Speckle tracking echocardiography could detect the difference of pressure overload-induced myocardial remodelling between young and adult rats

Abstract

The assessment by speckle tracking echocardiography (STE) provides useful information on regional and global left ventricular (LV) functions. The aim of the study is to investigate if STE-based strain analysis could detect the difference of pressure overload-induced myocardial remodelling between young and adult rats. Physiological, haemodynamic, histological measurements were performed post-operatively in young and adult rats with transverse aortic constriction (TAC) as well as the age-matched shams. Two-way ANOVA was used to detect the statistical difference of various measured parameters. Pressure overload decreased the ejection fraction, fractional shortening, dp/dt_max and $|\text{d}p/\text{d}{t}_{\min }|$, but increased the LV end-diastolic (ED) pressure in adult rat hearts for nine weeks after TAC operation than those in young rat hearts. Pressure overload also resulted in different changes of peak strain and strain rate in the free wall, but similar changes in the interventricular septum of young and adult rat hearts. The changes in myocardial remodelling were confirmed by the histological analysis including the increased apoptosis rate of myocytes and collagen area ratio in the free wall of adult rat hearts of LV hypertrophy when compared with the young. Pressure overload alters myocardial components in different degrees between young and adult animals. STE-based strain analysis could detect the subtle difference of pressure overload-induced myocardial remodelling between young and adult rats.

1. Introduction

Left ventricular (LV) hypertrophy (LVH) is known as a risk factor for cardiovascular morbidity and mortality [1]. Patients with LVH often have the high-blood pressure [2,3]. The prevalence of hypertension is increasing in both children and adolescents [4,5]. LV mass (LVM) and LVM index (LVMI), mainly determined by echocardiography, are often recommended as the diagnostic method of LVH [6,7]. Abnormal LVMI, however, occurs frequently in children with newly diagnosed hypertension or pre-hypertension and cannot predict LVH [8]. Hence, new diagnostic tools are needed to enhance the predictive diagnosis of LVH by echocardiography.

The assessment of myocardial deformation through the speckle tracking echocardiography (STE) has already been shown to provide useful information on regional and global ventricular functions [9]. These techniques were applied to various coronary heart diseases (CHD), e.g. LV diastolic dysfunctions [10], ischaemic heart disease (IHD) [11], heart failure with preserved ejection fraction (HFpEF) [12,13], and so on [14–16]. The prevalence of LVH increases with rising blood pressure and body mass index in ageing adults [17–19], which can alter both regional and global ventricular functions. The changes of LV functions in children with LVH may be different from those in adult LVH given the compensatory mechanisms in the young [20]. Although the STE-based strain analysis was applied to the diagnosis of patients with CHD, there is still a lack of studies to compare the differences in myocardial remodelling and biomechanical parameters by echocardiography between children and adult LVH.

Small animal models have been developed to investigate novel therapies in CHD [21,22]. Echocardiography has been a standard non-invasive method for the assessment of cardiac functions in those small animals [23]. Echocardiographic analysis was enhanced by the speckle tracking technique to predict the later development of LV remodelling [24]. Longitudinal strain was suggested as a new candidate for a prognostic indicator of LV systolic remodelling as hypertrophy progresses [25]; peak systolic strain rate was also found to be a more relevant parameter to assess myocardial contractile function [26,27]; and STE-based strain analysis could assess regional and global functions in the remodelling murine LV after myocardial infarction (MI) [28]. This study applies these techniques to analyse subtle changes in cardiac deformation in the rat model of LVH.

The objective of the study was to identify subtle changes in LV remodelling in young and adult rat hearts for 12 weeks after transverse aortic constriction (TAC) operation as well as in the age-matched shams. Our hypothesis is based on: (1) in response to pressure overload, there will be a difference in morphometric and haemodynamic parameters and more severely impaired systolic and diastolic functions in adult rat as compared with the young; (2) STE-based strain and stress analyses could detect the subtle differences in physiological, haemodynamic and cardiac stress parameters. The significance and implications of the study could potentially improve clinical diagnosis of LVH in children.

2. Methods

2.1. Experimental protocol

Studies were carried out in 133 male Wistar rats, including six-week-old rats (70 young animals, weighing 268 ± 26 g) and greater than eight month old rats (63 adult animals, weighing 551 ± 40 g). The experimental protocol consisted of four groups: 35 young animals of LVH induced by TAC (i.e. YM group) and 34 adult animals of LVH induced by TAC (i.e. AM group) as well as 35 young sham animals (i.e. YS group) and 29 adult sham animals (i.e. AS group). Various experimental measurements were performed in the four groups for 3, 9 or 12 weeks post-operatively, as shown in figure 1a. The animal was terminated by injecting an overdose of pentobarbital sodium via the tail vein. All animal experiments were performed in accordance with Chinese National and Peking University ethical guidelines regarding the use of animals in research, consistent with the NIH guidelines (Guide for the care and use of laboratory animals) on the protection of animals used for scientific purposes. The experimental protocols were approved by the Animal Care and Use Committee of Peking University, China.

