout (ApoE-KO) mouse and an age-matched wild-type control. Acceleration occurs in two phases in ApoE-KO mice, and the second peak (A2) is higher than the first (A1). From Hartley CJ, Reddy AK, Madala S, Martin-McNulty B, Vergona R, Sullivan ME, Halks-Miller M, Taffet GE, Michael LH, Entman ML, Wang YX. 2000. Hemodynamic changes in apolipoprotein E-knockout mice. Am J Physiol Heart Circ Physiol 279:H2326-H2334.

One of the applications for peripheral Doppler measurements of blood velocity is in the determination of arterial pulse wave velocity (PWV¹) as an index of arterial stiffness (Hartley et al. 1997). Stiffer vessels propagate pressure and velocity waves faster than more compliant vessels. By recording velocity signals from two sites separated by a known distance and measuring the difference in pulse arrival times, we can determine the pulse transit time and calculate PWV, as shown in Figure 6. In the past, sequential recordings were made at the two sites with a single probe and the arrival times were measured with respect to the ECG (Hartley et al. 1997). Currently, simultaneous measurements of velocity are made at two sites by displaying one signal up and the other down on the bidirectional display. ApoE-KO mice have elevated PWV compared with wild-type mice (Hartley et al. 2000), but in other models, the changes are not always apparent during rest. For instance, in alpha smooth muscle actin KO (αSMA^-/-) mice (Schildmeyer et al. 2000), the resting PWV is only slightly lower than in control mice, as shown in Figure 7. However, the response to a bolus intravenous injection of phenylephrine is markedly reduced in αSMA^-/- mice versus controls. In contrast, matrix GLA protein KO mice (Luo et al. 1997), which have calcified arteries, have elevated PWV even at rest.

Figure 6 Example of how aortic pulse wave velocity (PWV) can be measured in a mouse by timing the arrival of the velocity pulse from two sites a known distance apart. When two probes are used simultaneously, one is displayed up and the other down on the spectral display. Alternatively, pulses from two sites measured sequentially using a single probe can be timed with respect to the electrocardiogram (ECG). A pressure wave from a catheter in the carotid artery used in a validation study is also shown. Note that the time and amplitude scales have been expanded to improve the detection and resolution of the start or foot of the velocity waveforms.

Figure 7 Pulse-wave velocity in 19 normal, 10 alpha smooth muscle actin knockout (αSMA^-/-), and three matrix GLA^-/- protein knockout mice. The response to an intravenous injection of phenylephrine is shown for the normal and αSMA^-/- mice.

Another common method for stressing the cardiovascular system in mutant mice is a pressure overload model produced by constricting the aortic arch between the origins of the carotid arteries (Rockman et al. 1993). This model is known to produce cardiac hypertrophy in normal mice, but it is difficult to measure or predict the degree of stenosis, the pressure drop, or the resulting hypertrophy. It is possible at the time of sacrifice to cannulate (and occlude) both carotid arteries to measure the pressure difference, but this act is expected to alter flow (and the pressure drop) significantly. Using noninvasive Doppler methods, we have found differences in carotid velocity waveforms such that the ratio of peak velocities in the right and left carotid arteries measured at surgery is highly correlated with the degree of cardiac hypertrophy after 2 wk. We would also expect to find a high-velocity jet at the stenosis, although the jet is very small and difficult to locate sternally. To position a Doppler transducer closer to the stenosis, we developed an esophageal probe by mounting a 0.5-mm 20-MHz crystal to the side of a 22-gauge needle. This probe easily slides down the esophagus of a mouse and can sense velocity from the right and left carotid arteries and from the aortic arch at the stenosis and distal to it, as shown in Figure 8. We find that the pulsations in flow are increased in the right carotid artery proximal to the stenosis and decreased in the left carotid artery distal to the stenosis. In this mouse, the peak velocity at the stenotic jet is 3.5 m/sec, which corresponds to a pressure drop of 49 mmHg using the simplified Bernoulli equation (ΔP = 4V²). Distal to the stenosis, we often detect disturbed flow characterized by rapid fluctuations in velocity (250 Hz in this example), consistent with vorticity. Thus, the hemodynamic perturbations produced by this common model are significant and largely unstudied.

