Artificial Intelligence-Enabled Assessment of the Heart Rate Corrected QT Interval Using a Mobile Electrocardiogram Device.

BACKGROUND: Heart rate-corrected QT interval (QTc) prolongation, whether secondary to drugs, genetics including congenital long QT syndrome, and/or systemic diseases including SARS-CoV-2-mediated coronavirus disease 2019 (COVID-19), can predispose to ventricular arrhythmias and sudden cardiac death. Currently, QTc assessment and monitoring relies largely on 12-lead electrocardiography. As such, we sought to train and validate an artificial intelligence (AI)-enabled 12-lead ECG algorithm to determine the QTc, and then prospectively test this algorithm on tracings acquired from a mobile ECG (mECG) device in a population enriched for repolarization abnormalities. METHODS: Using >1.6 million 12-lead ECGs from 538 200 patients, a deep neural network (DNN) was derived (patients for training, n = 250 767; patients for testing, n = 107 920) and validated (n = 179 513 patients) to predict the QTc using cardiologist-overread QTc values as the "gold standard". The ability of this DNN to detect clinically-relevant QTc prolongation (eg, QTc ≥500 ms) was then tested prospectively on 686 patients with genetic heart disease (50% with long QT syndrome) with QTc values obtained from both a 12-lead ECG and a prototype mECG device equivalent to the commercially-available AliveCor KardiaMobile 6L. RESULTS: In the validation sample, strong agreement was observed between human over-read and DNN-predicted QTc values (-1.76±23.14 ms). Similarly, within the prospective, genetic heart disease-enriched dataset, the difference between DNN-predicted QTc values derived from mECG tracings and those annotated from 12-lead ECGs by a QT expert (-0.45±24.73 ms) and a commercial core ECG laboratory [10.52±25.64 ms] was nominal. When applied to mECG tracings, the DNN's ability to detect a QTc value ≥500 ms yielded an area under the curve, sensitivity, and specificity of 0.97, 80.0%, and 94.4%, respectively. CONCLUSIONS: Using smartphone-enabled electrodes, an AI DNN can predict accurately the QTc of a standard 12-lead ECG. QTc estimation from an AI-enabled mECG device may provide a cost-effective means of screening for both acquired and congenital long QT syndrome in a variety of clinical settings where standard 12-lead electrocardiography is not accessible or cost-effective.

Validation and Disease Risk Assessment of Previously Reported Genome-Wide Genetic Variants Associated With Brugada Syndrome: SADS-TW BrS Registry.

BACKGROUND: Brugada syndrome (BrS) is an oligogenic arrhythmic disease with increased risk of sudden cardiac arrest. Several BrS or ECG traits-related single-nucleotide polymorphisms (SNPs) were identified through previous genome-wide association studies in white patients. We aimed to validate these SNPs in BrS patients in the Taiwanese population, assessing the cumulative effect of risk alleles and the BrS-polygenic risk score in predicting cardiac events. METHODS: We genotyped 190 unrelated BrS patients using the TWB Array, and Taiwan Biobank was used as controls. SNPs not included in the array were imputed by IMPUTE2. Cox proportional hazards model was used to evaluate the associations between each particular SNP, the collective BrS-polygenic risk score, and clinical outcomes. RESULTS: Of the 88 previously reported SNPs, 22 were validated in Taiwanese BrS patients (P<0.05). Of the 22 SNPs, 2 (rs10428132 and rs9388451) were linked with susceptibility to BrS, 10 were SNPs previously reaching genome-wide significance, and 10 were SNPs associated with ECG traits. For the 3 most commonly reported SNPs, disease risk increased consistently with the number of risk alleles (odds ratio, 3.54; Ptrend=1.38×10-9 for 5 risk alleles versus 1). Similar patterns were observed in both SCN5A mutation+ (odds ratio, 3.66; Ptrend=0.049) and SCN5A mutation- (odds ratio, 3.75; Ptrend=8.54×10-9) subgroups. Furthermore, BrS patients without SCN5A mutations had more risk alleles than BrS patients with SCN5A mutations regardless of the range of polygenic risk scores. Three SNPs (rs4687718, rs7784776, and rs2968863) showed significant associations with the composite outcome (sudden cardiac arrest plus syncope, hazard ratio, 2.13, 1.48, and 0.41; P=0.02, 0.006, and 0.008, respectively). CONCLUSIONS: Our findings suggested that some SNPs associated with BrS or ECG traits exist across multiple populations. The cumulative risk of the BrS-related SNPs is similar to that in white BrS patients, but it appears to correlate with the absence of SCN5A mutations.

