Quantitative Evaluation On Cardiac Remodeling Mechanism After Intensive Exercise: A Preliminary Study of Non-enhanced Cardiac Cine MRI

Purpose: High intensity and longtime aerobic exercise may lead to the remodeling of both left and right ventricles with increased myocardial mass and cavity dilatation,which is mainly reected in the changes of traditional cardiac function parameters.Feature tracking myocardial strain allows quantitative strain analysis of myocardial functionThe purpose was to quantitatively evaluate traditional cardiac function and feature tracking myocardial strain of exercise-induced ,and Materials and methods: The study included 67 healthy volunteers (21 ± 2 years of age). The exercise group (n=43) who fullled our dened exercise criteria. The control group (n=23) who maintained a basic daily life .Noncontrast enhancement CMR scanning were performed on all the subjects using a 3T MRI scanner .Cvi42 software was used for post-processing . Left ventricular cardiac function and overall globle stress were measured. Results: Cardiac function parameters in the exercise group were signicantly higher than those of the control group except for the ejection fractions (EFs) and heart rates (HRs). The GRS peak strain and GLS peak diastolic strain rates of both groups were signicantly different (P(cid:0)0.05).The GRS peak strains and EFs were partly correlated (R=0.61). The GRS peak diastolic strains and cardiac Indices (CIs) were signicantly correlated (R=0.68). The GCS and GRS Peak Strains showed highly negative correlations (R=–0.96). The GCS and GRS time to peak values were also highly correlated (R= 0.87). Conclusion:The initial results showed that Changes in the functional parameters were more obvious than in the myocardial strain parameters, and some strain indices were correlated with the cardiac functional parameters,when the remodeling of the heart occurs.This is a new attempt to quantitatively assessment of Cardiac function and strain by Non-contrast-enhanced magnetic resonance.


Introduction
In 1899, the Swedish doctor, Hensehen [1] proposed the athlete heart concept. Since then, many scholars have con rmed that a certain amount of exercise can indeed cause changes, including the enlargement of cardiac chambers and alterations in cardiac function [2][3][4]. These exercise-induced adaptations are generally considered benign, but excessive exercise could lead to exercise-induced cardiac damage (exercise-induced myocardial ischemia, EIMI), and sudden cardiac death (SCD) [5][6], Thus, research on the mechanisms of cardiac remodeling in the early stage is needed.We need to assess whether the mechanisms of normal heart remodeling and pathological heart remodeling are consistent.
Traditionally, left ventricular remodeling has been primarily evaluated with ultrasound. However, due to the small eld of vision and poor repeatability, left ventricular reconstructions were not as accurate as those of cardiovascular magnetic resonance (CMR) [7]. Recent studies suggest that CMR continues to have a unique advantage for detecting cardiac function ,and assessing focal and diffuse myocardial brosis [8]. Late gadolinium enhancements (LGEs) of CMR imaging for detecting myocardial brosis have a distinct advantage over other methods; however, a recent study has shown that gadolinium might be deposited in the skin, dentate nucleus, and globus pallidus of patients with normal renal function [9] .. Studies have shown that non-contrast-enhanced cardiac cine MRI was able to identify myocardial strain, a sensitive indicator of early cardiovascular dysfunction [10][11]. Non-contrast-enhanced cardiac cine MRI is a standard cardiac MRI examination method that can obtain end-diastolic volume index (EDVi) evaluations of left ventricular volume changes, myocardial mass index (Myo Mass) evaluations of overall quality changes, and mass-to-volume rate (MVR) evaluations of concentric circular remodeling of left ventricular myocardia [12]. Feature tracking (FT), which has been developed in recent years, tracks tissue motion between the inner and outer membrane boundaries of lm sequences in the cardiac cycle. FT analyzes displacements and velocities for obtaining strains and strain rates. Based on longitudinal strain and the strain rates of central four-chamber lms, radial and circumferential strain and strain rates were obtained for short-axis llms [13]. The strain represents the rate of change in tissue lengths during myocardial movement: where S t is strain, L 0 is the initial length, and L is the length at a certain point. Since the length of longitudinal and circumferential motion is shorter, longitudinal and circumferential strains are negative.
Conversely, the thickness of the radial motion is increased, making radial strain positive. The immediate strain at the end of a contraction is usually recorded as the peak strain.
To this end, there are still no relevant researches about the relationship between cardiac function and strain after exercise using cardiac MRI lms. In order to further explore the occurrence,outcome of the structure and function of the exercise heart, we aimed to investigate the relationship between cardiac functional and myocardial strain after intense exercises.

