Machine Learning for Prediction of Ventricular Arrhythmia Episodes from Intracardiac Electrograms of Automatic Implantable Cardioverter-Defibrillators.

BACKGROUND: Despite the implantable cardioverter defibrillator's (ICD) effectiveness in saving patients with life-threatening ventricular arrhythmias (VAs), the temporal occurrence of VA following ICD implantation is unpredictable. OBJECTIVE: Apply machine learning (ML) to intracardiac electrograms (IEGMs) recorded by ICDs as a unique biomarker for predicting impending VAs. METHODS: The study included 13,516 patients who received BIOTRONIK ICDs and enrolled in the CERTITUDE registry between 01/01/2010 to 12/31/2020. Database extraction included IEGMs from standard quarterly transmissions and VA event episodes. The processed IEGM data were pulled from device transmissions stored in a centralized Home Monitoring Service Center and reformatted into an analyzable format. Long- (baseline or first scheduled remote recording), mid-(scheduled remote recording every 90 days), or short-range predictions (IEGM within 5 seconds before the VA onset) were used to determine whether ML-processed IEGMs predicted impending VA events. Convolutional neural network classifiers using ResNet architecture were employed. RESULTS: Of 13,516 patients (male 72%, age 67.5 ± 11.9 years), 301,647 IEGM recordings were collected; 27,845 episodes of sustained VT/VF were observed in 4,467 patients (33.0%). Neural networks based on CNN using ResNet-like architectures on far-field IEGMs yielded an AUC of 0.83 with a 95% confidence interval of [0.79, 0.87] in the short-term, while the long- and mid-range analyses had minimal predictive value for VA events. CONCLUSION: In this study, applying ML to ICD-acquired IEGMs predicted impending VT/VF events seconds before they occurred, whereas mid- to long-term predictions were not successful. This could have important implications for future device therapies.

Cardiac Resynchronization Therapy for Patients with Mild to Moderately Reduced Ejection Fraction and Left Bundle Branch Block.

BACKGROUND: It is unknown if CRT would improve or halt the progression of heart failure (HF) in patients with mild-moderately reduced LVEF (HFmmrEF) and LBBB. OBJECTIVE: To investigate the outcomes of CRT in patients with HFmmrEF and LV conduction delay. METHODS: A prospective, randomized clinical trial sponsored by NHLBI included 76 patients who met the study inclusion criteria (LVEF 36-50% and LBBB). Patients received CRTP and were randomized to CRT-OFF (RV pacing 40bpm) or CRT-ON (BIV pacing 60-150bpm). At a 6-month follow-up, pacing programming was changed to the opposite settings. NYHA class, NT pro-BNP levels, and echocardiography were collected at baseline, 6-month, and 12-month. The primary study endpoint was the LV end-systolic volume (LVESV) change from baseline, and the primary randomized comparison was the comparison of 6-month to 12-month changes between randomized groups. RESULTS: The mean patient age was 68.4±9.8 years (male 71%). Baseline characteristics were similar between 2 randomized groups (all P>0.05). In patients randomized to CRT-OFF first, then ON, LVESV was reduced from baseline only after CRT-ON (baseline 116.1±36.5 mL, CRT-ON 87.6±26.0 mL, P<0.0001). The randomized analysis of LVEF showed a significantly better change from 6 to 12 months in the "OFF-ON" group, P=0.003. LVEF was improved by CRT (baseline 41.3±4.7%, CRT-ON 46.0±8.0%, P=0.002). In patients randomized to CRT-ON first, then OFF, LVESV was reduced both after CRT-ON and CRT-OFF (baseline 109.8±23.5 mL, CRT-ON 91.7±30.5 mL, P<0.0001; CRT-OFF 99.3±28.9 mL, P=0.012). However, the LVESV reduction effect became smaller between CRT-ON and OFF (P=0.027). LVEF improved both after CRT-ON and CRT-OFF (baseline 42.7±4.3%, CRT-ON 48.5±8.6%, P<0.001; CRT-OFF 45.9±7.7%, P=0.025). CONCLUSION: CRT for patients with HFmmrEF significantly improves LVEF and ventricular remodeling after 6 months of CRT. The study provides novel evidence that early CRT benefits patients with HFmmrEF with LBBB.

Unlocking the potential of artificial intelligence in electrocardiogram biometrics: age-related changes, anomaly detection, and data authenticity in mobile health platforms.

