Bos, J. Martijn

Abstract According to the 2020 report of the American Heart Association’s Heart & Stroke Statistics report, nearly 1,000 people are dying daily because of sudden out-of-hospital cardiac arrests and unfortunately, their survival rate is as low as 10%. Hypertrophic Cardiomyopathy (HCM), a relatively rare genetic heart disease is one of these diseases but finding the right patient for the implantation of ICD is still a research question. Implantation of cardioverter-defibrillator (ICD) can save the life of some of these patients. Due to the complexity of the identification of HCM patients, financial burdens, and the clinical risks involved in the ICD implantation procedure, HCM patients will go into a monitoring state before reaching the implantation trigger. Our study cohort shows about 82% of HCM deaths, did not have an ICD, which highlights the need to improve the pre-screening algorithms. In the current paper, we have proposed a new deep learning-based unsupervised clustering technique to facilitate the prioritization of patients to undergo ICD device implantation. This model uses over 900 echocardiographic measurements to find patients who benefit more from the ICD implantation procedure. Our model was trained and tested over 6 years of echo reports collected at Mayo Clinic. This model can be used as a decision support assistant for cardiologists in finding the right HCM patient when decision-making is hard.

Introduction: Hypertrophic cardiomyopathy (HCM) is one of the most common genetic heart diseases characterized by asymmetric left ventricular hypertrophy (LVH) and vast clinical and genetic heterogeneity. Several transcriptional studies in HCM heart tissues have shown severe alterations in gene expression. However, these earlier studies relied on microarray technology or were limited by small sample size or lack of genotype-specific analyses. Methods: We obtained flash-frozen cardiac tissue from patients with HCM undergoing surgical myectomy/septal reduction therapy for relief of left ventricular outflow tract obstruction. RNA-sequencing (RNAseq) was performed using paired-end index reading on 97 myectomy tissues (44 female [45%]; mean age at surgery of 47 ± 18 years) and 23 control tissues (10 female [43%]; mean age at death of 43 ± 12 years) to identify differentially expressed genes (DEGs) using edgeR and pathways in HCM using Ingenuity Pathway Analysis. Additionally, genome sequencing (GS) was performed and HCM patients were divided into genotype-based subgroups to compare the transcriptional profile across different genotypes. Results: Overall, 43% had a pathogenic, HCM-associated variant including 23% with MYBPC3-HCM and 15% with MYH7-HCM. Among the 15,634 genes expressed in the cardiac tissue,1,246 genes (8%) were upregulated and 738 genes (3%) were downregulated significantly in HCM vs. controls. Pathway analysis showed alterations in 126 pathways with downregulation of 44 pathways and upregulation of 1 pathway. There were notable transcriptional alterations of most canonical, pro-hypertrophic pathways, including statistically and biologically significant downregulation of five pathways [-log (BH p-value)≥ 1.3; Z-score≤ -2]: ERK/MAPK, p38 MAPK, ERK5, IL6, and NF-κB signaling. Far less DEGs were observed between genotypic subgroups, with only 37 DEGSs identified between sarcomere positive- and genotype negative-HCM. Conclusions: At the time of surgical myectomy, the ventricular myocardium’s transcriptome, independent of genotype, reveals widespread downregulation of pro-hypertrophic pathways. Whether these counter regulatory measures can be targeted therapeutically warrants further investigation.

Introduction: The main goal of Clinical Decision Support (CDS) is to provide timely and useful information for providers to deliver high-quality and safe care. Issues, like alert fatigue and provider burnout are commonly cited as unintended consequences of the CDS systems. Herein, we describe leveraging electronic health record (EHR) self-service reporting tools to gain insights on a CDS rule designed to avoid providers from ordering QT-prolonging medication in patients with history of QT-prolongation. Methods: Our EHR (Epic) SlicerDicer reporting tool was used to identify the frequency of QT-prolongation Best Practice Advisory (BPA) triggerings from 1/1/2020 to 12/31/2020 across Mayo Clinic sites. Additionally, associated patient encounter and BPA-specific data were queried. Text from acknowledgement comments to the BPA were reviewed and central themes and comments were categorized. Epic Clarity Data Warehouse was queried to identify the BPA-triggering medications and frequencies. Comparisons of percent action taken on the BPA were compared internally to Mayo Clinic’s BPA benchmark, and with national benchmarks using Epic BestPractice Pulse dashboard. Results: A total of 17,538 BPAs of interest were triggered across the Mayo Clinic sites in 2020, of which 3,696 (20.1%) were acted upon through either monitoring or medication discontinuation. This is below Mayo Clinic’s median of 29.2% for inpatient BPAs and 45.6% BPAs for outpatient, but above the national median of 10% (IQR 6.1-13.1%) for inpatient and 10% (IQR 4.3-19.5%) for outpatient BPAs actioned upon. A total of 30 different medications triggered the BPA, with the most common being ondansetron (41.5%), amiodarone (11.3%), and haloperidol (10.1%). User comments were left on 2,382 orders (13.5%) that acknowledged the BPA, but provided argument to prescribe drug. Three most common themes identified from comments on overridden BPAs: patient previously tolerated medication (19.0%), QT interval not prolonged on recent ECG (14.2%), and patient on hospice care (5.2%). Conclusion: Analysis of free text comments of CDS can be and should be performed to reveal opportunities for further improvement of CDS. Providers should be encouraged to provide comments when encountering CDS interventions.

