Cardiac toxicity from QT-prolonging drugs can precipitate malignant ventricular arrhythmias in susceptible individuals, and family screening may clarify inherited risk. We report a 33-year-old woman with a history of postpartum cardiac arrest treated with a secondary-prevention implantable cardioverter-defibrillator (ICD) who developed an electrical storm after self-administration of a single low dose of amitriptyline (12.5 mg). ICD interrogation documented 176 episodes of ventricular fibrillation requiring repeated shocks, followed by complete battery depletion, hemodynamic collapse, and the need for venoarterial extracorporeal membrane oxygenation and continuous renal replacement therapy. The admission electrocardiogram showed marked QT prolongation (QTc 651 ms), with previously documented prolonged baseline QTc values. Targeted next-generation sequencing identified a novel SCN5A missense variant (NM_000335.5:c.5738G > A) and a rare pathogenic KCNQ1 splice variant (NM_000218.3:c.1032G > C), cascade testing across the family demonstrated variable expressivity among carriers. Given a suspected contribution of late sodium current, a mechanism-based strategy was implemented with mexiletine added to propranolol and overdrive pacing (90 bpm). This case underscores the risk of malignant ventricular arrhythmias after exposure to QT-prolonging agents even at low doses, and supports genotype-informed, mechanism-based therapy to mitigate arrhythmic risk in patients with marked QT prolongation.

Objective. Cardiovascular diseases remain the leading cause of death worldwide, highlighting the need for non-invasive and cost-effective risk assessment tools. Biological systems, including the heart, exhibit complex nonlinear dynamics arising from interactions between their subsystems. Information-theoretic measures, particularly entropy-based methods, provide a framework to quantify these interactions. Using ECG recordings, we investigate information flow between heart rhythm and ventricular repolarization to identify potential markers of pathological alterations in cardiac electrical activity.Approach. Entropy-based measures of information transfer were derived from beat-to-beat ECG time series using a window-based approach and subsequently averaged at the subject level. These features were used as inputs to supervised machine learning models to discriminate patients with congenital long QT syndrome (LQTS) from healthy controls. Model performance was evaluated using repeated stratified train-test splits, and classification robustness was assessed across multiple runs using standard performance metrics, including the area under the receiver operating characteristic curve. The explainable artificial intelligence techniques were applied. SHapley Additive exPlanations were used to quantify the contribution of entropy-based features to the model predictions. This post-hoc explainability analysis enabled systematic assessment of feature importance while preserving the predictive performance of the models.Results. The proposed approach achieved high and stable classification performance across repeated validation runs. Both random forest (RF) and support vector machine (SVM) classifiers demonstrated high discrimination between LQTS patients and healthy controls, with consistently high area under the curve. For RF a mean accuracy of 95.9%, mean sensitivity of 95.9%, and mean specificity of 92.9% were achieved across repeated runs. For SVM the corresponding mean values were 93.1%, 93.1%, and 92.0%, respectively.Conclusions.Explainability analysis revealed a dominant contribution of multivariate and conditional information flow features compared with single-source entropy measures, highlighting the relevance of joint and conditional interactions in the classification patterns.

Prolongation of the QT interval is a known precursor to serious arrhythmias and sudden cardiac death, often triggered by medication use. Current medication risk evaluation platforms rely on literature-based synthesis and may lag behind real-world developments. We aimed to evaluate whether a machine learning (ML) model trained on real-world genomic and medication data can identify associations between drug use and QTc duration, potentially enabling automated risk detection in clinical workflows. We included 10,208 individuals from the FinnGen biobank Expansion Area 3 substudy, integrating prescription records, clinical variables, and genetic information. We applied a nested-cross-validation approach to develop an ML framework to predict QTc duration using clinical characteristics, recent medication purchases, and polygenic score for QTc duration. We performed conventional linear regression analyses to estimate the robustness of the findings. Only a minority of ML-detected drug-QTc associations aligned with known effects listed in expert-curated reference. Several apparent false positives were observed, and effect sizes for true positives, such as amiodarone, were small and likely interpreted as clinically not meaningful (+1 ms in ML vs. +49 ms in linear regression). These findings highlight challenges in using ML to detect meaningful drug effects on ECG. ML models did not reliably identify medications associated with QT-interval prolongation. Consequently, risk quantification using QTc as an intermediate marker of electrophysiological vulnerability was limited in this framework. While new approaches continue to develop in medication safety assessment, a systematic evidence review conducted by clinical pharmacology experts is unlikely to be supplanted in the foreseeable future.