Aims: Mobile devices such as smartphones and watches can now record single-lead electrocardiograms (ECGs), making wearables a potential screening tool for cardiac and wellness monitoring outside of healthcare settings. Because friends and family often share their smart phones and devices, confirmation that a sample is from a given patient is important before it is added to the electronic health record. Methods and results: We sought to determine whether the application of Siamese neural network would permit the diagnostic ECG sample to serve as both a medical test and biometric identifier. When using similarity scores to discriminate whether a pair of ECGs came from the same patient or different patients, inputs of single-lead and 12-lead medians produced an area under the curve of 0.94 and 0.97, respectively. Conclusion: The similar performance of the single-lead and 12-lead configurations underscores the potential use of mobile devices to monitor cardiac health.

Mitral valve prolapse: arrhythmic risk during pregnancy and postpartum.

BACKGROUND AND AIMS: Arrhythmic mitral valve prolapse (AMVP) is linked to life-threatening ventricular arrhythmias (VAs), and young women are considered at high risk. Cases of AMVP in women with malignant VA during pregnancy have emerged, but the arrhythmic risk during pregnancy is unknown. The authors aimed to describe features of women with high-risk AMVP who developed malignant VA during the perinatal period and to assess if pregnancy and the postpartum period were associated with a higher risk of malignant VA. METHODS: This retrospective international multi-centre case series included high-risk women with AMVP who experienced malignant VA and at least one pregnancy. Malignant VA included ventricular fibrillation, sustained ventricular tachycardia, or appropriate shock from an implantable cardioverter defibrillator. The authors compared the incidence of malignant VA in non-pregnant periods and perinatal period; the latter defined as occurring during pregnancy and within 6 months after delivery. RESULTS: The authors included 18 women with AMVP from 11 centres. During 7.5 (interquartile range 5.8-16.6) years of follow-up, 37 malignant VAs occurred, of which 18 were pregnancy related occurring in 13 (72%) unique patients. Pregnancy and 6 months after delivery showed increased incidence rate of malignant VA compared to the non-pregnancy period (univariate incidence rate ratio 2.66, 95% confidence interval 1.23-5.76). CONCLUSIONS: The perinatal period could impose increased risk of malignant VA in women with high-risk AMVP. The data may provide general guidance for pre-conception counselling and for nuanced shared decision-making between patients and clinicians.

Artificial intelligence-enhanced electrocardiogram for arrhythmogenic right ventricular cardiomyopathy detection.

Aims: ECG abnormalities are often the first signs of arrhythmogenic right ventricular cardiomyopathy (ARVC) and we hypothesized that an artificial intelligence (AI)-enhanced ECG could help identify patients with ARVC and serve as a valuable disease-detection tool. Methods and results: We created a convolutional neural network to detect ARVC using a 12-lead ECG. All patients with ARVC who met the 2010 task force criteria and had disease-causative genetic variants were included. All case ECGs were randomly assigned in an 8:1:1 ratio into training, validation, and testing groups. The case ECGs were age- and sex-matched with control ECGs at our institution in a 1:100 ratio. Seventy-seven patients (51% male; mean age 47.2 ± 19.9), including 56 patients with PKP2, 7 with DSG2, 6 with DSC2, 6 with DSP, and 2 with JUP were included. The model was trained using 61 case ECGs and 5009 control ECGs; validated with 7 case ECGs and 678 control ECGs and tested in 22 case ECGs and 1256 control ECGs. The sensitivity, specificity, positive and negative predictive values of the model were 77.3, 62.9, 3.32, and 99.4%, respectively. The area under the curve for rhythm ECG and median beat ECG was 0.75 and 0.76, respectively. Conclusion: Our study found that the model performed well in excluding ARVC and supports the concept that the AI ECG can serve as a biomarker for ARVC if a larger cohort were available for network training. A multicentre study including patients with ARVC from other centres would be the next step in refining, testing, and validating this algorithm.

First-in-man report of transsubclavian venous implantation of the Aveir leadless cardiac pacing system.

INTRODUCTION: Transsubclavian venous implantation of the Aveir leadless cardiac pacemaker (LCP) has not been previously reported. METHODS AND RESULTS: Three cases of transsubclavian implantation of the Aveir LCP are reported. Two cases were postbilateral orthotopic lung transplant, without appropriate femoral or jugular access due to recent ECMO cannulation and jugular central venous catheters. In one case, there was strong patient preference for same-day discharge. Stability testing confirmed adequate fixation and electrical testing confirmed stable parameters in all cases. All patients tolerated the procedure well without significant immediate complications. CONCLUSIONS: We demonstrate the feasibility of transsubclavian implantation of the Aveir LCP.

Long-term outcomes with abandoning versus extracting sterile leads: A 10-year population-based study.

