Association of electrocardiographic markers with myocardial fibrosis as assessed by cardiac magnetic resonance in different clinical settings

The Role of Machine Learning in the Detection of Cardiac Fibrosis in Electrocardiograms: Scoping Review

BACKGROUND

Cardiovascular disease remains the leading cause of mortality worldwide. Cardiac fibrosis impacts the underlying pathophysiology of many cardiovascular diseases by altering structural integrity and impairing electrical conduction. Identifying cardiac fibrosis is essential for the prognosis and management of cardiovascular disease; however, current diagnostic methods face challenges due to invasiveness, cost, and inaccessibility. Electrocardiograms (ECGs) are widely available and cost-effective for monitoring cardiac electrical activity. While ECG-based methods for inferring fibrosis exist, they are not commonly used due to accuracy limitations and the need for cardiac expertise. However, the ECG shows promise as a target for machine learning (ML) applications in fibrosis detection.

OBJECTIVE

This study aims to synthesize and critically evaluate the current state of ECG-based ML approaches for cardiac fibrosis detection.

METHODS

We conducted a scoping review of research in ECG-based ML applications to identify cardiac fibrosis. Comprehensive searches were performed in PubMed, IEEE Xplore, Scopus, Web of Science, and DBLP databases, including publications up to October 2024. Studies were included if they applied ML techniques to detect cardiac fibrosis using ECG or vectorcardiogram data and provided sufficient methodological details and outcome metrics. Two reviewers independently assessed eligibility and extracted data on the ML models used, their performance metrics, study designs, and limitations.

RESULTS

We identified 11 studies evaluating ML approaches for detecting cardiac fibrosis using ECG data. These studies used various ML techniques, including classical (8/11, 73%), ensemble (3/11, 27%), and deep learning models (4/11, 36%). Support vector machines were the most used classical model (6/11, 55%), with the best-performing models of each study achieving accuracies of 77% to 93%. Among deep learning approaches, convolutional neural networks showed promising results, with one study reporting an area under the receiver operating characteristic curve (AUC) of 0.89 when combined with clinical features. Notably, a large-scale convolutional neural network study (n=14,052) achieved an AUC of 0.84 for detecting cardiac fibrosis, outperforming cardiologists (AUC 0.63-0.66). However, many studies had limited sample sizes and lacked external validation, potentially impacting the generalizability of the findings. Variability in reporting methods may affect the reproducibility and applicability of these ML-based approaches.

CONCLUSIONS

ML-augmented ECG analysis shows promise for accessible and cost-effective detection of cardiac fibrosis. However, there are common limitations with respect to study design and insufficient external validation, raising concerns about the generalizability and clinical applicability of the findings. Inconsistencies in methodologies and incomplete reporting further impede cross-study comparisons. Future work may benefit from using prospective study designs, larger and more clinically and demographically diverse datasets, advanced ML models, and rigorous external validation. Addressing these challenges could pave the way for the clinical implementation of ML-based ECG detection of cardiac fibrosis to improve patient outcomes and health care resource allocation.

QTc Interval: A frequently unrecognized electrocardiographic interval

The QT interval, an electrocardiographic temporal representation of the ventricular depolarization and repolarization, is an integral parameter that must be carefully evaluated to gather critical information regarding electrical instability that may cause malignant ventricular dysrhythmias or sudden cardiac death. The QT interval is affected by several inheritable and acquired factors, such as genetic mutations, electrolyte disturbances, and medication interactions. We strongly believe that prompt and accurate recognition of any QT interval abnormalities is critical in many clinical settings. This concise review article highlights the importance of accurate measurement of the QT interval, enhances understanding of the most prevalent factors yielding abnormalities within the QT interval and the prognostic value of the QT interval, as well as provides several key practical reminders for healthcare professionals to strengthen our clinical practice.

