Late-Gadolinium Enhancement Interface Area and Electrophysiological Simulations Predict Arrhythmic Events in Patients With Nonischemic Dilated Cardiomyopathy

The Uncertain Benefit from Implantable Cardioverter-Defibrillators in Nonischemic Cardiomyopathy: How to Guide Clinical Decision-Making?

Life-threatening dysrhythmias remain a significant cause of mortality in patients with nonischemic cardiomyopathy (NICM). Implantable cardioverter-defibrillators (ICD) effectively reduce mortality in patients who have survived a life-threatening arrhythmic event. The evidence for survival benefit of primary prevention ICD for patients with high-risk NICM on guideline-directed medical therapy is not as robust, with efficacy questioned by recent studies. In this review, we summarize the data on the risk of life-threatening arrhythmias in NICM, the recommendations, and the evidence supporting the efficacy of primary prevention ICD, and highlight tools that may improve the identification of patients who could benefit from primary prevention ICD implantation.

Prediction of sudden cardiac death using artificial intelligence: Current status and future directions

Sudden cardiac death (SCD) remains a pressing health issue, affecting hundreds of thousands each year globally. The heterogeneity among SCD victims, ranging from individuals with severe heart failure to seemingly healthy individuals, poses a significant challenge for effective risk assessment. Conventional risk stratification, which primarily relies on left ventricular ejection fraction, has resulted in only modest efficacy of implantable cardioverter-defibrillators (ICD) for SCD prevention. In response, artificial intelligence (AI) holds promise for personalised SCD risk prediction and tailoring preventive strategies to the unique profiles of individual patients. Machine and deep learning algorithms have the capability to learn intricate non-linear patterns between complex data and defined endpoints, and leverage these to identify subtle indicators and predictors of SCD that may not be apparent through traditional statistical analysis. However, despite the potential of AI to improve SCD risk stratification, there are important limitations that need to be addressed. We aim to provide an overview of the current state-of-the-art of AI prediction models for SCD, highlight the opportunities for these models in clinical practice, and identify the key challenges hindering widespread adoption.

Predicting postinfarct ventricular tachycardia by integrating cardiac MRI and advanced computational reentrant pathway analysis

BACKGROUND

Implantable cardiac defibrillator (ICD) implantation can protect against sudden cardiac death after myocardial infarction. However, improved risk stratification for device requirement is still needed.

OBJECTIVE

The purpose of this study was to improve assessment of postinfarct ventricular electropathology and prediction of appropriate ICD therapy by combining late gadolinium enhancement (LGE) and advanced computational modeling.

METHODS

ADAS 3D LV (ADAS LV Medical, Barcelona, Spain) and custom-made software were used to generate 3-dimensional patient-specific ventricular models in a prospective cohort of patients with a myocardial infarction (N = 40) having undergone LGE imaging before ICD implantation. Corridor metrics and 3-dimensional surface features were computed from LGE images. The Virtual Induction and Treatment of Arrhythmias (VITA) framework was applied to patient-specific models to comprehensively probe the vulnerability of the scar substrate to sustaining reentrant circuits. Imaging and VITA metrics, related to the numbers of induced ventricular tachycardias and their corresponding round trip times (RTTs), were compared with ICD therapy during follow-up.

RESULTS

Patients with an event (n = 17) had a larger interface between healthy myocardium and scar and higher VITA metrics. Cox regression analysis demonstrated a significant independent association with an event: interface (hazard ratio [HR] 2.79; 95% confidence interval [CI] 1.44-5.44; P < .01), unique ventricular tachycardias (HR 1.67; 95% CI 1.04-2.68; P = .03), mean RTT (HR 2.14; 95% CI 1.11-4.12; P = .02), and maximum RTT (HR 2.13; 95% CI 1.19-3.81; P = .01).

CONCLUSION

A detailed quantitative analysis of LGE-based scar maps, combined with advanced computational modeling, can accurately predict ICD therapy and could facilitate the early identification of high-risk patients in addition to left ventricular ejection fraction.