Materials And Methods
Study Population 72 participants were recruited.After excluding invalid information(( hypertension (n = 2), arrhythmia (n = 1)), no heart-related diseases were found in rest of the volunteers during routine physical examinations before entering a group). 69 healthy volunteers were assessed. And all participants were divided into two groups( the exercise group and the control group).According to the criterias, the exercise group selfreported on the ful llment of prespeci ed exercise criteria, as listed in the appendix( Table 1). The control group received daily basic life and exercise standards (using the exercise software, 5000-8000 steps/day only, and no weekly long-distance running, ball games, horizontal bar training, skipping rope, push-up, and sit-up training lasting for at least 4 years .All the recruited participants received CMR imaging. At last, the study included 67 healthy volunteers (the exercise group(n = 43) and the control group(n = 23)) as showed in the owchart (Fig. 1).

Data collection
The MRI scan was performed on all the subjects using a 3T MRI scanner (MAGNETOM Trio a Tim system, Siemens Healthcare, Erlangen, Germany) with a 6-channel body matrix coil plus 2 rows of the spine array coils, used to improve eld uniformity with a target cardiac shimming mode. We compared the differences in the cardiac function parameters and myocardial strain values between the controls and exercise group. We acquired nonenhanced cardiac cine images with 25 reconstructed phases using balanced steady-state free-precession (bSSFP) sequence with retrospective electrocardiogram (ECG) gating. The collected subject information included the data from cardiac short axes, 2-chamber view, and 4-chamber view. The scan parameters were as the following: slice thickness = 6 mm, time of echo = 1.7 ms, eld of view (FOV) = 325 x 400 mm², matrix = 256 x 256, and slice thickness = 1.5 mm. The commercial software, cvi42 (Circle Cardiovascular Imaging Inc., Calgary, Alberta, Canada), was used for post-processing using the 3D short module and 3D tissue tracking module. Endo-and epicardial contours were manually traced to measure the LV cavity at end-diastole and end-systole. The inner membrane contour was distanced from the blood pool, and the outer membrane contour was distanced from the fat outside of the heart and right ventricular blood pool to avoid the partial volume effects. The papillary muscles were included in the LV volume. The LV end-diastolic volume (EDV), end-systolic volume (ESV), stroke volume (SV), ejection fraction (EF), cardiac index (CI), myocardial mass index (Myo Mass ), global radial strain (GRS), global circumferential strain (GCS), and global longitudinal strain (GLS) parameters were obtained.

Statistical Analyses
The SPSS software version 19.0 (IBM Corp., Armonk/NY, USA) was used for data analysis. We compared the differences in each value using independent sample t tests. P-values < 0.05 were considered statistically signi cant. The data were expressed as the means ± the standard deviations. Mean differences between the groups were compared using paired t-tests and the Bland-Altman methods. The Pearson's method was used to analyze the relationships between the myocardial strain and cardiac function. The person that analyzed the MRIs was blinded to the study groups. For inter-observer reproducibility, two radiologists (each with > 10 years of experience) independently analyzed the images from ve randomly selected participants. Furthermore, one radiologist (with > 5 years of experience) reanalyzed the images of ten participants after one month. The sample sizes were also calculated.   The results comparing cardiac strain parameters in the participants are shown in Table 4. No signi cant differences were found in any strain parameters except for Global Radial Strain (GRS) peak strain and Global Longitudinal Strain (GLS) diastolic strain rates. The directional changes of strain in the heart caused by exercise, were primarily axial and longitudinal. The GRS peak strain and GLS peak diastolic strain rates were lower in the exercise group compared with the control group (p <0.05), as shown in Fig. 4. The correlations between cardiac function parameters and myocardial strain measurements are shown in Fig. 5. Signi cant correlations between GRS peak strains and EF percentages were found. The correlation coe cient between the GRS peak diastolic strain and the CI was signi cant A highly negative correlation was found between the GCS and GRS Peak Strains, and the correlation coe cient of the GCS and GRS time-to-peak measurements was high..

