Classification of Bundle Branch Blocks with QRS Templates Extracted from 12-lead ECGs

Being non-invasive, cheap and widely available, the 12-lead electrocardiogram (ECG) is a standard method to assess cardiac function. Still, its reliable interpretation requires specialized knowledge and experience, rendering a second opinion valuable. We evaluated the performance of machine learning based classification of 11,705 healthy and bundle branch block 12-lead ECGs from 3 open databases. For each lead of the ECG signal, a representative QRS-complex template was extracted automatically. Principal component analysis (PCA) was applied to the concatenated, normalized and rescaled QRS signals to reduce their dimensionality. Multilayer perceptron and support-vector machine classifiers were trained using the principal components of weighted and non-weighted QRS template signals as input data. Classifiers achieved F1 scores between 0.92 and 0.96 on the test set for different input configurations. Anomaly based weighting slightly improved the performance of the classifiers. Neither class-wise PCA for feature extraction nor adding information on sex, gender and electrical heart axis to the input data yielded considerable improvement of the F1 scores. The achieved classification accuracy is similar to deep learning classifier performances and should generalize robustly to other ECG datasets. Our results suggest that this simple and well interpretable approach based on morphological signal characteristics is suitable for automatically and non-invasively identifying bundle branch block pathologies in clinical or smart electronics contexts.

The Evolving Concept of Cardiac Conduction System Pacing

Cardiac pacing is an established treatment option for patients with bradycardia and heart failure. In the recent decade, there is an increasing scientific and clinical interest in the topic of direct His bundle pacing (HBP) and left bundle branch pacing (LBBP) as options for cardiac conduction system pacing (CSP). The concept of CSP started evolving from the late 1970s, passing several historical landmarks. HBP and LBBP used in CSP proved to be successful in small cohorts of patients with various clinical conditions, including binodal disease, atrioventricular blocks, and in patients with bundle branch blocks with indications for cardiac resynchronization therapy. The scope of this chapter is synthesis and analysis of works devoted to this subject, as well as representation of the author’s experience in this topic. The chapter includes historical background, technical, anatomical, and clinical considerations of CSP, covers evidence base, discusses patient outcomes in line with the pros and cons of the abovementioned methods. The separate part describes practical aspects of different pacing modalities, including stages of the operation and pacemaker programming. The textual content of the chapter is accompanied by illustrations, ECGs, and intracardiac electrograms.

Characteristic Electrocardiography Findings of COVID-19 Patients

Background: Determining a relationship between coronavirus disease 2019 (COVID-19) and the ECG findings of the patients with this disease can assist in early diagnosis and patient management based on these findings. This study aimed to investigate whether COVID-19 patients had characteristic ECG findings in the acute period. Methods: A total of 124 patients were divided into two groups as those diagnosed with COVID-19 and controls. The ECGs of these patients were evaluated in terms of rate, rhythm, presence of ST changes, PR interval, QRS width, QTc and QT interval, and presence of right and left bundle branch blocks. Results: On the ECG, the median heart rate of the COVID-19 patients was 104/min (IQR: 99–114), and there was a significant difference compared to the control group (P<0.001). The median PR interval was 157/ms, the QRS width was 86 ± 9/ms in the COVID-19 patients, with no significant difference compared to the controls (P = 0.161 and P = 0.631, respectively). The median QT interval of the COVID-19 patients was normal (400/ms), but a significant difference was detected compared to the controls (P = 0.005). The QTc, ST change, AF, and presence of right and left bundle branch blocks were not significantly different between the two groups. Conclusion: Considering the importance of ECG findings in order to diagnose COVID-19 disease early, we can state that sinus tachycardia is very common in COVID-19 patients, but there is no characteristic ECG finding for COVID-19, including tachycardia.