Assessment and Validation of a Phenotype-Enhanced Variant Classification Framework to Promote or Demote RYR2 Missense Variants of Uncertain Significance.

Background Many rare, potentially pathogenic, RYR2 variants identified in individuals with clinically definite catecholaminergic polymorphic ventricular tachycardia are classified ambiguously as variants of uncertain significance (VUS). We aimed to determine if a phenotype-enhanced variant classification approach could reduce the burden of RYR2 VUS encountered during clinical genetic testing. Methods This retrospective study was conducted in 84 RYR2-positive individuals from the Mayo Clinic (Rochester, MN) and validated in 149 RYR2-positive individuals from Amsterdam University Medical Center (Amsterdam, NL). Using a newly developed diagnostic scorecard, the pretest clinical probability of catecholaminergic polymorphic ventricular tachycardia was determined for all RYR2-positive individuals. Each RYR2 variant was then readjudicated using a phenotype-enhanced American College of Medical Genetics approach that incorporates new criteria that reflect the phenotypic strength associated with each individual RYR2 variant. Results Overall, 72 distinct RYR2 variants were identified among the 84 Mayo Clinic (39 unique) and 149 Amsterdam University Medical Center (30 unique) cases. Three variants were present in both cohorts. American College of Medical Genetics guidelines classified 47% of all RYR2 variants as VUS. In the Mayo Clinic cohort, readjudication using amended phenotype-enhanced American College of Medical Genetics standards dropped the VUS rate significantly (20/42 [48%] versus 3/42 [7%]; P<0.001) with 13/20 (65%) RYR2 VUS promoted to likely pathogenic and 4/20 (20%) demoted to likely benign. A similar drop in VUS rate (14/33 [42%] versus 3/33 [9%]; P=0.001) was observed in the Amsterdam University Medical Center validation cohort with 10/14 (71%) RYR2 VUS promoted to likely pathogenic and 1/14 (7%) demoted to likely benign. Conclusions This multicenter study illustrates the potential utility of phenotype-enhanced variant classification in catecholaminergic polymorphic ventricular tachycardia.

Cost Efficacy of α-Galactosidase A Enzyme Screening for Fabry Disease.

The prevalence of Fabry disease (FD) in adult patients with suspected hypertrophic cardiomyopathy (HCM) has been reported between 0.3% and 4%. Fabry disease-specific therapy necessitates early diagnosis; however, the optimal screening strategy and cost efficacy of routine α-galactosidase A (α-gal A) vs comprehensive galactosidase alpha gene (GLA) testing remain poorly understood. We identified 1192 patients who underwent routine α-gal A screening between January 1, 2011, and December 31, 2017, for suspected HCM. Cost efficacy was explored using prevalence and cost estimates. Ten patients had reduced α-gal A enzyme activity, and 5 (3 women) were ultimately diagnosed with FD (prevalence estimate, 0.42%). An alternative cardiac diagnosis was made in 3 patients with mildly reduced enzyme activity. Two women with reduced borderline enzyme levels did not undergo confirmatory testing, but FD was not suspected. The number needed to screen to diagnose 1 patient with FD in a similar cohort is estimated at 238 (5 new cases per 1192 at-risk individuals) at a cost of approximately US $24,000 per diagnosis. We identified a 0.42% prevalence of FD using routine α-gal A screening in adult patients referred to a dedicated HCM center in the United States. Compared with more comprehensive genetic testing strategies, we identified a similar prevalence of FD at a lower cost per diagnosis.

Noninvasive assessment of dofetilide plasma concentration using a deep learning (neural network) analysis of the surface electrocardiogram: A proof of concept study.

BACKGROUND: Dofetilide is an effective antiarrhythmic medication for rhythm control in atrial fibrillation, but carries a significant risk of pro-arrhythmia and requires meticulous dosing and monitoring. The cornerstone of this monitoring, measurement of the QT/QTc interval, is an imperfect surrogate for plasma concentration, efficacy, and risk of pro-arrhythmic potential. OBJECTIVE: The aim of our study was to test the application of a deep learning approach (using a convolutional neural network) to assess morphological changes on the surface ECG (beyond the QT interval) in relation to dofetilide plasma concentrations. METHODS: We obtained publically available serial ECGs and plasma drug concentrations from 42 healthy subjects who received dofetilide or placebo in a placebo-controlled cross-over randomized controlled clinical trial. Three replicate 10-s ECGs were extracted at predefined time-points with simultaneous measurement of dofetilide plasma concentration We developed a deep learning algorithm to predict dofetilide plasma concentration in 30 subjects and then tested the model in the remaining 12 subjects. We compared the deep leaning approach to a linear model based only on QTc. RESULTS: Fourty two healthy subjects (21 females, 21 males) were studied with a mean age of 26.9 ± 5.5 years. A linear model of the QTc correlated reasonably well with dofetilide drug levels (r = 0.64). The best correlation to dofetilide level was achieved with the deep learning model (r = 0.85). CONCLUSION: This proof of concept study suggests that artificial intelligence (deep learning/neural network) applied to the surface ECG is superior to analysis of the QT interval alone in predicting plasma dofetilide concentration.

