Asirvatham, S J

Abstract Background The right ventricular moderator band-papillary muscle (RV MB-PM) complex has a rich subendocardial network of purkinje fibers which can lead to ventricular arrhythmias (VA)1. Catheter ablation of such arrhythmias is procedurally challenging and there remains limited studies on long-term procedural success. Purpose We aimed to systematically study RV MB-PM complex VA. Methods A single center retrospective study from January 2013 to December 2023 identified 3249 patients who underwent VA ablation. The RV MB-PM complex was targeted in 47 patients. Acute success was defined as suppression of the targeted VA >=30 minutes post ablation. Long-term success was defined as the absence of symptoms, total PVC burden <5% or improvement in EF in patients presenting with isolated PVCs and as freedom from appropriate ICD therapy in patients presenting with malignant PVCs inducing VF, sustained VT or VT storm. Results Mean age was 51.9 ± 16.7 years and 64% were male. 40 (85%) patients had VA targeted from the MB and 7 (15%) from the PM complex (2 anterior, 2 inferior and 3 septal PM). 34 (72%) had no underlying structural heart disease, 3 (6%) ischemic cardiomyopathy and 10 (21%) non-ischemic cardiomyopathy. No patient had delayed gadolinium enhancement in the RV MB-PM complex. VA were inducible in 42 (89%) of patients and activation mapping was used in 9 (19%), pace mapping in 5 (11%) and a combination of activation and pace mapping used in 33 (70%) patients. Intracardiac echocardiography was used in all cases. Pre-QRS duration was 34.0 ± 10.4 ms. Radiofrequency ablation was used in 37 (78.7%) cases, cryotherapy in 4 (8.51%) cases and a combination of the 2 energy sources in 6 (12.8%) cases. In all cases, acute peri-procedural success was achieved. 24 (51.1%) patients presented with isolated PVCs and were followed up for 4.72 ± 3.55 years after their index procedure. 3 patients underwent repeat catheter ablation targeting the same PVC focus but all patients had long term success with a PVC burden of 16.1 ± 12.5% prior to ablation and 2.40 ± 2.44% at last follow-up. 16 patients underwent ablation for recurrent ICD shocks after initially presenting with PVC induced VF. Patients were followed up for 3.22 ± 2.09 years and 14 (88%) patients remained free from ICD therapy. 7 patients had MB ablation for sustained VT and were followed up for 4.06 ± 3.77 years. 5 (71%) patients remained free from ICD therapy. A purkinje potential was seen in 8% (2/24) cases of isolated PVCs and in 39% (9/23) cases of PVC induced VF or VT (p=0.13). Conclusions RV MB-PM VA commonly present with malignant VF and a purkinje potential is more likely to be seen in these cases suggesting different VA mechanisms to isolated PVCs and catheter ablation is an effective long-term strategy.

Abstract Background The determination of left ventricular diastolic dysfunction (LVDD) and filling pressure in patients with significant (>moderate) mitral regurgitation (MR) poses a complex challenge. We recently validated an artificial intelligence-enabled electrocardiogram (AI-ECG) algorithm to detect LVDD and estimate LV filling pressures. Methods This retrospective study sought to examine the risk of all-cause mortality across AI-ECG LVDD grades. Patients from the AI-ECG LVDD study test cohort who underwent transthoracic echocardiography confirming significant MR and an ECG within fourteen days of each other at Mayo Clinic between 09/2001 and 06/2022 were included. LVDD status was determined based on the index ECG, and patients were categorized into groups representing normal diastolic function (NDF) or LVDD grades 1 (G1), 2 (G2), or 3 (G3). Results Of 4,019 patients with significant MR (mean age 69.8, 49.0% women), 1,175 (29.2%), 1,881 (46.8%), and 963 (24.0%) were classified by AI-ECG as NDF/G1-LVDD, G2-LVDD, and G3-LVDD, respectively. The median mitral effective regurgitant orifice area was 26 mm² (interquartile range 20-36 mm²). Over a median follow-up of 3.5 years, 1,636 (40.7%) patients died. In multivariable survival analysis adjusted for multiple risk factors, a worse AI-ECG LVDD grade was independently associated with an increased death risk [G2DD: adjusted hazard ratio (aHR) 1.98 (95% confidence interval (95% CI) 1.63-2.40), G3DD: aHR 2.59 (95% CI 2.09-3.20)]. These findings were consistent when accounting for mitral valve intervention and among patients with >moderate MR or after mitral valve intervention. Conclusion In patients with significant MR, the grading of LVDD by AI-ECG is independently associated with all-cause mortality.

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