Eidem, Benjamin W.

Background Early detection of left and right ventricular systolic dysfunction (LVSD and RVSD respectively) in children can lead to intervention to reduce morbidity and death. Existing artificial intelligence algorithms can identify LVSD and RVSD in adults using a 12‐lead ECG; however, its efficacy in children is uncertain. We aimed to develop novel artificial intelligence–enabled ECG algorithms for LVSD and RVSD detection in pediatric patients. Methods and Results We identified 10 142 unique pediatric patients (age≤18) with a 10‐second, 12‐lead surface ECG within 14 days of a transthoracic echocardiogram, performed between 2002 and 2022. LVSD was defined quantitatively by left ventricular ejection fraction (LVEF). RVSD was defined semiquantitatively. Novel pediatric models for LVEF ≤35% and LVEF <50% achieved excellent test areas under the curve of 0.93 (95% CI, 0.89–0.98) and 0.88 (95% CI, 0.83–0.94) respectively. The model to detect LVEF <50% had a sensitivity of 0.85, specificity of 0.80, positive predictive value of 0.095, and negative predictive value of 0.995. In comparison, the previously validated adult data‐derived model for LVEF <35% achieved an area under the curve of 0.87 (95% CI, 0.84–0.90) for LVEF ≤35% in children. A novel pediatric model for any RVSD detection reached a test area under the curve of 0.90 (0.87–0.94). Conclusions An artificial intelligence–enabled ECG demonstrates accurate detection of both LVSD and RVSD in pediatric patients. While adult‐trained models offer good performance, improvements are seen when training pediatric‐specific models.

Background: Impact of right ventricular (RV) overload on left ventricular (LV) performance has been demonstrated in Ebstein’s anomaly (EA). We have shown that LV radial strain (RS) in 83 EA patients was significantly less than in controls. We subsequently sought to evaluate LV strain after EA surgery. Methods: Systolic strain, strain rate (SR) and early diastolic strain rate (SRe) were retrospectively performed, using velocity vector imaging, in 24 repaired EA patients (age 35 ±12 years; 67% women) with LV ejection fraction >50%, preoperatively (median 5 days [1-145 days]) and postoperatively within 5 years at sequential follow-up (6 months-3 years). Repair of EA included tricuspid valve repair, valve replacement and bidirectional cavopulmonary anastomosis. A slope of change postoperatively was estimated for each subject using linear regression. A one-sample t-test of the mean slope=0 assessed average change per year. Results: Mean (standard deviation) and median (range) of preoperative LV RS was 36.5 +/- 11.4% and 36.3% (16.8-59.8) respectively. LV strain analyses are summarized in tables 1 and 2. Conclusions: LV RS improved significantly after EA surgery whereas circumferential strain (CS), SR and radial SR did not. This may reflect the change in RV and subsequently LV geometry after resolution of tricuspid regurgitation with EA surgery. Ongoing evaluation with deformational imaging may help to characterize if these changes are compensatory or maladaptive and may determine optimal timing for surgery. In the subgroup with the lowest RS, significant improvement postoperatively was not seen. This may suggest that surgery was performed too late to allow favorable remodeling.