Figure 1.

(a) A schematic of the experimental protocol, (b) the number of rats terminated at 3, 9 or 12 weeks after the operation in the four groups. All of the following data in the figures are from these samples in each group. The four groups include young sham (YS), young model (YM), adult sham (AS) and adult model (AM), which are showed by a dotted line with dot, solid line with dot, dotted line with hollow box and solid line with hollow box, respectively, in the following figures. (c) Parasternal long-axis echocardiographic view with the 2D guided M-mode and B-mode tracings, and (d) a schematic of long-axis and short-axis echocardiographic views and a schematic of deformation analysis in six segments along the long-axis echocardiographic view, where segments 1–6 refer to LV basal posterior wall, LV middle posterior wall, inferior apex, basal septum, middle septum and anterior apex, respectively.

2.2. Animal preparation for transverse aortic constriction

In a sterile environment, surgical anaesthesia was maintained with approximately 2% isoflurane and animals were intubated and ventilated with room air and oxygen using a Harvard ventilator (Inspira). Electrocardiography (ECG) signals were monitored during the entire surgical operation. After the chest was shaved and sterilized, a left thoracotomy was performed between the second and third intercostal spaces. LVH was induced through constriction of the ascending aorta near the sinus of Valsalva with a 16-gauge needle for young rats and a 2.4 mm-diameter stick for adult rats as previously described [20,29], which resulted in approximately 60% stenosis. An animal was included in the statistical analysis only when the systolic pressure was increased by 40–50 mmHg in both the YM and AM groups at the time of termination (see the following section ‘Haemodynamic measurements before the termination'). Alternatively, the suture was placed but removed in sham-operated animals. After the chest was closed, animals were intramuscularly administered a dose of penicillin (400 000 U) and 1 ml dezocine (50 µg ml⁻¹) and allowed to recover from the surgery. Animals were given an intramuscular injection of penicillin (400 000 U) and 1 ml dezocine (50 µg ml⁻¹) for 3 consecutive days.

2.3. Echocardiographic measurements

ECG measurements of rat hearts, as shown in figure 1b, were carried out in the four groups for 3, 9 and 12 weeks post-operatively. M-mode measurements of the left ventricle, left atrium, and aorta and B-mode measurements of strain and strain rate were recorded in rats, similar to a previous study [30]. The images were obtained at 21 MHz using a MS-250 transducer operated by a Vevo2100 Colour Doppler Ultrasound Scanner (FUJIFILM VisualSonics Inc). Based on M-mode tracings, morphometric parameters, e.g. LVID;d, LVID;s, LVFW;s, LVFW;d, IVS;s and IVS;d, were measured according to the American Society of Echocardiography leading edge rule [31]. These parameters were averaged based on five measurements. Moreover, FS (%) and EF (%) were calculated from the measured parameters as: $((\text{LVID ; d-LVID ; s})/\text{LVID};\text{d})\times 100\text{\%};$ and $((\text{LVID; }{\text{d}}^3-\text{LVID; }{\text{s}}^3)/\text{LVID; }{\text{d}}^3)\times 100\text{\%}$, respectively, in Vevo LAB image analysis workstation.

2.4. STE measurements

In B-mode tracings, 2D greyscale images were obtained from the standard parasternal longitudinal view [10]. Frame rate is 133 Hz, gain is 20–25 dB, depth is approximately 20 mm, width is approximately 23 mm, and three cardiac cycles are recorded. Myocardial deformation measurements were demonstrated using the Vevo LAB image analysis workstation with advanced STE, which tracks natural acoustic markers (called speckles) across the cardiac cycle and estimates velocity vectors. Furthermore, longitudinal and radial stains ($=\mathrm{\Delta }L/L_0$, where $L_0$ and ΔL refer to the baseline length at the R-Wave and the absolute change in length, respectively) and strain rates ($=\mathrm{\Delta }V/L_0$, where $\mathrm{\Delta }V$ is the velocity gradient in the segment) were determined by the software across the entire left ventricle over a selected period of cardiac cycles.

2.5. Haemodynamic measurements before the termination

Animals were anaesthetized, intubated and ventilated with room air and oxygen for 3, 9 or 12 weeks post-operatively. A 1.4F micromanometer-tipped catheter (Millar Instruments) was inserted through the right carotid artery into the left ventricle to record pressure waves in 30 cardiac cycles, which was repeated three times. The zero-pressure baseline of the catheter was calibrated in 37°C saline. The catheter was monitored with a BIOPAC MP150. Heart rate, LV systolic pressure (LVSP), LV ED pressure (LVEDP), and rate of maximum positive and negative LV pressure development $((\text{d}p/\text{d}{t}_{\max })\hspace{0.17em}\text{and}\hspace{0.17em}(d\hspace{0.17em}\text{p/}d{\text{t}}_{min}))$ were determined from the measured pressure waves [32].