Imagenetics for Precision Medicine in Dilated Cardiomyopathy

Dilated cardiomyopathy (DCM) is a common heart muscle disorder of nonischemic etiology associated with heart failure development and the risk of malignant ventricular arrhythmias and sudden cardiac death. A tailored approach to risk stratification and prevention of sudden cardiac death is required in genetic DCM given its variable presentation and phenotypic severity. Currently, advances in cardiogenetics have shed light on disease mechanisms, the complex genetic architecture of DCM, polygenic contributors to disease susceptibility and the role of environmental triggers. Parallel advances in imaging have also enhanced disease recognition and the identification of the wide spectrum of phenotypes falling under the DCM umbrella. Genotype-phenotype associations have been also established for specific subtypes of DCM, such as DSP (desmoplakin) or FLNC (filamin-C) cardiomyopathy but overall, they remain elusive and not readily identifiable. Also, despite the accumulated knowledge on disease mechanisms, certain aspects remain still unclear, such as which patients with DCM are at risk for disease progression or remission after treatment. Imagenetics, that is, the combination of imaging and genetics, is expected to further advance research in the field and contribute to precision medicine in DCM management and treatment. In the present article, we review the existing literature in the field, summarize the established knowledge and emerging data on the value of genetics and imaging in establishing genotype-phenotype associations in DCM and in clinical decision making for DCM patients.

Fibrosis modeling choice affects morphology of ventricular arrhythmia in non-ischemic cardiomyopathy

Introduction: Patients with non-ischemic cardiomyopathy (NICM) are at risk for ventricular arrhythmias, but diagnosis and treatment planning remain a serious clinical challenge. Although computational modeling has provided valuable insight into arrhythmic mechanisms, the optimal method for simulating reentry in NICM patients with structural disease is unknown. Methods: Here, we compare the effects of fibrotic representation on both reentry initiation and reentry morphology in patient-specific cardiac models. We investigate models with heterogeneous networks of non-conducting structures (cleft models) and models where fibrosis is represented as a dense core with a surrounding border zone (non-cleft models). Using segmented cardiac magnetic resonance with late gadolinium enhancement (LGE) of five NICM patients, we created 185 3D ventricular electrophysiological models with different fibrotic representations (clefts, reduced conductivity and ionic remodeling). Results: Reentry was induced by electrical pacing in 647 out of 3,145 simulations. Both cleft and non-cleft models can give rise to double-loop reentries meandering through fibrotic regions (Type 1-reentry). When accounting for fibrotic volume, the initiation sites of these reentries are associated with high local fibrotic density (mean LGE in cleft models: p< 0.001, core volume in non-cleft models: p = 0.018, negative binomial regression). In non-cleft models, Type 1-reentries required slow conduction in core tissue (non-cleftsc models) as opposed to total conduction block. Incorporating ionic remodeling in fibrotic regions can give rise to single- or double-loop rotors close to healthy-fibrotic interfaces (Type 2-reentry). Increasing the cleft density or core-to-border zone ratio in cleft and non-cleftc models, respectively, leads to increased inducibility and a change in reentry morphology from Type 2 to Type 1. Conclusions: By demonstrating how fibrotic representation affects reentry morphology and location, our findings can aid model selection for simulating arrhythmogenesis in NICM.

The Uncertain Benefit from Implantable Cardioverter-Defibrillators in Nonischemic Cardiomyopathy: How to Guide Clinical Decision-Making?

Life-threatening dysrhythmias remain a significant cause of mortality in patients with nonischemic cardiomyopathy (NICM). Implantable cardioverter-defibrillators (ICD) effectively reduce mortality in patients who have survived a life-threatening arrhythmic event. The evidence for survival benefit of primary prevention ICD for patients with high-risk NICM on guideline-directed medical therapy is not as robust, with efficacy questioned by recent studies. In this review, we summarize the data on the risk of life-threatening arrhythmias in NICM, the recommendations, and the evidence supporting the efficacy of primary prevention ICD, and highlight tools that may improve the identification of patients who could benefit from primary prevention ICD implantation.

Impact of Cardiac Magnetic Resonance to Arrhythmic Risk Stratification in Nonischemic Cardiomyopathy

Left ventricular ejection fraction-based arrhythmic risk stratification in nonischemic cardiomyopathy (NICM) is insufficient and has led to the failure of primary prevention implantable cardioverter defibrillator trials, mainly due to the inability of selecting patients at high risk for sudden cardiac death (SCD). Cardiac magnetic resonance offers unique opportunities for tissue characterization and has gained a central role in arrhythmic risk stratification in NICM. The presence of myocardial scar, denoted by late gadolinium enhancement, is a significant, independent, and strong predictor of ventricular arrhythmias and SCD with high negative predictive value. T1 maps and extracellular volume fraction, which are able to quantify diffuse fibrosis, hold promise as complementary tools but need confirmatory results from large studies.

