Concordant versus discordant left bundle branch block in heart failure patients: novel clinical value of an old electrocardiographic diagnosis

Prediction of heart failure patients with distinct left ventricular ejection fraction levels using circadian ECG features and machine learning

Heart failure (HF) encompasses a diverse clinical spectrum, including instances of transient HF or HF with recovered ejection fraction, alongside persistent cases. This dynamic condition exhibits a growing prevalence and entails substantial healthcare expenditures, with anticipated escalation in the future. It is essential to classify HF patients into three groups based on their ejection fraction: reduced (HFrEF), mid-range (HFmEF), and preserved (HFpEF), such as for diagnosis, risk assessment, treatment choice, and the ongoing monitoring of heart failure. Nevertheless, obtaining a definitive prediction poses challenges, requiring the reliance on echocardiography. On the contrary, an electrocardiogram (ECG) provides a straightforward, quick, continuous assessment of the patient's cardiac rhythm, serving as a cost-effective adjunct to echocardiography. In this research, we evaluate several machine learning (ML)-based classification models, such as K-nearest neighbors (KNN), neural networks (NN), support vector machines (SVM), and decision trees (TREE), to classify left ventricular ejection fraction (LVEF) for three categories of HF patients at hourly intervals, using 24-hour ECG recordings. Information from heterogeneous group of 303 heart failure patients, encompassing HFpEF, HFmEF, or HFrEF classes, was acquired from a multicenter dataset involving both American and Greek populations. Features extracted from ECG data were employed to train the aforementioned ML classification models, with the training occurring in one-hour intervals. To optimize the classification of LVEF levels in coronary artery disease (CAD) patients, a nested cross-validation approach was employed for hyperparameter tuning. HF patients were best classified using TREE and KNN models, with an overall accuracy of 91.2% and 90.9%, and average area under the curve of the receiver operating characteristics (AUROC) of 0.98, and 0.99, respectively. Furthermore, according to the experimental findings, the time periods of midnight-1 am, 8-9 am, and 10-11 pm were the ones that contributed to the highest classification accuracy. The results pave the way for creating an automated screening system tailored for patients with CAD, utilizing optimal measurement timings aligned with their circadian cycles.

Investigating automated regression models for estimating left ventricular ejection fraction levels in heart failure patients using circadian ECG features

Heart Failure (HF) significantly impacts approximately 26 million people worldwide, causing disruptions in the normal functioning of their hearts. The estimation of left ventricular ejection fraction (LVEF) plays a crucial role in the diagnosis, risk stratification, treatment selection, and monitoring of heart failure. However, achieving a definitive assessment is challenging, necessitating the use of echocardiography. Electrocardiogram (ECG) is a relatively simple, quick to obtain, provides continuous monitoring of patient's cardiac rhythm, and cost-effective procedure compared to echocardiography. In this study, we compare several regression models (support vector machine (SVM), extreme gradient boosting (XGBOOST), gaussian process regression (GPR) and decision tree) for the estimation of LVEF for three groups of HF patients at hourly intervals using 24-hour ECG recordings. Data from 303 HF patients with preserved, mid-range, or reduced LVEF were obtained from a multicentre cohort (American and Greek). ECG extracted features were used to train the different regression models in one-hour intervals. To enhance the best possible LVEF level estimations, hyperparameters tuning in nested loop approach was implemented (the outer loop divides the data into training and testing sets, while the inner loop further divides the training set into smaller sets for cross-validation). LVEF levels were best estimated using rational quadratic GPR and fine decision tree regression models with an average root mean square error (RMSE) of 3.83% and 3.42%, and correlation coefficients of 0.92 (p<0.01) and 0.91 (p<0.01), respectively. Furthermore, according to the experimental findings, the time periods of midnight-1 am, 8-9 am, and 10-11 pm demonstrated to be the lowest RMSE values between the actual and predicted LVEF levels. The findings could potentially lead to the development of an automated screening system for patients with coronary artery disease (CAD) by using the best measurement timings during their circadian cycles.

Remodeling of the Purkinje Network in Congestive Heart Failure in the Rabbit

Supplemental Digital Content is available in the text.

Background:

Purkinje fibers (PFs) control timing of ventricular conduction and play a key role in arrhythmogenesis in heart failure (HF) patients. We investigated the effects of HF on PFs.

Methods:

Echocardiography, electrocardiography, micro-computed tomography, quantitative polymerase chain reaction, immunohistochemistry, volume electron microscopy, and sharp microelectrode electrophysiology were used.