2.6. Histological evaluation

As shown in figure 1a, animals were separately terminated for the histological analysis at post-operative 3, 9 or 12 weeks. After hearts were harvested, plugs of myocardial tissues were removed from different positions of the left ventricle. These plugs were fixed in 4% paraformaldehyde (PFA)/PBS solution overnight at room temperature and then processed for paraffin sectioning. Masson's trichrome and Picro-Sirius Red (PSR) staining was carried out according to standard procedures [33,34]. Tunel staining was performed according to manufacturer's recommendations (in situ Cell Death Detection Kit, Fluorescein, catalogue number 11684795910, Roche) [33]. Sections were also detected via wheat germ agglutinin (WGA) conjugated to Alexa Fluor 594 (50 ug ml⁻¹, Invitrogen), similar to a previous study [35]. Moreover, nuclear morphology was assessed by Hoechst 33258 dye (Molecular Probes) as reported [36].

2.7. Myocardial stress computation

Similar to a previous study [37], endocardial and epicardial contours at the ED state in long-axis view of LV wall of 20 randomly selected rats at 12 weeks (five from the AM group, five from the AS group, five from the YM group and five from the YS group) were manually segmented from 2D echocardiographic images by the Vevo LAB image analysis workstation (FUJIFILM VisualSonics Inc, Amsterdam, The Netherlands). The 3D geometrical model was generated based on the endocardial and epicardial contours and meshed with linear hexahedral finite elements using the Matlab (The MathWorks, Natick, MA, USA). The estimation of cubic Hermite derivatives was performed at nodes using the Blender software (Blender Foundation, Amsterdam, The Netherlands) plugin. Finally, the 3D model with cubic-Hermite finite elements (82 nodes, 36 elements) was built using the estimated derivatives. The fibre direction on the LV wall was modelled by the coordinate frame interpolation [38] and myofibre angles of the endocardium and epicardium were prescribed as 53° and −52° with respect to the circumferential direction and varied linearly through the LV wall [39]. The FE meshes were input to Continuity software (UCSD CMRG Group, San Diego, CA, USA) for computation of stress.

Material properties of the left ventricle obey Fung's transversely isotropic constitutive law [40], which was prescribed to the entire left ventricle as

$$\begin{array}{rl}W& =\frac{C}{2}(e^Q-1)\\ Q& =b_f{E}_{ff}^2+b_t(E_{cc}^2+E_{tt}^2+E_{ct}^2+E_{tc}^2)+b_{fs}(E_{fc}^2+E_{cf}^2+E_{ft}^2+E_{tf}^2),\end{array}\}$$ 2.1

where E_ij with subscripts $\{i,j\}\in \{f,c,t\}$ are the components of the Green strain tensor E (f, c and t represent the fibre, cross-fibre and transverse-fibre directions, respectively). Since the heart is continually loaded in vivo, a method developed by Krishnamurthy based on the multiplicative decomposition of the deformation gradient tensor [41] was adopted to estimate the unloaded left ventricle from the ED left ventricle. Moreover, Klotz et al. proposed an ED pressure–volume relationship to eliminate the difference between individuals and across species [42].

The stress scaling coefficient C was determined using the pattern search method [43] in the Matlab. Briefly, the stress scaling coefficient C was first set to 1.1 kPa from previous studies [39], based on which the ED pressure–volume relationship was computed. The difference between the computed ED pressure–volume relationship and the one defined by Klotz was set to the objective function. The pattern search method updated the stress scaling coefficient iteratively to minimize the objective function. According to the result of estimation, the stress scaling coefficient equals to 5.1 ± 0.5, 2.7 ± 0.3, 4.2 ± 0.6 and 2.5 ± 0.1 kPa for AM, AS, YM and YS rats, respectively. Resting material parameters, b_f, b_t and b_fs, were set to 9.2, 2 and 3.7 consistent with a previous study [37]. Based on the measured pressure waves, the LVEDP was set to 14.7 ± 1.0, 7.4 ± 0.6, 12.3 ± 1.4 and 6.5 ± 0.6 mmHg for AM, AS, YM and YS rats, respectively.

2.8. Data analysis

Morphometric, haemodynamic and myocardial stress data were expressed as mean ± s.e.m. (standard error of the mean) in the four groups. A two-way ANOVA (SigmaStat 3.5) was used to detect the statistical difference of those parameters between the sham and TAC groups and between young and adult groups, where p < 0.05 was indicative of a significant difference between two populations.

