Subjects

This case-control study included 100 men: 25 competitive top-level athletes (rugby players) with moderate LVH (13–15 mm), 25 competitive top-level athletes (rugby players) without LVH, 25 patients with HCM and moderate LVH (13–15 mm) and 25 sedentary healthy subjects (control group).

Top-level athletes were recruited from professional players of the Top 14 French National Rugby League. All athletes were screened annually by ECG and echocardiography and none had a history of cardiovascular disease. They had all undertaken endurance (running with interval training) as well as power (bodybuilding) training for > 10 hours/week for at least the past 8 months. The only specific criterion for the inclusion of athletes was septal thickness for athletes with moderate LVH.

HCM was defined by a wall thickness of ≥ 13 mm plus one of the following: a family history of HCM, a positive genetic screening or when a systolic anterior movement of the anterior mitral valve leaflet was present (with or without left ventricular outflow tract [LVOT] gradient). Patients with HCM were selected from our local cohort of patients with proven HCM without LVOT obstruction at rest before alcohol septal ablation or myomectomy. According to the Maron classification, 80% of patients with HCM were type 3, 12% were type 4 and 8% were type 2. Patients with atrial fibrillation, wide QRS duration (≥ 120 ms), an implanted pacemaker, with a suspected cause of HCM other than sarcomeric aetiology or with LV dysfunction (left ventricular ejection fraction [LVEF] ≤ 50%) were excluded.

The control group consisted of sedentary healthy subjects without a history of cardiac disease, with no specific criteria for inclusion. All subjects provided written informed consent and the study was approved by our local ethics committee.

Standard echocardiographic examination

Comprehensive 2D and 3D echocardiography was performed using a commercially available ultrasound system (Vivid 9, General Electrics, Horten, Norway). 2D echocardiography assessment included parasternal long- and short-axis views, apical 4-, 3- and 2-chamber views and conventional and tissue spectral Doppler imaging. LV, left atrial and aortic diameters were measured by M-mode from the parasternal long-axis view. Relative wall thickness (RWT) was defined as the sum of the septal and posterior wall thicknesses divided by the LV end-diastolic diameter. LV mass was determined using Devereux’s formula. Ejection fraction and LV volumes were calculated using Simpson’s biplane method. Pulsed mitral flow and tissue spectral Doppler imaging of the lateral and septal walls were used to assess diastolic function.

3D echocardiography assessment included an LV full-volume acquisition using multi-beat modality (four cardiac cycles) from the apical view. LV volumes and LVEF were computed using semiautomated software (3D auto left ventricular quantification, EchoPAC, GE).

Speckle tracking and dyssynchrony measurements

2D and 3D strain components were computed from apical views using speckle-tracking analysis. The temporal resolutions for 2D and 3D data were 53 ± 3 and 22 ± 7 frames per cardiac cycle, respectively. Analysis was performed offline using dedicated 2D and 3D Echo-PAC software (GE, version 110.1.2). Automated function imaging for 2D imaging and auto left ventricular quantification for 3D imaging used a block-matching model to compute strain data. Endocardial delineation was obtained after manual positioning of the mitral valve plane and LV apex. The region of interest was adjusted manually to provide optimal wall tracking and segments that were inadequately tracked were discarded. For strain processing, the peak of the R wave was used as the end-diastolic reference point. 2D global longitudinal strain (GLS) was obtained by averaging the 16 regional longitudinal strain curves computed from the 2D apical views (4 C, 2 C, 3 C) and 3D GLS and circumferential strain components by averaging the 17 regional strain curves separately. 3D global radial and area strain components were derived by the software from longitudinal and circumferential curves. For better comprehensibility, all strain data are expressed as absolute values. LV dyssynchrony was defined as the standard deviation (16-segment model) (“2D-16SD”) of time to peak 2D longitudinal strain as previously described . Time to peak strain was automatically computed using Excel from exported regional strain curves.

Statistical analysis

Continuous variables with a normal distribution are expressed as mean ± standard deviation (SD), those with a non-normal distribution as median (interquartile range [IQR]). Dichotomous data are expressed as numbers and percentages. To compare numerical data between different groups, paired Student’s t test, analysis of variance (ANOVA), Mann–Whitney and Kruskal–Wallis tests were used as appropriate. Nominal variables were compared using either the χ ² or Fisher tests. Differentiation between athletes and patients with HCM was performed in the subgroup of subjects with septal or posterior wall thickness 13–15 mm. Receiver operator characteristic (ROC) analysis was performed to evaluate the ability of different echocardiographic parameters to identify patients with HCM and the area under the curve (AUC) expressed accuracy. During echocardiographic analysis in the subgroup of patients with moderate LVH, the investigator (CB) was blinded to the aetiology. Reproducibility of 2D and 3D GLS and LV dyssynchrony was performed in 10 random patients by an independent observer (JT). Reproducibility is expressed as the absolute difference and the absolute difference divided by the mean value (reported as a percentage). Two-tailed P -values < 0.05 were considered statistically significant.