Associations between scar characteristics by cardiac magnetic resonance and changes in left ventricular ejection fraction in primary prevention defibrillator recipients

Arrhythmic risk stratification in ischemic, non-ischemic and hypertrophic cardiomyopathy: A two-step multifactorial, electrophysiology study inclusive approach

Annual arrhythmic sudden cardiac death ranges from 0.6% to 4% in ischemic cardiomyopathy (ICM), 1% to 2% in non-ischemic cardiomyopathy (NICM), and 1% in hypertrophic cardiomyopathy (HCM). Towards a more effective arrhythmic risk stratification (ARS) we hereby present a two-step ARS with the usage of seven non-invasive risk factors: Late potentials presence (≥ 2/3 positive criteria), premature ventricular contractions (≥ 30/h), non-sustained ventricular tachycardia (≥ 1episode/24 h), abnormal heart rate turbulence (onset ≥ 0% and slope ≤ 2.5 ms) and reduced deceleration capacity (≤ 4.5 ms), abnormal T wave alternans (≥ 65μV), decreased heart rate variability (SDNN < 70ms), and prolonged QT_c interval (> 440 ms in males and > 450 ms in females) which reflect the arrhythmogenic mechanisms for the selection of the intermediate arrhythmic risk patients in the first step. In the second step, these intermediate-risk patients undergo a programmed ventricular stimulation (PVS) for the detection of inducible, truly high-risk ICM and NICM patients, who will benefit from an implantable cardioverter defibrillator. For HCM patients, we also suggest the incorporation of the PVS either for the low HCM Risk-score patients or for the patients with one traditional risk factor in order to improve the inadequate sensitivity of the former and the low specificity of the latter.

Delayed Improvement of Left Ventricular Function in Newly Diagnosed Heart Failure Depends on Etiology-A PROLONG-II Substudy

In patients with newly diagnosed heart failure with reduced ejection fraction (HFrEF), three months of optimal therapy are recommended before considering a primary preventive implantable cardioverter-defibrillator (ICD). It is unclear which patients benefit from a prolonged waiting period under protection of the wearable cardioverter-defibrillator (WCD) to avoid unnecessary ICD implantations. This study included all patients receiving a WCD for newly diagnosed HFrEF (n = 353) at our center between 2012 and 2017. Median follow-up was 2.7 years. From baseline until three months, LVEF improved in patients with all peripartum cardiomyopathy (PPCM), myocarditis, dilated cardiomyopathy (DCM), or ischemic cardiomyopathy (ICM). Beyond this time, LVEF improved in PPCM and DCM only (10 ± 8% and 10 ± 12%, respectively), whereas patients with ICM showed no further improvement. The patients with newly diagnosed HFrEF were compared to 29 patients with a distinct WCD indication, which is an explantation of an infected ICD. This latter group had a higher incidence of WCD shocks and poorer overall survival. All-cause mortality should be considered when deciding on WCD prescription. In patients with newly diagnosed HFrEF, the potential for delayed LVEF recovery should be considered when timing ICD implantation, especially in patients with PPCM and DCM.

CinE caRdiac magneTic resonAnce to predIct veNTricular arrhYthmia (CERTAINTY)

Better models to identify individuals at low risk of ventricular arrhythmia (VA) are needed for implantable cardioverter-defibrillator (ICD) candidates to mitigate the risk of ICD-related complications. We designed the CERTAINTY study (CinE caRdiac magneTic resonAnce to predIct veNTricular arrhYthmia) with deep learning for VA risk prediction from cine cardiac magnetic resonance (CMR). Using a training cohort of primary prevention ICD recipients (n = 350, 97 women, median age 59 years, 178 ischemic cardiomyopathy) who underwent CMR immediately prior to ICD implantation, we developed two neural networks: Cine Fingerprint Extractor and Risk Predictor. The former extracts cardiac structure and function features from cine CMR in a form of cine fingerprint in a fully unsupervised fashion, and the latter takes in the cine fingerprint and outputs disease outcomes as a cine risk score. Patients with VA (n = 96) had a significantly higher cine risk score than those without VA. Multivariate analysis showed that the cine risk score was significantly associated with VA after adjusting for clinical characteristics, cardiac structure and function including CMR-derived scar extent. These findings indicate that non-contrast, cine CMR inherently contains features to improve VA risk prediction in primary prevention ICD candidates. We solicit participation from multiple centers for external validation.