3. Results

Figure 2a shows representative rat hearts for 3, 9 and 12 weeks post-operatively. The ratio of HW and LVM to BW in rats after nine weeks of TAC significantly increases and has a higher value in the YM group, as shown in figure 2b,c. Echocardiographic measurements show a significant decrease of EF and FS in rats after nine weeks of TAC when compared with the shams, as shown in figure 3a,b. The ratios of $(\text{E}{\text{F}}_{\mathrm{YM}}-\text{E}{\text{F}}_{\mathrm{YS}})/(\text{E}{\text{F}}_{\mathrm{AM}}-\text{E}{\text{F}}_{\mathrm{AS}})$ and $(\text{F}{\text{S}}_{\mathrm{YM}}-\text{F}{\text{S}}_{\mathrm{YS}})/(\text{F}{\text{S}}_{\mathrm{AM}}-\text{F}{\text{S}}_{\mathrm{AS}})$ are < 1 for 12 weeks after TAC, which indicates a higher decrease of EF and FS in adult rat hearts with TAC than young rat hearts with TAC. EF and FS in the AM, YM, AS and YS groups have values of 65.3 ± 3.1 and 19.4 ± 2.1%, 77.0 ± 2.2 and 22.7 ± 3.1%, 81.4 ± 3.1 and 34.3 ± 5.0%, and 82.3 ± 1.8 and 33.6 ± 3.2%, respectively, at post-operative 12 weeks. Figure 3d,e shows a gradual increase of LV internal diameter in diastole and systole (i.e. LVID;d and LVID;s) in adult rats after TAC (p-value < 0.05 at post-operative weeks 9 and 12) in comparison with the adult shams despite no statistical difference between the YM and YS groups. Moreover, figure 3f shows a gradual increase of LVEDP in the YM and AM groups. The LVSP in the YM and AM groups is significantly higher than that in the YS and AS groups, which is relatively unchanged after TAC, as shown in figure 3g. Figure 3h,i shows a gradual decrease of $\text{d}p/\mathrm{d}{t}_{max}$ and $|\text{d}p/\mathrm{d}{t}_{min}|$ in rat hearts after TAC. LVEDP and LVSP are higher and $\text{d}p/\mathrm{d}{t}_{max}$_, and $|\text{d}p/\mathrm{d}{t}_{\min }|$ are lower in the AM group when compared with the YM group. Table 1 lists post-operative changes (mean ± s.e.) of morphometric and haemodynamic parameters in the four groups. The LV wall thickness of both the YM and AM groups is greater than the shams.

Figure 2.

(a) Rat hearts in the four groups for 3, 9 and 12 weeks after operation (long-axis and short-axis views; scale bar: 5 mm), (b) post-operative changes of the ratio of HW to BW, and (c) post-operative changes of the ratio of LVM to BW. Data are mean ± s.e.m. *p < 0.05, YM versus YS. ^#p < 0.05, AM versus AS. Symbol keys of each group are listed in figure 1 legend.

Figure 3.

(a) Post-operative changes of EF (%); (b) FS (%); (c) the ratios of EF (Y/A) and FS (Y/A) (i.e. $(\text{E}{\text{F}}_{\mathrm{YM}}-\text{E}{\text{F}}_{\mathrm{YS}})/(\text{E}{\text{F}}_{\mathrm{AM}}-\text{E}{\text{F}}_{\mathrm{AS}})$ and $(\text{F}{\text{S}}_{\mathrm{YM}}-\text{F}{\text{S}}_{\mathrm{YS}})/(\text{F}{\text{S}}_{\mathrm{AM}}-\text{F}{\text{S}}_{\mathrm{AS}})$); (d) post-operative changes of LVID;d; (e) LVID;s; (f) post-operative changes of LVEDP; (g) post-operative changes of LVSP; (h) dp/dt_max and (i) dp/dt_min. Data are mean ± s.e.m. *p < 0.05, YM versus YS. ^#p < 0.05, AM versus AS. ^♦p < 0.05, AM versus YM.

Table 1.

Post-operative changes of echocardiography parameters.

	young sham			young model
measurement	3 W	9 W	12 W	3 W	9 W	12 W
HR (bmp)	365 ± 41	342 ± 49	368 ± 21	391 ± 33	468 ± 43	342 ± 41
BW (g)	355 ± 26	466 ± 46	505 ± 32	354 ± 37	407 ± 58§	450 ± 26||
HW (g)	1.15 ± 0.10	1.35 ± 0.10	1.71 ± 0.17	1.46 ± 0.14*	2.17 ± 0.35§	2.52 ± 0.19||
SV (ml)	0.20 ± 0.04	0.27 ± 0.03	0.26 ± 0.03	0.26 ± 0.01*	0.25 ± 0.02	0.26 ± 0.02
CO (ml/min)	72.6 ± 7.64	81.9 ± 3.86	87.2 ± 8.32	113.3 ± 12.2*	84.5 ± 12.8	83.2 ± 10.1
LVFW;s (mm)	2.64 ± 0.38	2.77 ± 0.27	2.95 ± 0.26	2.87 ± 0.24	3.27 ± 0.56§	3.52 ± 0.47||
LVFW;d (mm)	1.63 ± 0.28	1.67 ± 0.14	1.93 ± 0.13	1.77 ± 0.26	2.17 ± 0.37§	2.35 ± 0.22||
IVS;s (mm)	2.82 ± 0.39	3.02 ± 0.32	3.13 ± 0.25	3.23 ± 0.06*	3.62 ± 0.29§	3.81 ± 0.26||
IVS;d (mm)	1.69 ± 0.16	1.84 ± 0.13	2.11 ± 0.18	1.89 ± 0.27	2.33 ± 0.23§	2.50 ± 0.20||