Introduction: Congenital long QT syndrome (LQTS) is a sudden cardiac death-predisposing genetic disorder that has a characteristic electrocardiographic signature. Specific LQTS genotypes have been associated classically with different qualitative electrocardiographic ST - T wave patterns; however, clinical recognition of these patterns could be improved by an automated and quantitative method. We therefore aimed to use a novel, quantitative T wave morphological analysis technique from 12-lead surface ECGs to identify genotype specific LQTS ECG patterns. Methods: We analyzed a genotyped cohort of 420 patients (22 ± 15.7years, 43% male) with either LQT1 (61%) or LQT2 (39%). ECG analysis was conducted using a custom T wave program to automatically detect subtle changes in T wave morphology across all 12 Leads. Classification was performed using the top three T wave features in a linear discriminant classifier by 10x10 cross-validation. Results: The top three discriminant features were (1) a steeper T wave left slope for LQT1 in lead II (2950 vs 1700 mV/ms, p < 0.0001), (2) a greater center of gravity in 1st segment of T wave in lead aVR in LQT1 (0.248 vs 0.209, p<0.0001) and (3) a smaller Tpeak - Tend/QT ratio in lead aVR in LQT1 (0.20 vs 0.24, p<0.0001). Using these top three features, we were able to successfully differentiate LQT1 and LQT2 in 79% of cases (figure 1). Conclusions: An automated morphological T wave analysis was able to discriminate between the two most common LQTS genotypes. This novel method enhances phenotypic characterization of LQT and could have a significant clinical impact on the diagnosis and management of patients with suspected LQTS.

Background: Symptomatic children with catecholaminergic polymorphic ventricular tachycardia (CPVT) are at risk for recurrent arrhythmic events. β-Blockers decrease this risk, but studies comparing individual β-blockers in sizeable cohorts are lacking. We aimed to assess the association between risk for arrhythmic events and type of β-blocker in a large cohort of symptomatic children with CPVT. Methods: From 2 international registries of patients with CPVT, RYR2 variant–carrying symptomatic children (defined as syncope or sudden cardiac arrest before β-blocker initiation and age at start of β-blocker therapy <18 years), treated with a β-blocker were included. Cox regression analyses with time-dependent covariates for β-blockers and potential confounders were used to assess the hazard ratio (HR). The primary outcome was the first occurrence of sudden cardiac death, sudden cardiac arrest, appropriate implantable cardioverter-defibrillator shock, or syncope. The secondary outcome was the first occurrence of any of the primary outcomes except syncope. Results: We included 329 patients (median age at diagnosis, 12 [interquartile range, 7–15] years, 35% females). Ninety-nine (30.1%) patients experienced the primary outcome and 74 (22.5%) experienced the secondary outcome during a median follow-up of 6.7 (interquartile range, 2.8–12.5) years. Two-hundred sixteen patients (66.0%) used a nonselective β-blocker (predominantly nadolol [n=140] or propranolol [n=70]) and 111 (33.7%) used a β1-selective β-blocker (predominantly atenolol [n=51], metoprolol [n=33], or bisoprolol [n=19]) as initial β-blocker. Baseline characteristics did not differ. The HRs for both the primary and secondary outcomes were higher for β1-selective compared with nonselective β-blockers (HR, 2.04 [95% CI, 1.31–3.17]; and HR, 1.99 [95% CI, 1.20–3.30], respectively). When assessed separately, the HR for the primary outcome was higher for atenolol (HR, 2.68 [95% CI, 1.44–4.99]), bisoprolol (HR, 3.24 [95% CI, 1.47–7.18]), and metoprolol (HR, 2.18 [95% CI, 1.08–4.40]) compared with nadolol, but did not differ from propranolol. The HR of the secondary outcome was only higher in atenolol compared with nadolol (HR, 2.68 [95% CI, 1.30–5.55]). Conclusions: β1-selective β-blockers were associated with a significantly higher risk for arrhythmic events in symptomatic children with CPVT compared with nonselective β-blockers, specifically nadolol. Nadolol, or propranolol if nadolol is unavailable, should be the preferred β-blocker for treating symptomatic children with CPVT.

