The value of CMR Left ventricular strain analysis in evaluating ICM

The purpose of this article is to investigate the value of cardiac magnetic resonance imaging (CMR) derived left ventricular strain parameters in evaluation of ischemic cardiomyopathy (ICM). Thirty-one ICM patients and nineteen non-cardiomyopathy (non-CM) patients who performed CMR examinations during the same period were selected for this retrospective study. The basic clinical data, CMR left ventricular function parameters, left ventricular strain parameters were compared among the left ventricular ejection fraction (LVEF) preserved ICM group, the LVEF impaired ICM group and the non-CM group. The differences of MyoGCS (-21.9 ± 1.9 vs. -18.9 ± 2.7 P<0.001), MyoGLS (-20.8 ± 2.3 vs. -17.0 ± 2.9 P<0.001) and EndoGLS (-22.2 ± 3.1 vs. -17.6 ± 3.7 P<0.001) between LVEF preserved ICM group and non-CM group were statistically significant, while the differences of left heart function parameters between the two groups were not statistically significant (P > 0.05). The left ventricular strain analysis can be used to assess cardiac functional and morphological alterations in ICM patients prior to changes of left ventricular function parameters, which has high clinical significance.


Introduction
ICM often evolved from coronary artery disease (CAD). Perennial and repetitive myocardial ischemia can cause myocardial fibrosis, impaired cardiac function, cardiac enlargement, arrhythmia, myocardial infarction, heart failure and series of other pathological changes, which can be very harmful to the health of patients [1,2]. Therefore, early diagnosis of ICM has great clinical significance. CMR which can provide imaging evidence for myocardial ischemia has become the gold standard for evaluating cardiac structure and function due to its advantages of high soft tissue resolution, multi-parameter imaging and no radiation [3, history of previous cardiac surgery. In addition, 19 patients without cardiomyopathy were selected as non-CM group.

CMR acquisition
All subjects were examined on a 3.0 T MR scanner (Discovery MR 750 W, GEHC. Waukesha, WI, USA). Beta-blockers and anti-angina drugs were stopped for at least 24 h before the examination, and smoking, alcohol, tea and coffee were prohibited. Data for cardiac function and strain analysis were collected using balanced steady-state free precession (Balance-SSFP) cine CMR sequence with a retrospective ECG-gated breath-hold technique (end-expiratory) in 2-, 3-, and 4-chamber long-axis views as well as short-axis views. Scan parameters were as follows: slice thickness 8 mm, slice spacing 3 mm; TE shortest, TR, 3.3ms, flip Angle 50°, echo chain 256, matrix 256 × 113, FOV 300 ~ 370mm 2 .

CMR imaging analysis
Image post-processing was performed by two MRI physicians with more than 5 years of working experience, and the average value of parameters was taken as the final result. In case of a parameter with significant difference, the third MRI physician with more than 8 years of working experience was added, and the average value of two groups was taken as the final result.

Left ventricular strain analysis
The collected CMR images were imported into commercial software (Medis Suite 4.0) on an offline workstation for left ventricular strain analysis (Fig. 1). The left ventricular endocardial and epicardial contour were drawn by the joint method of artificial intelligence automatic identification and manual calibration. The software automatically generates strain parameters of the left ventricle as follows: MyoGCS

Statistical analysis
Statistical analysis was performed using SPSS 25.0. Measurement data were tested for Shapiro Wilk normality test and homogeneity of variance test. Data with normal distribution and homogeneity of variance were expressed as means and standard deviations ( − x ±s), and analysis of variance (ANOVA) was used for comparison among three groups, and independent sample t-test was used for comparison between two groups. The skew distribution data were expressed as P50 (P25, P75), and Mann-Whitney test was used for group comparisons and Bonferroni method was used for multiple comparisons. Dichotomous data were presented as percentages and were compared with chi-square test. All tests were significant as P < 0.05.