Background: The first pediatric appropriate use criteria (PAUC) were recently published for initial outpatient transthoracic echocardiography (TTE). We sought to determine the effect of release of the PAUC on the appropriateness of TTE ordering patterns of pediatric cardiologists. Methods: Data were prospectively collected from patients having initial outpatient TTE ordered prior to (Phase I, 6 mo) and 3 months after release (Phase II, 4 mo) of PAUC. Site-investigators determined the indication of the study and assigned appropriateness rating based on the PAUC document [Appropriate (A), May Be Appropriate (M), or Rarely Appropriate (R)] or "Unclassifiable" (U). Results: A total of 4562 TTEs (2655 Phase I, 1907 Phase II) were ordered by 103 physicians at 6 sites. Overall comparison of appropriateness rating for Phase I and II showed no change in the rate for A or M, but a decline in the rate of R and increase in U. Similar results were noted when comparisons were made for physicians that ordered at least 20 TTEs during each phase (N = 30/103, 29%), Table 1. There was no change in any appropriateness rating at 3 sites, a decline in R at 2, an increase in A at 1, and an increase in U at 2 sites. Overall and site-specific comparisons of rate of R TTEs are shown in Fig 1. Conclusions: The release of PAUC document had only a small impact on physician ordering behavior for initial TTEs, including a small decrease in R. Overall, these data suggest that more focused interventions are required to fully implement guideline recommendations and improve the appropriate use of pediatric TTE. This information should be helpful in designing educational interventions to reduce the rate of TTEs ordered for R indications.

Background: Atrial septal defects (ASD) and ventricular septal defects (VSD) are among the most common forms of congenital heart disease (CHD). Early detection and intervention for these defects are crucial to prevent complications. Given advancements in AI technology, the development of AI-enabled ECG models for detecting these defects is of significant interest. Aims: This study aimed to create AI models using pediatric ECG data for the detection of ASD, VSD, and a combined model detecting either ASD or VSD. Methods: We analyzed a dataset of 7,795 pediatric patients from Mayo Clinic who had undergone echocardiography between 2002 and 2022, with corresponding 10-second, 12-lead surface ECGs taken within 14 days of echocardiography. This dataset included 259 patients with ASDs and 139 patients with VSDs (single or multiple). The control group was comprised of 7,397 patients without significant or complex CHD. For cases and controls, we included patients with all forms of electrocardiographic abnormalities and those with valvular lesions who had less than moderate stenosis or regurgitation. The dataset was divided into training, validation, and testing subsets to develop convolutional neural network (CNN) models for ASD, VSD, and either ASD or VSD detection. Results: Our ASD and VSD detection models achieved test AUCs of 0.82 (95% CI 0.76, 0.88) and 0.80 (95% CI 0.70, 0.90), respectively. Meanwhile, a combined model that detected either ASD or VSD demonstrated a test AUC of 0.81 (95% CI, 0.76, 0.86), sensitivity of 0.73, specificity of 0.76, PPV of 0.14, and NPV of 0.98 (Figure 1). Conclusions: AI-enabled ECG models demonstrate promising performance in detecting ASD and VSD in pediatric patients. A combined model that detects either ASD or VSD also shows robust results. Future research should focus on multi-center validation of these models, refining their diagnostic capacities, and investigating their potential for early intervention in CHDs.

Introduction: Left and right ventricular systolic dysfunction (LVSD and RVSD respectively) contribute to considerable pediatric morbidity. Both LVSD and RVSD can occur in children before symptom onset, emphasizing the need for early detection. Existing deep-learning algorithms can identify LVSD in adults using 12-lead ECG; however, their efficacy in children is uncertain. Aims: We aimed to create a deep-learning model for LVSD and RVSD detection in pediatric patients using an AI-enabled ECG and compare novel pediatric models to a previously validated adult LVSD model. Methods: We identified 10,214 pediatric patients at the Mayo Clinic with a transthoracic echocardiogram (TTE) and a 10-second, 12-lead surface ECG within 14 days of TTE, performed from 2002 -2022. Our cohort comprised the first TTE-ECG pair from each patient, totaling 138 with EF ≤ 35%, 312 with EF ≤ 50%, 482 with RVSD, and 9,567 without any LVSD or RVSD. The cohort was split into training, validation, and testing datasets to develop neural networks for systolic dysfunction detection. Results: The cohort was 52.3% male, with an average age of 10.3 years. Novel models generated in the pediatric cohort demonstrated excellent performance in systolic dysfunction detection (Figure 1). Pediatric models for EF ≤ 35% and EF ≤ 50% achieved test AUCs of 0.92 (95% CI 0.89, 0.95) and 0.89 (95% CI 0.85, 0.94) respectively. The RVSD model reached a test AUC of 0.91 (95% CI 0.88, 0.94). A combined model for EF ≤ 35% or RVSD exhibited a test AUC of 0.90 (95% CI 0.87, 0.93), sensitivity 0.88, specificity 0.82, PPV 0.20, and NPV 0.95. A combined model for EF ≤ 50% or RVSD exhibited a test AUC of 0.88 (95% CI 0.85, 0.91). The previously validated adult data-derived model achieved an AUC of 0.86 (95% CI 0.83, 0.89) for EF ≤ 50% in children. Conclusions: AI-enabled ECGs demonstrate efficient detection of both LVSD and RVSD in pediatric patients. Future multi-site validation is needed to further establish the reliability of these models.