Ultrasound-guided left-sided or bilateral stellate ganglion block (SGB) has emerged as an effective bedside intervention for malignant ventricular arrhythmias (VA) refractory to conventional therapies, with demonstrated safety and efficacy in multiple studies. We report a medically complex patient with presumed genotype-negative Long QT Syndrome (LQTS) who experienced a sudden, transient increase in arrhythmic events shortly after an uncomplicated left-sided ultrasound-guided SGB. To our knowledge, this is the first documented case of increased VA burden following left-sided SGB, highlighting the potential for paradoxical effects and the need for cautious application in patients with comorbidities.

The long QT syndrome type 1 (LQT1) is caused by loss-of-function mutations affecting the potassium channel current IKs important for repolarization adaptation at heart rate (HR) increase. We therefore compared changes in three global VR dispersion parameters following a rapid, atropine-induced HR increase in LQT1 patients versus healthy controls applying Frank vectorcardiography. The adaptation patterns, magnitudes, and changing rates of T amplitude, T area, and the ventricular gradient were assessed during 5 min following injection. Twenty-one LQT1 patients and 31 controls were enrolled. HR increased similarly within ~23 s. The majority had a tri-phasic VR dispersion adaptation pattern: a rapid decrease to a minimum, a slower return to a slightly higher level, and a very slow almost horizontal decrease. In the LQT1 group, all three dispersion parameters had a longer lasting rapid phase and larger overshoot reaction, although statistically significant only for T amplitude. In conclusion, LQT1 was associated with an augmented adaptation of global dispersion parameters to a rapid HR increase versus healthy controls. Further studies are needed to determine how these observations relate to arrhythmogenesis and if the atropine test of QT interval and dispersion adaptation can be used in risk assessment on the individual level.

Electrocardiography (ECG) plays a vital role in the diagnosis of cardiovascular diseases by analyzing the electrical activity of the heart. ECG semantic segmentation is a subfield focused on sample-wise delineation of ECG waveforms by assigning a physiological label to each time sample, enabling explicit estimation of clinically meaningful onset and offset boundaries. Recent advancements in deep learning have significantly improved ECG classification accuracy; however, the same has not yet been observed in automatic ECG segmentation. Existing models often lack explainability and adaptability to patient-specific variations, thereby reducing their generalizability. This study proposes a personalized deep neural network approach for enhanced ECG processing. This method incorporates convolutional neural networks (CNNs) and bidirectional long short-term memory (BiLSTM) networks, incorporating an attention mechanism to refine segmentation accuracy. A novel loss function is introduced to ensure smoother temporal transitions and better classification accuracy. The model was evaluated using the QT Database, demonstrating substantial improvements in P-wave and QRS delineation and in T-wave offset localization segmentation when fine-tuned for individual patients when fine-tuned and evaluated on held-out data from the same patient, demonstrating the benefit of intra-patient adaptation. Our results indicate that personalization improves delineation accuracy for challenging waveforms (notably P and T waves), supporting the potential of deep learning to better capture patient-specific morphology and providing a stronger basis for waveform-level, clinically interpretable ECG analysis.