Computer Modeling of the Heart for ECG Interpretation—A Review

Computer modeling of the electrophysiology of the heart has undergone significant progress. A healthy heart can be modeled starting from the ion channels via the spread of a depolarization wave on a realistic geometry of the human heart up to the potentials on the body surface and the ECG. Research is advancing regarding modeling diseases of the heart. This article reviews progress in calculating and analyzing the corresponding electrocardiogram (ECG) from simulated depolarization and repolarization waves. First, we describe modeling of the P-wave, the QRS complex and the T-wave of a healthy heart. Then, both the modeling and the corresponding ECGs of several important diseases and arrhythmias are delineated: ischemia and infarction, ectopic beats and extrasystoles, ventricular tachycardia, bundle branch blocks, atrial tachycardia, flutter and fibrillation, genetic diseases and channelopathies, imbalance of electrolytes and drug-induced changes. Finally, we outline the potential impact of computer modeling on ECG interpretation. Computer modeling can contribute to a better comprehension of the relation between features in the ECG and the underlying cardiac condition and disease. It can pave the way for a quantitative analysis of the ECG and can support the cardiologist in identifying events or non-invasively localizing diseased areas. Finally, it can deliver very large databases of reliably labeled ECGs as training data for machine learning.

Anatomical localization and classification of bundle branch blocks using novel CineECG

Funding Acknowledgements Type of funding sources: Public grant(s) – National budget only. Main funding source(s): This work was supported by the Dutch Heart Foundation Background Ventricular conduction disorders can induce arrhythmias and impair cardiac function. Bundle branch blocks are diagnosed by 12-lead ECG, but discrimination between complete bundle branch blocks, incomplete bundle branch blocks and normal tracings can be challenging. CineECG computes the mean temporo-spatial isochrone (mTSI) trajectory of activation waveforms in a 3D-heart model from 12-lead ECGs. This trajectory represents the mean trajectory of the ventricular electrical activation at any time interval directly related to ventricular anatomy. In Brugada patients, CineECG has localized the terminal components of ventricular depolarization to right ventricle outflow tract (RVOT). Also, for the localization of bundle branch blocks, the region of latest activation contains the most information. Using CineECG, subject specific anatomically related information about the location of bundle branch blocks is obtained. Purpose This study aimed at exploring whether CineECG can improve the discrimination between complete left/right bundle branch blocks (LBBB/RBBB), and incomplete RBBB (iRBBB). Methods We utilized 400 12-lead ECGs from the online Physionet-XL-PTB-Diagnostic ECG Database with a certified ECG diagnosis. The mTSI trajectory was calculated and projected into the anatomical 3D-heart model. Five CineECG classes were established: "Normal", "iRBBB", "RBBB", "LBBB" and "Undetermined", to which each tracing was allocated. We determined the accuracy of CineECG classification with the gold standard diagnosis. Results A total of 391 ECGs were analyzed (9 ECGs were excluded for noise) and 240/266 were correctly classified as "normal", 14/17 as "iRBBB", 55/55 as "RBBB", 51/51 as "LBBB" and 31 as "undetermined". Average mTSI trajectories were calculated according to ECG diagnosis (Figure). The terminal mTSI contained most information about the BBB localization, as that part directs to the site of latest activation (Figure, red arrow). Conclusion CineECG provided the anatomical localization of different BBBs and accurately differentiated between normal, LBBB and RBBB, and iRBBB. CineECG may aid clinical diagnostic work-up, potentially also contributing to the difficult discrimination between normal, iRBBB and Brugada patients. Abstract Figure. Average CineECG trajectories

AV-node isolation in atrial fibrillation vs AV-node ablation by pace and ablate strategy

