Association of hypercholesterolemia and cardiac function evaluated by speckle tracking echocardiography in a rabbit model

Background Although hypercholesterolemia is a major risk factor for coronary artery disease (CAD), only limited data are available regarding its direct effect on myocardial function apart from CAD. The aim of this study was to evaluate LV systolic function using speckle-tracking echocardiography and investigate the relationship between hypercholesterolemia and myocardial function. Methods Twenty-eight rabbits were randomly divided into three groups: 8 were fed normal chow for 3 months (group 1) and the remaining 20 were fed an atherogenic diet for 2 (group 2) or 3 months (group 3). Global systolic radial, circumferential and longitudinal peak strain were calculated. Serum total cholesterol (TC), low density lipoprotein cholesterol (LDL-C) and myocardial cholesterol levels were measured. Results Global systolic longitudinal strain were both decreased in the group 2 and 3 (P < 0.001), whereas radial strain were increased (P < 0.001) compared with group 1. Global circumferential strain in the group 3 was significantly reduced (P < 0.001). Serum and myocardial cholesterol concentration markedly increased in the group 2 and group 3 (P < 0.001). There was a significant inverse correlation between longitudinal strain and serum TC, LDL-C as well as myocardial cholesterol levels (r = - 0.723, r = - 0.794, r = - 0.700, P both < 0.001). A significant negative correlation was also noted between circumferential strain and serum TC, LDL-C as well as myocardial cholesterol levels (r = - 0.518, P = 0.007; r = - 0.691, P < 0.001; r = - 0.659, P < 0.001). A significant positive correlation was found between radial strain and serum TC, LDL-C as well as myocardial cholesterol levels (r = 0.432, P = 0.028; r = 0.602, P = 0.001; r = 0.469, P = 0.016). Conclusion Although LV morphology and ejection fractions were not different among the three groups, elevated concentration of cholesterol, especially in serum LDL-C, was significantly associated with LV systolic dysfunction. The findings also indicate that reductions in longitudinal was the first appeared, followed by circumferential, and was compensated for by increasing radial strain.


Background
Although hypercholesterolemia has emerged as a strong risk factor for coronary artery disease (CAD) [1][2][3], only limited data are available regarding its direct effect on myocardial function apart from CAD [4][5][6]. The metabolic derangement of hypercholesterolemia can result in abnormalities of cardiac function that are likely independent of effects on the vasculature [5]. While single left ventricular (LV) myocytes isolated from hypercholesterolemic rabbits demonstrated a significant reduction in systolic function without any change in blood pressure or LV morphology [4], few data are available from in vivo investigations.
Speckle tracking echocardiography (STE), a relatively new echocardiographic imaging modalities, offers an objective and quantitative evaluation of global and regional myocardial deformation in longitudinal, radial and circumferential directions [7][8][9]. A large amount of published data has described that STE could detect subtle changes in LV function at an early subclinical stage [10][11][12][13].
The aim of the present study was to elucidate whether dietary hypercholesterolemia alters LV systolic function independently of CAD using STE in rabbits model and investigate their relationship.

Animal model
The experimental protocol was approved by a local ethical committee (First Affiliated Hospital, Xinjiang Medical University, Xinjiang, China). Twenty-eight male New Zealand rabbits (1.9-2.3 kg) were housed in separate cages in an environmentally controlled facility (AAALAC accredited) and were given water ad libitum and received humane care in compliance with institutions guidelines. The rabbits were acclimatized to laboratory conditions for 7 days prior to treatment. Eight rabbits were fed normal chow for 3 months as control (group 1) and the remaining 20 accepted an atherogenic diet for 2 months (group 2) or 3 months (group 3). The atherogenic diet contained 84% standard chow diet, 5% lard, 5% egg yolk powder and 2% cholesterol [13]. Diarrhea, appetite and coat color were observed during the experimental period.