Figure 8 Doppler signals from four sites in a mouse with a transverse aortic band using a probe mounted to the side of a 22-gauge needle and passed down the esophagus. The pulsatility index (PI = maximum-minimum/mean velocity) is much higher in the right than in the left carotid artery. The peak velocity (V) at the stenotic jet is 3.5 m/sec, which results in a calculated pressure drop (ΔP) of 49 mmHg using the simplified Bernoulli equation. Distal to the band are transient 250-Hz oscillations in velocity.

One of the common criticisms of noninvasive Doppler measurements of velocity and ultrasonic measurements of LV dimensions is that the effect of loading conditions (blood pressure) cannot be determined (Hoit et al. 1997). In response, we are developing an improved tail-cuff system to measure both systolic and diastolic arterial pressures noninvasively in anesthetized mice. Standard tail-cuff systems can determine only systolic pressure (Krege et al. 1995); however, by incorporating a Doppler probe to sense tail artery flow distal to the pressure cuff, diastolic pressure can also be determined, as shown in Figure 9. To obtain a pressure reading, the pressure cuff is placed (deflated) on the proximal tail, and a 3.5-mm 20-MHz Doppler flow cuff (Hartley and Cole 1974) is positioned distal to the pressure cuff and oriented to detect flow in the tail artery. If no flow signal can be found, the board temperature is increased until flow is detected. As soon as flow becomes continuous, the pressure cuff is inflated until flow ceases and is then slowly deflated, as shown in Figure 9. Flow reappears when the cuff pressure reaches systolic pressure and becomes continuous throughout the cardiac cycle when the cuff pressure reaches diastolic pressure. Thus it is also possible to assess afterload in mice noninvasively.

Figure 9 Illustration of how systolic and diastolic blood pressure can be measured using a pressure cuff and a Doppler cuff placed on the tail of a mouse. The arterial pressure signal is shown for illustration purposes only and is not required. Flow in the tail artery occurs when arterial pressure exceeds cuff pressure, and the velocity waveform can be used to determine both systolic and diastolic blood pressures.

A Typical Noninvasive Study

In a typical study, the mouse is anesthetized and placed on the ECG/heater board (5-7 min). The board temperature is adjusted to 37°C, the quality of the ECG is assessed, and the electrode contact is optimized if needed. Next, cardiac Doppler signals are obtained from the mitral inflow track and from the aortic root using a 10-MHz probe placed as shown in Figure 1 (2-5 min). Then any peripheral Doppler signals are obtained from carotid or other vessels or from two aortic sites for pulse wave velocity measurements (3-5 min). If needed, M-mode and 2-D echocardiograms are taken (5-10 min). Finally, blood pressure is taken with the Doppler tail-cuff method (5-10 min). With experience, the entire process takes from 15 to 30 min per mouse, depending on the number of measurements required. Subsequent analysis of the signals may require another 15 to 30 min. It thus becomes feasible to screen and phenotype large numbers of mice using these relatively sophisticated but simple-to-apply noninvasive techniques.

Summary

A significant advantage of the measurement of velocity rather than volume flow is that it is not necessary to normalize the values (like blood pressure and ejection fraction) to body size if the cross-sectional area is considered normal (Dawson 1991; Hartley et al. 2000). In fact, cardiac index (which is cardiac output normalized to body surface area) has units of velocity. Thus, so-called normal values for pressure, velocity, and ejection fraction are similar in all mammals. The magnitudes and waveforms of pressure and velocity in mice are almost indistinguishable from signals recorded from similar sites in humans and other species with the exception of the time scale (Nichols and O'Rourke 1998). Indeed, many of the indices (Taffet et al. 1996) used to evaluate systolic (peak aortic velocity and acceleration) and diastolic function (peak E and A filling velocities and E:A ratio) in mice are derived from human indices. Arterial pulse wave velocity in mice is also similar to values reported for humans and other species (Nichols and O'Rourke 1998). Thus, the similarity to humans and the ability to make noninvasive, accurate, high-fidelity measurements of pressure and velocity in mice makes the genetically engineered mouse a viable and useful model for evaluating the cardiovascular consequences of human diseases and for developing new methods for diagnosis and treatment.

