Electrocardiogram with Artificial Intelligence Assessment as a Predictor of Cardiac Events and Overall Survival in Patients Receiving Radiotherapy for Esophageal Cancer

PURPOSE/OBJECTIVE(S)

Neoadjuvant (chemo)radiotherapy (RT) has demonstrated an overall survival (OS) benefit in esophageal cancer and constitutes part of the standard of care trimodality therapy. Unfortunately, subsequent cardiac toxicity can reduce the benefit of treatment. Our group aimed to study whether data from electrocardiograms (ECGs) could predict clinical outcomes and cardiac events after RT for esophageal cancer, allowing for identification of and early intervention for patients at high risk for cardiac toxicity.

MATERIALS/METHODS

Included patients received at least 41.4 Gy of pre-operative or definitive photon or proton RT for esophageal cancer from 2015 through July 2022. All ECGs were assessed using a previously validated artificial intelligence assessment for atrial fibrillation (AF) and reduced ejection fraction (rEF) (Noseworthy et al. Lancet 2022). The model determined propensities for the development of multiple cardiac events, including AF and heart failure (HF). Medical records were reviewed for cardiac events and conditions prior to and after RT.

RESULTS

A cohort of 491 patients was assembled, with 301, 121, and 364 patients having an ECG prior to, during, and after RT, respectively. Of these, 84% had malignancy in the lower third of the esophagus and 48% underwent esophagectomy. At last follow-up relative to baseline assessment, patients had increased propensity for rEF (median 0.013, interquartile range (IQR): 0.001-0.038 vs. median 0.022, IQR: 0.011-0.074, p < 0.0001) and AF (median 0.16, IQR: 0.04-0.40 vs. median 0.048, IQR: 0.01-0.19, p < 0.0001). Increases in AF propensity were associated with reduced OS (hazard ratio (HR) = 1.10 per 0.1 increase, 95% confidence interval (CI): 1.03-1.17, p = 0.0071). Baseline rEF propensity was predictive of future HF events (HR = 1.14, 95% CI: 1.07-1.22, p < 0.001) for all patients or after excluding the 172 (35%) patients with baseline HF (HR = 1.45, 95% CI: 1.19-1.76, p < 0.001). Among patients who did not have HF prior to radiotherapy, the development of HF was associated with reduced OS (HR = 1.60, 95% CI: 1.10-2.32, p = 0.014). Currently available cardiac dosimetric parameters, including heart mean/max doses, did not significantly correlate with cardiac outcomes. Patients who underwent esophagectomy had improved OS (HR = 0.62, 95% CI: 0.47-0.82, p = 0.0008) and were not more likely to develop cardiac toxicity.

CONCLUSION

This analysis suggests that chemoradiotherapy for esophageal cancer can have significant impacts on a patient's propensity for cardiac events, which are associated with reduced OS. ECGs carry the potential to identify patients at greater risk for such events, and baseline ECGs with artificial intelligence assessment could select patients for increased surveillance or early intervention to further optimize the therapeutic ratio of RT.

Patient-Level Artificial Intelligence-Enhanced Electrocardiography in Hypertrophic Cardiomyopathy: Longitudinal Treatment and Clinical Biomarker Correlations

BACKGROUND

Artificial intelligence (AI) applied to 12-lead electrocardiographs (ECGs) can detect hypertrophic cardiomyopathy (HCM).

OBJECTIVES

The purpose of this study was to determine if AI-enhanced ECG (AI-ECG) can track longitudinal therapeutic response and changes in cardiac structure, function, or hemodynamics in obstructive HCM during mavacamten treatment.

METHODS

We applied 2 independently developed AI-ECG algorithms (University of California-San Francisco and Mayo Clinic) to serial ECGs (n = 216) from the phase 2 PIONEER-OLE trial of mavacamten for symptomatic obstructive HCM (n = 13 patients, mean age 57.8 years, 69.2% male). Control ECGs from 2,600 age- and sex-matched individuals without HCM were obtained. AI-ECG output was correlated longitudinally to echocardiographic and laboratory metrics of mavacamten treatment response.