Prevalence and Outcome of High-Risk QT Prolongation Recorded in the Emergency Department from an Institution-Wide QT Alert System.

BACKGROUND: QT prolongation is an independent risk factor for sudden death, stroke, and all-cause mortality. However, additional studies have shown that in certain settings, QT prolongation may be transient and a result of external factors. OBJECTIVE: In this study, we evaluated the clinical characteristics and outcomes of patients seen in the emergency department (ED) with QT prolongation. METHODS: Between November 2010 and June 2011, 7522 patients had an electrocardiogram (ECG) obtained during their evaluation in the ED. Clinical, laboratory, and therapeutic information was collected for all patients with QT prolongation (i.e., ≥ 500 ms and QRS < 120 ms). Potential QT-inciting factors (drugs, electrolyte disturbances, and comorbidities) were synthesized into a pro-QT score. RESULTS: Among the 7522 patients with an ECG obtained in the ED, a QT alert was activated in 93 (1.2%; mean QTc 521 ± 34 ms). The majority of ED patients (64%) had more than one underlying condition associated with QT prolongation, with electrolyte disturbances in 51%, a QT prolonging condition in 56%, and QT-prolonging drugs in 77%. Thirty-day mortality was 13% for patients with QT prolongation noted in the ED. CONCLUSIONS: One percent of patients evaluated with an ECG in the ED activated our prolonged QTc warning system, with most demonstrating > 1 QT-prolonging condition. Thirty-day mortality was significant, but it requires further investigation to determine whether the QTc simply provided a non-invasive indicator of increased risk or heralded the presence of a vulnerable host at risk of a QT-mediated sudden dysrhythmic death.

Whole exome sequencing with genomic triangulation implicates CDH2-encoded N-cadherin as a novel pathogenic substrate for arrhythmogenic cardiomyopathy.

BACKGROUND: Arrhythmogenic cardiomyopathy (ACM) is a heritable disease characterized by fibrofatty replacement of cardiomyocytes, has a prevalence of approximately 1 in 5000 individuals, and accounts for approximately 20% of sudden cardiac death in the young (≤35 years). ACM is most often inherited as an autosomal dominant trait with incomplete penetrance and variable expression. While mutations in several genes that encode key desmosomal proteins underlie about half of all ACM, the remainder is elusive genetically. OBJECTIVE: Here, whole exome sequencing (WES) was performed with genomic triangulation in an effort to identify a novel explanation for a phenotype-positive, genotype-negative multi-generational pedigree with a presumed autosomal dominant, maternal inheritance of ACM. METHODS: WES and genomic triangulation was performed on a symptomatic 14-year-old female proband, her affected mother and affected sister, and her unaffected father to elucidate a novel ACM-susceptibility gene for this pedigree. Following variant filtering using Ingenuity® Variant Analysis, gene priority ranking was performed on the candidate genes using ToppGene and Endeavour. The phylogenetic and physiochemical properties of candidate mutations were assessed further by 6 in silico prediction tools. Species alignment and amino acid conservation analysis was performed using the Uniprot Consortium. Tissue expression data was abstracted from Expression Atlas. RESULTS: Following WES and genomic triangulation, CDH2 emerged as a novel, autosomal dominant, ACM-susceptibility gene. The CDH2-encoded N-cadherin is a cell-cell adhesion protein predominately expressed in the heart. Cardiac dysfunction has been demonstrated in prior CDH2 knockout and over-expression animal studies. Further in silico mutation prediction, species conservation, and protein expression analysis supported the ultra-rare (minor allele frequency <0.005%) p.Asp407Asn-CDH2 variant as a likely pathogenic variant. CONCLUSIONS: Herein, it is demonstrated that genetic mutations in CDH2-encoded N-cadherin may represent a novel pathogenetic basis for ACM in humans. The prevalence of CDH2-mediated ACM in heretofore genetically elusive ACM remains to be determined.

Genetic Insurance Discrimination in Sudden Arrhythmia Death Syndromes: Empirical Evidence From a Cross-Sectional Survey in North America.

