INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

THE DEVELOPMENT OF LEFT VENTRICULAR (LV) hypertrophy is a basic adaptive response to increased hemodynamic load. LV wall stress is believed to be an important determinant of the degree and type of LV hypertrophy induced by pressure overload (5, 6). LV systolic performance at early time points after aortic constriction has not been studied extensively, and very little information is available concerning the time course for the development of LV hypertrophy. The mouse has become an increasingly important subject for hemodynamic studies with the availability of transgenic models, but invasive measurements in mice are fraught with significant morbidity and mortality. Echocardiography is well suited for serial noninvasive measurements of LV mass to determine the rate of development of hypertrophy and its relationship to changes in wall stress as well as the acute effects on systolic performance.

With the use of transverse aortic constriction (TAC), an effective technique for inducing LV hypertrophy in mice (15), this study was designed to assess continuous-wave (CW) Doppler measurements across the constriction coupled with noninvasively measured blood pressure for an adequate estimation of peak LV systolic pressure, which together with simultaneous echocardiographic measurement of LV mass (LVM), wall thickness, and chamber diameter could be used for calculation of the LV peak systolic wall stress index (WSI). Our findings support the hypothesis that LV peak systolic wall stress is inversely related to systolic performance during the development of hypertrophy induced by acute LV pressure overload, and we define the time course for the development of LV hypertrophy in the pressure-overload model.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

This study was approved by the committee on Animal Research of the University of California at San Francisco, and the animals were maintained in accordance with the guidelines of the University of California Animal Use Committee and the National Institutes of Health Guide for the Care and Use of Laboratory Animals (NIH Publication No. 85-23, Revised 1996).

Animal preparations. Twenty-three male mice (C57BL/6, age 8-9 wk, 24-26 g body wt, Charles River Laboratories; Gilroy, CA), housed in an air-conditioned room with a 12:12-h light-dark cycle and fed standard mouse chow and tap water ad libitum, underwent TAC and survived the initial 24 h. Anesthesia was induced with 3% isoflurane gas and maintained with 1.5% isoflurane in room air supplemented with 100% O₂ (3). The surgical procedure for producing TAC in mice was performed as described by Rockman et al. (15). In brief, the aortic arch was isolated by entering the extrapleural space above the first rib, and the transverse aorta was isolated between the right and left carotid arteries. A 7-0 nylon suture ligature was tied around the transverse aorta against a 27-gauge needle to produce a 65-70% constriction (outer diameter, ~0.3 mm) after the removal of the needle. Nine age-matched animals underwent the same surgical procedure except for TAC (sham). The overall 20-day mortality rates, excluding deaths in the initial 24 h, were 39% (9 of 23) for TAC mice; none of the sham-operated mice died. Four of the TAC mice died within 1-3 days after the procedure, and five mice died within 3-20 days due to heart failure.

Transthoracic echocardiographic Doppler studies. Mice were anesthetized with 3% isoflurane and maintained with 1.5% isoflurane in room air supplemented with 100% O₂. Anesthetized animals underwent transthoracic echocardiographic Doppler studies at baseline and on postoperative days 1, 3, 10, and 20 with a commercially available system (Acuson Sequoia c256; Mountain View, CA) using a 15-MHz linear array transducer (15L8). After the anterior chest was shaved, the animals were placed on a warming pad to maintain normothermia. Two-dimensional (2D) imaging was performed at a minimum depth setting of 2 cm, and the enhanced resolution imaging function was activated with a region of interest adjusted to the heart size. Gain was set for best imaging, and compression was 60 dB. Animals were imaged in a shallow left lateral decubitus position. The echocardiographic gel was warmed before use to avoid hypothermia. Care was taken to avoid excessive pressure on the thorax, which can induce bradycardia. 2D long-axis images of the LV were obtained at the plane of the aortic and mitral valves with adequate visualization of the LV apex, and a short-axis image was recorded at the level of the papillary muscles. A 2D guided M-mode echocardiogram was recorded through the anterior and posterior LV walls at a sweep speed of 200 mm/s. Images were digitally acquired and stored on magnetooptical disk (SONY EDM-230C, Sony; Tokyo, Japan). All primary measurements were made from digital images captured on cine loops at the time of study with the use of a specialized software package (Acuson Sequoia). For each study, measurements were made from 3 beats [average beats, 3.8 ± 0.4 beats (mean ± SD); range, 3-6 beats].