RESULTS

In the validation cohorts, both algorithms exhibited similar performance for HCM diagnosis, and exhibited mean HCM score decreases during mavacamten treatment: patient-level score reduction ranged from approximately 0.80 to 0.45 for Mayo and 0.70 to 0.35 for USCF algorithms; 11 of 13 patients demonstrated absolute score reduction from start to end of follow-up for both algorithms. HCM scores were significantly associated with other HCM-relevant parameters, including left ventricular outflow tract gradient at rest, postexercise, and with Valsalva, and NT-proBNP level, independent of age and sex (all P < 0.01). For both algorithms, the strongest longitudinal correlation was between AI-ECG HCM score and left ventricular outflow tract gradient postexercise (slope estimate: University of California-San Francisco 0.70 [95% CI: 0.45-0.96], P < 0.0001; Mayo 0.40 [95% CI: 0.11-0.68], P = 0.007).

CONCLUSIONS

AI-ECG analysis longitudinally correlated with changes in echocardiographic and laboratory markers during mavacamten treatment in obstructive HCM. These results provide early evidence for a potential paradigm for monitoring HCM therapeutic response.

Postdevelopment Performance and Validation of the Artificial Intelligence-Enhanced Electrocardiogram for Detection of Cardiac Amyloidosis

BACKGROUND

We have previously applied artificial intelligence (AI) to an electrocardiogram (ECG) to detect cardiac amyloidosis (CA).

OBJECTIVES

In this validation study, the authors observe the postdevelopment performance of the AI-enhanced ECG to detect CA with respect to multiple potential confounders.

METHODS

Amyloid patients diagnosed after algorithm development (June 2019-January 2022) with a 12-lead ECG were identified (n = 440) and were required to have CA. A 15:1 age- and sex-matched control group was identified (n = 6,600). Area under the receiver operating characteristic (AUC) was determined for the cohort and subgroups.

RESULTS

The average age was 70.4 ± 10.3 years, 25.0% were female, and most patients were White (91.3%). In this validation, the AI-ECG for amyloidosis had an AUC of 0.84 (95% CI: 0.82-0.86) for the overall cohort and between amyloid subtypes, which is a slight decrease from the original study (AUC 0.91). White, Black, and patients of "other" races had similar algorithm performance (AUC >0.81) with a decreased performance for Hispanic patients (AUC 0.66). Algorithm performance shift over time was not observed. Low ECG voltage and infarct pattern exhibited high AUC (>0.90), while left ventricular hypertrophy and left bundle branch block demonstrated lesser performance (AUC 0.75 and 0.76, respectively).

CONCLUSIONS

The AI-ECG for the detection of CA maintained an overall strong performance with respect to patient age, sex, race, and amyloid subtype. Lower performance was noted in left bundle branch block, left ventricular hypertrophy, and ethnically diverse populations emphasizing the need for subgroup-specific validation efforts.

Exploring Factors Influencing ECG Interpretation Proficiency of Medical Professionals

The electrocardiogram (ECG) is a crucial diagnostic tool in medicine with concerns about its interpretation proficiency across various medical disciplines. Our study aimed to explore potential causes of these issues and identify areas requiring improvement. A survey was conducted among medical professionals to understand their experiences with ECG interpretation and education. A total of 2515 participants from diverse medical backgrounds were surveyed. A total of 1989 (79%) participants reported ECG interpretation as part of their practice. However, 45% expressed discomfort with independent interpretation. A significant 73% received less than 5 hours of ECG-specific education, with 45% reporting no education at all. Also, 87% reported limited or no expert supervision. Nearly all medical professionals (2461, 98%) expressed a desire for more ECG education. These findings were consistent across all groups and did not vary between primary care physicians, cardiology FIT, resident physicians, medical students, APPs, nurses, physicians, and nonphysicians. This study reveals substantial deficiencies in ECG interpretation training, supervision, and confidence among medical professionals, despite a strong interest in increased ECG education.