BACKGROUND: Long-term outcomes of sterile lead management strategies of lead abandonment (LA) or transvenous lead extraction (TLE) remain unclear. METHODS: We performed a retrospective study of a population residing in southeastern Minnesota with follow-up at the Mayo Clinic and its health systems. Patients who underwent LA or TLE of sterile leads from January 1, 2000, to January 1, 2011, and had follow-up for at least 10 years or until their death were included. RESULTS: A total of 172 patients were included in the study with 153 patients who underwent LA and 19 who underwent TLE for sterile leads. Indications for subsequent lead extraction arose in 9.1% (n = 14) of patients with initial LA and 5.3% (n = 1) in patients with initial TLE, after an average of 7 years. Moreover, 28.6% of patients in the LA cohort who required subsequent extraction did not proceed with the extraction, and among those who proceeded, 60% had clinical success and 40% had a clinical failure. Subsequent device upgrades or revisions were performed in 18.3% of patients in the LA group and 31.6% in the TLE group, with no significant differences in procedural challenges (5.2% vs. 5.3%). There was no difference in 10-year survival probability among the LA group and the TLE group (p = .64). CONCLUSION: An initial lead abandonment strategy was associated with more complicated subsequent extraction procedures compared to patients with an initial transvenous lead extraction strategy. However, there was no difference in 10-year survival probability between both lead management approaches.

Cardiac implantable electronic devices in female patients: Esthetic, breast implant, and anatomic considerations.

INTRODUCTION: The implantation of a cardiac implantable electronic device (CIED) can have esthetic and psychological consequences on patients. We explore a heart team model for care coordination and discuss esthetic approaches for improved cosmetic outcomes in patients undergoing (CIED)-related procedures or de novo implantation. METHODS: Patients undergoing CIED surgery for approved indications between June 2015 and June 2022 were identified. Patients were included when surgical care was provided by a collaborative relationship between the primary electrophysiologist and the plastic surgeon. Patient demographics, details of the surgical procedure, information on breast implants, complications, and outcomes related to cosmesis were recorded. RESULTS: Twenty-two female patients were included in this study. The mean age was 50.2 ± 18.2 years. The mean follow-up duration was 2.2 ± 5.5 months. The top two indications for the procedure included CIED generator change (n = 9, 41%) and implantable cardioverter-defibrillator (ICD) implantation (n = 7, 32%). The most common reasons for involving plastic surgery in the procedure included surgery near breast implants (n = 10, 45%) and device displacement or discomfort (n = 8, 36%). CIED pocket position was prepectoral in 10 cases (45%), subpectoral in 11 patients (50%), and intramuscular in one patient (4.5%). The majority of the patients (20, 91%) had cosmetically acceptable results postprocedure. One patient (4.5%) had breast asymmetry on the CIED side, and another continued to have skin erosion over the CIED and leads. CONCLUSION: A heart team approach incorporating the expertize of cardiac electrophysiology and plastic surgery is essential for providing optimal care for patients with breast implants and patients requesting esthetic appeal.

Artificial intelligence-enabled ECG for left ventricular diastolic function and filling pressure.

Assessment of left ventricular diastolic function plays a major role in the diagnosis and prognosis of cardiac diseases, including heart failure with preserved ejection fraction. We aimed to develop an artificial intelligence (AI)-enabled electrocardiogram (ECG) model to identify echocardiographically determined diastolic dysfunction and increased filling pressure. We trained, validated, and tested an AI-enabled ECG in 98,736, 21,963, and 98,763 patients, respectively, who had an ECG and echocardiographic diastolic function assessment within 14 days with no exclusion criteria. It was also tested in 55,248 patients with indeterminate diastolic function by echocardiography. The model was evaluated using the area under the curve (AUC) of the receiver operating characteristic curve, and its prognostic performance was compared to echocardiography. The AUC for detecting increased filling pressure was 0.911. The AUCs to identify diastolic dysfunction grades ≥1, ≥2, and 3 were 0.847, 0.911, and 0.943, respectively. During a median follow-up of 5.9 years, 20,223 (20.5%) died. Patients with increased filling pressure predicted by AI-ECG had higher mortality than those with normal filling pressure, after adjusting for age, sex, and comorbidities in the test group (hazard ratio (HR) 1.7, 95% CI 1.645-1.757) similar to echocardiography and in the indeterminate group (HR 1.34, 95% CI 1.298-1.383). An AI-enabled ECG identifies increased filling pressure and diastolic function grades with a good prognostic value similar to echocardiography. AI-ECG is a simple and promising tool to enhance the detection of diseases associated with diastolic dysfunction and increased diastolic filling pressure.