(1)

(2)

(3)

(4)

Measurement of systolic blood pressure. Systolic blood pressure was measured in the conscious state (TAC mice, n = 7; sham-operated mice, n = 5) using a noninvasive computerized tail-cuff system (BP-2000, Visitech Systems; Apex, NC) as described by others (13). The reported values are the means of at least three recordings of 20 cardiac cycles per recording taken on the same time of day at baseline and at postoperative days 1, 3, 10, and 20.

Catheterization studies. To validate the peak gradient measured by CW Doppler, TAC mice (n = 14) underwent catheterization studies at postoperative day 20, just before death. Two of the mice died at catheterization. In brief, the right (upstream of the constriction) and left carotid arteries (downstream of the constriction) of anesthetized animals were exposed and cannulated with polyethylene-10 catheters connected to calibrated fluid-filled transducers (COBE transducer). Left and right carotid pressures were measured simultaneously and recorded digitally at 500 Hz (MP100 Workstation, BIOPAC Systems; Santa Barbara, CA).

Autopsy. The mice were immediately killed after catheterization studies at postoperative day 20. In 12 of the mice (TAC mice, n = 7; sham-operated mice, n = 5), the hearts were excised, and the right ventricular free wall, major blood vessels, and both atria were removed. The LV was opened and blotted once with gauze to remove intracavitary blood. The isolated LV, including the entire septum, was then weighed.

Measurement variability. To assess the variability of echocardiographic Doppler parameters, a total of 60 measurements were read independently by two observers (A. Nakamura and M. Paccanaro) in 12 animals. A single observer (A. Nakamura) repeated the measurements several weeks later. Inter- and intraobserver errors were calculated as the difference between the two observations divided by the means of the observations and expressed as a percentage (23). The agreement for both inter- and intraobserver variabilities was evaluated by correlation coefficients.

Statistical analysis. All data are expressed as means ± SD. Statistical analyses were performed with commercially available software (StatView version 4.02, Abacus Concepts; Berkeley, CA). Comparisons of values were done between sham-operated and TAC animals using an unpaired Student's t-test. Statistical comparisons of serial changes were performed by a repeated-measures ANOVA followed by Fisher's protected least significant difference test. The relationship between variables was examined by linear regression analysis. A value of P < 0.05 was considered to be statistically significant.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The results of echocardiographic and Doppler studies in surviving mice are summarized in Table 1, and heart rate and systolic blood pressure during tail-cuff measurements are summarized in Table 2. Representive 2D long- and short-axis images and M-mode tracings of the LV in a TAC mouse are shown in Fig. 1.

View larger version (119K): [in this window] [in a new window]   Fig. 1.   Representative echocardiograms of the left ventricle (LV) in a mouse with transverse aortic constriction (TAC). Two-dimensional long- (top) and short-axis images (middle) and M-mode tracings (bottom) were recorded from the same TAC mouse. LV anterior wall (AW), posterior wall (PW), and internal dimension (LVID) are indicated on the M-mode tracing. Scale bars for size on the long- and short-axis images and for both size and time on the M-mode tracing are shown. On day 1, the TAC mouse shows no significant increase in LV wall thickness compared with baseline, and LV wall motion recorded by M-mode is significantly lower than that at baseline. On day 10, LV wall thickness is significantly greater than that on baseline or day 1, and LV wall motion shows a recovery to the baseline level.

LV hypertrophy develops rapidly after TAC. The time course of LV hypertrophy acutely after TAC has not been described previously. To assess this, we banded the transverse aorta of adult male mice and studied the LV wall thickness, LVID, and LVM index from day 1 to day 20 by echocardiography. The LV wall thickness was not significantly different between TAC and sham-operated mice at postoperative day 1 but progressively increased in TAC mice by day 10 and showed no further increase at day 20 (Fig. 2A). The LVID at end diastole was not significantly different between TAC and sham-operated mice at any time point (Fig. 2A). At postoperative day 1, the LVM index was not significantly different between TAC and sham-operated mice [3.24 ± 0.42 mg/g in TAC mice vs. 3.13 ± 0.34 mg/g in sham-operated mice, P = not significant (NS)] but was 34% greater in TAC mice than sham-operated mice by day 3 and 75% greater by day 10 (Fig. 2B). There was no further increase in the LVM index between postoperative day 10 and day 20 (5.33 ± 0.71 vs. 5.52 ± 0.83 mg/g, P = NS). Autopsy LV weight (x) in surviving mice showed a strong correlation with echocardiographic LVM (y): y = 0.966x  7.903, r = 0.928, P < 0.0001, n = 12. Body weight was not significantly different between TAC and sham-operated mice at baseline and day 20 (baseline: 23.8 ± 0.6 vs. 23.6 ± 0.6 g, P > 0.2; day 20: 24.6 ± 0.8 vs. 24.2 ± 1.2 g, P = NS; Table 1).