Screening for peripartum cardiomyopathies using artificial intelligence in Nigeria (SPEC-AI Nigeria): Clinical trial rationale and design

BACKGROUND

Artificial intelligence (AI), and more specifically deep learning, models have demonstrated the potential to augment physician diagnostic capabilities and improve cardiovascular health if incorporated into routine clinical practice. However, many of these tools are yet to be evaluated prospectively in the setting of a rigorous clinical trial-a critical step prior to implementing broadly in routine clinical practice.

OBJECTIVES

To describe the rationale and design of a proposed clinical trial aimed at evaluating an AI-enabled electrocardiogram (AI-ECG) for cardiomyopathy detection in an obstetric population in Nigeria.

DESIGN

The protocol will enroll 1,000 pregnant and postpartum women who reside in Nigeria in a prospective randomized clinical trial. Nigeria has the highest reported incidence of peripartum cardiomyopathy worldwide. Women aged 18 and older, seen for routine obstetric care at 6 sites (2 Northern and 4 Southern) in Nigeria will be included. Participants will be randomized to the study intervention or control arm in a 1:1 fashion. This study aims to enroll participants representative of the general obstetric population at each site. The primary outcome is a new diagnosis of cardiomyopathy, defined as left ventricular ejection fraction (LVEF) < 50% during pregnancy or within 12 months postpartum. Secondary outcomes will include the detection of impaired left ventricular function (at different LVEF cut-offs), and exploratory outcomes will include the effectiveness of AI-ECG tools for cardiomyopathy detection, new diagnosis of cardiovascular disease, and the development of composite adverse maternal cardiovascular outcomes.

SUMMARY

This clinical trial focuses on the emerging field of cardio-obstetrics and will serve as foundational data for the use of AI-ECG tools in an obstetric population in Nigeria. This study will gather essential data regarding the utility of the AI-ECG for cardiomyopathy detection in a predominantly Black population of women and pave the way for clinical implementation of these models in routine practice.

TRIAL REGISTRATION

Clinicaltrials.gov: NCT05438576.

Design and rationale of the MODULAR ATP global clinical trial: A novel intercommunicative leadless pacing system and the subcutaneous implantable cardioverter-defibrillator

Background

The subcutaneous implantable cardioverter-defibrillator (S-ICD) has demonstrated safety and efficacy for the treatment of malignant ventricular arrhythmias. However, a limitation of the S-ICD lies in the inability to either pace-terminate ventricular tachycardia or provide prolonged bradycardia pacing support.

Objective

The rationale and design of a prospective, single-arm, multinational trial of an intercommunicative leadless pacing system integrated with the S-ICD will be presented.

Methods

A technical description of the modular cardiac rhythm management (mCRM) system (EMPOWER leadless pacemaker and EMBLEM S-ICD) and the implantation procedure is provided. MODULAR ATP (Effectiveness of the EMPOWER™ Modular Pacing System and EMBLEM™ Subcutaneous ICD to Communicate Antitachycardia Pacing) is a multicenter, international trial enrolling up to 300 patients at risk of sudden cardiac death at up to 60 centers trial design. The safety endpoint of freedom from major complications related to the mCRM system or implantation procedure at 6 months and 2 years are significantly higher than 86% and 81%, respectively, and all-cause survival is significantly >85% at 2 years.

Results

Efficacy endpoints are that at 6 months mCRM communication success is significantly higher than 88% and the percentage of subjects with low and stable thresholds is significantly higher than 80%. Substudies to evaluate rate-responsive features and performance of the pacing module are also described.

Conclusion

The MODULAR ATP global clinical trial will prospectively test the safety and efficacy of the first intercommunicating leadless pacing system with the S-ICD. This trial will allow for robust validation of device-device communication, pacing performance, rate responsiveness, and system safety.