Machine learning analysis of complex late gadolinium enhancement patterns to improve risk prediction of major arrhythmic events

Background

Machine learning analysis of complex myocardial scar patterns affords the potential to enhance risk prediction of life-threatening arrhythmia in stable coronary artery disease (CAD).

Objective

To assess the utility of computational image analysis, alongside a machine learning (ML) approach, to identify scar microstructure features on late gadolinium enhancement cardiovascular magnetic resonance (LGE-CMR) that predict major arrhythmic events in patients with CAD.

Methods

Patients with stable CAD were prospectively recruited into a CMR registry. Shape-based scar microstructure features characterizing heterogeneous ('peri-infarct') and homogeneous ('core') fibrosis were extracted. An ensemble of machine learning approaches were used for risk stratification, in addition to conventional analysis using Cox modeling.

Results

Of 397 patients (mean LVEF 45.4 ± 16.0) followed for a median of 6 years, 55 patients (14%) experienced a major arrhythmic event. When applied within an ML model for binary classification, peri-infarct zone (PIZ) entropy, peri-infarct components and core interface area outperformed a model representative of the current standard of care (LVEF<35% and NYHA>Class I): AUROC (95%CI) 0.81 (0.81-0.82) vs. 0.64 (0.63-0.65), p = 0.002. In multivariate cox regression analysis, these features again remained significant after adjusting for LVEF<35% and NYHA>Class I: PIZ entropy hazard ratio (HR) 1.88, 95% confidence interval (CI) 1.38-2.56, p < 0.001; number of PIZ components HR 1.34, 95% CI 1.08-1.67, p = 0.009; core interface area HR 1.6, 95% CI 1.29-1.99, p = <0.001.

Conclusion

Machine learning models using LGE-CMR scar microstructure improved arrhythmic risk stratification as compared to guideline-based clinical parameters; highlighting a potential novel approach to identifying candidates for implantable cardioverter defibrillators in stable CAD.

Predicting arrhythmia recurrence following catheter ablation for ventricular tachycardia using late gadolinium enhancement magnetic resonance imaging: Implications of varying scar ranges

BACKGROUND

Thresholding-based analysis of late gadolinium enhancement cardiac magnetic resonance (LGE-CMR) can create scar maps and identify corridors that might provide a reentrant substrate for ventricular tachycardia (VT). Current recommendations use a full-width-at-half-maximum approach, effectively classifying areas with a pixel signal intensity (PSI) >40% as border zone (BZ) and >60% as core.

OBJECTIVE

The purpose of this study was to investigate the impact of 4 different threshold settings on scar and corridor quantification and to correlate this with postablation VT recurrence.

METHODS

Twenty-seven patients with ischemic cardiomyopathy who had undergone catheter ablation for VT were included for retrospective analysis. LGE-CMR images were analyzed using ADAS3D LV. Scar maps were created for 4 PSI thresholds (40-60, 35-65, 30-70, and 45-55), and the extent of variation in BZ and core, as well as the number and weight of conduction corridors, were quantified. Three-dimensional representations were reconstructed from exported segmentations and used to quantify the surface area between healthy myocardium and scar (BZ + core), and between BZ and core.

RESULTS

A wider PSI threshold was associated with an increase in BZ mass and decrease in scar (P <.001). No significant differences were observed for the total number of corridors and their mass with increasing PSI threshold. The best correlation in predicting arrhythmia recurrence was observed for PSI 45-55 (area under the curve 0.807; P = .001).

CONCLUSION

Varying PSI has a significant impact on quantification of LGE-CMR parameters and may have incremental clinical value in predicting arrhythmia recurrence. Further prospective investigation is warranted to clarify the functional implications of these findings for LGE-CMR-guided ventricular ablation.

Left ventricular shape predicts arrhythmic risk in fibrotic dilated cardiomyopathy

AIMS

Remodelling of the left ventricular (LV) shape is one of the hallmarks of non-ischaemic dilated cardiomyopathy (DCM) and may contribute to ventricular arrhythmias and sudden cardiac death. We sought to investigate a novel three dimensional (3D) shape analysis approach to quantify LV remodelling for arrhythmia prediction in DCM.