¹Abbreviations used in this article: ApoE, apolipoprotein-E; 2-D, two-dimensional; ECG, electrocardiogram; KO, knockout; LV, left ventricle; PWV, pulse-wave velocity.

Acknowledgments

The work described in this report was supported in part by grants HL22512, HL42550, HL57068, HL52364, HL42550, HL42267, AG15568, HL42313, HL13870, and AG13251 from the National Institutes of Health. The authors wish to acknowledge J. Pocius, T. Pham, L. Ochoa, K. Gould, Y. Wang, S. Madala, and J. Lacy, who contributed scientifically and technically to the studies described.

References

Agduhr F, Stenstrom N. 1929. The appearance of the electrocardiogram in heart lesions produced by cod liver oil treatment. Acta Paediatr 8:493-510.

Banda M, Lefer DJ, Granger DN. 1997. Postischemic endothelium-dependent vascular reactivity is preserved in adhesion molecule-deficient mice. Am J Physiol Heart Circ Physiol 273:H2721-H2725.

Berul CI, Aronovitz MJ, Wang PJ, Mendelsohn ME. 1996. In vivo cardiac electrophysiology studies in the mouse. Circulation 94:2641-2648.

Berul CI, Christe ME, Aronovitz MJ, Seidman CE, Seidman JG, Mendelsohn ME. 1997. Electrophysiological abnormalities and arrhythmias in aMHC mutant familial hypertrophic cardiomyopathy mice. J Clin Invest 99:570-576.

Briaud S, Ding ZM, Michael LH, Entman ML, Daniel SL, Ballantyne CM. 2001. Leukocyte trafficking and myocardial reperfusion injury in ICAM-1/P-selectin-knockout mice. Am J Physiol Heart Circ Physiol 280:H60-H67.

Chacko VP, Aresta F, Chacko SM, Weiss RG. 2000. MRI/MRS assessment of in vivo murine cardiac metabolism, morphology, and function at physiological heart rates. Am J Physiol Heart Circ Physiol 279:H2218-H2224.

Chien KR. 1996. Genes and physiology: Molecular physiology in genetically engineered animals. J Clin Invest 97:901-909.

Cowley AW Jr. 1997. Genomics to physiology and beyond: How do we get there? Physiologist 40:205-211.

Dawson TH. 1991. Engineering Design of the Cardiovascular System of Mammals. Englewood Cliffs: Prentice Hall.

Desai KH, Sato R, Schauble E, Barsh GS, Kobilka BK, Bernstein D. 1997. Cardiovasuclar indexes in the mouse at rest and with exercise: New tools to study models of cardiac disease. Am J Physiol Heart Circ Physiol 272:H1053-H1061.

Doevendans PA, Daemen MJ, DeMuinck ED, Smits JF. 1998. Cardiovascular phenotyping in mice. Cardiovasc Res 39:34-49.

Feldman MD, Erikson JM, Korcarz CE, Lang RM, Freeman GL. 2000a. Validation of a mouse conductance system to determine LV volume: Comparison to echocardiography and crystals. Am J Physiol Heart Circ Physiol 279:H1698-H1707.

Feldman MD, Mao Y, Valvano JW, Pearce JA, Freeman GL. 2000b. Development of a multifrequency conductance catheter-based system to determine LV function in mice. Am J Physiol Heart Circ Physiol 279:H1411-H1420.

Fentzke RC, Korcarz CE, Shroff SG, Lin H, Leiden JM, Lang RM. 2001. The left ventricular stress-velocity relation in transgenic mice expressing a dominant negative CREB transgene in the heart. J Am Soc Echocardiogr 14:209-218.

Gardin JM, Siri FM, Kitsis RN, Leinwand LA. 1996. Intravascular ultrasound catheter evaluation of the left ventricle in mice. Echocardiography 13:609-612.