Baseline and Dynamic Risk Predictors of Appropriate Implantable Cardioverter Defibrillator Therapy

Background Current approaches fail to separate patients at high versus low risk for ventricular arrhythmias owing to overreliance on a snapshot left ventricular ejection fraction measure. We used statistical machine learning to identify important cardiac imaging and time-varying risk predictors. Methods and Results Three hundred eighty-two cardiomyopathy patients (left ventricular ejection fraction ≤35%) underwent cardiac magnetic resonance before primary prevention implantable cardioverter defibrillator insertion. The primary end point was appropriate implantable cardioverter defibrillator discharge or sudden death. Patient characteristics; serum biomarkers of inflammation, neurohormonal status, and injury; and cardiac magnetic resonance-measured left ventricle and left atrial indices and myocardial scar burden were assessed at baseline. Time-varying covariates comprised interval heart failure hospitalizations and left ventricular ejection fractions. A random forest statistical method for survival, longitudinal, and multivariable outcomes incorporating baseline and time-varying variables was compared with (1) Seattle Heart Failure model scores and (2) random forest survival and Cox regression models incorporating baseline characteristics with and without imaging variables. Age averaged 57±13 years with 28% women, 66% white, 51% ischemic, and follow-up time of 5.9±2.3 years. The primary end point (n=75) occurred at 3.3±2.4 years. Random forest statistical method for survival, longitudinal, and multivariable outcomes with baseline and time-varying predictors had the highest area under the receiver operating curve, median 0.88 (95% CI, 0.75-0.96). Top predictors comprised heart failure hospitalization, left ventricle scar, left ventricle and left atrial volumes, left atrial function, and interleukin-6 level; heart failure accounted for 67% of the variation explained by the prediction, imaging 27%, and interleukin-6 2%. Serial left ventricular ejection fraction was not a significant predictor. Conclusions Hospitalization for heart failure and baseline cardiac metrics substantially improve ventricular arrhythmic risk prediction.

Updating the Risk Stratification for Sudden Cardiac Death in Cardiomyopathies: The Evolving Role of Cardiac Magnetic Resonance Imaging. An Approach for the Electrophysiologist

The prevention of sudden cardiac death (SCD) in cardiomyopathies (CM) remains a challenge. The current guidelines still favor the implantation of devices for the primary prevention of SCD only in patients with severely reduced left ventricular ejection fraction (LVEF) and heart failure (HF) symptoms. The implantation of an implantable cardioverter-defibrillator (ICD) is a protective barrier against arrhythmic events in CMs, but the benefit does not outweigh the cost in low risk patients. The identification of high risk patients is the key to an individualized prevention strategy. Cardiac magnetic resonance (CMR) provides reliable and reproducible information about biventricular function and tissue characterization. Furthermore, late gadolinium enhancement (LGE) quantification and pattern of distribution, as well as abnormal T1 mapping and extracellular volume (ECV), representing indices of diffuse fibrosis, can enhance our ability to detect high risk patients. CMR can also complement electro-anatomical mapping (EAM), a technique already applied in the risk evaluation and in the ventricular arrhythmias ablation therapy of CM patients, providing a more accurate assessment of fibrosis and arrhythmic corridors. As a result, CMR provides a new insight into the pathological substrate of CM. CMR may help identify high risk CM patients and, combined with EAM, can provide an integrated evaluation of scar and arrhythmic corridors in the ablative therapy of ventricular arrhythmias.