Echocardiographic imaging
On the day of the study, rabbits fast for approximately 4 h to reduce abdominal distention and to facilitate obtaining the images. Echocardiographic images were acquired after lightly sedated with 10 mg/kg ketamine (Fujian Gutian Pharmaceutical Co., Ltd, China), 1 mg/kg Diazepam (Tianjin Jinyao Amino Acid Co., Ltd, China) and 0.025 mg/kg Atropine (Tianjin Pharmaceutical Group Co., Ltd, China) administered intravenously. The rabbits were placed in prone position without restraint. All images were obtained using a commercial ultrasound machine (Vivid 7 Dimension; GE Vingmed Ultrasound AS, Horten, Norway) with an M5S probe. M-mode images of parasternal long-axis view, B-mode images of apical three-chamber, four-chamber, and two-chamber views, short-axis views at the level of the mitral valve,  papillary muscles, and apex were obtained and digitally stored in cine-loop format for off line analysis [14]. LV end-diastolic diameter (LVEDd), LV end-systolic diameter (LVEDs), septal and LV posterior wall thickness, and left atrial anteroposterior diameter (LAD) were measured from standard planes. LV ejection fraction (EF) was calculated with the Teicholz formula [15].

Strain analysis
Two-dimensional B-mode images were captured with a frame rate of 50-80 fps and five beats were recorded for analysis. Blinded offline analyses of the short-axis views and apical long-axis views were performed using EchoPAC PC version 6.1.1 (GE Vingmed Ultrasound AS, Horten, Norway). After selecting the bestquality image of the cardiac cycle, the LV endocardial border was manually traced at the end-systolic frame, from which a speckle-tracking region of interest was automatically selected to approximate the myocardium between the endocardium and epicardium [16]. The workstation then computed and generated strain curves.   were averaged for the apical views or the short-axis views.

Blood pressure measurement
After echocardiographic imaging, blood pressure measurements were made from the central ear artery (CEA) of rabbits using a 20G vascular catheter (Johnson and Johnson, Belgium). The arterial catheter was connected to a pressure transducer (MLT0699, AD Instruments, Pty Ltd, Australia) and an analog-to-digital converter (PowerLab, ML866, AD Instruments, Pty Ltd, Australia). Arterial systolic and diastolic pressure were recorded.

Cholesterol analysis
Peripheral blood was collected from ear veins with a 25-gauge needle and syringe at baseline, 2 months and 3 months. Total cholesterol (TC) and low density lipoprotein cholesterol (LDL-C) were measured with an automated clinical chemistry analyzer (AU680, Beckman Instruments, USA). All rabbits were euthanized at correspnding time and the hearts removed. A segment of myocardial tissue (10 × 3 mm) from the LV free wall was excised. Cholesterol levels were measured in the tissue segment using a cholesterol Kit (EnzyChrom Cholesterol Assay Kit, BioAssay Systems, Hayward, CA).
Statistical analysis SPSS 16.0 (SPSS inc., Chicago, Illinois, USA) was used for statistical analysis. The data were tested for normality and homogeneity of variance. Data are expressed as mean ± standard deviation (SD). One way ANOVA was used to compare the echocardiographic parameters, strain parameters, myocardial and serum cholesterol levels for all three groups. Pearson correlation analysis was done between STE variables and cholesterol levels. For all analyses, a P value < 0.05 was considered significant. Interobserver and intraobserver variability for strain measurements were examined using both Pearson's bivariate two-tailed correlations and Bland-Altman analysis from 10 randomly selected rabbits.

Animal
Of the 28 experimental rabbits, 1 died in the group 3 due to diarrhea. Heart rates and blood pressure of three groups were similar (P > 0.05, Table 1).  Data are expressed as mean ± SD. *p < 0.05 for group 3 and group 2 vs. group 1, ▲ p < 0.05 for group 3 vs. group 2.

Figure 2
Global peak systolic longitudinal, circumferential and radial strain in the three groups. *p < 0.001 for group 3 vs. group 2 and group 1. ▲ p < 0.05 for group 2 vs group 1.

Conventional echocardiography
Echocardiographic measurements of the different groups were shown in Table 2. There were no significant differences in LVEF, LVEDd, LVEDs, IVS, PW and LAD among three groups (P > 0.05).

Strain measurements
From a total of 972 analyzed segments, 28 segments were excluded owing to suboptimal myocardial tracking and poor image quality. Regional longitudinal (Table 3) and circumferential (Table 4) strain of LV were significantly reduced in group 3 compared with group 1 and 2. But regional radial strain of LV were significantly increased in group 3 (Table 5). Global longitudinal myocardial deformation of the LV was significantly impaired both in group 2 and 3, while radial deformation was increased in group 3 compared with group 1 and 2. In addition, global circumferential strain was also reduced in group 3 compared with group 1 and 2 ( Figure 2).