Funding Acknowledgements Type of funding sources: None. Introduction AV-node ablation (AVNA) is a common therapy option for rate control strategy of permanent atrial fibrillation. However, this therapy has numerous side-effects as bundle branch blocks and absence of escape rhythm. We have invented a new technique that isolates the AV node instead of ablation. Methods Our study includes 60 patients with 40 being treated with AVNA and 20 being treated with AV-node isolation (AVNI). In AVNI patient´s AV-node region was mapped using 3D navigation system. Ablation was performed around the previously mapped HIS-cloud regions isolating the atrium from the AV-node. In all cases in AVNI group ablation was performed with irrigated tip ablation catheter. Procedure time, ablation points, fluoroscopy time and total DAP, escape rhythm and delta QRS among other data were measured and compared in these two methods. Statistical analysis was performed by SPSS software 27.0. Results No complications appeared in this study. In the AVNI group more ablation points were used (p &lt; 0.01) but there was no significant difference in total procedure time (p = 0.730). With AVNI highly significantly less fluoroscopy time (1.4 vs. 4.7 minutes) and total DAP (40.75 vs. 382.85 µGym²) was achieved (p &lt; 0.01). Moreover, the median change in QRS width was 0 ms in the AVNI group vs. 26 ms in the AVNA group (p &lt; 0.01). Immediately postoperative the escape rhythm of the AVNI group was significantly higher than in the AVNA group (MRang = 37.38 vs. MRang = 27.06, p = 0.023). Conclusion AV-node isolation using 3D navigation mapping system is a safe and effective alternative to classic AVNA procedure with less radiation exposure and potential preservation of escape rhythm, which is a major benefit in case of a pacemaker failure. Figure 1. 3D Navigation pictures after successful AVNI with NavX-System (left panel) and CARTO 3-System (right panel). His cloud with yellow points. The Tricuspid annulus could be defined with different color (green on the left panel) or with the tag points (white points on the right panel). Modified left lateral to posterior view. Abstract Figure.

Abstract 14740: Sympathetic Down-regulation With Zamicastat Reduces Arrhythmogenicity in Isolated Hearts From Rats With Pulmonary Hypertension

Introduction: Pulmonary arterial hypertension (PAH) is a disease often causing right ventricular failure and death. During disease progression, structural and electrical remodeling of the right ventricle creates proarrhythmic substrates and triggers for arrhythmias. Studies of PAH patients show a hyperstimulation of the sympathetic nervous system (SNS) indicating that PAH may be mediated by SNS hyperactivation. Zamicastat (ZAM) is a reversible dopamine β-hydroxylase inhibitor that modulates SNS by reducing the biosynthesis of norepinephrine in peripheral sympathetic nerves. ZAM improves survival in the monocrotaline (MCT) rat model of PAH. Hypothesis: To evaluate the effect of ZAM on electrophysiological parameters in isolated hearts from MCT-treated rats. Methods: Male Wistar Han rats were randomized in 3 groups, Sham (n=6), MCT (n=15) and MCT+ZAM (n=17). On day zero, MCT and MCT+ZAM groups received MCT (60 mg/kg, SC) and Sham group received saline. Starting on day 12, MCT+ZAM rats were daily administered ZAM (30 mg/kg, PO) while all other animals were given vehicle. On day 28, hearts were excised and mounted on a Langendorff system. A bipolar circuit of two electrodes was placed on the epicardium for ECG recording. Provoked arrhythmias were induced with a burst pacing protocol (50Hz, 1 second) and the recovery time was measured as the time needed to reach 3 consecutive stable R-R intervals. Results: MCT increased the proportion of hearts exhibiting bundle branch blocks, 3 rd degree atrioventricular (AV) blocks, atrial fibrillation (AF), ventricular fibrillation (VF) and premature ventricular contractions (PVCs) when compared to Sham group. ZAM treatment suppressed 3 rd degree AV blocks, AF and VF. In addition, PVCs were less frequent in hearts from MCT+ZAM animals as compared to MCT group. Isolated hearts naturally entered a pause following overpacing-induced VF. MCT hearts showed a non-significant 1.7-fold increase in recovery time compared to Sham group. Hearts from MCT+ZAM animals had a significant ≈50% reduction in post-fibrillation pause. Conclusions: These data suggest that treatment with Zamicastat yields a lower proarrhythmic cardiac tissue decreasing the incidence of cardiac arrhythmias.