Ventricular Tachycardia Ablation in Patients With Desmoplakin Cardiomyopathy.

BACKGROUND: Desmoplakin (DSP) pathogenic variants are rare causes of arrhythmogenic cardiomyopathy and often involve the right and left ventricles. Ventricular tachycardia (VT) ablations may be required in these patients, but procedural characteristics have not been reported. OBJECTIVES: In this study, the authors sought to report a multicenter experience of VT ablation in patients with DSP pathogenic variants. METHODS: VT ablations performed in patients with known DSP pathogenic variants were analyzed across 6 centers in 3 countries. Patient characteristics and acute and long-term procedural outcomes were reported. RESULTS: A total of 20 patients (13 men, median age 43 years [Q1-Q3: 41.5-53.0 years], left ventricular ejection fraction 43.0% [Q1-Q3: 41.5%-53.0%], 11 previous failed ablations) were referred for VT ablation procedures. All patients had symptomatic VTs, with ICD therapy in 19 patients. Epicardial procedures were performed in 16 of the 20 patients. VT target sites were located in the right ventricular (RV) endocardium (n = 11), the RV epicardium (n = 4), the left ventricular (LV) endocardium (n = 2) and the LV epicardium (n = 7). In 3 patients, the VT target sites were in close proximity to coronary arteries, limiting ablation. During follow-up, VTs recurred in 11 patients, and repeated ablations were performed in 9 patients. Allowing for multiple procedures, 19 of the 20 patients remained free of VT recurrence after a median follow-up of 18 months [Q1-Q3: 5-60 months]. CONCLUSIONS: Patients with DSP cardiomyopathy often have biventricular involvement, and ablation procedures often require ablation in both ventricles and the epicardium. Recurrences are not uncommon, and the pathologic substrate can be located in close proximity to epicardial coronary arteries, limiting the success rate of ablations.

Incidence and Outcomes of New-Onset Right Bundle Branch Block Following Transcatheter Aortic Valve Replacement.

BACKGROUND: The incidence and prognosis of right bundle branch block (RBBB) following transcatheter aortic valve replacement (TAVR) are unknown. Hence, we sought to characterize the incidence of post-TAVR RBBB and determine associated risks of permanent pacemaker (PPM) implantation and mortality. METHODS: All patients 18 years and above without preexisting RBBB or PPM who underwent TAVR at US Mayo Clinic sites and Mayo Clinic Health Systems from June 2010 to May 2021 were evaluated. Post-TAVR RBBB was defined as new-onset RBBB in the postimplantation period. The risks of PPM implantation (within 90 days) and mortality following TAVR were compared for patients with and without post-TAVR RBBB using Kaplan-Meier analysis and Cox proportional hazards modeling. The risks of PPM implantation (within 90 days) and mortality following TAVR were compared for patients with and without post-TAVR RBBB using Kaplan-Meier analysis and Cox proportional hazards modeling. RESULTS: Of 1992 patients, 15 (0.75%) experienced new RBBB post-TAVR. There was a higher degree of valve oversizing among patients with new RBBB post-TAVR versus those without (17.9% versus 10.0%; P=0.034). Ten patients (66.7%) with post-TAVR RBBB experienced high-grade atrioventricular block and underwent PPM implantation (median 1 day; Q1, 0.2 and Q3, 4), compared with 268/1977 (13.6%) without RBBB. Following propensity score adjustment for covariates (age, sex, balloon-expandable valve, annulus diameter, and valve oversizing), post-TAVR RBBB was significantly associated with PPM implantation (hazard ratio, 8.36 [95% CI, 4.19-16.7]; P<0.001). No statistically significant increase in mortality was seen with post-TAVR RBBB (hazard ratio, 0.83 [95% CI, 0.33-2.11]; P=0.69), adjusting for age and sex. CONCLUSIONS: Although infrequent, post-TAVR RBBB was associated with elevated PPM implantation risk. The mechanisms for its development and its clinical prognosis require further study.

Anatomy of the Ventricular Outflow Tracts: An Electrophysiology Perspective.

Outflow tract ventricular arrhythmias are the most common type of idiopathic ventricular arrhythmia. A systematic understanding of the outflow tract anatomy improves procedural efficacy and enables electrophysiologists to anticipate and prevent complications. This review emphasizes the three-dimensional spatial relationships between the ventricular outflow tracts using seven anatomical principles. In turn, each principle is elaborated on from a clinical perspective relevant for the practicing electrophysiologist. The developmental anatomy of the outflow tracts is also discussed and reinforced with a clinical case.

Mayo Clinic VT calculator: A practical tool for accurate wide complex tachycardia differentiation.