ECG is the most used tools in cardiology. Some life-threatening arrhythmias and cardiac abnormalities can be diagnosed with ECG. The aim is to assess the role of ECG for the detection of cardiac disease in children, from September 2014 to January 2020, ECG was performed in 52 primary school in Como. 5793 children (49.4% female) aged 8.1 (± 1.3) years had an ECG; 79 (1.3%) were abnormal due to the following findings: 34 (43%) right bundle branch block (RBBB), 9 (11%) abnormal ventricular repolarization, 7 (9%) long QT, 7 atrio-ventricular block (AVB), 7 (9%) accessory pathway, 5 (5%) premature ventricular beat, 5 (5%) anterior hemiblock, 4 (5%) left ventricular hypertrophy, 1 (2%) ST alteration. Long QT syndrome was excluded in all children with long QT. Among those with RBBB, 2 had severe inter-atrial septal defect (one underwent to closure), while one had moderate mitral valve regurgitation and bicuspid aortic, and 2 had moderate mitral regurgitation. One child with AVB had pacemaker implantation. One child with accessory pathway had Wolf-Parkinson syndrome and undergone to catheter ablation. Among apparently healthy children, ECG may be useful to detect cardiac disease with potential consequence in the adulthood.

Long QT Syndrome (LQTS) is an inherited cardiac disorder characterized by dysfunctional cardiac ion channels, which result in prolonged QT intervals on electrocardiograms (ECGs). LQTS can lead to severe clinical manifestations, including syncope, ventricular arrhythmias, and sudden cardiac death. Effective genotype-specific management strategies are essential to mitigate the risk of life-threatening arrhythmias. This study aims to achieve automatic discrimination among the LQT1, LQT2, and LQT3 genotypes to enable targeted treatment and prevention strategies. Utilizing ECG data from the Telemetric and Holter ECG Warehouse's LQTS database, our methodology involves an automated extraction process of short ECG signals, geometric parameterization techniques, and classification using a two-stage cascade of binary support vector machine classifiers. The input features for the classifiers are derived from Lead I ECG signals sampled at 200 Hz, highlighting the potential application in developing single-lead ECG monitoring devices and applications, such as widely used smartwatches, for which short recording duration, low sampling frequency, and arm-to-arm lead measurements are fundamental prerequisites for practical use. The proposed classifier achieved 71% weighted accuracy on out-of-sample data (LQT1: 65% recall, 58% precision; LQT2: 79% recall, 82% precision; and LQT3: 71% recall, 77% precision). Our findings demonstrate the feasibility of noninvasive genotype differentiation for LQTS based on the morphological analysis of ECG signals, providing an advancement in the field of personalized cardiology and the development of portable diagnostic tools.

Sudden cardiac death in athletes, though uncommon, remains a major concern in sports cardiology. Many responsible cardiovascular conditions, including cardiomyopathies, inherited channelopathies, valvular disease, and congenital coronary anomalies, may remain asymptomatic until intense physical exertion. Current pre-participation screening relies on clinical history, physical examination, electrocardiography, and selective cardiac imaging. While effective, these tools are limited by interobserver variability, dependence on specialist expertise, and difficulty distinguishing physiological athletic remodeling from pathological disease. These limitations have prompted growing interest in artificial intelligence (AI) as an adjunct to cardiovascular screening in athletes. This review summarizes current evidence on AI applications in sports cardiology, with a focus on electrocardiography, digital auscultation, transthoracic echocardiography, and selected imaging modalities. AI-enhanced electrocardiographic analysis has demonstrated improved sensitivity compared with traditional criteria for detecting left ventricular hypertrophy, long QT syndrome, including concealed forms, Brugada syndrome, electrolyte abnormalities, and aortic stenosis. Several deep learning models identify disease patterns even when conventional electrocardiographic parameters appear normal, addressing a key limitation of standard screening approaches. AI-assisted digital auscultation improves the detection of pathological murmurs and differentiation from benign flow murmurs, supporting earlier identification of valvular disease. In echocardiography, AI-guided image acquisition and automated analysis improve access, workflow efficiency, and measurement consistency, with diagnostic performance approaching expert interpretation. This review proposes a pragmatic AI-integrated screening framework that complements clinician judgment rather than replacing it. Although promising, limitations remain, including limited athlete-specific training data, false-positive risk related to physiological remodeling, and the need for external validation. When thoughtfully integrated into clinical workflows, AI may enhance early detection of occult cardiac disease and improve cardiovascular risk stratification in athletes.