	adult sham			adult model
measurement	3 W	9 W	12 W	3 W	9 W	12 W
HR (bmp)	339 ± 18	345 ± 18	345 ± 18	330 ± 31	328 ± 31	342 ± 19
BW (g)	618 ± 45	670 ± 43	707 ± 31	561 ± 43†	584 ± 63‡	616 ± 35#
HW (g)	1.78 ± 0.15	1.72 ± 0.23	1.82 ± 0.22	2.05 ± 0.15†	3.03 ± 0.27‡	3.19 ± 0.23#
SV (ml)	0.25 ± 0.06	0.24 ± 0.05	0.25 ± 0.04	0.25 ± 0.03	0.29 ± 0.03‡	0.26 ± 0.02
CO (ml/min)	94.0 ± 11.9	81.9 ± 11.9	85.4 ± 21.6	93.5 ± 7.43	98.3 ± 19.7	89.5 ± 5.44
LVFW;s (mm)	2.96 ± 0.39	3.04 ± 0.38	2.87 ± 0.37	3.35 ± 0.48	3.51 ± 0.18‡	3.63 ± 0.45#
LVFW;d (mm)	1.90 ± 0.28	1.82 ± 0.26	2.01 ± 0.15	2.19 ± 0.19†	2.24 ± 0.21‡	2.33 ± 0.39#
IVS;s (mm)	3.03 ± 0.18	3.09 ± 0.13	3.02 ± 0.20	3.43 ± 0.17†	3.63 ± 0.34‡	3.72 ± 0.51#
IVS;d (mm)	2.12 ± 0.18	2.07 ± 0.15	2.17 ± 0.13	2.46 ± 0.17†	2.50 ± 0.18‡	2.63 ± 0.28#

*p < 0.05 versus young sham 3 W; §p < 0.05 versus young sham 9 W; ||p < 0.05 versus young sham 12 W.

†p < 0.05 versus adult sham 3 W; ‡p < 0.05 versus adult sham 9 W; #p < 0.05 versus adult sham 12 W.

Figure 4a,b shows representative images of radial strain curves in the four groups, where the black line refers to the averaged strain curve over the free wall (a combination of segments 1–3 in figure 1d) or interventricular septum (a combination of segments 4–6 in figure 1d) and the arrow points to the peak value (Sr_PK). Imposition of pressure overload on the myocardium results in an increase and decrease of Sr_PK in the interventricular septum and free wall, respectively, of adult rats while it increases the values in young rat hearts, as shown in figure 4c–f. Accordingly, figure 5a–h shows post-operative changes of peak longitudinal strains (Sl_PK) on endocardium and epicardium of the four groups. Young rats with TAC have a higher value of $|\text{S}{\text{l}}_{\mathrm{PK}}|$ at post-operative weeks 3 and 9 than the shams, but show no statistical difference of $|\text{S}{\text{l}}_{\mathrm{PK}}|$ at 12 weeks post-operatively. There is a decreased $|\text{S}{\text{l}}_{\mathrm{PK}}|$ in the free wall of adult rats of LVH after three weeks post-operatively despite no changes in the interventricular septum. Peak radial and longitudinal strain rates (SRr_PK and SRl_PK) in figure 6 show similar post-operative changes to the strains in the AM and YM groups when compared with the AS and YS groups.

Figure 4.

(a,b) Typical images of radial strain curves from one cardiac cycle in the four groups. The green, pink, pale blue, mazarine, yellow and purple lines represent the radial strain curves of segment 1–6, respectively (corresponding to the six myocardial segments in figure 1c), and the black line refers to the averaged strain curve of the other three curves, where the arrows point to the peak value; the x-axis in (a) and (b) are time. (c,d) Post-operative changes of peak radial strains in the free wall; and (e,f) post-operative changes of peak radial strains in the interventricular septum. Data are mean ± s.e.m. *p < 0.05, YM versus YS. ^#p < 0.05, AM versus AS. Symbol keys of each group are listed in figure 1 legend.

Figure 5.