View larger version (25K): [in this window] [in a new window]   Fig. 2.   Effects of TAC on diastolic LV posterior wall thickness (PWTd) and diastolic LVID (LVIDd) (A), LV hypertrophy (B), and LV systolic function as assessed by LV ejection fraction (LVEF; C) and percent fractional shortening (%FS; D). Values are presented as means ± SD. Sham, sham-operated mice (n = 9); TAC, mice with TAC (n = 14). *P < 0.01 and **P < 0.05 vs. sham-operated mice at the same time point; #P < 0.01, P < 0.01, and P < 0.01 vs. TAC mice at baseline, day 1, and day 3, respectively; §P < 0.01 and ¶ P < 0.05 vs. TAC mice at day 10.

LV systolic function deteriorates immediately after TAC. To assess LV systolic function after TAC, we performed 2D and M-mode measurements beginning on day 1 in the same mice. On day 1, LVEF and %FS were significantly lower in TAC mice than sham-operated mice (LVEF: 41 ± 7 vs. 61 ± 6%, P < 0.01; %FS: 16 ± 4 vs. 27 ± 4%, P < 0.01; Fig. 2, C and D). The difference narrowed by day 3, and by day 10 LVEF and %FS in TAC mice had recovered to baseline levels. LVEF and %FS in TAC mice were not significantly different between postoperative days 10 and 20 (LVEF: 62 ± 6 vs. 64 ± 5%, P > 0.7; %FS: 28 ± 4 vs. 29 ± 4%, P = NS). On day 1, the peak flow velocity of the ascending aorta in TAC mice remarkably decreased to 69% of that at baseline. Thereafter, it increased with time, coinciding with the improvement of LV systolic function, and normalized by day 10; there was no significant difference between postoperative days 10 and 20 (0.75 ± 0.15 vs. 0.76 ± 0.16 m/s, P = NS; Table 1).

Transconstriction CW flow pattern as an index of gradient. To assess the degree of afterload imposed by TAC, we measured peak velocities across the constriction by CW Doppler. The peak gradient by CW Doppler was measurable in all TAC mice and progressively increased (day 1: 28 ± 7 mmHg vs. day 3: 46 ± 11 mmHg, P < 0.01; day 3 vs. day 10: 68 ± 11 mmHg, P < 0.01) concordant with the increase in aortic flow velocity (Table 1). The peak gradient in TAC mice was not significantly different between days 10 and 20 (68 ± 11 vs. 62 ± 9 mmHg, P = NS; Fig. 3B). The peak gradient measured by CW Doppler (x) showed a good correlation with echocardiographc LVM (y): y = 0.057x + 2.084, r = 0.922, P < 0.0001 (Fig. 3C).

View larger version (65K): [in this window] [in a new window]   Fig. 3.   A: representive continuous-wave (CW) Doppler flow pattern across the TAC on postoperative days 3, 10, and 20 in a mouse with TAC. On day 3, the peak velocity (V) was 3.46 m/s, which (according to the modified Bernoulli equation) is equivalent to a peak instantaneous gradient (P) of 48 mmHg. It increased to 4.33 m/s (peak gradient: 75 mmHg) on day 10 and 4.21 m/s (71 mmHg) on day 20. B: effects of TAC on the peak gradient measured by CW Doppler echocardiography. NS, not significant. C: correlation between the peak gradient measured by CW Doppler echocardiography (x) and LV mass index (y).

View larger version (52K): [in this window] [in a new window]   Fig. 4.   A: representative CW Doppler flow pattern across the TAC on postoperative day 20. Peak velocity is 3.69 m/s, which (according to the modified Bernoulli equation) is equivalent to a peak instantaneous gradient of 54.5 mmHg. B: analog recordings of simultaneous pressures measured in the left (top) and right carotid arteries (bottom) from the same mouse as in A. The difference in peak systolic pressures from the right and left carotid arteries is 57 mmHg. C: correlation between the CW Doppler gradient (y) and catheter gradient (x) across the constriction in mice with TAC. D: correlation between the value of the CW Doppler gradient added to the tail-cuff systolic pressure (y) and pressure upstream to the constriction measured by the catheterization method (x).

