Intracardiac electrogram analysis may allow for prediction of lesion transmurality after pulsed field ablation of atria in a porcine model

Figures

• 

  Figure 1. Raw unprocessed signals at 4 different time points relative to the energy delivery. Electrocardiography, bipolar intracardiac electrogram (iEGM) (bandpass 30–500 Hz) and unipolar iEGM (bandpass 0.05–500 Hz) signals are shown in the upper, middle, and bottom rows, respectively. A: Pulsed field ablation (PFA) lesion (1500 V, 4 pulse trains). B: Radiofrequency ablation (RFA) lesion (50 W at 60 °C for 10 seconds).

• 

  Figure 2. Comparison of changes in bipolar intracardiac electrograms (iEGMs) (A) and unipolar iEGMs (B) between pulsed field ablation (PFA) and radiofrequency ablation (RFA). The ablation energy was delivered at time 0. Absolute peak-to-peak (PP) values are shown (median and interquartile range for 5 lesion sites per treatment modality). Before averaging across the lesion sites, the individual values for each lesion for all time points (placed 10 seconds apart) were determined as the median values of PP values for all heartbeats within each 10-second interval centered at the chosen time point for the preablation period (the first 2 time points) and for the period of 30 minutes postablation.

• 

  Figure 3. Isolation of the high-frequency (HF) and low-frequency (LF) content from bipolar and unipolar intracardiac electrograms (iEGMs), respectively, following the discrete wavelet transform (DWT) decomposition of iEGM at 4 different time points relative to the energy delivery. A: Pulsed field ablation (PFA) lesion; B: radiofrequency ablation (RFA) lesion. HF content: 63–500 Hz of bipolar iEGM. LF content: 1–16 Hz of unipolar iEGM. The same signals as in Figure 1 are shown. ECG = electrocardiography.

• 

  Figure 4. Comparison of changes in high-frequency (HF) content (bandpass: 63–500 Hz) of bipolar intracardiac electrograms (iEGMs) (A, C) and low-frequency (LF) content (bandpass: 1–16 Hz) of unipolar iEGMs (B, D) between pulsed field ablation (PFA) and radiofrequency ablation (RFA). The ablation energy was delivered at time 0. Absolute peak-to-peak (PP) values are shown (median and interquartile range) in panels A and B, representing the median values for the heartbeats within 10-second intervals for the preablation period (the first 2 data points) and for a period of 10 minutes after ablation. C: Data from panel A normalized to the first preablation value for each individual lesion before averaging. D: Data from panel B normalized to the first valid postablation value (30 seconds postablation) for each individual lesion before averaging. ∗P < .001, linear mixed-effect models—analysis of variance (time-by-group interaction) for the difference between the slopes on the interval between 30 seconds and 3 minutes postablation.

• 

  Figure 5. Comparison of changes in high-frequency content (bandpass: 63–500 Hz) of bipolar intracardiac electrograms with respect to the lesion status. The ablation energy was delivered at time 0. A: Pulsed field ablation (PFA); B: radiofrequency ablation (RFA). Individual absolute peak-to-peak values were normalized to the last preablation value before averaging. Medians with interquartile ranges are shown, representing the median values for the heartbeats within 10-second intervals for the preablation period (the first 2 data points) and for a period of 10 minutes after ablation. ∗P < .001, linear mixed-effect models—analysis of variance (time-by-group interaction) for the difference between the slopes on the interval between 30 seconds and 3 minutes postablation. ∗∗P < .05, Mann-Whitney test. T+ = transmural lesions, T– = nontransmural lesions.

• 

  Figure 6. Comparison of changes in low-frequency content (bandpass: 1–16 Hz) of unipolar intracardiac electrograms with respect to the lesion status. The ablation energy was delivered at time 0. A: Pulsed field ablation (PFA); B: radiofrequency ablation (RFA). Individual absolute peak-to-peak values were normalized to the first valid postablation value (30 seconds postablation). Medians with interquartile ranges are shown, representing the median values for the heartbeats within 10-second intervals for the preablation period (the first 2 data points) and for a period of 10 minutes after ablation. ∗P < .001, linear mixed-effect models—analysis of variance (time-by-group interaction) for the difference between the slopes on the interval between 30 seconds and 3 minutes postablation. ∗∗P < .05, t test. T+ = transmural lesions, T– = nontransmural lesions.