Progressive electrocardiographic changes in parallel with cardiac magnetic resonance findings in fabry disease

Background Cardiac Magnetic Resonance (CMR) allows to detect progressive stages of cardiac involvement in Fabry Disease (FD). A systematic description of electrocardiographic (ECG) alterations occurring in FD is currently missing. Purpose To explore ECG changes in progressive stages of FD cardiomyopathy. Methods 71 FD patients and 17 healthy controls underwent CMR with T1 mapping and 12-lead ECG. ECG analysis included the duration of the P-wave and the interval between the end of P-wave and the beginning of QRS (PendQ). FD patients in the test cohort were divided into 3 groups with increasing severity of cardiac involvement: A) normal T1, no LVH; B) low T1, no LVH; C) low T1, LVH. Results An increase of Pwave/PendQ ratio was observed in Group A compared to Controls (1.08 vs. 0.75, p&lt;0.0001). Higher Pwave/PendQ ratio (1.50 vs. 1.08, p&lt;0.0001), shorter PQc interval (127.9 vs. 159.5, p=0.0007), increased Sokolow-Lyon Index (SLI) (3.2 vs. 2.4, p&lt;0,001) and T wave amplitude (0.6 vs. 0.4 mV, p=0.002) characterized Group B in comparison with Group A. A higher prevalence of left bundle branch blocks (13.6% vs. 0%, p=0.03) and repolarization abnormalities (77.3% vs. 5.7%), wider QRS (120 vs 95 msec, p&lt;0.0001) and QT (460 vs 400 msec, p=0.003) intervals were found in Group C compared to Group B. SLI (AUC 0.769), Pwave/PendQ (AUC 0.778), QRS (AUC 0.703) and QT (AUC 0.769) durations resulted to be independent predictors of low T1 values on CMR at stepwise multivariate analysis. Conclusion FD is characterized by progressive ECG changes. The identification of ECG parameters able to predict a lowering of myocardial T1 values on CMR may promote early detection of cardiac involvement, helping to target the therapeutic approach. Progressive ECG and CMR changes in FD Funding Acknowledgement Type of funding source: Other. Main funding source(s): This study was partially supported by Ricerca Corrente funding from the Italian Ministry of Health to IRCCS Policlinico San Donato.

Implications of pacemaker implantation after TAVR: insights from the Nationwide Readmissions Database

Introduction Conduction disturbances after transcatheter aortic valve replacement (TAVR) requiring permanent pacemaker (PPM) implantation are a well-known procedural complication. Data on the effect of post-TAVR PPM implantation on 30-day readmissions is scarce. Methods The Nationwide Readmissions Database (NRD) is a nationally representative sample of all US hospitalizations, representing over 35 million discharges. The 2016 NRD was used to identify patients who underwent TAVR and PPM implantation from January to November 2016. Propensity matching was used to balance baseline clinical characteristics. Results Of the 44,607 patients who underwent TAVR, 4,878 (10.9%) required permanent pacemaker implantation during their index hospitalization. Patients requiring PPM during their index admission for TAVR had a higher crude median length of stay (LOS) (5d vs. 3d, p&lt;0.001), median cost of index admission ($61,604 vs. $45,513, p&lt;0.001) and rate of 30-day readmission (14.5% vs 11.2%, p&lt;0.001). After 2:1 propensity matching, PPM patients still had a higher median LOS (5d vs. 3d, p&lt;0.001), median cost of index admission ($61,902 vs. $41,162, p&lt;0.001), and rate of 30-day readmission (13.8% vs 11.1%, p=0.003). Patients who received PPM were more likely to be older (81.1 vs 80.3, OR 1.01, CI 1.01–1.02), diabetic (OR 1.27, 1.13–1.44), obese (OR 1.22, 1.05–1.43), and have right (OR 4.35, 3.72–5.09) or left (OR 1.80, 1.51–2.15) bundle branch blocks on multivariate analysis. Causes of readmission in patients with and without PPM were predominantly non-cardiac (62.9% vs 68.0%). Heart failure was the most common cause of readmission for both groups (18.4% vs 14.6%). Median cost of readmissions ($8716 vs $8250, p=0.34) and LOS (4d vs 4d, p=1) were not significantly different during readmissions. Conclusions Based on a nationally representative sample, 10.9% of patients undergoing TAVR required PPM implantation during the index hospitalization. Age, diabetes, obesity, and right or left bundle branch blocks were significant predictors of PPM implantation. PPM implantation resulted in significantly higher LOS, costs, and 30-day readmissions on propensity-matched analysis. Funding Acknowledgement Type of funding source: None