High frame rate speckle tracking echocardiography for the mapping of cardiac mechanical activation sequence

Abstract Funding Acknowledgements Type of funding sources: Public Institution(s). Main funding source(s): KU Leuven. Background Cardiac arrhythmias are a leading cause of worldwide morbidity and mortality, with catheter ablation being an effective treatment. Current clinical practice includes performing an electrophysiological study in order to detect the focal source or the re-entrant circuit of the arrhythmia, i.e. the ablation target. This technique is invasive, with increased patient related risk and variable success rate. Conversely, echocardiography could be used to determine the correct ablation target. Indeed, high frame rate (HFR) ultrasound provides an adequately high temporal resolution, has proven helpful in detecting the onset of myocardial deformation and could be used for accurate localization of the arrhythmia sources. Purpose To test whether HFR speckle tracking echocardiography (STE) can be employed to construct cardiac mechanical activation maps. Methods 16 healthy volunteers (age: 30±6y; 69% males) and 11 heart failure patients treated with cardiac resynchronisation therapy (CRT) (age: 69±10y; 73% males) were included. All participants were scanned with a research high frame rate ultrasound scanner (frame rate: 848±101fps) and apical 4-, 2- and 3-chamber views were acquired. Patients were scanned with CRT on and after turning the device off, in order to allow native ventricular conduction. All patients had a left bundle branch block (LBBB) pattern of intrinsic activation on the surface ECG. For each apical view, a manually placed contour was tracked during the cardiac cycle by a custom-made 2D HFR STE algorithm; the contours were divided in a standard 16 segment model and segmental strain rate (SR) curves were computed and used to measure the temporal distance between electrical and mechanical activation (i.e. distance between the beginning of QRS and the first zero crossing in the SR curve representing the onset of segmental shortening). Finally, an activation map for each subject was created by placing the extracted timings in the middle of the corresponding segment of a bull’s-eye plot and interpolating values between them. Results Tracking was feasible in 96% of the segments; extracted curves showed a physiological pattern (Fig. 1A). For the healthy participants, mechanical activation started from the mid-anteroseptal (69%) or mid-inferoseptal (31%) segment at 22±5ms, spreading to the basal posterolateral segment at 50±8ms from the start of QRS, in all but two cases, where activation ended in the basal inferoseptal and basal inferior segment respectively (Fig. 1B). During CRT off, the septal wall was activated 31±9ms and the lateral wall 73±20ms after the beginning of QRS (p<0.01) (Fig. 1C), whereas during CRT on, the septal wall was activated 40±11ms and the lateral wall 37±7ms after the beginning of QRS (p = 0.3) (Fig. 1D). Conclusion Our findings comply with left ventricular activation as described in the literature and show that HFR STE could be a useful tool in defining the ablation target for arrhythmia treatment.