Results:

Congestive HF was induced in rabbits by left ventricular volume- and pressure-overload producing left ventricular hypertrophy, diminished fractional shortening and ejection fraction, and increased left ventricular dimensions. HF baseline QRS and corrected QT interval were prolonged by 17% and 21% (mean±SEMs: 303±6 ms HF, 249±11 ms control; n=8/7; P=0.0002), suggesting PF dysfunction and impaired ventricular repolarization. Micro-computed tomography imaging showed increased free-running left PF network volume and length in HF. mRNA levels for 40 ion channels, Ca²⁺-handling proteins, connexins, and proinflammatory and fibrosis markers were assessed: 50% and 35% were dysregulated in left and right PFs respectively, whereas only 12.5% and 7.5% changed in left and right ventricular muscle. Funny channels, Ca²⁺-channels, and K⁺-channels were significantly reduced in left PFs. Microelectrode recordings from left PFs revealed more negative resting membrane potential, reduced action potential upstroke velocity, prolonged duration (action potential duration at 90% repolarization: 378±24 ms HF, 249±5 ms control; n=23/38; P<0.0001), and arrhythmic events in HF. Similar electrical remodeling was seen at the left PF-ventricular junction. In the failing left ventricle, upstroke velocity and amplitude were increased, but action potential duration at 90% repolarization was unaffected.

Conclusions:

Severe volume- followed by pressure-overload causes rapidly progressing HF with extensive remodeling of PFs. The PF network is central to both arrhythmogenesis and contractile dysfunction and the pathological remodeling may increase the risk of fatal arrhythmias in HF patients.

Silencing miR-370-3p rescues funny current and sinus node function in heart failure

Bradyarrhythmias are an important cause of mortality in heart failure and previous studies indicate a mechanistic role for electrical remodelling of the key pacemaking ion channel HCN4 in this process. Here we show that, in a mouse model of heart failure in which there is sinus bradycardia, there is upregulation of a microRNA (miR-370-3p), downregulation of the pacemaker ion channel, HCN4, and downregulation of the corresponding ionic current, If, in the sinus node. In vitro, exogenous miR-370-3p inhibits HCN4 mRNA and causes downregulation of HCN4 protein, downregulation of If, and bradycardia in the isolated sinus node. In vivo, intraperitoneal injection of an antimiR to miR-370-3p into heart failure mice silences miR-370-3p and restores HCN4 mRNA and protein and If in the sinus node and blunts the sinus bradycardia. In addition, it partially restores ventricular function and reduces mortality. This represents a novel approach to heart failure treatment.

Comparison of Quantitative Perfusion and Function Parameters of Gated-SPECT Myocardial Perfusion Imaging in Patients With Concordant and Discordant Left Bundle-Branch Block

BACKGROUND

Patients with left bundle-branch block (LBBB) can be classified to discordant LBBB (dLBBB) and concordant LBBB (cLBBB) according to T-wave orientation in lateral leads. This classification indicates different functional left ventricular (LV) parameters with worse prognosis in dLBBB patients. However, there are no data regarding the impact of this classification on perfusion status of the left ventricle. The aim of this study is to evaluate and compare the SPECT myocardial perfusion imaging (MPI) findings of LV perfusion and function between dLBBB and cLBBB patients.

METHODS

All patients who were referred for SPECT MPI during an 11 months' period were evaluated. Patients with evidence of LBBB on standard baseline 12-lead ECG were included, and their demographic, medical history, and imaging data were recorded. Quantitative perfusion and function parameters of LV included summed stress score, summed rest score, summed difference score, total perfusion deficit (TPD) at both phases with delta TPD, ejection fraction, end-diastolic volume, end-systolic volume, summed motion score, summed thickening score, phase SD, and phase histogram bandwidth. All baseline ECGs were further assessed by a cardiologist to categorize patients as the cLBBB or dLBBB group according to concordance of the T wave with QRS complex in lateral leads.

RESULTS

Finally, 97 patients with 46 cLBBB and 51 dLBBB cases were included. Baseline characteristics and cardiovascular risk factors including diabetes mellitus, hypertension, hyperlipidemia, history of coronary artery disease (CAD), family history of CAD, and smoking were not significantly different between the 2 groups. However, summed stress score (12.2 vs 6.7), summed difference score (4.0 vs 2.6), stress TPD (11.0 vs 6.4), and delta TPD (4.8 vs 3.9) were significantly higher in dLBBB patients. In addition, functional parameters were also significantly worse in dLBBB patients with lower ejection fraction and higher end-diastolic volume, end-systolic volume, summed motion score, and summed thickening score in these patients. Mean phase SD and phase histogram bandwidth were also significantly higher in dLBBB patients.

CONCLUSIONS

This study revealed that LBBB patients with discordant T wave in lateral leads have significantly higher ischemic scores and worse functional parameters with more dyssynchrony in gated SPECT MPI.

Re-evaluating the electro-vectorcardiographic criteria for left bundle branch block

The criteria for left bundle branch block have gained growing interest in the last few years. In this overview, we discuss diagnostic and prognostic aspects of different criteria. It was already shown that stricter criteria, including longer QRS duration and slurring/notching of the QRS, better identify responders to cardiac resynchronization therapy. We also include aspects of ST/T concordance and discordance and vectorcardiography, which could further improve in the fine-tuning of the left bundle branch criteria.