Introduction: Biologic sex and hormonal concentrations shape ECG parameters due to sex hormone effects on cardiac function. AI-enabled ECGs have previously been shown to accurately predict patient sex in adults and correlate with sex hormone levels. Aims: We aimed to test the ability of AI-enabled ECGs to predict pediatric patient sex and explore the influence of pubertal development. Methods: Two AI-enabled ECG models were created using a convolutional neural network trained on pediatric 10-second, 12-lead surface ECGs. The first model was trained de novo using pediatric data. The second model used transfer learning from a previously validated adult data-derived algorithm. We analyzed the first ECG from 89,063 pediatric patients (aged <19 years) recorded from 1988-2022, and divided the cohort into training, validation, and testing datasets. Subgroup analysis was performed on prepubertal (0-7 years), pubertal (8-14 years), and post-pubertal (15-18 years) patients. Results: The cohort consisted of 46.6% males, with 21,434 prepubertal, 26,300 pubertal, and 41,329 post-pubertal children. No significant performance difference was observed between the two models. Both models demonstrated 81% accuracy and an overall AUC of 0.91 in the testing cohorts, with high discriminatory ability in post-pubertal teenagers (AUC = 0.98). (Figure 1) Conclusions: AI-ECG predicts pediatric patient sex with varying performance across pubertal stages. The highest discriminatory ability is observed post-puberty, decreasing in pubertal and prepubertal children. The results suggest hormonal and physiological changes during puberty may influence cardiac electrophysiology measurably. More research is required to interpret CNN-predicted sex discordance with actual sex and its implications for future cardiovascular risk.

Importance Whether vigorous intensity exercise is associated with an increase in risk of ventricular arrhythmias in individuals with hypertrophic cardiomyopathy (HCM) is unknown. Objective To determine whether engagement in vigorous exercise is associated with increased risk for ventricular arrhythmias and/or mortality in individuals with HCM. The a priori hypothesis was that participants engaging in vigorous activity were not more likely to have an arrhythmic event or die than those who reported nonvigorous activity. Design, Setting, and Participants This was an investigator-initiated, prospective cohort study. Participants were enrolled from May 18, 2015, to April 25, 2019, with completion in February 28, 2022. Participants were categorized according to self-reported levels of physical activity: sedentary, moderate, or vigorous-intensity exercise. This was a multicenter, observational registry with recruitment at 42 high-volume HCM centers in the US and internationally; patients could also self-enroll through the central site. Individuals aged 8 to 60 years diagnosed with HCM or genotype positive without left ventricular hypertrophy (phenotype negative) without conditions precluding exercise were enrolled. Exposures Amount and intensity of physical activity. Main Outcomes and Measures The primary prespecified composite end point included death, resuscitated sudden cardiac arrest, arrhythmic syncope, and appropriate shock from an implantable cardioverter defibrillator. All outcome events were adjudicated by an events committee blinded to the patient’s exercise category. Results Among the 1660 total participants (mean [SD] age, 39 [15] years; 996 male [60%]), 252 (15%) were classified as sedentary, and 709 (43%) participated in moderate exercise. Among the 699 individuals (42%) who participated in vigorous-intensity exercise, 259 (37%) participated competitively. A total of 77 individuals (4.6%) reached the composite end point. These individuals included 44 (4.6%) of those classified as nonvigorous and 33 (4.7%) of those classified as vigorous, with corresponding rates of 15.3 and 15.9 per 1000 person-years, respectively. In multivariate Cox regression analysis of the primary composite end point, individuals engaging in vigorous exercise did not experience a higher rate of events compared with the nonvigorous group with an adjusted hazard ratio of 1.01. The upper 95% 1-sided confidence level was 1.48, which was below the prespecified boundary of 1.5 for noninferiority. Conclusions and Relevance Results of this cohort study suggest that among individuals with HCM or those who are genotype positive/phenotype negative and are treated in experienced centers, those exercising vigorously did not experience a higher rate of death or life-threatening arrhythmias than those exercising moderately or those who were sedentary. These data may inform discussion between the patient and their expert clinician around exercise participation.