METHODS AND RESULTS

We created 3D LV shape models from end-diastolic cardiac magnetic resonance images of 156 patients with DCM and late gadolinium enhancement (LGE). Using the shape models, principle component analysis, and Cox-Lasso regression, we derived a prognostic LV arrhythmic shape (LVAS) score which identified patients who reached a composite arrhythmic endpoint of sudden cardiac death, aborted sudden cardiac death, and sustained ventricular tachycardia. We also extracted geometrical metrics to look for potential prognostic markers. During a follow-up period of up to 16 years (median 7.7, interquartile range: 3.9), 25 patients met the arrhythmic endpoint. The optimally prognostic LV shape for predicting the time-to arrhythmic event was a paraboloidal longitudinal profile, with a relatively wide base. The corresponding LVAS was associated with arrhythmic events in univariate Cox regression (hazard ratio = 2.0 per quartile; 95% confidence interval: 1.3-2.9), in univariate Cox regression with propensity score adjustment, and in three multivariate models; with LV ejection fraction, New York Heart Association Class III/IV (Model 1), implantable cardioverter-defibrillator receipt (Model 2), and cardiac resynchronization therapy (Model 3).

CONCLUSION

Biomarkers of LV shape remodelling in DCM can help to identify the patients at greatest risk of lethal ventricular arrhythmias.

Determining anatomical and electrophysiological detail requirements for computational ventricular models of porcine myocardial infarction

BACKGROUND

Computational models of the heart built from cardiac MRI and electrophysiology (EP) data have shown promise for predicting the risk of and ablation targets for myocardial infarction (MI) related ventricular tachycardia (VT), as well as to predict paced activation sequences in heart failure patients. However, most recent studies have relied on low resolution imaging data and little or no EP personalisation, which may affect the accuracy of model-based predictions.

OBJECTIVE

To investigate the impact of model anatomy, MI scar morphology, and EP personalisation strategies on paced activation sequences and VT inducibility to determine the level of detail required to make accurate model-based predictions.

METHODS

Imaging and EP data were acquired from a cohort of six pigs with experimentally induced MI. Computational models of ventricular anatomy, incorporating MI scar, were constructed including bi-ventricular or left ventricular (LV) only anatomy, and MI scar morphology with varying detail. Tissue conductivities and action potential duration (APD) were fitted to 12-lead ECG data using the QRS duration and the QT interval, respectively, in addition to corresponding literature parameters. Paced activation sequences and VT induction were simulated. Simulated paced activation and VT inducibility were compared between models and against experimental data.

RESULTS

Simulations predict that the level of model anatomical detail has little effect on simulated paced activation, with all model predictions comparing closely with invasive EP measurements. However, detailed scar morphology from high-resolution images, bi-ventricular anatomy, and personalized tissue conductivities are required to predict experimental VT outcome.

CONCLUSION

This study provides clear guidance for model generation based on clinical data. While a representing high level of anatomical and scar detail will require high-resolution image acquisition, EP personalisation based on 12-lead ECG can be readily incorporated into modelling pipelines, as such data is widely available.

Anatomically informed deep learning on contrast-enhanced cardiac magnetic resonance imaging for scar segmentation and clinical feature extraction

Background

Visualizing fibrosis on cardiac magnetic resonance (CMR) imaging with contrast enhancement (late gadolinium enhancement; LGE) is paramount in characterizing disease progression and identifying arrhythmia substrates. Segmentation and fibrosis quantification from LGE-CMR is intensive, manual, and prone to interobserver variability. There is an unmet need for automated LGE-CMR image segmentation that ensures anatomical accuracy and seamless extraction of clinical features.

Objective

This study aimed to develop a novel deep learning solution for analysis of contrast-enhanced CMR images that produces anatomically accurate myocardium and scar/fibrosis segmentations and uses these to calculate features of clinical interest.

Methods

Data sources were 155 2-dimensional LGE-CMR patient scans (1124 slices) and 246 synthetic "LGE-like" scans (1360 slices) obtained from cine CMR using a novel style-transfer algorithm. We trained and tested a 3-stage neural network that identified the left ventricle (LV) region of interest (ROI), segmented ROI into viable myocardium and regions of enhancement, and postprocessed the segmentation results to enforce conforming to anatomical constraints. The segmentations were used to directly compute clinical features, such as LV volume and scar burden.

