An in silico guide for ventriculo-ventricular delay programming for left bundle branch-optimized cardiac resynchronization therapy

Figures

• 

  Figure 1. In silico study pipeline. Twenty-four publicly available heart failure anatomies were used in this study (top). Two LV conduction substrates were simulated: LV diffuse conduction system delay and LV intra-myocardial delay, in the presence or in the absence of proximal LBBB. We simulated CRT delivery methods (BVP, LBBP, and LOT-CRT), and BVP and LOT-CRT were simulated with different VVDs. We validated the model against clinically measured response, and we quantified LV synchrony with different VVDs for different LV conduction substrates. BVP, biventricular pacing; CRT, cardiac resynchronization therapy; LBBB, left bundle branch pacing; LOT-CRT, left bundle branch pacing–optimized cardiac resynchronization therapy; LV, left ventricular; VVD, ventriculo-ventricular delay.

• 

  Figure 2. The in silico model was validated against response from electrocardiographic (ECGi) data during baseline (proximal LBBB), biventricular pacing (BVP) and left bundle branch pacing (LBBP). On the left, we show a schematic of the modelled scenarios with the corresponding simulated epicardial activation maps. On the right, we compare the simulated and the measured electrical epicardial response by plotting the mean biventricular and LV activation time during baseline, and their reduction in response to BVP and LBBP. The black lines show ± 1 SD, and blue and orange bars show the measured and the simulated data, respectively. LV, left ventricular.

• 

  Figure 3. Response to LOT-CRT for different VVDs for different LV conduction substrates. The top row shows the LVAT95 for proximal LBBB combined with different LV conduction substrates. From left to right: mild and severe LV diffuse conduction system delay, simulated as LV His-Purkinje system CV slowed to 60% and 40% of healthy CV; mild and severe LV intra-myocardial delay, simulated as LV myocardium CV slowed to 60% and 40% of healthy CV. The bottom row shows the same results with LV conduction substrates alone, without proximal LBBB (with healthy proximal His-Purkinje system, HPS), where mild and severe conduction diseases were simulated using 40% and 20% of healthy CV, respectively. All results are shown for VVDs ranging from −100 ms (LBBP ahead) to +100 ms (LVepiP ahead). The grey bars and the stars indicate the LVAT95 at baseline and the best VVD on average, respectively, while the error bars indicate ± the standard deviation. CV, conduction velocity; HPS, His-Purkinje system; LBBB, left bundle branch block; LBBP, left bundle branch pacing; LOT-CRT, left bundle branch pacing–optimized cardiac resynchronization therapy; VVD, ventriculo-ventricular delay.

• 

  Figure 4. Comparison between LOT-CRT and BVP for different VVDs and different LV conduction diseases. Mean LVAT95 is shown as a function of VVDs during LOT-CRT (top row) and BVP (bottom row). Different columns represent different conduction substrates in combination with proximal LBBB. From left to right: mild and severe LV conduction system delay, simulated as LV His-Purkinje system CV slowed to 60% and 40% of healthy CV; mild and severe LV intra-myocardial delay, simulated as LV myocardium CV slowed to 60% and 40% of healthy CV. All results are shown for VVDs ranging from −100 ms (LBBP or RV pacing ahead) to +100 ms (LVepiP ahead). The grey bars and the stars indicate the LVAT95 at baseline and the best VVD on average, respectively, while the error bars indicate ± the standard deviation. CV, conduction velocity; LBBB, left bundle branch block; LBBP, left bundle branch pacing; LOT-CRT, left bundle branch pacing–optimized cardiac resynchronization therapy; VVD, ventriculo-ventricular delay.

• 

  Figure 5. Comparison between response to LOT-CRT with non-selective and myocardial capture for different VVDs and different conduction substrates. Mean LVAT95 is shown as a function of VVDs during LOT-CRT with non-selective (top row) and purely myocardial (bottom row) capture (LV septal pacing). Different columns represent different LV conduction substrates in combination with LBBB. From left to right: mild and severe LV conduction system delay, simulated as LV His-Purkinje system CV slowed to 60% and 40% of healthy CV; mild and severe LV intra-myocardial delay, simulated as LV myocardium CV slowed to 60% and 40% of healthy CV. All results are shown for VVDs ranging from −100 ms (LBBP ahead) to +100 ms (LVepiP ahead). The grey bars and the stars indicate the LVAT95 at baseline and the best VVD on average, respectively, while the error bars indicate ± the standard deviation. CV, conduction velocity; LBBB, left bundle branch block; LBBP, left bundle branch pacing; LOT-CRT, left bundle branch pacing–optimized cardiac resynchronization therapy; VVD, ventriculo-ventricular delay.