Contemporary trends and barriers to oral anticoagulation therapy in Non-valvular atrial fibrillation during DOAC predominant era

There is a need to reassess contemporary oral anticoagulation (OAC) trends and barriers against guideline directed therapy in the United States. Most previous studies were performed before major guideline changes recommended direct oral anticoagulant (DOAC) use over warfarin or have otherwise lacked patient level data. Data on overuse of OAC in low-risk group is also limited. To address these knowledge gaps, we performed a nationwide analysis to analyze current trends. This is a retrospective cohort study assessing non-valvular AF identified using a large United States de-identified administrative claims database, including commercial and Medicare Advantage enrollees. Prescription fills were assessed within a 90-day follow-up from the patient's index AF encounter between January 1, 2016, and December 31, 2020. Among the 339,197 AF patients, 4.4%, 8.0%, and 87.6% were in the low-, moderate-, and high-risk groups (according to CHA₂DS₂-VASc score). An over (29.6%) and under (52.2%) utilization of OAC was reported in low- and high-risk AF patients. A considerably high frequency for warfarin use was also noted among high-risk group patients taking OAC (33.1%). The results suggest that anticoagulation use for stroke prevention in the United States is still comparable to the pre-DOAC era studies. About half of newly diagnosed high-risk non-valvular AF patients remain unprotected against stroke risk. Several predictors of OAC and DOAC use were also identified. Our findings may identify a population at risk of complications due to under- or over-treatment and highlight the need for future quality improvement efforts.

Physiological Age by Artificial Intelligence-Enhanced Electrocardiograms as a Novel Risk Factor of Mortality in Kidney Transplant Candidates

BACKGROUND

Mortality risk assessment before kidney transplantation (KT) is imperfect. An emerging risk factor for death in nontransplant populations is physiological age as determined by the application of artificial intelligence to the electrocardiogram (ECG). The aim of this study was to examine the relationship between ECG age and KT waitlist mortality.

METHODS

We applied a previously developed convolutional neural network to the ECGs of KT candidates evaluated 2014 to 2019 to determine ECG age. We used a Cox proportional hazard model to examine whether ECG age was associated with waitlist mortality.

RESULTS

Of the 2183 patients evaluated, 59.1% were male, 81.4% were white, and 11.4% died during follow-up. Mean ECG age was 59.0 ± 12.0 y and mean chronological age at ECG was 53.3 ± 13.6 y. After adjusting for chronological age, comorbidities, and other characteristics associated with mortality, each increase in ECG age of >10 y than the average ECG age for patients of a similar chronological age was associated with an increase in mortality risk (hazard ratio 3.59 per 10-y increase; 95% confidence interval, 2.06-5.72; P  < 0.0001).

CONCLUSIONS

ECG age is a risk factor for KT waitlist mortality. Determining ECG age through artificial intelligence may help guide risk-benefit assessment when evaluating candidates for KT.

Trends in the 30-year span of Noninfectious Cardiovascular Implantable Electronic Device Complications in Olmsted County

Background

Cardiovascular implantable electronic devices (CIEDs) such as permanent pacemakers, implantable cardioverter-defibrillators, and cardiac resynchronization therapy devices alleviate morbidity and mortality in various diseases. There is a paucity of real-world data on CIED complications and trends.

Objectives

Describe trends in noninfectious CIED complications over the past three decades in Olmsted County.

Methods

The Rochester Epidemiology Project is a medical records linkage system comprising records of over 500,000 residents of Olmsted County from 1966-current. CIED implants between 1988-2018 were determined. Trends in noninfectious complications within 30 days of implant were analyzed.

Results

175 out of 2536 (6.9%) patients who received CIED experienced device complications. 3.8% of the implants had major complications requiring intervention. Lead dislodgement was the most common (2.9%), followed by hematoma (2.1%). Complications went up from 1988 to 2005, then showed a downtrend until 2018, driven by a decline in hematomas in the last decade (p<0.01). Those with complications were more likely to have prosthetic valves. Obesity appeared to have a protective effect in a multivariate regression model. The mean Charlson comorbidity score has trended up over the 30 years.

