Woelber, Tiffany J.

Introduction: Accurate pre-procedure localization of the premature ventricular contractions (PVCs) is critical in preparing for the ablation. So far, no tools exist for artificial intelligence guided PVC-localization. Hypothesis: Trained deep learning model will be able to predict the site of origin (SOO) of PVCs using 12 lead electrocardiogram (ECG). Methods: We analyzed 707 ECG-ablation pairs from 2403 patients who had at least one ECG flagged for PVCs in the 6 months prior to their procedure. First, we used a cohort of 100,000 ECGs without PVCs and 10,000 ECGs with PVCs to train a convolutional neural network (CNN) to isolate the ECG segments with PVCs in them from a 12-lead ECG (Figure 1). We then trained two models, one utilizing a 10-second 12-lead ECG, and the other with just the isolated PVC beats as input. The models were trained using the Keras framework with TensorFlow (Google). Results: The mean age of the patients in the cohort was 61.2 ± 15.0 years. 66.3 percent of patients underwent left ventricle ablation, 54.6 percent had outflow ablation, and 55.3 percent had ablation of sites with leftward axis (septal location for midline and left ventricle PVCs). The model trained on 10-second ECGs yielded validation and test AUCs of 0.941 and 0.931 for outflow, 0.898 and 0.891 for ventricle, and 0.863 and 0.846 for leftward axis (Figure 1A). The model trained on isolated segments containing PVC beats yielded validation and test AUCs of 0.966 and 0.957 for outflow, 0.879 and 0.907 for ventricle, and 0.846 and 0.829 for SOO for leftward axis (Figure 1B). Conclusions: We have trained an accurate convolutional neural network that can predict SOO using the surface ECG. Further fine-tuning would be needed to refine the model and develop prediction of specific sites.

Abstract Background Catheter ablation of ventricular tachycardia (VT) may be performed under general anesthesia (GA) or monitored anesthesia care (MAC). There is a paucity of data studying these two sedation strategies with respect to intraprocedural characteristics, and outcomes of VT ablation. Purpose The purpose of this study was to assess the effects of anesthesia techniques on VT inducibility and outcomes. Methods A single-center, retrospective study of patients undergoing catheter ablation for scar-related VT, from 2016-2022, was conducted. Data collected included demographics, comorbidities, medications, relevant laboratory abnormalities, electrocardiograms, echocardiograms, detailed procedural characteristics, and outcomes. Results One hundred sixty-eight patients who underwent a total of 200 VT ablation procedures were included. The patients undergoing GA had a greater burden of comorbidities including congestive heart failure (77.2% vs. 60.9%), hypertension (66.9% vs. 60.0%), atrial arrhythmias (50.7% vs. 43.8%), diabetes mellitus (21.3% vs. 9.4%), chronic kidney disease (44.1% vs. 31.3%), prior stroke (11.8% vs. 10.9%), myocardial infarction (31.6% vs. 29.7%), pre-existing implantable cardioverter-defibrillator (86.8% vs. 71.9%) and had a higher median number of antiarrhythmic drugs at baseline (2 vs. 1) [Figure 1]. There was a greater percentage of patients with non-ischemic cardiomyopathy (58.1% vs. 51.6%) in the GA cohort. The mean left ventricular ejection fraction was not different among the two groups (39.3 ± 13.9 vs. 42.5 ± 15.3). The GA group had more hemodynamic instability at baseline (p < 0.01), and during VT induction (p < 0.01). There was no significant difference in the number of VT induced, or tachycardia cycle length, in each group. Despite higher radiofrequency energy delivery time in the GA group as compared to the MAC group, the GA group had higher VT inducibility post-procedure and were more likely to be discharged on antiarrhythmic drugs. Even though the recurrence of VT was higher in the GA group compared to the MAC group (41.9% vs. 31.3%), the VT-free survival curves were not significantly different by Kaplan-Meier analysis (p=0.19) [Figure 2]. In a multivariate regression analysis, the sedation strategy was not an independent predictor of VT recurrence (p=NS). However, baseline hemodynamic instability was an independent predictor of VT recurrence after accounting for all intraprocedural characteristics (p=0.008). Conclusion Patients undergoing VT ablation under GA have more comorbidities and tend to be on more antiarrhythmic drugs. The GA group had more hemodynamic instability during the procedure and had greater overall VT recurrence, however, not significant in a time-based analysis. The greater overall recurrence in the GA group could be because they represent a sicker population, or the hemodynamics during the procedure preclude adequate ablation of the VT circuit elements.Figure 1Figure 2