Improving Clinical Translation of Machine Learning Approaches Through Clinician-Tailored Visual Displays of Black Box Algorithms: Development and Validation

BACKGROUND

Despite the promise of machine learning (ML) to inform individualized medical care, the clinical utility of ML in medicine has been limited by the minimal interpretability and black box nature of these algorithms.

OBJECTIVE

The study aimed to demonstrate a general and simple framework for generating clinically relevant and interpretable visualizations of black box predictions to aid in the clinical translation of ML.

METHODS

To obtain improved transparency of ML, simplified models and visual displays can be generated using common methods from clinical practice such as decision trees and effect plots. We illustrated the approach based on postprocessing of ML predictions, in this case random forest predictions, and applied the method to data from the Left Ventricular (LV) Structural Predictors of Sudden Cardiac Death (SCD) Registry for individualized risk prediction of SCD, a leading cause of death.

RESULTS

With the LV Structural Predictors of SCD Registry data, SCD risk predictions are obtained from a random forest algorithm that identifies the most important predictors, nonlinearities, and interactions among a large number of variables while naturally accounting for missing data. The black box predictions are postprocessed using classification and regression trees into a clinically relevant and interpretable visualization. The method also quantifies the relative importance of an individual or a combination of predictors. Several risk factors (heart failure hospitalization, cardiac magnetic resonance imaging indices, and serum concentration of systemic inflammation) can be clearly visualized as branch points of a decision tree to discriminate between low-, intermediate-, and high-risk patients.

CONCLUSIONS

Through a clinically important example, we illustrate a general and simple approach to increase the clinical translation of ML through clinician-tailored visual displays of results from black box algorithms. We illustrate this general model-agnostic framework by applying it to SCD risk prediction. Although we illustrate the methods using SCD prediction with random forest, the methods presented are applicable more broadly to improving the clinical translation of ML, regardless of the specific ML algorithm or clinical application. As any trained predictive model can be summarized in this manner to a prespecified level of precision, we encourage the use of simplified visual displays as an adjunct to the complex predictive model. Overall, this framework can allow clinicians to peek inside the black box and develop a deeper understanding of the most important features from a model to gain trust in the predictions and confidence in applying them to clinical care.

Heterogeneous Outcomes of Heart Failure with Better Ejection Fraction

We evaluated the heterogeneity of outcomes among heart failure patients with ventricular recovery. The BEST trial studied patients with left ventricular ejection fraction (LVEF) ≤ 35%. Serial LVEF assessment was performed at baseline, 3 months, and 12 months. Heart failure with better ejection fraction (HFbEF) was defined as an LVEF > 40% at any point. Of the patients who survived to 1 year, 399 (21.3%) had HFbEF. Among subjects with HFbEF, 173 (43.4%) had "extended" recovery, 161 (40.4%) had "late" recovery, and 65 (16.3%) patients had "transient" recovery. Subjects with HFbEF had an improved event-free survival from death or first HF hospitalization compared to subjects without recovery (HR 0.50, 95% CI, 0.39-0.64, p < 0.001). Compared to "transient" recovery, "late" and "extended" recovery were associated with an improved event-free survival from all-cause death and HF hospitalization (HR 0.55, 95% CI, 0.34-0.90, p = 0.016). Our study shows patients with HFbEF to be a heterogeneous population with differing prognoses.

Clinical risk prediction with random forests for survival, longitudinal, and multivariate (RF-SLAM) data analysis

Background

Clinical research and medical practice can be advanced through the prediction of an individual’s health state, trajectory, and responses to treatments. However, the majority of current clinical risk prediction models are based on regression approaches or machine learning algorithms that are static, rather than dynamic. To benefit from the increasing emergence of large, heterogeneous data sets, such as electronic health records (EHRs), novel tools to support improved clinical decision making through methods for individual-level risk prediction that can handle multiple variables, their interactions, and time-varying values are necessary.

Methods

We introduce a novel dynamic approach to clinical risk prediction for survival, longitudinal, and multivariate (SLAM) outcomes, called random forest for SLAM data analysis (RF-SLAM). RF-SLAM is a continuous-time, random forest method for survival analysis that combines the strengths of existing statistical and machine learning methods to produce individualized Bayes estimates of piecewise-constant hazard rates. We also present a method-agnostic approach for time-varying evaluation of model performance.