Background: Catecholaminergic polymorphic ventricular tachycardia (CPVT) is a genetic arrhythmia syndrome characterized by adrenergically-triggered ventricular arrhythmias, syncope, and sudden cardiac death. Several small studies suggest that atrial arrhythmias (AAs) are common in patients with CPVT. Objective: To determine the incidence and type of AAs observed within a large, single-center cohort of CPVT cases as well as the efficacy and durability of AA-directed management. Methods: In this retrospective study, the electronic medical record of 129 patients (52% female; average age at diagnosis 20.8  15.3 years) with CPVT (95% with a putative CPVT1-causative RYR2 variant) between 01/2000 and 09/2019 were reviewed for electrocardiographic evidence of AAs. Clinical features and efficacy of pharmacologic and ablation therapy were assessed. Results: Overall, 10/129 (7.8%) CPVT patients, all RYR2 variant-positive, had evidence of an AA (atrial fibrillation/flutter in 6, atrial tachycardia in 3, and supraventricular tachycardia in 1). The median age at AA diagnosis was 23 (14.2-35.5) years. 8/10 of patients experienced symptoms attributed to their AA, including inappropriate shocks. All patients were trialed on anti-arrhythmics, including -blockers, and/or flecainide. Owing to drug failure (1/10), drug intolerance (1/10), or patient preference (2/10); 4/10 patients received an ablation. Over a median follow-up of 23.5 (4.5-63) months, no AA recurrences were observed. Conclusion: Compared to prior studies, the incidence of AAs in this large, single-center referral cohort of CPVT patients was substantially lower (7.8% vs. 26%-35%). Although larger multi-center studies are needed to confirm, this study suggests that ablation is efficacious and durable in CPVT-associated AAs.

Abstract Background Interval duration measurements (IDMs) were compared between standard 12‐lead electrocardiograms (ECGs) and 6‐lead ECGs recorded with AliveCor's KardiaMobile 6L, a hand‐held mobile device designed for use by patients at home. Methods Electrocardiograms were recorded within, on average, 15 min from 705 patients in Mayo Clinic's Windland Smith Rice Genetic Heart Rhythm Clinic. Interpretable 12‐lead and 6‐lead recordings were available for 685 out of 705 (97%) eligible patients. The most common diagnosis was congenital long QT syndrome (LQTS, 343/685 [50%]), followed by unaffected relatives and patients (146/685 [21%]), and patients with other genetic heart diseases, including hypertrophic cardiomyopathy (36 [5.2%]), arrhythmogenic cardiomyopathy (23 [3.4%]), and idiopathic ventricular fibrillation (14 [2.0%]). IDMs were performed by a central ECG laboratory using lead II with a semi‐automated technique. Results Despite differences in patient position (supine for 12‐lead ECGs and sitting for 6‐lead ECGs), mean IDMs were comparable, with mean values for the 12‐lead and 6‐lead ECGs for QTcF, heart rate, PR, and QRS differing by 2.6 ms, −5.5 beats per minute, 1.0 and 1.2 ms, respectively. Despite a modest difference in heart rate, intervals were close enough to allow a detection of clinically meaningful abnormalities. Conclusions The 6‐lead hand‐held device is potentially useful for a clinical follow‐up of remote patients, and for a safety follow‐up of patients participating in clinical trials who cannot visit the investigational site. This technology may extend the use of 12‐lead ECG recordings during the current COVID‐19 pandemic as remote patient monitoring becomes more common in virtual or hybrid‐design clinical studies.

Background We aimed to provide personalized risk estimates for cardiac events (CEs) and life‐threatening events in women with either type 1 or type 2 long QT. Methods and Results The prognostic model was derived from the Rochester Long QT Syndrome Registry, comprising 767 women with type 1 long QT (n=404) and type 2 long QT (n=363) from age 15 through 60 years. The risk prediction model included the following variables: genotype/mutation location, QTc‐specific thresholds, history of syncope, and β‐blocker therapy. A model was developed with the end point of CEs (syncope, aborted cardiac arrest, or long QT syndrome–related sudden cardiac death), and was applied with the end point of life‐threatening events (aborted cardiac arrest, sudden cardiac death, or appropriate defibrillator shocks). External validation was performed with data from the Mayo Clinic Genetic Heart Rhythm Clinic (N=467; type 1 long QT [n=286] and type 2 long QT [n=181]). The cumulative follow‐up duration among the 767 enrolled women was 22 243 patient‐years, during which 323 patients (42%) experienced ≥1 CE. Based on genotype‐phenotype data, we identified 3 risk groups with 10‐year projected rates of CEs ranging from 15%, 29%, to 51%. The corresponding 10‐year projected rates of life‐threatening events were 2%, 5%, and 14%. C statistics for the prediction model for the 2 respective end points were 0.68 (95% CI 0.65–0.71) and 0.71 (95% CI 0.66–0.76). Corresponding C statistics for the model in the external validation Mayo Clinic cohort were 0.65 (95% CI 0.60–0.70) and 0.77 (95% CI 0.70–0.84). Conclusions This is the first risk prediction model that provides absolute risk estimates for CEs and life‐threatening events in women with type 1 or type 2 long QT based on personalized genotype‐phenotype data. The projected risk estimates can be used to guide female‐specific management in long QT syndrome.