The discrimination of ventricular tachycardia (VT) versus supraventricular wide complex tachycardia (SWCT) via 12-lead electrocardiogram (ECG) is crucial for achieving appropriate, high-quality, and cost-effective care in patients presenting with wide QRS complex tachycardia (WCT). Decades of rigorous research have brought forth an expanding arsenal of applicable manual algorithm methods for differentiating WCTs. However, these algorithms are limited by their heavy reliance on the ECG interpreter for their proper execution. Herein, we introduce the Mayo Clinic ventricular tachycardia calculator (MC-VTcalc) as a novel generalizable, accurate, and easy-to-use means to estimate VT probability independent of ECG interpreter competency. The MC-VTcalc, through the use of web-based and mobile device platforms, only requires the entry of computerized measurements (i.e., QRS duration, QRS axis, and T-wave axis) that are routinely displayed on standard 12-lead ECG recordings.

Targeted ablation of epicardial ganglionated plexi during cardiac surgery with pulsed field electroporation (NEURAL AF).

BACKGROUND: Modulation of the cardiac autonomic nervous system (ANS) is a promising adjuvant therapy in the treatment of atrial fibrillation (AF). In pre-clinical models, pulsed field (PF) energy has the advantage of selectively ablating the epicardial ganglionated plexi (GP) that govern the ANS. This study aims to demonstrate the feasibility and safety of epicardial ablation of the GPs with PF during cardiac surgery with a primary efficacy outcome of prolongation of the atrial effective refractory period (AERP). METHODS: In a single-arm, prospective analysis, patients with or without a history of AF underwent epicardial GP ablation with PF during coronary artery bypass grafting (CABG). AERP was determined immediately pre- and post- GP ablation to assess cardiac ANS function. Holter monitors were performed to determine rhythm status and heart rate variability (HRV) at baseline and at 1-month post-procedure. RESULTS: Of 24 patients, 23 (96%) received the full ablation protocol. No device-related adverse effects were noted. GP ablation resulted in a 20.7 ± 19.9% extension in AERP (P < 0.001). Post-operative AF was observed in 7 (29%) patients. Holter monitoring demonstrated an increase in mean heart rate (74.0 ± 8.7 vs. 80.6 ± 12.3, P = 0.01). There were no significant changes in HRV. There were no study-related complications. CONCLUSIONS: This study demonstrates the safety and feasibility of epicardial ablation of the GP using PF to modulate the ANS during cardiac surgery. Large, randomized analyses are necessary to determine whether epicardial PF ablation can offer a meaningful impact on the cardiac ANS and reduce AF. TRIAL REGISTRATION: Clinical trial registration: NCT04775264.

Catheter Ablation of Atrial Fibrillation in Adult Congenital Heart Disease: Procedural Characteristics and Outcomes.

BACKGROUND: The outcomes of catheter ablation for atrial fibrillation in adults with congenital heart disease are not well described. METHODS: In a retrospective study of adult patients with congenital heart disease who underwent catheter ablation for atrial fibrillation between 2000 and 2020 at Mayo Clinic, procedural characteristics and outcomes were collected. The primary outcomes were atrial arrhythmia (AA) recurrence following a 3-month blanking period and repeat ablation. An arrhythmia clinical severity score was assessed pre- and post-ablation based on the duration of arrhythmia episodes, symptoms, cardioversion frequency, and antiarrhythmic drug use. RESULTS: One hundred forty-five patients (age, 57±12 years; 28% female; 63% paroxysmal atrial fibrillation) underwent 198 ablations with a median follow-up of 26 months (interquartile range, 14-69). One hundred ten, 26, and 9 patients had simple, moderate, and complex congenital heart disease, respectively. All patients underwent pulmonary vein isolation, and non-pulmonary vein targets were ablated in 79 (54%). AA recurrence at 12 months was 37% (95% CI, 29%-45%). On univariate analysis, increasing left atrial volume index was associated with higher odds of AA recurrence (odds ratio, 1.03 [1.00-1.06] per 1 mL/m2 increment; P=0.05). Noninducibility of atrial flutter was predictive of decreased odds of AA recurrence (odds ratio, 0.43 [0.21-0.90]; P=0.03). A second ablation was performed in 43 patients after a median of 20 (interquartile range, 8-37) months. Arrhythmia clinical severity scores improved following ablation, reflecting a decrease in symptoms, cardioversions, and antiarrhythmic drugs. CONCLUSIONS: Catheter ablation of atrial fibrillation is feasible and effective in patients with adult congenital heart disease and reduces symptoms. Recurrence of AA frequently requires repeat ablation.