ECG-only explainable deep learning algorithm predicts the risk for malignant ventricular arrhythmia in phospholamban cardiomyopathy.

BACKGROUND: Phospholamban (PLN) p.(Arg14del) variant carriers are at risk for development of malignant ventricular arrhythmia (MVA). Accurate risk stratification allows timely implantation of intracardiac defibrillators and is currently performed with a multimodality prediction model. OBJECTIVE: This study aimed to investigate whether an explainable deep learning-based approach allows risk prediction with only electrocardiogram (ECG) data. METHODS: A total of 679 PLN p.(Arg14del) carriers without MVA at baseline were identified. A deep learning-based variational auto-encoder, trained on 1.1 million ECGs, was used to convert the 12-lead baseline ECG into its FactorECG, a compressed version of the ECG that summarizes it into 32 explainable factors. Prediction models were developed by Cox regression. RESULTS: The deep learning-based ECG-only approach was able to predict MVA with a C statistic of 0.79 (95% CI, 0.76-0.83), comparable to the current prediction model (C statistic, 0.83 [95% CI, 0.79-0.88]; P = .054) and outperforming a model based on conventional ECG parameters (low-voltage ECG and negative T waves; C statistic, 0.65 [95% CI, 0.58-0.73]; P < .001). Clinical simulations showed that a 2-step approach, with ECG-only screening followed by a full workup, resulted in 60% less additional diagnostics while outperforming the multimodal prediction model in all patients. A visualization tool was created to provide interactive visualizations (https://pln.ecgx.ai). CONCLUSION: Our deep learning-based algorithm based on ECG data only accurately predicts the occurrence of MVA in PLN p.(Arg14del) carriers, enabling more efficient stratification of patients who need additional diagnostic testing and follow-up.

Value of genetic testing in the diagnosis and risk stratification of arrhythmogenic right ventricular cardiomyopathy.

BACKGROUND: Arrhythmogenic right ventricular cardiomyopathy (ARVC) is characterized by risk of malignant ventricular arrhythmia (VA). ARVC is diagnosed using an array of clinical tests in the consensus-based Task Force Criteria (TFC), one of which is genetic testing. OBJECTIVE: The purpose of this study was to investigate the value of genetic testing in diagnosing ARVC and its relation to the occurrence of first malignant VA. METHODS: A multicenter cohort of patients with ARVC was scored using the revised 2010 TFC with and without genetic criterion, analyzing any resulting loss or delay of diagnosis. Malignant VA was defined as sustained VA (≥30-second duration at ≥100 beats/min or requiring intervention). RESULTS: We included 402 subjects (221 [55%] male; 216 [54%] proband; 40 [27-51] years old at presentation) who were diagnosed with definite ARVC. A total of 232 subjects (58%) fulfilled genetic testing criteria. Removing the genetic criterion caused loss of diagnosis in 18 patients (4%) (11 of 216 probands [5%] and 7 of 186 relatives [4%]) and delay of diagnosis by ≥30 days in 22 patients (5%) (21 of 216 probands [10%] and 1 of 186 relative [0.5%]). A first malignant VA occurred in no patients who lost diagnosis and in 3 patients (3 of 216 probands [1%] and no relatives) during their diagnosis delay, none fatal. Time-to-event analysis showed no significant difference in time from diagnosis to malignant VA between pathogenic variant carriers and noncarriers. CONCLUSION: Disregarding the genetic criterion of the TFC caused loss or delay of diagnosis in 10% of patients with ARVC (40 of 402). Malignant VA occurred in 1% of cases with lost or delayed diagnosis (3 of 402), none fatal.

Optimal echocardiographic assessment of myocardial dysfunction for arrhythmic risk stratification in phospholamban mutation carriers.

AIMS: Phospholamban (PLN) p.Arg14del mutation carriers are at risk of developing malignant ventricular arrhythmias (VAs) and/or heart failure. Currently, left ventricular ejection fraction (LVEF) plays an important role in risk assessment for VA in these individuals. We aimed to study the incremental prognostic value of left ventricular mechanical dispersion (LVMD) by echocardiographic deformation imaging for prediction of sustained VA in PLN p.Arg14del mutation carriers. METHODS AND RESULTS: We included 243 PLN p.Arg14del mutation carriers, which were classified into three groups according to the '45/45' rule: (i) normal left ventricular (LV) function, defined as preserved LVEF ≥45% with normal LVMD ≤45 ms (n = 139), (ii) mechanical LV dysfunction, defined as preserved LVEF ≥45% with abnormal LVMD >45 ms (n = 63), and (iii) overt LV dysfunction, defined as reduced LVEF <45% (n = 41). During a median follow-up of 3.3 (interquartile range 1.8-6.0) years, sustained VA occurred in 35 individuals. The negative predictive value of having normal LV function at baseline was 99% [95% confidence interval (CI): 92-100%] for developing sustained VA. The positive predictive value of mechanical LV dysfunction was 20% (95% CI: 15-27%). Mechanical LV dysfunction was an independent predictor of sustained VA in multivariable analysis [hazard ratio adjusted for VA history: 20.48 (95% CI: 2.57-162.84)]. CONCLUSION: LVMD has incremental prognostic value on top of LVEF in PLN p.Arg14del mutation carriers, particularly in those with preserved LVEF. The '45/45' rule is a practical approach to echocardiographic risk stratification in this challenging group of patients. This approach may also have added value in other diseases where LVEF deterioration is a relative late marker of myocardial dysfunction.

Prediction of ventricular arrhythmia in phospholamban p.Arg14del mutation carriers-reaching the frontiers of individual risk prediction.

AIMS: This study aims to improve risk stratification for primary prevention implantable cardioverter defibrillator (ICD) implantation by developing a new mutation-specific prediction model for malignant ventricular arrhythmia (VA) in phospholamban (PLN) p.Arg14del mutation carriers. The proposed model is compared to an existing PLN risk model. METHODS AND RESULTS: Data were collected from PLN p.Arg14del mutation carriers with no history of malignant VA at baseline, identified between 2009 and 2020. Malignant VA was defined as sustained VA, appropriate ICD intervention, or (aborted) sudden cardiac death. A prediction model was developed using Cox regression. The study cohort consisted of 679 PLN p.Arg14del mutation carriers, with a minority of index patients (17%) and male sex (43%), and a median age of 42 years [interquartile range (IQR) 27-55]. During a median follow-up of 4.3 years (IQR 1.7-7.4), 72 (10.6%) carriers experienced malignant VA. Significant predictors were left ventricular ejection fraction, premature ventricular contraction count/24 h, amount of negative T waves, and presence of low-voltage electrocardiogram. The multivariable model had an excellent discriminative ability {C-statistic 0.83 [95% confidence interval (CI) 0.78-0.88]}. Applying the existing PLN risk model to the complete cohort yielded a C-statistic of 0.68 (95% CI 0.61-0.75). CONCLUSION: This new mutation-specific prediction model for individual VA risk in PLN p.Arg14del mutation carriers is superior to the existing PLN risk model, suggesting that risk prediction using mutation-specific phenotypic features can improve accuracy compared to a more generic approach.