BACKGROUND: There is virtually no information assessing the insurability of families affected with Sudden Arrhythmia Death Syndromes (SADS) for the determination of the nonclinical implications of genetic screening. It is important to identify the barriers and challenges faced by families as a result of genetic screening for SADS to enable equitable access to insurance coverage. METHODS AND RESULTS: To explore the insurance coverage experiences of SADS-affected families, we administered a cross-sectional online survey across North America from April 28, 2012 to November 13, 2013. Participants included individuals with a SADS diagnosis and their relatives who have applied for insurance (health, life, travel, and disability) or have existing insurance coverage. Of 202 participants, 92% had a SADS diagnosis (92%) as either a proband (50%) or an affected relative (42%); 8% of participants were unaffected family members of a proband; and genetic confirmation was reported by 73%. Of the 54% of SADS respondents who applied for insurance, 60% were rejected by insurers. The preexisting SADS diagnosis was the major reason reported for rejection (57%). Most respondents (80%) had insurance coverage through a spouse/parent plan at the time of diagnosis; 14% experienced a subsequent negative effect on coverage. Thirty-nine percent of affected SADS respondents reported an increase in insurance premium rates. CONCLUSIONS: Increased genetic testing has negatively impacted insurability for SADS patients and affected family members. The challenges in obtaining life and health insurance are mainly because of the preexisting condition, even in the presence of protective laws in the United States.

Combined assessment of sex- and mutation-specific information for risk stratification in type 1 long QT syndrome.

BACKGROUND: Men and women with type 1 long QT syndrome (LQT1) exhibit time-dependent differences in the risk for cardiac events. OBJECTIVE: We hypothesized that sex-specific risk for LQT1 is related to the location and function of the disease-causing mutation in the KCNQ1 gene. METHODS: The risk for life-threatening cardiac events (comprising aborted cardiac arrest [ACA] or sudden cardiac death [SCD]) from birth through age 40 years was assessed among 1051 individuals with LQT1 (450 men and 601 women) by the location and function of the LQT1-causing mutation (prespecified as mutations in the intracellular domains linking the membrane-spanning segments [ie, S2-S3 and S4-S5 cytoplasmic loops] involved in adrenergic channel regulation vs other mutations). RESULTS: Multivariate analysis showed that during childhood (age group: 0-13 years) men had >2-fold (P < .003) increased risk for ACA/SCD than did women, whereas after the onset of adolescence the risk for ACA/SCD was similar between men and women (hazard ratio = 0.89 [P = .64]). The presence of cytoplasmic-loop mutations was associated with a 2.7-fold (P < .001) increased risk for ACA/SCD among women, but it did not affect the risk among men (hazard ratio 1.37; P = .26). Time-dependent syncope was associated with a more pronounced risk-increase among men than among women (hazard ratio 4.73 [P < .001] and 2.43 [P = .02], respectively), whereas a prolonged corrected QT interval (≥ 500 ms) was associated with a higher risk among women than among men. CONCLUSION: Our findings suggest that the combined assessment of clinical and mutation location/functional data can be used to identify sex-specific risk factors for life-threatening events for patients with LQT1.

Developing next-generation telehealth tools and technologies: patients, systems, and data perspectives.

The major goals of telemedicine today are to develop next-generation telehealth tools and technologies to enhance healthcare delivery to medically underserved populations using telecommunication technology, to increase access to medical specialty services while decreasing healthcare costs, and to provide training of healthcare providers, clinical trainees, and students in health-related fields. Key drivers for these tools and technologies are the need and interest to collaborate among telehealth stakeholders, including patients, patient communities, research funders, researchers, healthcare services providers, professional societies, industry, healthcare management/economists, and healthcare policy makers. In the development, marketing, adoption, and implementation of these tools and technologies, communication, training, cultural sensitivity, and end-user customization are critical pieces to the process. Next-generation tools and technologies are vehicles toward personalized medicine, extending the telemedicine model to include cell phones and Internet-based telecommunications tools for remote and home health management with video assessment, remote bedside monitoring, and patient-specific care tools with event logs, patient electronic profile, and physician note-writing capability. Telehealth is ultimately a system of systems in scale and complexity. To cover the full spectrum of dynamic and evolving needs of end-users, we must appreciate system complexity as telehealth moves toward increasing functionality, integration, interoperability, outreach, and quality of service. Toward that end, our group addressed three overarching questions: (1) What are the high-impact topics? (2) What are the barriers to progress? and (3) What roles can the National Institutes of Health and its various institutes and centers play in fostering the future development of telehealth?