Introduction: Safety of vigorous exercise for individuals with appropriately-treated HCM remains debated. Physical activity practices in this population have not been described. The objective of this report is to describe physical activity pattern in HCM patients. Long term follow up which is ongoing will determine safety of vigorous exercise. Methods: The NIH-funded LIVE-HCM study prospectively enrolled individuals age 8-60 years with clinically overt HCM or patients with a positive genetic test for sarcomeric HCM. Participants (or parents, for children) answered surveys describing activity patterns. Vigorous exercise was defined as > 6 METS for > 60 hours per year. Clinical and demographic data were derived from medical record review. Results: Among 1798 participants, 745 (41%) reported exercising vigorously, including 296 competitive athletes, with 59 at high-school or college varsity level, 770 (43%) were exercising at moderate, and 283 (16%) at low levels. Of those working, 6% described jobs more active than walking, including 16 in protective services, and 11 in sports/coaching. Clinical/demographic data including imaging characteristics and treatments, are shown in Table, and were mostly similar amongst the exercise-groups. Conclusions: Many patients with HCM are engaged in vigorous exercise including competitive athletics. Prospective follow-up of this cohort is ongoing. Comparison of arrhythmic outcomes outcomes in vigorous exercisers vs moderate/low level exercisers will help determine the safety of exercise in HCM patients.

Objectives: To longitudinally measure LV diastolic function in HIV-exposed but uninfected (HEU) children perinatally exposed to ART. Design: HEU children who were perinatally exposed to antiretroviral therapy (ART) may be at risk for adverse cardiac effects. We have previously reported that those children have decreased left ventricular (LV) mass, dimension, and septal thickness with increased contractility. Methods: Serial echocardiograms were obtained at specific times from birth to 48 months from two groups of HIV-uninfected children: 148 HIV-negative children who were perinatally exposed to ART and 130 non-ART-exposed HIV-unexposed healthy controls. The following LV diastolic indices were obtained: mitral valve early and late diastolic velocity (E and A), tissue Doppler-derived LV-free wall and septal early diastolic velocity (LV e′ and sep e′). Results: All echocardiographic indices were significantly different in ART-exposed children compared with ART-unexposed healthy controls. Both E and A were overall lower at all ages by 8.28 cm/s (P = 0.0002) and 13.46 cm/s (P < 0.0001) respectively. E/A ratio was higher by 0.27, 0.46, and 0.28 units at birth, 1 year and 2 years of age, respectively (all P ≤ 0.01). Moreover, LV e′ and sep e′ were overall lower at all ages by 0.84 cm/s (P = 0.01) and 0.47 cm/s (P = 0.02), respectively. Conclusion: Children who were exposed to ART in utero have subclinical yet significant differences in specific LV diastolic indices. Follow-up with serial echocardiograms are recommended in this population to further assess the potential cardiac toxicity of perinatal exposure to ART.