Introduction: Long QT syndrome (LQTS) is a genetic disorder associated with abnormal ventricular repolarization and a predisposition for torsade de pointes and sudden cardiac death. Identification of markers to elucidate high risk patients is of major practical importance. Our aim was to use a novel, automated quantitative T wave analysis program to distinguish symptomatic from asymptomatic LQTS patients. Methods: We analyzed a genotyped cohort of 420 patients (22 ± 15.7years, 43% male) with either LQT1 (61%) or LQT2 (39%). ECG analysis was conducted using novel software that we created to automatically detect subtle changes in T wave morphology. Symptomatic patients were defined as those with a history of syncope or resuscitated cardiac arrest that was suspected of being LQTS triggered. Classification was performed using the top three T wave features in a linear discriminant classifier by 10x10 cross-validation. Results: Symptomatic LQT1 patients have a greater QTc duration in lead aVF (437 msec vs 416 msec, p <0.0001), a greater T wave center of gravity on the x axis (COGx) in lead V2 (0.304 vs 0.280, p <0.0001) and longer Tpeak-Tend/QT in lead II (0.208 vs 0.188, p < 0.0001) than asymptomatic patients with LQT1. Use of these 3 features enabled classification of symptomatic status in 72% of LQT1 cases, compared to 68% when QTc was used to classify alone. Symptomatic LQT2 patients had a longer QTc in Lead 1 (460 vs 418 msec, p < 0.0001), COGx in 4th segment of T wave was greater in lead V5 (0.399 vs 0.350, p<0.0001), and the Tpeak-Tend interval in V2 was longer (126 vs 99 msec, P =0.0004). Use of these 3 features enabled correct classification in 73% of LQT2 cases, compared to 67% when QTc was used alone. Conclusions: Application of an automated T wave analysis program was able to identify T wave features that distinguished the patients who presented as symptomatic LQTS, associated with syncope or resuscitated cardiac arrest. Use of these features could help assess arrhythmic risk and guide management strategies for patients with LQTS.

Background: Mutations in sarcomeric/myofilament proteins are commonly implicated in hypertrophic cardiomyopathy (HCM). However, the mechanisms underlying disease pathogenesis are not completely understood. Utilizing flash-frozen surgically resected left ventricular (LV) tissue from patients with obstructive HCM, we investigated the effect of sarcomeric HCM on tissue expression of various myofilament proteins. Methods: Frozen surgical specimens from 47 patients with HCM who underwent septal myectomy at the Mayo Clinic between 1999 and 2003 were examined. Using PCR, DHPLC, and direct DNA sequencing, all were genotyped for mutations involving the 8 myofilament-encoding genes that currently comprise the clinically available genetic test for sarcomeric HCM. Whole protein lysate was extracted from the frozen myectomy tissues and levels of myosin binding protein C, myosin heavy chain, alpha actin and alpha actinin were analyzed by Western Blot; GAPDH was used as a loading control. Frozen LV tissue derived from normal hearts of accidental death victims served as a reference group. Results: Overall, 25/47 (53%) patients had sarcomeric HCM including 13 patients with myosin binding protein C (MYBPC3) mutations and 10 with mutations in the beta myosin heavy chain (MYH7). Compared to normal heart tissue, the expression of myosin heavy chain was up-regulated significantly regardless of the underlying disease-causing genotype (genotype positive = genotype negative). However, there was gene-specific upregulation of myosin binding protein C with marked increases detected in patients with MYH7-HCM. In contrast, normal levels of myosin binding protein C were seen in patients with either MYBPC3-HCM or genotype-negative HCM. Conclusions: This is the first study to examine the expression pattern of sarcomeric proteins in a large collection of flash frozen human HCM myectomy tissue. Up-regulated, genotype-specific cardiac tissue expression of myosin binding protein C among patients with MYH7 (beta myosin heavy chain)-HCM suggests a novel compensatory pathway underlying the development of cardiac hypertrophy for one of most common genetic subtypes of HCM.