(a,b) Post-operative changes of peak longitudinal strains in the endocardium of the free wall; (c,d) post-operative changes of peak longitudinal strains in the endocardium of the interventricular septum; (e,f) post-operative changes of peak longitudinal strains in the epicardium of the free wall; and (g,h) post-operative changes of peak longitudinal strains in the epicardium of the interventricular septum. Data are mean ± s.e.m. *p < 0.05, YM versus YS. ^#p < 0.05, AM versus AS. Symbol keys of each group are listed in figure 1 legend.

Figure 6.

(a,b) Post-operative changes of peak radial strain rates in the free wall; (c,d) post-operative changes of peak radial strain rates in the interventricular septum; (e,f) post-operative changes of peak longitudinal strain rates in the endocardium of the free wall; (g,h) post-operative changes of peak longitudinal strain rates in the endocardium of the interventricular septum; (i,j) post-operative changes of peak longitudinal strain rates in the epicardium of the free wall; and (k,l) post-operative changes of peak longitudinal strain rates in the epicardium of the interventricular septum. Data are mean ± s.e.m. *p < 0.05, YM versus YS. ^#p < 0.05, AM versus AS. Symbol keys of each group are listed in figure 1 legend.

Figure 7a shows a comparison of ED pressure–volume relationships between the present computational results and the estimations from the Klotz's method, which shows strong correlation (R² > 0.95 for all animals). Figure 7b shows the distribution of Cauchy stress on a long-axis cardiac surface at diastole corresponding to figure 7a. Figure 7c shows mean Cauchy stresses (averaged over elements) in both the free wall and interventricular septum of the four groups. There are similar Cauchy stresses in the free wall and interventricular septum of the YS and AS groups. In comparison with the shams, the elevated LVEDP in conjunction with the changes of cardiac wall thickness and LV volume results in a significant increase of stresses in the AM group, but a relatively slight increase in the YM group for 12 weeks after TAC. The Cauchy stress increases by 61% and 28% in the free wall of the AM and YM groups, respectively, while it only increases by 26% and 18% in the interventricular septum of the two groups.

Figure 7.

(a) A comparison of end-diastolic pressure–volume relationships between the present computational results and previous observations in [42], where AS, AM, YS and YM represent representative AS, AM, YS and YM rats at 12 weeks. (b) Typical Cauchy stress distribution on a cardiac surface of AS, AM, YS and YM rats corresponding to the long-axis echocardiographic view in figure 1b; and (c) mean Cauchy stress (averaged over all elements) in the interventricular septum and free wall of AS, AM, YS and YM rats (five rats in each group). Data are mean ± s.e.m. *p < 0.05, AM free wall versus AM interventricular septum.

Figure 8a shows representative immunofluorescence images of myocytes stained by WGA and Hoechst for 12 weeks after TAC operation. Imposition of pressure overload on the myocardium increases the size of myocytes (i.e. a decrease of pixels/area) and results in larger myocytes in the free wall than the interventricular septum, as shown in figure 8b. Figure 8c shows representative images of myocytes stained by Tunel and Hoechst and figure 8d shows a significant increase in apoptosis rate in the free wall of adult rat hearts for 12 weeks after TAC operation. Figure 8e shows representative images of myocardial tissues stained by PSR. Imposition of pressure overload on the myocardium increases collagen fibres and leads to a higher increase of collagen area ratio in the free wall of adult rat hearts than the young for 12 weeks after TAC operation, as shown in figure 8f,g.

Figure 8.

(a,b) Representative immunofluorescence images of myocytes stained by WGA and Hoechst (scale bar, 100 µm) (a) and cell size measured by the average number of pixels per area in the free wall and interventricular septum of animals in the four groups at 12 weeks post-operatively (b); (c,d) Representative images of myocytes stained by Tunel and Hoechst (scale bar, 1 mm) (c) and apoptosis rate (average number of pixels per area) in the free wall and interventricular septum of animals in the four groups at 12 weeks post-operatively (d); and (e–g) representative images of myocardial tissues stained by PSR (scale bar, 10 μm) (e) and collagen area ratio (%) in the free wall and interventricular septum of animals in the YS and YM groups (f) and in the AS and AM groups (g). Data are mean ± s.e.m. *p < 0.05, YM versus YS. ^♦p < 0.05, YM (IVS) versus YM (FW). ^#p < 0.05, AM versus AS. ^§p < 0.05, AM (FW) versus AM (IVS).

4. Discussion

The present study carried out a comparison of LV remodelling between young and adult rats of LVH resulting from pressure overload (an increased systolic pressure of about 40 mmHg caused by TAC). The major findings are reported as: (1) there are more deteriorated LV functions (i.e. a decrease of EF, FS, dp/dt_max and $|\text{d}p/\mathrm{d}{t}_{\min }|$ as well as an increase of LVEDP) in adult rat hearts for nine weeks after the TAC operation than those in young rat hearts, (2) pressure overload leads to different changes of strain and strain rate in the free wall, but similar changes in the interventricular septum of young and adult rat hearts and (3) a significant increase in the apoptosis rate of myocytes and collagen area ratio occurs in the free wall of adult rat hearts of LVH when compared with the young.