Clinical and electrocardiographic characteristics of immune checkpoint inhibitor-related myocarditis

BACKGROUND

Immune checkpoint inhibitor (ICI) has become a major breakthrough in the field of tumor therapy, leading to improved survival. This study evaluated the clinical and electrocardiographic characteristics of patients with ICI-related myocarditis.

METHODS

Patients with ICI-related myocarditis were enrolled from 4 centers in China until September 2023. Demographic data (age, sex, comorbidity), types of ICI, clinical manifestations, electrocardiogram (ECG) and treatment were analyzed retrospectively. Arrhythmia and characteristics of ECG were compared according to prognosis and grading.

RESULTS

A total of 29 participants (13 females with a median age of 63.25 years) with ICI-related myocarditis were enrolled. Lung cancer was the most, with a proportion of 31.03 % (9/29). The median time from the first administration of ICI to the diagnosis of myocarditis was 50 days. Camrelizumab was the main type of ICI (9/29). Most patients had non-specific symptoms, dyspnea (n = 16) and palpitation (n = 9) were common. The overall mortality rate was 37.93 % (11/29) with a median follow-up of 9(4,11) days. Compared with the survivors, P-wave abnormality was more common in participants who were dead (24.14 %vs6.90 %, p = 0.010). A total of 19 patients with severe ICI-related myocarditis were included in this study. The proportions of sinus tachycardia (34.48 %vs0.00 %, p = 0.005), premature ventricular complex (27.59 %vs0.00 %, p = 0.027) and atrioventricular block (34.48 %vs3.45 %, p = 0.044) were higher in severe ICI-related myocarditis.

CONCLUSIONS

Clinical manifestations of ICI-related myocarditis usually lacked specificity. ECGs can be manifested as new-onset arrhythmias, ST-T segment changes, fragmented QRS complex, abnormal P wave, prolonged QTc interval and multi‑lead low voltage.

A Cautionary Tale of Hypertrophic Cardiomyopathy-From "Benign" Left Ventricular Hypertrophy to Stroke, Atrial Fibrillation, and Molecular Genetic Diagnostics: A Case Report and Review of Literature

This case report concerns a 48-year-old man with a history of ischemic stroke at the age of 41 who reported cardiac hypertrophy, registered in his twenties when explained by increased physical activity. Family history was positive for a mother with permanent atrial fibrillation from her mid-thirties. At the age of 44, he had a first episode of persistent atrial fibrillation, accompanied by left atrial thrombosis while on a direct oral anticoagulant. He presented at our clinic at the age of 45 with another episode of persistent atrial fibrillation and decompensated heart failure. Echocardiography revealed a dilated left atrium, reduced left ventricular ejection fraction, and an asymmetric left ventricular hypertrophy. Cardiac magnetic resonance was positive for a cardiomyopathy with diffuse fibrosis, while slow-flow phenomenon was present on coronary angiography. Genetic testing by whole-exome sequencing revealed three variants in the patient, c.309C > A, p.His103Gln in the ACTC1 gene, c.116T > G, p.Leu39Ter in the PLN gene, and c.5827C > T, p.His1943Tyr in the SCN5A gene, the first two associated with hypertrophic cardiomyopathy and the latter possibly with familial atrial fibrillation. This case illustrates the need for advanced diagnostics in unexplained left ventricular hypertrophy, as hypertrophic cardiomyopathy is often overlooked, leading to potentially debilitating health consequences.

Prevalence of electrocardiographic markers associated with myocardial fibrosis in masters athletes: a cohort study

Objectives

Myocardial fibrosis (MF) is associated with an increased predisposition to adverse cardiac events. The accumulation of high-volume and high-intensity exercise over an extended duration potentially increases the risk of MF. Specific electrocardiographic markers have been correlated with the presence of MF. This study assessed the prevalence of MF-related electrocardiographic markers in a Track and Field Master Athletics Cohort (TaFMAC).