Analysis of clinical data
The differences in age, BMI, hypertension, systolic blood pressure, diastolic blood pressure, fasting blood glucose and total cholesterol between the ICM group and the non-CM group were statistically significant (P < 0.05), while the differences of other clinical data were not statistically significant (P > 0.05).The data is shown in Table 1.

Left ventricular strain parameter analysis
ANOVA among the three groups found significant differences in MyoGCS, MyoGLS, EndoGCS, EndoGLS, GRS, ESL, ESDbas, and EDDbas (P < 0.05), and no significant difference in EDL (P > 0.05). Bonferroni pairwise comparison showed that MyoGCS, MyoGLS, and EndoGLS were lower in LVEF preserved ICM group than those in non-CM group (P < 0.05); MyoGCS, MyoGLS, EndoGCS, Endo-GLS, and GRS were lower and ESL, ESDbas and EDDbas were higher in LVEF impaired ICM group than those in non-CM group (P < 0.05); MyoGCS, MyoGLS, EndoGCS, EndoGLS were lower and ESL, ESDbas, and EDDbas were higher in LVEF impaired ICM group than those in LVEF preserved ICM group (P < 0.05). The data are shown in Table 3. were LVEF preserved and their average value of LVEF was 50.6 ± 15.0(> 50%), indicating that ICM with preserved LVEF are not uncommon. There were not statistically significant differences of SV and CO among three groups, suggesting that myocardium of partial ICM patients were still at the compensatory stage. ICM patients suffering from heart overload in the long-term could trigger the myocardial hypertrophy featured by enlargement of myocardial both in cell volume and in mass. Cardiac hypertrophy in early ICM is mainly concentric hypertrophy, when the myocardial thickness increases while the cardiac cavity is not significantly expanded. With the progression of the disease, myocardial hypertrophy evolves into the stage of centrifugal hypertrophy, and the cardiac cavity also expands significantly. In this study, the inter-group differences of EDVI, ESVI, EDV, ESV, MAED, MA, MAES, EDEV and ESEV between the non-CM group and the LVEF impaired ICM group were statistically significant (P < 0.05), but there were no statistical significance between the non-CM group and the LVEF preserved ICM group (P > 0.05), indicating that LVEF impaired ICM had evolved into the stage of eccentric hypertrophy of myocardium, while LVEF preserved ICM may still be at the compensatory stage. The above results indicated that the left heart function parameters alone cannot make an accurate evaluation of left heart function in patients with LVEF preserve ICM, and more sensitive quantitative evaluation methods are needed to assess left ventricular deformation and motion dyskinesia in order to identify ICM with LVEF preserved earlier. The lack of statistically significant differences of SV, CO, CI and SVI among the three groups further indicated that most of ICM patients had not evolved into the stage of decompensation. If we could identify LVEF preserved ICM and interfere as soon as possible at the early stage, more ischemic myocardium could be saved, which can reduce the occurrence of cardiovascular risk events, and avoid or delay ICM from entering the decompensation stage. Therefore, more sensitive parameters are needed at the early stage in order to make up the Discussion ICM which consumes substantial medical resources is the major cause of cardiovascular death world-wide [10,11]. The discovery of ICM risk factors possesses significant clinical value in preventing ICM at the early stage and reducing the occurrence of cardiovascular risk events. With the change of diet structure and living standard, the aging of population, diabetes, hypertension and hyperlipidemia have become common chronic diseases, and also act as risk factors for many diseases, posing a great threat to human health [12]. In this study, the differences between ICM and non-CM patients in age, BMI, hypertension, systolic BP, diastolic BP, fasting blood glucose level, and total cholesterol level were statistically significant (P < 0.05), while there were no statistically significant differences relating to gender, weight and height between the two groups (P < 0.05). The result was similar to the previous studies and clinical consensus [13]. ICM belongs to a special high-risk subgroup of CAD, which could cause variation of left ventricular systolic function which is a valid predictor of cardiovascular mortality [14][15][16][17]. ANOVA in this study found significant differences among the non-CM and ICM(with or without LVEF impairment) groups in EDVI, ESVI, EDV, ESV, MAED, MA, MAES, EDEV, and ESEV (P < 0.05). Multiple comparisons showed that the above mentioned parameters in the LVEF impaired ICM group were higher than those in the non-CM group (P < 0.05), and EDVI, ESVI, EDV, ESV, EDEV and ESEV in the LVEF impaired ICM group were also higher than those in the LVEF preserved ICM group, while there were not statistically significant difference in all of the LV functional parameters between the LVEF preserved ICM group and the non-CM group (P > 0.05). LVEF is a commonly used parameter in clinical evaluation of left heart function and can reflect myocardial contractility [18]. The heart rate in patients with ICM at the early stage accelerates to increase cardiac output, but as ICM progresses, the acceleration of heart rate is insufficient to compensate so that cardiac output and ejection fraction decrease. Among  clinical value if the clinical intervention is initiated at this stage, which may save myocardium and prevent the patient from entering the decompensated stage. In this study, the variation of some left ventricular strain parameters in the LVEF preserved ICM patients are preliminarily discussed. The theoretical basis is provided for radiologists to evaluate changes of left heart function quantitatively in the ICM patients. The new assessment methods and the more accurate imaging data are provided for the ICM patients and clinicians.