Pressure overload can increase the LV wall stress [20]. To follow the uniform wall stress rule [44,45], a significant increase of LV wall thickness (i.e. LVFW;s, LVFW;d) and LV mass occurred after nine weeks of TAC operation. In comparison with the shams, a significant increase of LVID;d and LVID;s in adult rat hearts happened during 9 and 12 weeks after the TAC operation despite no statistical difference between the YM and YS groups. The compensation in young animals increased LV wall thickness to maintain the uniform wall stress given the pressure overload from TAC. Similarly, LV wall thickness was increased in adult rat hearts after TAC. On the other hand, the left ventricular wall becomes stiffened because of cellular and molecular mechanisms due to cardiac ageing [17], which still requires more studies. The internal diameter in the left ventricle of adult rats had to increase to maintain uniform wall stress in response to hardened myocardium and elevated pressure after TAC. This induced deteriorated LV functions in the AM group, i.e. a higher decrease of EF, FS, dp/dt_max and $|\text{d}p/\mathrm{d}{t}_{\min }|$, and a higher increase of LVEDP.

Myocardial measurements of radial and longitudinal peak strains ($\text{S}{\text{r}}_{\mathrm{PK}}$ and $\text{S}{\text{l}}_{\mathrm{PK}}$) and strain rates ($\text{SR}{\text{r}}_{\mathrm{PK}}$ and $\text{SR}{\text{l}}_{\mathrm{PK}}$) by the STE provide the quantitative assessment of LV myocardial systolic shortening [46–49]. Global myocardial function quantification illustrated a decrease in longitudinal and radial strains and strain rates in patients with diastolic heart failure [10,50], but others showed preserved radial values in patients of hypertension [49]. Here, the decrease in strain, $\text{S}{\text{r}}_{\mathrm{PK}}$ and $|\text{S}{\text{l}}_{\mathrm{PK}}|$, in the free wall of adult rat hearts with TAC as well as the increased LVSP shows systolic stiffening and the decrease in strain rate, $\text{SR}{\text{r}}_{\mathrm{PK}}$, demonstrates the severely impaired systolic contractile function despite negligible changes of $\text{SR}{\text{l}}_{\mathrm{PK}}$. Moreover, the increased $\text{S}{\text{r}}_{\mathrm{PK}}$, $\text{SR}{\text{r}}_{\mathrm{PK}}$ and $|\text{SR}{\text{l}}_{\mathrm{PK}}|$ occur in the interventricular septum of adult rats with TAC when compared with the shams, which implies the possibility of the reserved systolic contractility. Hence, different responses of regional systolic contractile functions to pressure overload can reasonably explain the inconsistent measurements of global myocardial function [10,49,50]. The interventricular septum is affected by the pressure and flow of both LV and RV while the free wall only undergoes the LV haemodynamic stimuli in a cardiac cycle. Moreover, myocardial fibres are of local anisotropy such that LV mechanical properties are nonlinear, time varying and spatially inhomogeneous [51–59]. The structure-functional difference between the free wall and interventricular septum results in different changes in systolic contractile functions in adult rats for 12 weeks after TAC.

On the other hand, the augmentation of $\text{S}{\text{r}}_{\mathrm{PK}}$, $|\text{S}{\text{l}}_{\mathrm{PK}}|$, $\text{SR}{\text{r}}_{\mathrm{PK}}$ and $|\text{SR}{\text{l}}_{\mathrm{PK}}|$ in both the free wall and interventricular septum of young rat hearts with TAC reveals acute compensatory remodelling of LV to maintain the Frank–Starling mechanism during the initial period after TAC operation, which is consistent with a previous study [20]. Furthermore, the augmented strains and strain rates in young rats with TAC gradually decrease with time extension and have similar values to those in the shams for 12 weeks after surgery. This is mainly attributed to the gradually altered myocardial components that increase the LV stiffness and impair the LV systolic shortening.

A comparison of the measured EF, FS, dp/dt_max and $|\text{d}p/\mathrm{d}{t}_{\min }|$, radial and longitudinal peak strains and strain rates between the YM and AM groups reveals that imposition of pressure overload on the myocardium results in more deteriorated systolic functions in the left ventricle of adult rats than the young. Multiple factors lead to LV remodelling of different degrees between young and adult rats. For example, the PSR staining shows a higher ratio of collagen fibres in the LV of adult rats that increases the LV stiffness. Although the free wall has a lower ratio of collagen fibres than the interventricular septum, there is significant augmentation of collagen fibres in the free wall of adult rats in response to pressure overload. This is an important risk factor to increase the stiffness of the free wall for impairing the contractile reserve. In particular, the higher apoptosis rate of myocytes significantly contributes to the impaired systolic shortening in the free wall of adult rats with TAC albeit the enlarged size of myocytes could enhance the LV contractility.