Conclusions

Our study describes a real-world trend of CIED complications over three decades. Lead dislodgements and hematomas were the most common complications. Complications have declined over the last decade due to safer practices and a better understanding of anticoagulant management.

Correlation between artificial intelligence-enabled electrocardiogram and echocardiographic features in aortic stenosis

Aims

An artificial intelligence-enabled electrocardiogram (AI-ECG) is a promising tool to detect patients with aortic stenosis (AS) before developing symptoms. However, functional, structural, or haemodynamic components reflected in AI-ECG responsible for its detection are unknown.

Methods and results

The AI-ECG model that was developed at Mayo Clinic using a convolutional neural network to identify patients with moderate-severe AS was applied. In patients used as the testing group, the correlation between the AI-ECG probability of AS and echocardiographic parameters was investigated. This study included 102 926 patients (63.0 ± 16.3 years, 52% male), and 28 464 (27.7%) were identified as AS positive by AI-ECG. Older age, atrial fibrillation, hypertension, diabetes, coronary artery disease, and heart failure were more common in the positive AI-ECG group than in the negative group (P < 0.001). The AI-ECG was correlated with aortic valve area (ρ = -0.48, R² = 0.20), peak velocity (ρ = 0.22, R² = 0.08), and mean pressure gradient (ρ = 0.35, R² = 0.08). The AI-ECG also correlated with left ventricular (LV) mass index (ρ = 0.36, R² = 0.13), E/e' (ρ = 0.36, R² = 0.12), and left atrium volume index (ρ = 0.42, R² = 0.12). Neither LV ejection fraction nor stroke volume index had a significant correlation with the AI-ECG. Age correlated with the AI-ECG (ρ = 0.46, R² = 0.22) and its correlation with echocardiography parameters was similar to that of the AI-ECG.

Conclusion

A combination of AS severity, diastolic dysfunction, and LV hypertrophy is reflected in the AI-ECG to detect AS. There seems to be a gradation of the cardiac anatomical/functional features in the model and its identification process of AS is multifactorial.

Safety of Magnetic Resonance Imaging in Patients with Surgically Implanted Permanent Epicardial Leads

BACKGROUND

Magnetic resonance imaging (MRI) safety in patients with an epicardial cardiac implantable electronic device (CIED) is uncertain.

OBJECTIVE

To assess the safety and adverse effects of MRI in patients who had surgically implanted epicardial CIED.

METHODS

Patients with surgically implanted CIEDs who underwent MRI with an appropriate Cardiology-Radiology collaborative protocol between January 2008 and January 2021 were prospectively studied in two clinical centers. All patients underwent close cardiac monitoring through MRI procedures. Outcomes were compared between the epicardial CIED group and matched the non-MRI-conditional transvenous CIED group.

RESULTS

Twenty-nine consecutive patients with epicardial CIED (male 41.4%, mean age of 43 years) underwent 52 MRIs in the 57 anatomic regions. Sixteen patients had pacemakers, 9 had cardiac defibrillators or cardiac resynchronization therapy defibrillators, and 4 had no device generators. There were no significant adverse events in epicardial or transvenous CIED groups. The battery life, pacing, sensing thresholds, lead impedance and cardiac biomarkers were not significantly changed, except one patient had a transient decrease in atrial lead sensing function.

CONCLUSION

MRI of CIEDs with epicardially implanted leads does not represent a greater risk than the transvenous CIEDs when performed with a multidisciplinary collaborative protocol centered on patient safety.

Ventricular Arrhythmias in Patients with Prior Aortic Valve Intervention: Characteristics, Ablation and Outcomes

BACKGROUND

Data regarding ventricular tachycardia (VT) or premature ventricular complex (PVC) ablation in patients with aortic valve intervention (AVI) is limited. Catheter ablation (CA) can be challenging given perivalvular substrate in the setting of prosthetic valves.

