Search

Sudden cardiac death from arrhythmia is a major cause of mortality worldwide. In this study, we developed a novel deep learning (DL) approach that blends neural networks and survival analysis to predict patient-specific survival curves from contrast-enhanced cardiac magnetic resonance images and clinical covariates for patients with ischemic heart disease. The DL-predicted survival curves offer accurate predictions at times up to 10 years and allow for estimation of uncertainty in predictions. The performance of this learning architecture was evaluated on multi-center internal validation data and tested on an independent test set, achieving concordance indexes of 0.83 and 0.74 and 10-year integrated Brier scores of 0.12 and 0.14. We demonstrate that our DL approach, with only raw cardiac images as input, outperforms standard survival models constructed using clinical covariates. This technology has the potential to transform clinical decision-making by offering accurate and generalizable predictions of patient-specific survival probabilities of arrhythmic death over time.

BACKGROUND: Emerging evidence has linked sleep behaviors with the risk of cardiac arrhythmias. The various sleep behaviors are typically correlated; however, most of the previous studies only focused on the individual sleep behavior, without considering the overall sleep patterns. OBJECTIVE: To prospectively investigate the associations between a healthy sleep pattern with the risks of cardiac arrhythmias. METHODS: A total of 403,187 participants from UK Biobank were included. A healthy sleep pattern was defined by chronotype, sleep duration, insomnia, snoring, and daytime sleepiness. Weighted genetic risk score (GRS) for atrial fibrillation was calculated. RESULTS: The healthy sleep pattern was significantly associated with lower risks of AF [hazard ratio (HR) comparing extreme categories: 0.71, 95% confidence interval (CI): 0.64–0.80] and bradyarrhythmia (HR: 0.65, 95% CI: 0.54–0.77), but not ventricular arrhythmias, after adjustment for demographic, lifestyle, and genetic risk factors. Compared with individuals with a healthy sleep score of 0–1 (poor sleep group), those with a healthy sleep score of 5 had a 29% and 35% lower risk of developing AF and bradyarrhythmia, respectively. Additionally, the genetic predisposition to AF significantly modified the association of the healthy sleep pattern with the risk of AF (p-interaction=0.017). The inverse association of the healthy sleep pattern with the risk of AF was stronger among those with a lower genetic risk of AF. CONCLUSIONS: Our results indicate that a healthy sleep pattern is associated with lower risks of AF and bradyarrhythmia, independent of the traditional risk factors; and the association with AF is modified by genetic susceptibility.

Background We conducted a prospective multicenter study to assess the prognostic value of combined baseline preimplant plasma levels of the biomarkers cardiac troponin T (TnT) and B-type natriuretic peptide (BNP) among cardiac resynchronization therapy (CRT) with or without defibrillator capability (CRT-D) recipients. Methods At CRT-D implant, patients were stratified based on detectable TnT (≥0.01 ng/mL) and elevated BNP (predefined as >440 pg/mL) levels. Patients were classified into three groups: high (both detectable TnT and high BNP), intermediate (either detectable TnT or high BNP), or low (nondetectable TnT and low BNP). Patients were followed for 12 months. Survival curves free from mortality or heart failure hospitalizations (HFH) were assessed. To assess the predictive value of biomarker category, we constructed a multivariate Cox regression model, including the covariates of age, New York Heart Association class, left ventricular ejection fraction (LVEF), and QRS duration. Results A total of 267 patients (age 66 ± 12 years, males 80%, LVEF 25% ± 8%, ischemic cardiomyopathy 52%, QRSd 155 ± 26 ms) were studied. After 1 year, there were 13 deaths and 25 HFH events. A significant difference in event-free survival among the three groups was observed, with high and intermediate categories having worse survival than low (log-rank test, P < 0.001). In the multivariate model, risk category was a significant predictor of outcome: hazard ratios were 7.34 (95% confidence interval [CI]: 2.48–21.69) and 2.50 (95% confidence interval [CI]: 1.04–6.04) for high-risk and intermediate-risk groups, respectively (P < 0.0001). Conclusion Among CRT-D recipients, baseline TnT and BNP values alone or in combination provide significant prognostic value for the outcome of mortality or HFH.