LV peak systolic WSI during development of LV hypertrophy. The peak LV systolic pressure was estimated by adding the tail-cuff pressure to the peak gradient across the constriction (Table 2) and was used to estimate the LV peak systolic WSI. As seen in Fig. 5A, on postoperative day 1, the LV peak systolic WSI in TAC mice was significantly greater than that in sham-operated mice: 259 ± 40 vs. 148 ± 13 g/cm², P < 0.01. The difference narrowed by day 3, and by day 10 the LV peak systolic WSI in TAC mice had recovered to baseline levels. The LV peak systolic WSI in TAC mice was not significantly different between postoperative days 10 and 20 (131 ± 24 vs. 136 ± 8 g/cm², P = NS; Fig. 5A). LV %FS was linearly and inversely related to the LV peak systolic WSI (Fig. 5B).

View larger version (17K): [in this window] [in a new window]   Fig. 5.   A: effects of TAC on the LV peak systolic wall stress index (WSI). Values are presented as means ± SD; n = 5 sham-operated mice and 7 TAC mice. *P < 0.05 and P < 0.01 vs. sham-operated mice at the same time point. B: correlation between the LV percent fractional shortening (%FS) (y) and LV peak systolic WSI (x).

Inadequate LV hypertrophy in TAC mice. Of the five TAC mice who died between days 3 and 20, all five had echocardiographic Doppler studies on day 3 and four had echocardiographic Doppler studies on day 10. On day 3, the LVM index and LVEF were not significantly different between the TAC mice who died (n = 5) and those who survived (n = 14) (LVM index: 3.92 ± 0.49 vs. 4.21 ± 0.71 mg/g, P = NS; LVEF: 45 ± 7 vs. 55 ± 9% , P = 0.2), although the LVM index and LVEF tended to be higher in the survivors. On day 10, the LVM index was lower in those who died, but the difference did not reach statistical signficance (4.58 ± 1.26 vs. 5.33 ± 0.71; P = NS). However, the LVEF in the TAC mice who died (n = 4) was significantly lower than in survivors (n = 14) (LVEF: 44 ± 10 vs. 62 ± 6%, P < 0.05). However, there was overlap among the animals who died and those who survived. The LV peak systolic WSI was not significantly different between TAC mice who died and those who survived on postoperative day 3 (206 ± 15 vs.175 ± 20 g/cm², P = NS), but it was significantly greater on day 10 in those mice who died prematurely (222 ± 43 vs. 126 ± 32 g/cm², P < 0.01). By day 10, the LV peak systolic WSI in TAC mice who died had not recovered to baseline levels (day 1: 166 ± 14; day 3: 206 ± 15 vs. day 10: 222 ± 43 g/cm²).

Inter- and intraobserver variability. The results of inter- and intraobserver variabilities for echocardiographic measurements are summarized in Table 3. Overall, inter- and intraobserver variabilities were low, and agreements were excellent for all echocardiographic Doppler parameters. The coefficients for inter- and intraobserver measurements ranged from 0.87 to 0.97 and 0.86 to 0.98, respectively. The inter- and intraobserver variabilities for the LV peak systolic WSI were 5.6 ± 5.2 and 4.6 ± 4.6%, respectively. The coefficients for the LV peak systolic WSI were 0.92 and 0.94, respectively.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The new finding of this study is that the LV peak systolic WSI rose and contractile performance deteriorated immediately after TAC in mice but recovered to baseline levels with the development of compensatory LV hypertrophy over 10-20 days. The LV peak systolic WSI was inversely related to systolic LV performance during the development of hypertrophy induced by acute LV pressure overload.

LV contractile function before the transition to stable compensated hypertrophy has been incompletely examined in mice because of difficulty in performing serial studies. In this study, we performed transthoracic echocardiography in mice using a newly developed high-frequency linear transducer, and this is the first report as to the time course of the LV peak systolic WSI, contractile function, and hypertrophy immediately after TAC. Our observations raise the question of what mechanisms are responsible for the myocardial dysfunction and its recovery. In cardiac muscle, contraction and relaxation are regulated by cyclic release and removal of Ca²⁺ by the sarcoplasmic reticulum, and many factors, including sarco(endo)plasmic reticulum Ca²⁺-ATPase (12, 14), phospholamban (2, 12), ryanodine (2), calsequestrin (22), and the Na⁺-Ca²⁺ exchanger (25), are involved in the calcium homeostasis that plays an important role in cardiac function. This methodology forms the basis for the future study of the relationship between molecular and physiological alterations in pressure-overload-induced LV hypertrophy.