Results

Predicted LV and scar segmentations achieved 96% and 75% balanced accuracy, respectively, and 0.93 and 0.57 Dice coefficient when compared to trained expert segmentations. The mean scar burden difference between manual and predicted segmentations was 2%.

Conclusion

We developed and validated a deep neural network for automatic, anatomically accurate expert-level LGE- CMR myocardium and scar/fibrosis segmentation, allowing direct calculation of clinical measures. Given the training set heterogeneity, our approach could be extended to multiple imaging modalities and patient pathologies.

Combined In-silico and Machine Learning Approaches Toward Predicting Arrhythmic Risk in Post-infarction Patients

Background: Remodeling due to myocardial infarction (MI) significantly increases patient arrhythmic risk. Simulations using patient-specific models have shown promise in predicting personalized risk for arrhythmia. However, these are computationally- and time- intensive, hindering translation to clinical practice. Classical machine learning (ML) algorithms (such as K-nearest neighbors, Gaussian support vector machines, and decision trees) as well as neural network techniques, shown to increase prediction accuracy, can be used to predict occurrence of arrhythmia as predicted by simulations based solely on infarct and ventricular geometry. We present an initial combined image-based patient-specific in silico and machine learning methodology to assess risk for dangerous arrhythmia in post-infarct patients. Furthermore, we aim to demonstrate that simulation-supported data augmentation improves prediction models, combining patient data, computational simulation, and advanced statistical modeling, improving overall accuracy for arrhythmia risk assessment.

Methods: MRI-based computational models were constructed from 30 patients 5 days post-MI (the “baseline” population). In order to assess the utility biophysical model-supported data augmentation for improving arrhythmia prediction, we augmented the virtual baseline patient population. Each patient ventricular and ischemic geometry in the baseline population was used to create a subfamily of geometric models, resulting in an expanded set of patient models (the “augmented” population). Arrhythmia induction was attempted via programmed stimulation at 17 sites for each virtual patient corresponding to AHA LV segments and simulation outcome, “arrhythmia,” or “no-arrhythmia,” were used as ground truth for subsequent statistical prediction (machine learning, ML) models. For each patient geometric model, we measured and used choice data features: the myocardial volume and ischemic volume, as well as the segment-specific myocardial volume and ischemia percentage, as input to ML algorithms. For classical ML techniques (ML), we trained k-nearest neighbors, support vector machine, logistic regression, xgboost, and decision tree models to predict the simulation outcome from these geometric features alone. To explore neural network ML techniques, we trained both a three - and a four-hidden layer multilayer perceptron feed forward neural networks (NN), again predicting simulation outcomes from these geometric features alone. ML and NN models were trained on 70% of randomly selected segments and the remaining 30% was used for validation for both baseline and augmented populations.

Results: Stimulation in the baseline population (30 patient models) resulted in reentry in 21.8% of sites tested; in the augmented population (129 total patient models) reentry occurred in 13.0% of sites tested. ML and NN models ranged in mean accuracy from 0.83 to 0.86 for the baseline population, improving to 0.88 to 0.89 in all cases.

Conclusion: Machine learning techniques, combined with patient-specific, image-based computational simulations, can provide key clinical insights with high accuracy rapidly and efficiently. In the case of sparse or missing patient data, simulation-supported data augmentation can be employed to further improve predictive results for patient benefit. This work paves the way for using data-driven simulations for prediction of dangerous arrhythmia in MI patients.

Electrical Ventricular Remodeling in Dilated Cardiomyopathy

Ventricular arrhythmias contribute significantly to morbidity and mortality in patients with heart failure (HF). Pathomechanisms underlying arrhythmogenicity in patients with structural heart disease and impaired cardiac function include myocardial fibrosis and the remodeling of ion channels, affecting electrophysiologic properties of ventricular cardiomyocytes. The dysregulation of ion channel expression has been associated with cardiomyopathy and with the development of arrhythmias. However, the underlying molecular signaling pathways are increasingly recognized. This review summarizes clinical and cellular electrophysiologic characteristics observed in dilated cardiomyopathy (DCM) with ionic and structural alterations at the ventricular level. Furthermore, potential translational strategies and therapeutic options are highlighted.