Introduction: Primary prevention implantable cardioverter defibrillators (ICDs) reduce mortality in select patients with left ventricular systolic dysfunction (LVSD). The occurrence of device therapies after ICD implantation in contemporary clinical practice is not well described, especially in patient subgroups designated in the Centers for Medicare and Medicaid Services (CMS) 2005 Coverage with Evidence Development (CED). Methods: The Longitudinal Study of ICDs assessed rates and correlates of device therapies (overall and those requiring shock) up to 3 years post-implant among 2540 patients with LVSD receiving first-time primary prevention ICDs in 7 US health care systems from 2006-2009. Implant data from the National Cardiovascular Data Registry ICD Registry were linked to electronic longitudinal health care system data and a novel centrally adjudicated repository of device therapies abstracted from medical records. Proportional hazard models evaluated associations with device therapies by appropriateness, adjusted for demographic and clinical factors. Results: Subjects were 26% women, 35% =65 years (1.40, 1.04-1.88), and lower among 2009 implants compared to 2006 (0.66, 0.46-0.95). Regarding the 3 CMS CED patient subgroups, neither ejection fraction nor heart failure symptom severity was associated with device therapy; for LVSD etiology, the rate of inappropriate therapy was nominally but not significantly higher among subjects with non-ischemic cardiomyopathy of < 9 months duration (1.38, 0.88-2.18). Conclusions: In a representative cohort of primary prevention ICD patients in usual care, rates of device therapies were somewhat lower than in landmark trials, differed by some patient characteristics, but did not differ meaningfully for CMS CED subgroups.

Electrocardiographic and electrophysiologic differences between men and women have long been noted. Women have a higher intrinsic heart rate than men, along with a longer corrected QT interval and a shorter sinus nodal recovery time. The incidence of and risk factors for a variety of arrhythmias differ between men and women. Atrioventricular nodal reentry tachycardia has a 2:1 female-to-male predominance, while accessory pathways are twice as frequent in men. Although atrial fibrillation is more prevalent in men of all age groups, the absolute numbers of men and women with atrial fibrillation are equal, and the associated morbidity and mortality experienced by women with atrial fibrillation appear to be worse. Women have a lower incidence of sudden cardiac death, and female survivors of sudden cardiac death have a lower frequency of spontaneous or inducible ventricular tachycardia. On the other hand, drug-induced torsade de pointes and symptomatic long QT syndrome have a female predominance. Therefore, greater caution should be used when prescribing QT-prolonging drugs in women. The incidence of arrhythmias is increased during pregnancy, and management of pregnant patients poses a significant challenge. The mechanisms of these gender differences are unclear but may be related to hormonal effects and the shorter QT interval in adult males. Pharmacologic and nonpharmacologic therapies are usually equally efficacious, but the risks of pharmacologic therapy are different in men and women. Atrial fibrillation may be more difficult to treat in women.

We conducted a prospective multicenter study to assess the prognostic value of combined baseline preimplant plasma levels of the biomarkers cardiac troponin T (TnT) and B-type natriuretic peptide (BNP) among cardiac resynchronization therapy (CRT) with or without defibrillator capability (CRT-D) recipients.At CRT-D implant, patients were stratified based on detectable TnT (≥0.01 ng/mL) and elevated BNP (predefined as440 pg/mL) levels. Patients were classified into three groups: high (both detectable TnT and high BNP), intermediate (either detectable TnT or high BNP), or low (nondetectable TnT and low BNP). Patients were followed for 12 months. Survival curves free from mortality or heart failure hospitalizations (HFH) were assessed. To assess the predictive value of biomarker category, we constructed a multivariate Cox regression model, including the covariates of age, New York Heart Association class, left ventricular ejection fraction (LVEF), and QRS duration.A total of 267 patients (age 66 ± 12 years, males 80%, LVEF 25% ± 8%, ischemic cardiomyopathy 52%, QRSd 155 ± 26 ms) were studied. After 1 year, there were 13 deaths and 25 HFH events. A significant difference in event-free survival among the three groups was observed, with high and intermediate categories having worse survival than low (log-rank test, P0.001). In the multivariate model, risk category was a significant predictor of outcome: hazard ratios were 7.34 (95% confidence interval [CI]: 2.48-21.69) and 2.50 (95% confidence interval [CI]: 1.04-6.04) for high-risk and intermediate-risk groups, respectively (P0.0001).Among CRT-D recipients, baseline TnT and BNP values alone or in combination provide significant prognostic value for the outcome of mortality or HFH.