Gardiner SM, Compton AM, Bennett T, Hartley CJ. 1990. Can the pulsed Doppler technique measure changes in aortic blood flow in conscious rats? Am J Physiol Heart Circ Physiol 259:H448-H456.

Gehrmann J, Hammer PE, Maguire CT, Wakimoto H, Triedman JK, Berul CI. 2000. Phenotypic screening for heart rate variability in the mouse. Am J Physiol Heart Circ Physiol 279:H733-H740.

Georgakopoulos D, Mitzner WA, Chen CH, Byrne BJ, Millar HD, Hare JM, Kass DA. 1998. In vivo murine left ventricular pressure-volume relations by miniaturized conductance micromanometry. Am J Physiol Heart Circ Physiol 274:H1414-H1422.

Goldbarg A, Hellerstein H, Bruell J, Daroczy A. 1968. Electrocardiogram of the normal mouse, Mus musculus: General considerations and genetic aspects. Cardiovasc Res 2:93-99.

Gui YH, Linask KK, Khowsathit P, Huhta JC. 1996. Doppler echocardiography of normal and abnormal embryonic mouse heart. Pediatr Res 40:633-642.

Hanley HG, Lewis RM, Hartley CJ, Franklin D, Schwartz A. 1975. Effects of an inotropic agent, X-537A (RO2-2985), on regional blood flow and myocardial function in chronically instrumented conscious dogs and anesthetized dogs. Circ Res 37:215-225.

Hartley CJ, Cole JS. 1974. Ultrasonic pulsed Doppler system for measuring blood flow in small vessels. J Appl Physiol 37:626-629.

Hartley CJ, Hanley HG, Lewis RM, Cole JS. 1978. Synchronized pulsed Doppler blood flow and ultrasonic dimension measurement in conscious dogs. Ultrasound Med Biol 4:99-110.

Hartley CJ, Lacy JL, Dai D, Nayak N, Taffet GE, Entman ML, Michael LH. 1999. Functional cardiac imaging in mice using Ta-178. Nature Med 5:237-239.

Hartley CJ, Michael LH, Entman ML. 1995. Noninvasive measurement of ascending aortic blood velocity in mice. Am J Physiol Heart Circ Physiol 268:H499-H505.

Hartley CJ, Reddy AK, Madala S, Martin-McNulty B, Vergona R, Sullivan ME, Halks-Miller M, Taffet GE, Michael LH, Entman ML, Wang YX. 2000. Hemodynamic changes in apolipoprotein E-knockout mice. Am J Physiol Heart Circ Physiol 279:H2326-H2334.

Hartley CJ, Taffet GE, Michael LH, Pham TT, Entman ML. 1997. Noninvasive determination of pulse-wave velocity in mice. Am J Physiol Heart Circ Physiol 273:H494-H500.

Hoit BD, Khoury SF, Kranias EG, Ball N, Walsh RA. 1995. In vivo echocardiographic detection of enhanced left ventricular function in gene-targeted mice with phospholamban dificiency. Circ Res 77:632-637.

Hoit BD, Khan ZU, Pawloski-Dahm CM, Walsh RA. 1997. In vivo determination of left ventricular wall stress-shortening relationship in normal mice. Am J Physiol Heart Circ Physiol 272:H1047-H1052.

Hoit BD, Walsh RA. 1998. Cardiovascular Physiology in the Genetically Engineered Mouse. Norwell MA: Kluwer Academic Publishers.

Hongo M, Ryoke T, Ross J Jr. 1997. Animal models of heart failure: Recent developments and perspectives. Trends Cardiovas Med 7:161-167.

James JF, Hewett TE, Robbins J. 1998. Cardiac physiology in transgenic mice. Circ Res 82:407-415.

Johansson C, Gothe S, Forrest D, Vennstrom B, Thoren P. 1999. Cardiovascular phenotype and temperature control in mice lacking thyroid hormone receptor-β or both α1 and β. Am J Physiol Heart Circ Physiol 276:H2006-H2012.