Abstract Background Catheter ablation has emerged as the first-line therapy for management of ventricular tachycardia (VT) over the last decade. There is a reluctance for older patients to be referred for catheter ablation with concerns for complications and poor outcomes. Purpose The purpose of the study was to assess the safety and outcomes of catheter ablation for VT in patients more than 74 years of age in comparison to younger patients. Methods A single-center retrospective study of 200 procedures in 168 patients undergoing catheter ablation for scar-related VT was conducted. Data collected included demographics, comorbidities, medications, relevant laboratory abnormalities, electrocardiograms, Holter, device electrograms, structural data including echocardiograms and cardiac magnetic resonance imaging, detailed procedural characteristics, and outcomes. Results The mean age in the older cohort was 77.7 ± 3.7 years compared to 57.6 ± 13.1 years in the younger group (p < 0.01) [Figure 1]. The geriatric patients had more males (93.6% vs. 75.8%) and had greater proportion of comorbidities including congestive heart failure (89.4% vs. 66.7%), hypertension (91.5% & 56.9%), atrial arrhythmias (66.0% vs. 43.1%), chronic kidney disease (68.1% vs. 31.4%), prior myocardial infarction (53.2% vs. 24.2%), and stroke (21.3% vs. 8.5%). The older patients were on more antiarrhythmic drugs at baseline. During the procedure, the number of cases with VT induction and hemodynamic stability was similar in older and younger groups. In cases where VT was induced, the older patients had a greater number of VTs induced (2.7 ± 3.0 vs. 1.9 ± 1.8), with slower cycle lengths (454.9 ± 122.6 vs. 379.3 ± 108.1 milliseconds). There was no difference noted in the procedural complications between the two groups. Older patients were more likely to be discharged on antiarrhythmic medications (87.2% vs. 71.2%) and had higher long-term mortality (29.8% vs. 7.8%). The older group was less likely to have a repeat ablation procedure (8.5% vs 24.2%). There was no difference in long-term VT-free survival rates among the two groups [Figure 2]. Conclusion Patients more than 74 years of age undergoing catheter ablation for structural VT have a greater comorbidity burden but demonstrate similar procedural complications and VT-free survival rates, when compared to the younger cohort.Figure 1Figure 2

Abstract Background Premature ventricular contractions (PVCs) are the most common arrhythmia affecting the ventricle. Frequent PVCs are associated with the development of PVC-cardiomyopathy (CM). It is known that the PVC burden has a direct correlation to the severity of left ventricular (LV) dysfunction. However, only 5% of patients with frequent PVCs (>10,000 per day) have associated PVC-CM. Thus, selecting patients for therapy options with drugs or catheter ablation is challenging, especially, when patients are asymptomatic with the PVCs. Purpose To train a deep learning model to predict the probability of low LV ejection fraction (LVEF) in patients with PVCs on a 12-lead electrocardiogram (ECG). Methods Using 69,440 paired 12-lead ECGs (containing at least one PVC), and echocardiogram data, including the left ventricular ejection fraction at Mayo Clinic, we trained a convolutional neural network to identify patients with ventricular dysfunction, defined as ejection fraction ≤35%, using the ECG data alone. ECGs were acquired at a sampling rate of 500 Hz using a Marquette ECG machine (GE Healthcare) and stored using the MUSE data management system for later retrieval. Comprehensive 2-D or 3-D and Doppler echocardiography were available for all patients. We trained a convolutional neural network using the Keras framework with TensorFlow (Google). The ECG-echocardiogram pairs were divided into a 60:20:20 training:validation:test split. Results The mean age of the patients used to train the model was 69.9 ± 13.0 years, and 65.2 percent of patients were male. The mean LVEF for this population was 52 ± 14.9 %, and the proportion of patients with low LVEF was 16.8 percent. The model yielded values for area under the curve, sensitivity, specificity, and accuracy of 0.897, 0.830, 0.813, and 0.816, respectively (Figure 1). A representative example of model prediction of normal LVEF and low LVEF is shown in Figure 2. Conclusion Using large-scale data, we have trained a highly accurate convolutional neural network with excellent predictive accuracy for LV systolic dysfunction using the surface ECG. Such a tool would be invaluable for cardiologists when risk-stratifying patients with frequent PVCs, especially those that are asymptomatic. Patients at risk will benefit from early referral to electrophysiologists for drug therapy and catheter ablation to mitigate the risk of PVC-CM.Figure 1Figure 2