Background: Prolonged QT offset dispersion (QToD), an index of heterogeneity of ventricular repolarization, is thought to be an independent predictor of all-cause and cardiovascular mortality. However the influence of gender and autonomic tone in healthy adults on age-related changes in measures of ventricular repolarization are not well characterized. Methods: QToD and T wave complexity were measured in 56 healthy subjects with no detectable heart disease (by echo and stress test)—38 young subjects with a mean age of 28 ± 4 years and 18 old subjects with a mean age of 71 ± 7 years. QToD and T wave complexity were computed from 12-lead ECGs using the GE Marquette QT Guard automated analysis program with manual over-reading at rest (baseline), following exercise, and double autonomic blockade with atropine and propranolol. Data was analyzed using factorial ANOVA. Results: Young males had a significantly greater QToD than young and old females at baseline (28 ± 5 ms, 23 ± 5 ms, and 22 ± 5 ms, respectively, P < 0.01), an intrinsic effect independent of changes in autonomic tone. In contrast, females had significantly greater T wave complexity than males following exercise and double autonomic blockade with a definite trend at baseline. There was no correlation between T wave complexity and QToD. Conclusions: Age and gender demonstrate a complex interaction on indices of myocardial repolarization with different measures behaving differently. These findings have implications for better understanding age and gender effects on myocardial electrophysiology. A.N.E. 2001;6(4):290–297

Peri-infarct border zone (BZ) as quantified by delayed-enhancement cardiac MRI (CMR) has been proposed as a risk stratification tool, and is associated with increased mortality, but has been measured by various methods in the literature. We assessed which BZ analysis best predicts inducible arrhythmia during electrophysiological study (EPS). Methods: CMR was performed in 47 patients with CAD referred for EPS to assess for ventricular tachyarrhythmias (VT). CMR data was analyzed for BZ quantification by three methods. Method I used pixels 2–3 standard deviations over the mean of normal tissue, expressed as % of LV mass. Method II (Yan) used the same pixels but is expressed as % of infarct size. Method III (Schmidt) used a full-width, half-max method. EPS results were classified as negative (non-inducible) or positive (monomorphic VT). Patients with indeterminate EPS outcomes (VF or polymorphic VT, n=9) were not included in this analysis. Results: There were 38 subjects-age 60.2 ± 10.8 yrs, 68% male. During EPS, 20 patients were non-inducible and 18 had induced monomorphic VT (MVT). EF was not significantly different between non-inducible patients and those with MVT (33.4 ± 11.9% vs. 28.1 ± 9.3, p=0.13). Infarct size was significantly different (15.0 ± 11.6% vs. 26.4 ± 12.0%, p = 0.005). BZ by method I was significantly different (1.4 ± 1.3% vs. 3.1 ± 1.5%, p = 0.001), but not by method II (15.2 ± 21.3% vs. 8.0 ± 4.8%, p = 0.17) or method III (4.1 ± 4.5g vs. 3.7 ± 3.7g, p = 0.756). Multivariate analysis demonstrated that BZ by method I was an independent predictor of EPS outcome after controlling for infarct size (OR 1.97 per % change, 95% CI 1.04 to 3.73, p = 0.04). Method I BZ remained an independent predictor when controlling for both infarct size and EF, (OR 2.12 per % change, 95% CI 1.06 to 34.22, p = 0.03). Correlation between techniques was variable (Method I vs. Method II R=0.77, p=0.001; Method I vs. Method III R=0.03, p=0.86; Method II vs. Method III R=0.15, p Conclusion: This study demonstrates significant variability between the published methods for measuring BZ. We also show that large BZ by Method I is a stronger predictor of inducible MVT during EPS than EF and infarct size. BZ may be yet another CMR marker of elevated risk of arrhythmia, but requires more rigorous definition.