Serum and tissue cholesterol profiles
The serum cholesterol profiles of the three groups after experiment were shown in Table 6. There was a statistically significant increase in serum TC, LDL-C and tissue cholesterol levels in animals fed with cholesterol enriched diet compared with the control group (p <0.05). Morerove, the concentration of cholesterol increased with feeding duration (p <0.05).
The Bland-Altman plots demonstrated acceptable interobserver variability for all strain parameters (Figure 3).

Discussion
To the best of our knowledge, the present study is the first to comprehensively compare strain parameters-derived STI with the levels of serum and myocardial cholesterol in diet-induced expeimental hypercholesterolemia. Although previous experimental studies have shown that dietinduced hypercholesterolemia resulted in contractile reduction of single ventricular myocyte without any change in pressure or LV morphology [4], few data are available from in vivo investigations. The present study displayed the application of STE as a noninvasive imaging technique to elucidate the direct effect of hypercholesterolemia on LV myocardial deformation in a rabbit model.
In our study, there were no significant differences in LV morphology, EF and blood pressure among groups, whereas LV strain was found to be reduced in the hypercholesterolemic rabbits. Previous studies failed to show abnormalities using EF, which may be due to EF reflects the whole LV systolic function, under the influence of pre and afterload [18]. With the application of advanced techniques, such as strain, strain rate, incipient systolic dysfunction has been detected in subclinical diseases [19]. Moreover, our analysis indicates that longitudinal dysfunction are the first appeared, followed by circumferential, which suggest the importance of longitudinal strain in the assessment of LV systolic dysfunction in subclinical stage.  Whether similar changes occur in humans with hypercholesterolemia can not confirm from our study. However, a recent human study demontrated longitudinal and circumferential deformations were both impaired in the children with heterozygous familiar hypercholesterolemia [20]. Thus, we believe that the abnormalities we found in rabbit models with hypercholesterolemia indicate an early sign of hypercholesterolemiainduced myocardial dysfunction, in agreement with the in vitro expriments [4].
Interestingly, our study demonstrates that the increased radial deformation make up for impaired longitudinal and circumferential strain in rabbit hypercholesterolemic models to maintain LVEF. This finding is consistent with prior reports in children with heterozygous familial hypercholesterolemia and other preclinical diseases [14,20,21]. The potential mechanism by which hypercholesterolemia causes the increase in radial deformation remains unclear. A possible explanation could be the realignments of myocardial fiber orientation in the outer half of the myocardium may contribute to "transmural compensation" by less impaired epicardial fibers [22].
In the present study, a significant negative correlation were found between global longitudinal strain and serum cholesterol level as well as myocardial cholesterol levels.
These results indicate that the cholesterol accumulated in the myocardium may be responsible for a reduction in myocardial strain. Similar to our study, Wang et al. [23] reported a positive correlation between serum HDL levels and LVEF in human subjects with serum hypercholesterolemia even in the absence of angiographic evidence of CAD.
The precise mechanism responsible for the association between cholesterol level and impaired myocardial deformation cannot be determined from our study. However, several mechanisms have been proposed to explain LV dysfunction induced by hypercholesterolemia: (1) increased cardiac oxidative stress [24], (2) alteration of the myocardial energy metabolism [22], (3) changes in myosin heavy-chain isoform expression patterns [4], (4) down-regulation and redistribution of connexin-43 expression in myocardium [25], and (5) impaired activation of myocardial adenosine triphosphate-sensitive potassium channels [19]. These mechanisms may represent the basis for a "hypercholesterolemic cardiomyopathy [26].

Study limitations
As a limitations of our study, administration of ketamine-Diazepam -Atropine combinations induces mild bradycardia, which slightly alters cardiac function. In addition, LV diastolic function, rotation and torsion mechanics are potentially very important features for the comprehensive understanding of myocardial tissue damage; therefore, lack of measurement of diastolic function, rotation and torsion was another limitation of the present study.

Conclusion
Hypercholesterolemia was significantly associated with LV myocardial functional alterations apart from CAD. The findings also indicate that decreases in longitudinal was the first appeared, followed by circumferential, and was compensated for by increasing radial strain. Thus, the application of STE may provide noninvasive functional insight into disease progression or recovery in reponse to therapeutic intervention.