Results

We derive and illustrate the method by predicting sudden cardiac arrest (SCA) in the Left Ventricular Structural (LV) Predictors of Sudden Cardiac Death (SCD) Registry. We demonstrate superior performance relative to standard random forest methods for survival data. We illustrate the importance of the number of preceding heart failure hospitalizations as a time-dependent predictor in SCA risk assessment.

Conclusions

RF-SLAM is a novel statistical and machine learning method that improves risk prediction by incorporating time-varying information and accommodating a large number of predictors, their interactions, and missing values. RF-SLAM is designed to easily extend to simultaneous predictions of multiple, possibly competing, events and/or repeated measurements of discrete or continuous variables over time.Trial registration: LV Structural Predictors of SCD Registry (clinicaltrials.gov, NCT01076660), retrospectively registered 25 February 2010

Relationship between Extension or Texture Features of Late Gadolinium Enhancement and Ventricular Tachyarrhythmias in Hypertrophic Cardiomyopathy

Purpose

To evaluate the relationship between extension or texture features of late gadolinium enhancement (LGE) and ventricular tachyarrhythmias in hypertrophic cardiomyopathy (HCM).

Materials and Methods

Twenty-three patients with HCM were enrolled in this IRB-approved study. The extension of LGE was determined based on the American Heart Association segments model. Texture analysis was performed for 43 myocardial LGE using an open-access software (MaZda, Technical University of Lodz, Institute of Electronics, Poland). The relationship between the extension or texture features of LGE and ventricular tachyarrhythmias was evaluated using unpaired test and receiver-operating characteristic (ROC) analysis.

Results

Six of 23 patients had a history of ventricular tachyarrhythmias, and 16 patients had LGE. All of the 6 patients with the arrhythmias had more than 4 LGE segments and more LGE segments than those without (p < 0.01). Among 4 texture features, entropy LL was the only discriminator between the 2 patient groups (p < 0.01; threshold, 19624; area under the curve [AUC], 0.72). An ROC analysis gave the number of segments showing LGE a better result (AUC, 0.96) for identification of HCM patients with ventricular tachyarrhythmias than the entropy LL of LGE.

Conclusion

Patients with HCM and a history of ventricular tachyarrhythmias had a wider extension of LGE, and their entropy LL of LGE was significantly lower than those without. The extension of LGE and texture analysis may provide information about LGE related to ventricular tachyarrhythmias in HCM.

Sudden Cardiac Death Substrate Imaged by Magnetic Resonance Imaging: From Investigational Tool to Clinical Applications

Sudden cardiac death (SCD) is a devastating event afflicting 350 000 Americans annually despite the availability of life-saving preventive therapy, the implantable cardioverter defibrillator. SCD prevention strategies are hampered by over-reliance on global left ventricular ejection fraction <35% as the most important criterion to determine implantable cardioverter defibrillator candidacy. Annually in the United States alone, this results in ≈130 000 implantable cardioverter defibrillator placements at a cost of >$3 billion but only a 5% incidence per year of appropriate firings. This approach further fails to identify individuals who experience the majority, as many as 80%, of SCD events, which occur in the setting of more preserved left ventricular ejection fraction. Better risk stratification is needed to improve care and should be guided by direct pathophysiologic markers of arrhythmic substrate, such as specific left ventricular structural abnormalities. There is an increasing body of literature to support the prognostic value of cardiac magnetic resonance imaging with late gadolinium enhancement in phenotyping the left ventricular to identify those at highest risk for SCD. Cardiac magnetic resonance has unparalleled tissue characterization ability and provides exquisite detail about myocardial structure and composition, abnormalities of which form the direct, pathophysiologic substrate for SCD. Here, we review the evolution and the current state of cardiac magnetic resonance for imaging the arrhythmic substrate, both as a research tool and for clinical applications.