Artificial intelligence for ventricular arrhythmia capability using ambulatory electrocardiograms.

Aims: European and American clinical guidelines for implantable cardioverter defibrillators are insufficiently accurate for ventricular arrhythmia (VA) risk stratification, leading to significant morbidity and mortality. Artificial intelligence offers a novel risk stratification lens through which VA capability can be determined from the electrocardiogram (ECG) in normal cardiac rhythm. The aim of this study was to develop and test a deep neural network for VA risk stratification using routinely collected ambulatory ECGs. Methods and results: A multicentre case-control study was undertaken to assess VA-ResNet-50, our open source ResNet-50-based deep neural network. VA-ResNet-50 was designed to read pyramid samples of three-lead 24 h ambulatory ECGs to decide whether a heart is capable of VA based on the ECG alone. Consecutive adults with VA from East Midlands, UK, who had ambulatory ECGs as part of their NHS care between 2014 and 2022 were recruited and compared with all comer ambulatory electrograms without VA. Of 270 patients, 159 heterogeneous patients had a composite VA outcome. The mean time difference between the ECG and VA was 1.6 years (⅓ ambulatory ECG before VA). The deep neural network was able to classify ECGs for VA capability with an accuracy of 0.76 (95% confidence interval 0.66-0.87), F1 score of 0.79 (0.67-0.90), area under the receiver operator curve of 0.8 (0.67-0.91), and relative risk of 2.87 (1.41-5.81). Conclusion: Ambulatory ECGs confer risk signals for VA risk stratification when analysed using VA-ResNet-50. Pyramid sampling from the ambulatory ECGs is hypothesized to capture autonomic activity. We encourage groups to build on this open-source model. Question: Can artificial intelligence (AI) be used to predict whether a person is at risk of a lethal heart rhythm, based solely on an electrocardiogram (an electrical heart tracing)? Findings: In a study of 270 adults (of which 159 had lethal arrhythmias), the AI was correct in 4 out of every 5 cases. If the AI said a person was at risk, the risk of lethal event was three times higher than normal adults. Meaning: In this study, the AI performed better than current medical guidelines. The AI was able to accurately determine the risk of lethal arrhythmia from standard heart tracings for 80% of cases over a year away-a conceptual shift in what an AI model can see and predict. This method shows promise in better allocating implantable shock box pacemakers (implantable cardioverter defibrillators) that save lives.

Distal His bundle pacing in a patient with surgically corrected complex Ebstein anomaly and symptomatic second-degree atrioventricular block: a case report.

Background: Ebstein's anomaly occurs when there is an apical displacement of the tricuspid valve with septal and posterior valve leaflets tethering. This condition often occurs in association with other congenital, structural, or conduction system diseases, including intracardiac shunts, valvular lesions, arrhythmias, accessory conduction pathways, and first-degree atrioventricular (AV) block. We present for the first time a case of a patient with Ebstein's anomaly who presented with second-degree Mobitz II AV block and was successfully treated with conduction system pacing (CSP) due to her young age and the likelihood of a long-term high percentage of pacing. Case summary: We present a case of a 42-year-old lady with a background of complex congenital heart disease, including severe pulmonary stenosis, Ebstein anomaly, and atrial septal defect (ASD). She required complex surgical intervention, including tricuspid valve (TV) repair and subsequently replacement, ASD closure, and pulmonary balloon valvuloplasty. She presented to our hospital with symptomatic second-degree Mobitz II AV block (dizziness, shortness of breath, and exercise intolerance) and right bundle branch block (RBBB) on her baseline ECG. Her echocardiogram showed dilated right ventricle (RV) and left ventricle (LV) with low normal LV systolic function. Due to her young age and the likelihood of a long-term high percentage of RV pacing, we opted for CSP after a detailed discussion and patient consent. The distal HIS position is the preferred pacing strategy at our centre. We could not cross the TV with the standard Medtronic C315 HIS catheter, so we had to use the deflectable C304 HIS catheter. Mapping and pacing of the distal HIS bundle were achieved by Medtronic Selectsecure 3830, 69 cm lead. HIS bundle pacing led to the correction of both second-degree Mobitz II AV block and pre-existing RBBB. The implantation was uneventful, and the patient was discharged home the next day without any acute complications. Discussion: Distal HIS pacing is feasible in patients with surgically treated complex Ebstein anomaly and heart block. This approach can normalize the QRS complex with a high probability of preserving or improving LV function.

Digitally enabled acute care for atrial fibrillation: conception, feasibility and early outcomes of an AF virtual ward.

