The study was performed in 36 male athletes engaged in wresting (average age 19.50 ± 1.38 years) and 32 control individuals (average age 20.03 ± 1.06 years). The inclusion criteria of the athlete group were training years ≥ 5, training time ≥ 30 hours per week and never stop training. The healthy controls with no records of participation in long-term intensive training were collected from the physical examination center of the First Affiliated Hospital of Zhengzhou University. Meanwhile, the participants having good image quality for myocardial speckle tracing analysis and without cardiovascular disease such as arrhythmia, valvular stenosis or regurgitation were necessary. Subjects that did not fulfill all the inclusion criteria were excluded. The study was authorized by the local ethics committee, and written informed consent was obtained.
Transthoracic echocardiography was performed to obtain images for analysis, employing a Vivid E95 ultrasound system equipped with a M5S 3.5 mHz transducer (GE Vingmed Ultrasound, Horten, Norway). The participants maintained a left lateral decubitus position when scanning under peaceful breath and connecting the electrocardiogram synchronously. Collect standard 2D gray-scale dynamic images consisting of three consecutive cardiac cycles in long-axis, apical two-chamber and four-chamber views, then put in a certain workstation for offline analysis. Ultrasonic recordings and measurements were processed in line with the recommendation of the American Society of Echocardiography and European Association of Cardiovascular Imaging.12
Left ventricular end-diastolic diameter (LVDd), diastolic interventricular septal thickness (IVSTd), and diastolic posterior wall thickness (PWTd) were obtained in the parasternal long-axis section of LV, and relative wall thickness (RWT) was computed using the ratio: (IVSTd + PWTd)/LVDd. Left ventricular mass (LVM) was obtained by the standard cube formula and normalized to body surface area (BSA). LV end-diastolic volume (EDV), end-systolic volume (ESV), stroke volume (SV) and cardiac output (CO) were measured and indexed to BSA using biplane Simpson’s method as recommended,13 and LV ejection fraction (EF) was calculated.
Strain analysis by 2D-STE
2D gray-scale images were obtained from the apical two-chamber, four-chamber and long-axis views at frame rates ≥ 60 frames/sec. By manually clicking on the mitral annulus and apex of three sections at the end-systolic frame, the region of interest between the endocardium and epicardium was automatically defined using 2D-STE and manually adjusted if necessary.14 Then, the assessment of global longitudinal strain (GLS) was acquired and peak strain dispersion (PSD) was subsequently determined. Aortic valve closure was automatically defined in the LV apical long-axis view.7 The GLS of LV was calculated from the average value of the three views, including 17 segments of the myocardium.
Myocardial work analysis
LVMW parameters were calculated from non-invasive pressure-strain loop (PSL) area as previously described (Fig. 2). Brachial cuff pressure measured before echocardiographic study was used to substitute for aortic pressure as peak systolic LV pressure. Then, the LV pressure curve was constructed by defining the isovolumetric and ejection time based on the valvular timing events via the software (EchoPAC ver. 202, GE Vingmed Ultrasound, Norway).15 Furthermore, the replicability of the non-invasive pressure curve was demonstrated in a dog model and in patients with diverse cardiac disorders.10
Global myocardial work index (GWI) was the total work derived from the area of LV PSL. Global constructive myocardial work (GCW), which represented positive work, was performed by contracting myocytes during systole and elongating myocytes during isovolumic relaxation. Global wasted myocardial work (GWW), representing the energy loss, was defined as myocardial lengthening during systole and shortening during isovolumic relaxation.9 Global myocardial work efficiency (GWE) was the ratio of GCW to sum of GCW and GWW.
All statistical data was processed using standard statistical software SPSS (ver. 24.0, IBM, Chicago, IL). Continuous variables were confirmed for normal distribution by the Kolmogorov-Smirnov test and expressed as mean values ± standard deviation (SD). Differences between the two groups in continuous variables were analyzed using independent t-test for normal distribution and Mann-Whitney U test for non-normal distribution. Receiver operating characteristic curve (ROC) was applied to find optimal parameters to predict LV systolic function or synchrony in athletes with values of sensitivity, specificity and the area under ROC (AUC). P-value༜0.05 were considered as statistically significance.