The TAC model is a stable microsurgical technique that has been developed as in vivo mouse model of myocardial hypertrophy to investigate cardiac gene expression during pressure-induced LV hypertrophy in normal and transgenic mice (4, 15, 16). Other alternative sites for aortic constriction include the ascending (7) and abdominal aorta (21). With the use of supravalvular constriction, it is difficult to reliably measure the peak gradients across the constriction. Moreover, ascending aorta constriction is more likely to cause an excessive and abrupt overload on the LV in mice. Abdominal aortic constriction leaves a considerable part of the circulation as a reactive vascular bed. Thus, with these alternative techniques, the precise hemodynamic afterload imposed on the LV in mice cannot be measured. With the use of the TAC model, we were able to measure the hemodynamic load on the ventricle in vivo by performing serial noninvasive measurements of the peak-systolic WSI.

With the use of CW Doppler echocardiography, the signals of continuous flow pattern across the constriction were detected in all TAC mice. Our results showed that the peak gradient across the constriction, which was similar to that measured by invasive technique, increased with time, coinciding with development of LV hypertrophy and improvement of LV systolic function. Although CW Doppler echocardiography was a relatively simple technique that allowed repetitive noninvasive detection of continuous flow pattern across the constriction, peak gradients measured by CW Doppler underestimated those measured invasively, probably due to angle dependency of the recorded frequency shift.

From the viewpoint of myocardial mechanics, myocyte systolic load can be estimated by LV systolic wall stress, which is an important determinant of LV systolic performance (10). Myocardial stress is believed to be the major mechanical determinant of LV hypertrophy rather than LV pressure or aortic impedance (1, 6, 8). One of the goals of this study was to assess the feasibility of longitudinal noninvasive measurements of the peak systolic WSI in TAC mice. Although a method of measuring peak LV systolic pressure by means of ventricular catheterization has been developed in TAC mice (9, 16), it requires instrumentation and microsurgical techniques and is difficult to perform repeatedly in the same animal, thereby limiting longitudinal studies. For this reason, we developed a novel noninvasive method by which upstream blood pressure across the constriction would be predictable by combination of CW Doppler echocardiography and tail-cuff blood pressure measurements.

Interestingly, several of the animals who died by postoperative day 20 appeared to have insufficient LV hypertrophy compared with surviving TAC mice. LV systolic performance in these mice had not recovered. Although the numbers are too small to allow a firm conclusion, these data suggest that inadequate hypertrophy may lead to persistent systolic dysfunction and may have contributed to mortality. Further studies, including those in transgenic models, may elucidate the maladaptive mechanisms to pressure overload.

For evaluation of LV systolic performance, the peak flow velocity of the ascending aorta was also longitudinally assessed using pulse-wave Doppler echocardiography and showed changes similar to LVEF and %FS (Table 1). Although we did not evaluate LV systolic performance invasively, it has been reported that the peak aortic velocity was highly correlated with the maximal change in pressure over time (17, 24).

A limitation of this study is that measurements of the peak gradients by CW Doppler were performed under anesthesia, whereas those of tail-cuff blood pressures were measured in the awake animals. These conditions could potentially underestimate the LV systolic pressure and, therefore, the calculated WSI. However, there was a very close correlation between pressures upstream across the constriction measured by invasive method under anesthesia and peak gradients by CW Doppler added to tail-cuff systolic blood pressures. In this study, the maximal "audible" signal was obtained for measurement of the peak pressure gradient across the constriction. The measurements in the conscious state may introduce artifacts related to fear or pain in the experimental animals. Isoflurane gas used in echocardiographic Doppler studies has been reported as an anesthetic agent with only minor hemodynamic effects (11), and our results did not show a marked slowing of heart rate in anesthetized animals. Although this study minimized the potential effects of anesthetic agent on cardiac function, repeat studies in conscious animals may provide additional insights into LV cardiac performance in the pressure-overload model.

With the advent of transgenic and knockout technology, genetically altered mice with remarkable cardiovascular phenotypes are now available. To benefit from the full potential of these genetically engineered mice, it is important to develop noninvasive approaches for accurate serial measurements of cardiac morphology and function in intact animals. The techniques developed in this study should facilitate investigation of the early effects of altered hemodynamic loading in intact mice.

In conclusion, serial changes in LV mass and systolic function during the adaptation to pressure overload were studied in a TAC murine model. This is the first demonstration that LV systolic performance declines immediately after TAC and recovers with the development of compensatory LV hypertrophy. Furthermore, our results demonstrate the technical feasibility of obtaining high-quality CW Doppler signals from the flow across the constriction in TAC mice and assessing peak LV systolic pressure for assessing the LV peak systolic WSI from measurements of peak gradients across the constriction and tail-cuff blood pressure. Because the LV peak systolic WSI is an important determinant of cardiac performance, our results provide insight into the hemodynamic mechanism of pressure-overload induced-hypertrophy.