Background: The sympathetic nervous system (SNS) is thought to play an important role in promoting atrial fibrillation (AF) initiation and maintenance, especially in the setting of congestive heart failure (CHF). However, it is unclear as to which components of the SNS are altered in atria isolated from failing hearts. These studies are essential, as they will delineate potential molecular targets that may be probed for therapeutic intervention in order to reduce AF pathogenesis. Methods : CHF was induced by ventricular tachypacing (240 beats/minute) for 4 weeks in hound dogs. A total of 4 CHF and 5 control dogs were used. Immunoblot analysis was performed on left atrial appendage (LAA) homogenates to examine the protein expression of Gsα, Gi2α, and the catalytic subunit of protein kinase A (PKA). Catalytic activities of PKA and protein phosphatase type 2a (PP2a) were determined using non-radioactive enzyme activity assay systems. Cyclic AMP (cAMP) accumulation was assessed by an enzyme linked immunoabsorbant assay and phosphorylation of cardiac troponin I (cTnI) was determined using a phosphospecific antibody against cTnI. Results: Gsαconcentrations were increased approximately 4-fold in failing left atrial preparations relative to control. In contrast, expression of Gi2αwas modestly increased ~1-fold in failing LAA relative to non-failing atrial muscles. PKA activity was significantly increased in failing LAA preparations compared to control (124.7 vs. 42.1 pmole phosphate/minute, P P Conclusions: Our data are the first to characterize sympathetic remodeling in left atrial muscles isolated from an experimental model of CHF. Additionally, our results support the hypothesis that augmented sympathetic signaling and activation may contribute to the AF substrate in a canine model of CHF characterized by increased AF susceptibility.

Editorial Comment The treatment of atrial fibrillation (AF) can be challenging, with a rate control strategy not being optimal for all patients and antiarrhythmic drugs frequently ineffective and possessing serious adverse effects. A fluoroscopic or electroanatomic catheter-based approach is rapidly becoming one of the preferred treatments of AF. 1-3 However, success rates are still not ideal and complications due to damage of extra thoracic structures remain a problem. 1 Also of concern is the significant ionizing radiation exposure during the prolonged fluoroscopy required to delineate the complex anatomy of the left atrium and pulmonary veins, even with advanced techniques to minimize exposure. Total fluoroscopy duration for AF ablations is typically 8-fold higher than atrioventricular nodal reentrant tachycardia ablations, averaging over 60 minutes of fluoroscopy time, with radiation exposure in the 1.0‐1.5 Gy range. 4 One of the major promises of using integratedmodality navigation systems such as this for AF ablation is faster, more accurate catheter navigation, reduced complications, and diminished fluoroscopy use. In view of the complex and variable anatomy of cardiac structures and pulmonary veins, many of which remain invisible to fluoroscopy, the addition of further, more accurate imaging modalities are likely to improve ablation success and safety. 5,6 While three-dimensional (3D) mapping techniques using magnetic or electrical localizing systems in the modern electrophysiology laboratory are extremely useful and are currently in widespread clinical use, they require catheter manipulation to identify geometry and do not provide independent anatomic information. Both magnetic resonance (MR) imaging 7,8 and computed tomography (CT) scanning 9,10 have been shown prior to be suitable for delineating the left atrial anatomy and assisting with catheter navigation. However, it is not yet clear which imaging modality is superior for mapping of the left atrium and other cardiac structures; certainly navigation and imaging systems that allow the use of either format will be advantageous.

Whether primary prevention implantable cardioverter-defibrillators (ICDs) are efficacious among older patients is uncertain. We merged patient-level data from 5 primary prevention trials randomizing patients to an ICD or usual care: the Multicenter Automatic Defibrillator Implantation Trial I, the