OBJECTIVE

To investigate the characteristics, safety, and outcomes of CA in patients with prior AVI and ventricular arrhythmias (VA).

METHODS

We identified consecutive patients with prior AVI (replacement or repair) who underwent CA for VT or PVC between 2013 and 2018. We investigated the mechanism of arrhythmia, ablation approach, perioperative complications, and outcomes.

RESULTS

We included 34 patients (88% men, mean age 64±10.4 years, left ventricular ejection fraction 35.2±15.0%) with prior AVI who underwent CA (22 VT; 12 PVC). LV access was obtained through trans-septal approach in all patients except one patient who had percutaneous transapical access. One patient had combined retrograde aortic and trans-septal approach. Scar-related reentry was the dominant mechanism of induced VTs. Two patients had bundle branch reentry VTs. In the VT group, substrate mapping demonstrated heterogeneous scar that involved the periaortic valve area in 95%. Despite that, the site of successful ablation included the periaortic region only in 6 (27%) patients. In the PVC group, signal abnormalities consistent with scar in the periaortic area were noted in 4 (33%) patients. In 8 (67%) patients, the successful site of ablation was unrelated to the periaortic area. No procedure-related complications occurred. The survival and recurrence-free survival rate at 1 year tended to be lower in VT group than in PVC group (P=0.06 and P=0.05, respectively) with a 1-year recurrence-free survival rate of 52.8% and 91.7%, respectively. No arrhythmia-related death was documented on long-term follow-up.

CONCLUSION

CA of VAs can be performed safely and effectively in patients with prior AVI. This article is protected by copyright. All rights reserved.

Nonthrombogenic Roles of the Left Atrial Appendage: JACC Review Topic of the Week

The atrial appendage (LAA) is a well-established source of cardioembolism in patients with atrial fibrillation. Therefore, research involving the LAA has largely focused on its thrombogenic attribute and the utility of its exclusion in stroke prevention. However, recent studies have highlighted several novel functions of the LAA that may have important therapeutic implications. In this paper, we provide a concise overview of the LAA anatomy and summarize the emerging data on its nonthrombogenic roles.

Artificial Intelligence-Derived Electrocardiogram Assessment of Cardiac Age and Molecular Markers of Senescence in Heart Failure

OBJECTIVE

To ascertain whether heart failure (HF) itself is a senescent phenomenon independent of age, and how this is reflected at a molecular level in the circulating progenitor cell niche, and at a substrate level using a novel electrocardiogram (ECG)-based artificial intelligence platform.

PATIENTS AND METHODS

Between October 14, 2016, and October 29, 2020, CD34⁺ progenitor cells were analyzed by flow cytometry and isolated by magnetic-activated cell sorting from patients of similar age with New York Heart Association functional classes IV (n = 17) and I-II (n = 10) heart failure with reduced ejection fraction and healthy controls (n = 10). CD34⁺ cellular senescence was quantitated by human telomerase reverse transcriptase expression and telomerase expression by quantitative polymerase chain reaction, and senescence-associated secretory phenotype (SASP) protein expression assayed in plasma. An ECG-based artificial intelligence (AI) algorithm was used to determine cardiac age and difference from chronological age (AI ECG age gap).

RESULTS

CD34⁺ counts and telomerase expression were significantly reduced and AI ECG age gap and SASP expression increased in all HF groups compared with healthy controls. Expression of SASP protein was closely associated with telomerase activity and severity of HF phenotype and inflammation. Telomerase activity was more closely associated with CD34⁺ cell counts and AI ECG age gap.

CONCLUSION

We conclude from this pilot study that HF may promote a senescent phenotype independent of chronological age. We show for the first time that the AI ECG in HF shows a phenotype of cardiac aging beyond chronological age, and appears to be associated with cellular and molecular evidence of senescence.