Jones SP, Girod WG, Palazzo AJ, Granger DN, Grisham MB, Jourd'heuil D, Huang Pl, Lefer DJ. 1999. Myocardial ischemia-reperfusion injury is exacerbated in absence of endothelial cell nitric oxide synthase. Am J Physiol Heart Circ Physiol 276:H1567-H1573.

Kass DA, Hare JM, Georgakopoulos D. 1998. Murine cardiac function: A cautionary tail. Circ Res 82:519-522.

Krege JH, Hodgin JB, Hagaman JR, Smithies O. 1995. A noninvasive computerized tail-cuff system for measuring blood pressure in mice. Hypertension 25:1111-1115.

Kubota T, Mahler CM, McTiernan CF, Wu CC, Feldman MD, Feldman AM. 1998. End-systolic pressure-dimension relationship of in situ mouse left ventricle. J Mol Cell Cardiol 30:357-363.

Kurrelmeyer KM, Michael LH, Baumgarten G, Taffet GE, Peschon JJ, Sivasubramanian N, Entman ML, Mann DL. 2000. Endogenous tumor necrosis factor protects the adult cardiac myocyte against ischemic-induced apoptosis in a murine model of acute myocardial infarction. Proc Natl Acad Sci U S A 97:5456-5461.

Lorenz JN, Robbins J. 1996. Measurement of intraventricular pressure and cardiac performance in the intact closed-chest anesthetized mouse. Am J Physiol Heart Circ Physiol 272:H1137-H1146.

Luo G, Ducy P, McKee MD, Pinero GJ, Loyer E, Behringer RR, Karsenty G. 1997. Spontaneous calcification of arteries and cartilage in mice lacking matrix GLA protein. Nature 386:78-81.

Michael LH, Entman ML, Hartley CJ, Youker KA, Zhu J, Hall SR, Hawkins HK, Ballantyne CM. 1995. Myocardial ischemia and reperfusion: A murine model. Am J Physiol Heart Circ Physiol 269:H2147-H2154.

Michael LH, Ballantyne CM, Zachariah JP, Gould KE, Pocius JS, Taffet GE, Hartley CJ, Pham TT, Daniel SL, Funk E, Entman ML. 1999. Myocardial infarction and remodeling in mice: Effect of reperfusion. Am J Physiol Heart Circ Physiol 277:H660-H668.

Milner DJ, Taffet GE, Wang X, Pham TT, Tamura T, Hartley CJ, Gerdes AM, Capetanaki Y. 1999. The absence of desmin leads to cardiac hypertrophy and dilation with compromised systolic function. J Mol Cell Cardiol 31:2063-2076.

Mitchell GF, Jeron A, Koren G. 1998. Measurement of heart rate and Q-T interval in the conscious mouse. Am J Physiol Heart Circ Physiol 274:H747-H751.

Nichols WW, O'Rourke MF. 1998. McDonald's Blood Flow in Arteries: Theoretical, Experimental, and Clinical Principles. London: Edward Arnold.

Patten RD, Aronovitz MJ, Deras-Mejia L, Pandian NG, Hanak GG, Smith JJ, Mendelsohn ME, Konstam MA. 1998. Ventricular remodeling in a mouse model of myocardial infarction. Am J Physiol Heart Circ Physiol 274:H1812-H1820.

Richards AG, Simonson E, Visscher MB. 1953. Electrocardiogram and phonogram of adult mice in normal conditions and under the effect of cooling, hypoxia and potassium. Am J Physiol 174:293-298.

Rockman HA, Knowlton KU, Ross J Jr, Chien KR. 1993. In vivo murine cardiac hypertrophy: A novel model to identify genetic signaling mechanisms that activate an adaptive physiological response. Circulation 87(Suppl VII):VII-14-VII-21.

Rockman HA, Ono S, Ross RS, Jones LR, Karimi M, Bhargava V, Ross J, Jr., Chen KR. 1994. Molecular and physiological alterations in murine ventricular dysfunction. Proc Natl Acad Sci U S A 91:2694-2698.