Discussion
Exercise could cause changes in cardiac shapes, structures and functions. Some studies demonstrated that exercise can not only increase heart cavity volumes, but can also cause ventricular wall thickening and improve myocardial mass. Short exercise periods do not cause obvious changes in cardiac morphology. Cardiac enlargement is more pronounced as exercise training periods increase. This mechanical stimulus process is caused by preload and afterload pressures inducing myocardial hypertrophy [14].In this study, we assessed the feasibility of nonenhanced CMR imaging to detect myocardial strain and cardiac function after exercise. We demonstrated that 1) The ejection fractions and heart rate of participants in the exercise group were lower than those in the control group, 2) The functional parameter changes were more obvious than those of myocardial strain, and 3) Some myocardial strain index changes were correlated with cardiac function parameters.
The results of our study were consistent with previous research on the cardiac function. The cardiac function parameters of the exercise participants were higher than those of the control group, except for EFs and HRs. LVEFs are widely used and accurate for the evaluation of left ventricular systolic function. Ginzton et al [15] believed that during high-intensity exercise plays a major role in muscle contraction functions. We found that EFs of the exercise group were lower than those of the control group, suggesting that long-term high-intensity exercise resulted in marked depletion of left ventricular systolic function. These ndings suggest that people who exercise with the intensity of those in our study should actually cut back on exercise regimens. As for decreased heart rates in the exercise group participants compared with the control group participants is consistent with previous research, which showed a reduction in the excitability of both the vagus and sympathetic nerve stimulations, however, sympathetic excitability was decreased more, causing reduced heart rates.
Previous studies have demonstrated that patients with Type 2 Diabetes Mellitus (T2DM) and hypertension have structural changes in the left ventricular myocardium and stress structures[16-18].
Myocardial strain, which re ects changes in myocardial mechanics, has been considered an important indicator of the myocardial deformation [19][20], and reductions in myocardial strain are found to obe earlier than those of EFs [21]. In our study, no signi cant statistical differences were found for all strain parameters, except for GRS peak strain and GLS diastolic strain rates, indicating that changes in strain directions occurred later than changes in cardiac function, inconsistent with previous research. This inconsistency might be because healthy enrolled volunteers had no pathologic cardiac changes, and the mechanisms behind cardiac changes were different, this might be the main difference between the normal heart and pathological heart. Cardiac muscle fatigue has been reported to be caused by extraordinary endurance exercises [22], however, recovery is quick after even very long events and does not appear to stimulate pathologic or biologic processes. These results stimulated our group to assess whether cardiac function parameters and myocardial strain would be reversible after the cessation of exercise and whether some association exist between the two. In addition, myocardial strain changes caused by exercise were primarily axial and longitudinal. The decrease in longitudinal and radial strain values in the exercise group indicated myocardial elasticity changes that could be responsible for the increased cardiac morphologies and decreased in EFs.
We also correlated cardiac function parameters with myocardial strain indicators. GRS peak strains and EFs were signi cantly correlated. GRS peak diastolic strain rates and CIs were also signi cantly correlated, suggesting that cardiac functional changes were closely related to radial strains of the myocardium during the early stage of remodeling. GCS and GRS Peak Strains were highly and negatively correlated, The GCS time-to-peak and GRS time-to-peak measurements were also highly and negatively correlated, suggesting that myocardial strain of both GRS and GCS was simultaneously affected and changed during cardiac remodeling.
There are some limitations which must be addressed. First, the sample size of this study was relatively small, and more volunteers should be included in future studies. Second, the volunteers will be recruited for longitudinal studies, although this was relatively complicated. The study results regarding the correlations between myocardial strain and cardiac function provided important ndings for the study behind myocardial remodeling mechanisms. These results should be further veri ed and replicated by experiments with large sample sizes.
In summary, This study has several strengths. First, to the best of our knowledge, the cardiac magnetic resonance with its unique advantage has become the strong methods of evaluation on the heart disease,this is a new attempt to quantitatively assessment the effect of Cardiac change by Non-contrastenhanced magnetic resonance ,which provides additional insight in the eld of Cardiac remodeling.
Second, this nd out the change mechanism and potential correlation between heart function and myocardial strain in healthy heart, and the study of the mechanism of early changes in normal heart remodeling has opened up new noninvasive and reliable methods. Lastly, to nd out the similarities and differences between normal and pathological heart change reporteded in recent literature., thereby expanding the scope of our ndings. Avalability of data material Data and material are available on request.

Con ict of interest
The authors declare that there was no con ict of interest concerning the conduct of the study or study outcomes.

Authors' contributions
Prof.WJ take responsibility for the conception and design of the study as the corresponding author. LQ designed the study, and wrote the manuscript. ZQ analyzed the datasets from the study and was in charge of statistics. They made equal contributions as the rst author. JB was in charge of scaning , SN and ZY collected the datasets. All authors read and approved the nal manuscript.  were higher in the exercise than the control group participants (BOTTOM). The means plus standard deviations are presented.

Figure 4
The strain degree and direction of longitudinal strain and axial strain were changed, compared with the control group,the strain degree of exercise group was increased and the myocardial synchronicity was decreased .

Figure 5
Correlation analyses of the cardiac function parameters and myocardial strain for all participants in the study. Global radial strain (GRS) peak strains and EF percentages were signi cantly correlated (TOP left; R = 0.61, p 0.05). The correlation between the global circumferential strain (GCS) and GRS time-to-peak measurements was high (TOP right; R = 0.87). The correlation between the cardiac index (CI) and GRS peak diastolic strain was negative and signi cant (BOTTOM left; (R = -0.68, p 0.05). A strong and negative correlation was found between the GCS and GRS Peak Strains (BOTTOM right; R = -0.96).