Limitations
There are some limitations in our study. Firstly, the sample of this study is small, so the results need to be verified in large sample study. Secondly, in this study, only the global left ventricular strain parameters were analyzed, segment analysis will be studied in the future.

Conclusion
The changes in quantitative parameters of left ventricular strain in the LVEF preserved ICM patients were earlier than the changes in left ventricular function parameters. The finding has high clinical application value.
insufficiency of assessment ability of traditional left heart function parameters.
Myocardial strain is a concept proposed by Mirsky in the 1970s to quantitatively evaluate left ventricular compliance, which theoretically refers to the degree of myocardial deformation without interference from traction of adjacent tissue and overall displacement of the heart [19][20][21]. Due to its advantages of high soft tissue resolution and multiparameter imaging, CMR has been recommended by the European Society of Cardiology Guidelines for the Diagnosis and Management of Chronic Coronary Syndrome 2019 (IIB) for the evaluation of cardiac anatomical and functional variation in CAD patients that cannot be determined by ultrasound [22][23][24][25][26]. CMR myocardial strain analysis can identify left cardiac dysfunction sooner than traditional CMR functional parameters [27][28][29], which can be used to identify the LVEF preserved ICM with higher sensitivity and specificity [30][31][32][33]. For CMR myocardial strain analysis, it is unnecessary to add extra scanning sequence and the operation is simply and highly repeatable [34,35]. The analysis of the myocardial strain is supposed to divide the myocardial motion in the cardiac cycle into mutually perpendicular motions in three directions: radial motion, circumferential motion and longitudinal motion, and then series of strain parameters are derived in three directions accordingly. In this study, the strain parameters (Table 4) with statistical differences among the three groups were mainly the reflection of the differences between the non-CM group and the LVEF impaired ICM group, and between the two ICM subgroups. The results suggested that these strain parameter changes like decrease of EndoGCS and increase of ESL, ESDbas and EDDbas had higher correlation with LVEF impairment, which provide us similar information as traditional left ventricular function analysis provided. However, the ability of these strain parameters to evaluate the changes of myocardial motion dyskinesia in the early ICM was weak.
This study mainly focuses on the exploration of strain parameters which has statistically significant differences between the non-CM group and the LVEF preserved ICM group so as to provide us with strain parameters which can identify early stage ICM patients with LVEF preserved. This study found that MyoGCS, MyoGLS and EndoGLS had decreased before LVEF impaired, indicating that the changes in myocardial motility had occurred in LVEF preserved ICM patients. Therefore MyoGCS, MyoGLS and EndoGLS can be used as useful parameters to detect changes in left heart function earlier than LVEF in ICM patients. In the early stage of ICM, the myocardial motion in the direction of circumferential and longitudinal has been weakened, but the myocardium is still in the compensatory stage due to its strong compensatory capacity. It is of great