Sah VP, Minamisawa S, Tam SP, Wu TH, Dorn II, GW, Ross Jr. J, Chien KR, Brown JH. 1999. Cardiac-specific overexpression of RhoA results in sinus and atrioventricular nodal dysfunction and contractile failure. J Clin Invest 103:1627-1634.

Scherrer-Crosbie M, Steudel W, Hunziker PR, Foster GP, Garrido L, Liel-Cohen N, Zapol WM, Picard MH. 1998. Determination of right ventricular structure and function in normoxic and hypoxic mice--A transesophageal echocardiographic study. Circulation 98:1015-1021.

Schildmeyer LA, Braun R, Taffet GE, DeBiasi M, Burns AE, Bradley A, Schwartz RJ. 2000. Impaired vascular contractility and blood pressure homeostasis in the smooth muscle α-actin null mouse. FASEB J 14:2213-2220.

Siri FM, Jedlicks LA, Leinwand LA, Gardin JM. 1997. Gated magnetic resonance imaging of normal and hypertrophied murine mearts. Am J Physiol Heart Circ Physiol 272:H2394-H2402.

Spencer KT, Collins K, Korcarz C, Fentzke RC, Lang RM, Leiden JM. 2000. Effects of exercise training on LV performance and mortality in a murine model of dilated cardiomyopathy. Am J Physiol Heart Circ Physiol 279:H210-H215.

Taffet GE, Hartley CJ, Wen X, Pham TT, Michael LH, Entman ML. 1996. Noninvasive indexes of cardiac systolic and diastolic function in hyperthyroid and senescent mouse. Am J Physiol Heart Circ Physiol 270:H2204-H2209.

Vaidya D, Morley GE, Samie FH, Jalife J. 1999. Reentry and fibrillation in the mouse heart: A challenge to the critical mass hypothesis. Circ Res 85:174-181.

Verdouw PD, Van den Doel MA, De Zeeuw S, Duncker DJ. 1998. Animal models in the study of myocardial ischaemia and ischaemic syndromes. Cardiovasc Res 39:121-135.

Wang YX, Halks-Miller M, Vergona R, Sullivan ME, Fitch R, Mallari C, Martin-McNulty B, Da Cuhna V, Freay A, Rubanyi GM, Kauser K. 2000. Increased aortic stiffness by pulse wave velocity in apolipoprotein E-deficient mice. Am J Physiol Heart Circ Physiol 278:H428-H434.

Wehrens XH, Kirchhoff S, Doevendans PA. 2000. Mouse electrocardiography: An interval of thirty years. Cardiovasc Res 45:231-237.

Wiesmann F, Ruff J, Hiller KH, Rommel E, Haase A, Neubauer S. 2000. Developmental changes of cardiac function and mass assessed with MRI in neonatal, juvenile, and adult mice. Am J Physiol Heart Circ Physiol 278:H652-H657.

Williams RV, Lorenz JN, Witt SA, Hellard DT, Khoury SF, Kimball TR. 1998. End-systolic stress-velocity and pressure-dimension relationships by transthoracic echocardiography in mice. Am J Physiol Heart Circ Physiol 274:H1828-H1835.

Yang B, Larson DF, Watson R. 1999a. Age-related left ventricular function in the mouse: Analysis based on in vivo pressure-volume relationships. Am J Physiol Heart Circ Physiol 277:H1906-H1913.

Yang XP, Liu YH, Rhaleb NE, Kurihara N, Kim HE, Carretero OA. 1999b. Echocardiographic assessment of cardiac function in conscious and anesthetized mice. Am J Physiol Heart Circ Physiol 277:H1967-H1974.

Youn HJ, Rokosh G, Lester SJ, Simpson P, Schiller NB, Foster E. 1999. Two-dimensional echocardiography with a 15-MHz transducer is a promising alternative for in vivo measurement of left ventricular mass in mice. J Am Soc Echocardiogr 12:70-75.

Zhang D, Gaussin V, Taffet GE, Belaguli NS, Yamada M, Schwartz RJ, Michael LH, Overbeek PA, Schneider MD. 2000. TAK1 is activated in the myocardium after pressure overload and is sufficient to provoke heart failure in transgenic mice. Nature Med 6:556-563.