BACKGROUND: Atrial fibrillation (AF) represents a growing healthcare challenge, mainly driven by acute hospitalisations. Virtual wards could be the way forward to manage acute AF patients through remote monitoring, especially with the rise in global access to digital telecommunication and the growing acceptance of telemedicine post-COVID-19. METHODS: An AF virtual ward was implemented as a proof-of-concept care model. Patients presenting acutely with AF or atrial flutter and rapid ventricular response to the hospital were onboarded to the virtual ward and managed at home through remote ECG-monitoring and 'virtual' ward rounds, after being given access to a single-lead ECG device, a blood pressure monitor and pulse oximeter with instructions to record daily ECGs, blood pressure, oxygen saturations and to complete an online AF symptom questionnaire. Data were uploaded to a digital platform for daily review by the clinical team. Primary outcomes included admission avoidance, readmission avoidance and patient satisfaction. Safety outcomes included unplanned discharge from the virtual ward, cardiovascular mortality and all-cause mortality. RESULTS: There were 50 admissions to the virtual ward between January and August 2022. Twenty-four of them avoided initial hospital admission as patients were directly enrolled to the virtual ward from outpatient settings. A further 25 readmissions were appropriately prevented during virtual surveillance. Patient satisfaction questionnaires yielded 100% positive responses among participants. There were three unplanned discharges from the virtual ward requiring hospitalisation. Mean heart rate on admission to the virtual ward and discharge was 122±26 and 82±27 bpm respectively. A rhythm control strategy was pursued in 82% (n=41) and 20% (n=10) required 3 or more remote pharmacological interventions. CONCLUSION: This is a first real-world experience of an AF virtual ward that heralds a potential means for reducing AF hospitalisations and the associated financial burden, without compromising on patients' care or safety.

Implantable cardioverter defibrillator for primary prevention in patients with non-ischemic cardiomyopathy in the era of novel therapeutic agents- meta-analysis.

Background: Evidence regarding the mortality benefit of implantable cardioverter defibrillator (ICD) non-ischemic dilated cardiomyopathy (NIDCM) is inconsistent. The most recent randomized study, the DANISH trial, did not find improved outcomes with ICD. However, based on previous studies and meta-analyses, current guidelines still highly recommend ICD implantation in NIDCM patients. The introduction of novel medications for heart failure improved the clinical outcome dramatically. We aimed in this study to evaluate the effect of Angiotensin Receptor-Neprilysin Inhibitors (ARNi) and sodium-glucose transport protein 2 inhibitors (SGLT2i) on the mortality benefit of ICD in NIDCM. Methods: We used a previous metanalysis algorithm and added an updated comprehensive literature search in PubMed for randomized control trials that examined the mortality benefit of ICD in NIDCM vs. optimal medical treatment. The primary outcome included death from any cause. We did a meta-regression analysis to search for a single independent factor affecting mortality. Using previous data, we evaluated the theoretical effect of ICD implementation on patients treated with SGLT2 inhibitors and ARNi. Results: No new articles were added to the results of the previous meta-analysis. 2,622 patients with NIDCM from 5 cohort studies published between 2002 and 2016 were included in the analysis. 50% of them underwent ICD implantation for primary prevention of sudden cardiac death, and 50% did not. ICD was associated with a significantly decreased risk for death from any cause compared to control (OR = 0.79, 95%CI: 0.66-0.95, p = 0.01, I2 = 0%). The theoretical addition of ARNi and the SGLT2 inhibitor dapagliflozin did not change the significant mortality effect of ICD (OR = 0.82, 95%CI: 0.7-0.9, p = 0.001, I2 = 0%) and (OR = 0.82, 95%CI: 0.7-0.9, p = 0.001, I2 = 0%). A meta-regression revealed no association between death from any cause and left bundle branch block (LBBB), use of amiodarone, use of angiotensin-converting enzyme inhibitors (ACEi) or angiotensin receptor blockers, year initiated enrollment, and the year ended enrollment (R2 = 0.0). Conclusion: In patients with NIDCM, the addition of ARNi and SGLT2i did not affect the mortality advantages of ICD for primary prevention. PROSPERO registry number: https://www.crd.york.ac.uk/prospero/, identifier: CRD42023403210.

Dialysis Therapy for Volume Overload: A Feasible Option to Reduce Heart Failure Hospitalizations in Advanced Heart Failure.

BACKGROUND: Advanced heart failure (HF) carries a high rate of recurrent HF hospitalizations and a very high mortality rate. Mechanical devices and heart transplantation are limited to a select few. Dialysis may be a good alternative for advanced HF patients with volume overload despite maximal pharmacological therapy. OBJECTIVES: To assess the net clinical outcome of peritoneal dialysis or hemodialysis in patients with advanced HF. METHODS: We analyzed all advanced HF patients who were referred for dialysis due to volume overload in our institution. Patients were followed for complications, HF hospitalizations, and survival. RESULTS: We assessed 35 patients; 10 (29%) underwent peritoneal dialysis and 25 (71%) underwent hemodialysis; 71% were male; median (interquartile range) age was 74 (67-78) years. Estimated glomerular filtration rate was 20 (13-32) ml/min per 1.73 m2. New York Heart Association functional capacity was III. Median follow-up time was 719 days (interquartile range 658-780). One-year mortality rate was 8/35 (23%) and overall mortality rate was 16/35 (46%). Three patients (9%) died during the first year due to line or peritoneal dialysis related sepsis, and 6 (17%) died during the entire follow-up. The median number of HF hospitalizations was significantly reduced during the year on dialysis compared to the year prior to dialysis (0.0 [0.0-1.0] vs. 2.0 [0.0-3.0], P < 0.001). CONCLUSIONS: Dialysis is reasonably safe and significantly reduced HF hospitalization in advanced HF patients. Dialysis could be a good alternative for advanced HF patients with intractable volume overload.