Association between T-wave discordance and the development of heart failure in left bundle branch block patients: Results from the Copenhagen ECG study

BACKGROUND

In left bundle branch block (LBBB), discrepancies between depolarization and repolarization of the heart can be assessed by similar direction (concordant) or opposite direction (discordant) of the lateral T-waves compared to the direction of the QRS complex and by the QRS-T angle. We examined the association between discordant T-waves and high QRS-T angles for heart failure development in primary care LBBB patients.

METHODS

Between 2001 and 2011, we identified 2540 patients from primary care with LBBB without overt heart failure. We examined the development of heart failure in relation to two ECG measures: (1) LBBB as either discordant (two or three monophasic T-waves in the opposite direction of the QRS complex in leads I, V5 or V6) or concordant, and (2) the frontal plane QRS-T angle in quartile groups.

RESULTS

In total, 244 of 913 patients (26.7%) with discordant LBBB developed heart failure compared to 302 of 1627 patients (16.7%) with concordant LBBB. Multivariable Cox regression comparing discordant with concordant LBBB showed a hazard ratio (HR) of 2.58 (95% Confidence interval [CI] 1.71-3.89) for heart failure development within 30 days of follow-up and a HR of 1.45 (95%CI 1.19-1.77) after 30 days. For QRS-T angle, comparing the highest quartile (160°-180°) with the lowest quartile (0°-110°) we found a HR of 2.25 (95%CI 1.26-4.02) within 30 days and a HR of 1.67 (95%CI 1.25-2.23) after 30 days.

CONCLUSION

T-wave discordance in lateral ECG leads and a high QRS-T angle are associated with heart failure development in primary care LBBB patients.

Risk factors and outcomes associated with the development of myocardial ischemic events in patients who receive cardiac resynchronization therapy

There are limited data regarding risk factors for the development of ischemic events (IEs) among patients with ischemic cardiomyopathy (IC) who receive cardiac resynchronization therapy with a defibrillator (CRT-D) and their effect on the efficacy of the device. The present study population comprised 1,045 patients with IC enrolled in the Multicenter Automatic Defibrillator Implantation Trial-Cardiac Resynchronization Therapy. We used multivariate Cox proportional hazards regression modeling to identify risk factors for the development of IE (comprising hospitalization for acute coronary syndromes and/or coronary interventions during the trial) among study patients. Time-dependent analysis was performed to identify the effect of IE on the risk for subsequent heart failure (HF) or death in CRT-D recipients. Independent predictors for the development of IE among study patients included previous revascularization (coronary artery bypass surgery: hazard ratio [HR] 1.88, p = 0.003; percutaneous coronary intervention: HR 3.21, p <0.001) and increased systolic blood pressure (HR 1.67, p = 0.02), whereas a left bundle branch block pattern on the baseline electrocardiogram was associated with reduced risk for IE (HR 0.62, p = 0.02). Treatment with CRT-D did not have a significant effect on IE risk compared with defibrillator-only therapy (HR 0.87, p = 0.51). Time-dependent analysis showed that the development of IEs among CRT-D recipients was associated with more than twofold (p = 0.01) increased risk for subsequent heart failure or death. In conclusion, our data suggest that treatment with CRT-D does not reduce the risk of IE in patients with IC and that the benefit of CRT-D is attenuated after the development of IEs in this population.

Structure, function and clinical relevance of the cardiac conduction system, including the atrioventricular ring and outflow tract tissues

It is now over 100years since the discovery of the cardiac conduction system, consisting of three main parts, the sinus node, the atrioventricular node and the His-Purkinje system. The system is vital for the initiation and coordination of the heartbeat. Over the last decade, immense strides have been made in our understanding of the cardiac conduction system and these recent developments are reviewed here. It has been shown that the system has a unique embryological origin, distinct from that of the working myocardium, and is more extensive than originally thought with additional structures: atrioventricular rings, a third node (so called retroaortic node) and pulmonary and aortic sleeves. It has been shown that the expression of ion channels, intracellular Ca(2+)-handling proteins and gap junction channels in the system is specialised (different from that in the ordinary working myocardium), but appropriate to explain the functioning of the system, although there is continued debate concerning the ionic basis of pacemaking. We are beginning to understand the mechanisms (fibrosis and remodelling of ion channels and related proteins) responsible for dysfunction of the system (bradycardia, heart block and bundle branch block) associated with atrial fibrillation and heart failure and even athletic training. Equally, we are beginning to appreciate how naturally occurring mutations in ion channels cause congenital cardiac conduction system dysfunction. Finally, current therapies, the status of a new therapeutic strategy (use of a specific heart rate lowering drug) and a potential new therapeutic strategy (biopacemaking) are reviewed.