The computational results of cardiac stress show an increase of Cauchy stresses by 61% and 28% in the free wall of the AM and YM groups, respectively, when compared with the shams and an increase by 26% and 18% in the interventricular septum of the two groups at the ED time instance. The stress is approximately equal to $(\text{LV}\hspace{0.17em}\text{pressur}{\text{e}}_{\mathrm{ED}}\times \text{LV}\hspace{0.17em}\text{diamete}{\text{r}}_{\mathrm{ED}})/\text{LV}\hspace{0.17em}\text{wall}\hspace{0.17em}\text{thicknes}{\text{s}}_{\mathrm{ED}}$. The increase in LV pressure and diameter is higher than that of LV wall thickness, which increases the ED stresses in rats for 12 weeks after TAC. At diastole, the free wall thickness is lower than the interventricular septum in adult rats for 12 weeks after TAC. Hence, pressure overload leads to significantly regional difference of diastolic cardiac stresses in the LV of aged rats. The elevation of ED stress was accompanied by an increase in myocardial oxygen consumption [60]. The long-term elevation of ED stress could lead to fibre elongation, chamber enlargement, and hypertrophy [61] and destroy the supply–demand balance in the myocardium [62]. Hence, pressure overload could lead to more deteriorated diastolic function in the free wall of adult rats than the young, which remains to be validated against experimental measurements in future studies.

4.1. Clinical implications of the study

This study shows the increased LVMI, unchanged haemodynamic parameters (i.e. EF, FS, LVEDP, dp/dt_max and $|\text{d}p/\mathrm{d}{t}_{\min }|$), and augmented systolic shortening in young rat hearts for three weeks after TAC operation. Hence, abnormal LVMI does not denote the occurrence of LVH in children with newly diagnosed hypertension. By contrast, STE-based strain analysis shows the impaired systolic functions in adult rat hearts with the increased LVMI for three weeks after TAC operation despite no statistical difference of EF and FS between the AS and AM groups. Cardiac stress computation with the increased LVEDP also denotes the impaired diastolic function in the AM group. Therefore, abnormal LVMI indicates the occurrence of LVH in adolescents with newly diagnosed hypertension. Abnormal LVMI in conjunction with the deteriorated haemodynamic environment and impaired systolic shortening obtained from STE-based strain analysis may be a reasonable diagnostic indicator to predict LVH in children with newly diagnosed hypertension or pre-hypertension, which requires further clinical trials. Different compensatory mechanisms of heart different regions also implies a new direction in therapy whereby cardiologists could monitor the motion of the free wall of patients in order to discover abnormal heart symptoms as early as possible.

4.2. Critique of the study

The present study only considered animals with reduced EF, but excluded animals with preserved EF. The throughout study considering both reduced and preserved EF should be performed to improve the clinical diagnosis. Regional cardiac stresses were computed at the ED time instance. The distribution of cardiac stress in a cardiac cycle will be determined when further measurements are carried out to demonstrate material parameters in the AM, AS, YM and YS groups. The following study also needs to carry out more experimental measurements to reveal molecular and cellular mechanisms though histological measurements were used here to explain the findings from haemodynamic and STE measurements.

5. Conclusion

This study investigated the use of STE-based strain and stress analyses to detect the subtle difference in LV remodelling degrees between young and adult rat hearts in response to pressure overload, which is consistent with the histological measurements. Imposition of pressure overload on the myocardium results in more deteriorated morphometric and haemodynamic parameters and impairs systolic and diastolic functions of the free wall more severely in adult rat hearts than the young. This is could be detected by the STE-based analyses. Hence, our study provides a potential non-invasive tool to monitor the changes in adverse cardiac remodelling.

Ethics

All animal experiments were performed in accordance with Chinese National and Peking University ethical guidelines regarding the use of animals in research, consistent with the NIH guidelines (Guide for the care and use of laboratory animals) on the protection of animals used for scientific purposes. The experimental protocols were approved by the Animal Care and Use Committee of Peking University, China.

Data accessibility

Original data are provided as electronic supplementary material.

Authors' contributions

P.N. and L.L. performed the experiments; Z.Y. performed the computation; P.N. and Y.H. drafted the manuscript; Y.H., W.T. and J.D. reviewed the manuscript; and all authors approved it for publication.

Competing interests

We declare we have no competing interests.

Funding

This research is supported in part by National Natural Science Foundation of China grant nos. 11672006 (Y.H.) and 11732001 (W.T.) and Shenzhen Science and Technology Program KQTD20180411143400981 (W.T. and Y.H.) and Leading Talents of Guangdong Province Program 2016LJ06S686 (W.T.).