Purpose: Deformation imaging can be used to detect an impairment of left ventricular (LV) function by abnormalities of circumferential and radial strain as well as twist- and untwisting. The detection of abnormal subepicardial deformation by 2D speckle tracking might be a promising approach using echocardiography, especially in athletes with suspected myocarditis. However, limitations of 2D speckle tracking due to artefacts have not been described. The aim of the present study was to test the feasibility of deformation analysis of LV rotation.

Methods and Results: 2D speckle tracking was performed in athletes (n=50) and healthy volunteers (n=20, control group) with excellent image quality to determine radial and circumferential strain as well as basal and apical LV rotation. Parasternal basal, mid and apical short axis views were acquired by biplane scanning to verify the level of the sectional planes in relation to maximum diastolic and systolic LV length. The susceptibility due to artefacts was defined by varying peak differences within circumferential graphs.

Typical monophasic radial and circumferential strain curves as well as counter-clockwise apical and clockwise basal LV rotation curves were only observed when the corresponding basal parasternal views have been acquired at a level above 85% of the maximum LV length at end-diastole and above 75% at end-systole. At these levels, it was possible to acquire apical parasternal short axis views in 97% (n=68/70). The extent of apical rotational deformation correlated with the level of the short axis views of the LV. The more apical, the higher the maximum of rotational deformation indicating the importance of LV levels of the analysed parasternal views. The basal lateral (endocardial: 6%, n=4/70)and basal posterior (endocardial: 8%, n=5/70) segment were highly susceptible to artefacts, e.g. rip shadowing, blooming by the pericardium or lung superimpositions, even after manual optimisation of the tracking area. Artefacts could not be avoided in 4% of the athletes (Fig. 1a - before optimisation; Fig. 1b - after optimisation) but in all subjects of the control group (Fig. 1c - before optimisation; Fig. 1d - after optimisation).