Methods

Twelve-lead resting electrocardiograms (ECGs) were conducted on 155 athletes (90 males and 65 females) participating in the World Masters Athletics 2022. The ECG markers associated with MF, including pathological Q waves, inverted T waves, fragmented QRS complex, and prolonged QRS complex, were compared among different athletic specialities (endurance athletes n=51, sprinters n=69 and strength and power n=35).

Results

Overall, 71 instances of MF-related markers were identified from 155 ECG recordings (46%). Fragmented QRS emerged as the most common marker, with a prevalence of 29% in endurance and strength and power athletes, and 35% in sprinters. No significant group differences were observed in the prevalence of MF markers, whether analysed collectively (p=0.467) or individually (pathological Q waves p=0.367, inverted T waves p=0.309, fragmented QRS complex p=0.747 and prolonged QRS complex p=0.132).

Conclusions

The prevalence of MF markers, as determined by resting ECG, was evident in nearly half of masters athletes, irrespective of sex and sporting specialisation. These findings suggest resting ECG as a promising non-invasive method for the early identification of MF in athlete's hearts.

Associations between electrocardiographic findings and echocardiographic profiles in patients with hypertrophic cardiomyopathy

BACKGROUND

The relationships between electrocardiography (ECG) findings and echocardiographic profiles in patients with hypertrophic cardiomyopathy (HCM) are not fully understood.

METHODS

One hundred forty patients (mean age: 62.9 ± 15.3 years, 96 men) with HCM were studied. We assessed the associations between ECG findings and echocardiographic findings including maximum left ventricular wall thickness, HCM subtypes and distribution of left ventricular hypertrophy (LVH): the LV was divided into basal, mid, and apical segments by dividing it into thirds along the long axis.

RESULTS

In ECG, LVH by voltage criteria, abnormal Q wave, negative T wave, and giant negative T wave (GNT) were observed in 74 (53 %), 30 (21 %), 132 (94 %), and 25 (18 %) of the patients, respectively. In two groups with and without an LVH pattern according to voltage criteria in ECG, there were no significant differences in maximum LV wall thickness, subtype of HCM, and distribution of LVH. Regarding an abnormal Q wave, the proportion of patients with LVH in the basal segment was significantly higher in patients with an abnormal Q wave than in patients without an abnormal Q wave (87 % vs 61 %, p = 0.008). An abnormal Q wave was not observed in patients with LVH confined to the apex. Patients with a GNT included patients with LVH located at only the apex (apical HCM), LVH from the mid segment to apex, and LVH from the base to apex. No GNT was found in patients with hypertrophy located in the upper region from the base to mid segment of the LV.

CONCLUSIONS

In patients with HCM, there was no significant correlation between the presence of LVH by voltage criteria in ECG and echocardiographic findings. An abnormal Q wave was associated with disproportionate hypertrophy of the basal wall and a GNT reflected the presence of LVH in the apical segment.

Fragmentation of the QRS Complex Is Associated with Right Ventricular Dilatation and Mortality in Critically Unwell Coronavirus Disease 2019 Patients

BACKGROUND

QRS fragmentation (fQRS) is a depolarization disorder that can be detected on routine electrocardiography (ECG). Current evidence suggests that fQRS is a prognosticator of adverse cardiovascular events. This study aimed to assess the relationship between fQRS and all-cause mortality in critically unwell coronavirus disease 2019 (COVID-19) patients and to investigate the significance of associated abnormalities on echocardiography.

METHODS

A retrospective cohort study of COVID-19 patients in a critical care setting was performed. Electrocardiography was performed on presentation to hospital, admission to the critical care unit, and at subsequent points according to clinical need. Transthoracic echocardiography was performed at clinical discretion to assess for structural and functional cardiac abnormalities. Primary outcome was in-hospital mortality and secondary outcome was the need for mechanical invasive ventilation.

RESULTS

Totally, 212 consecutive patients were included of which 120 (57%) exhibited fQRS and inferior leads were involved in 88% of the patients. Overall, fQRS was a significant predictor of mortality [65% vs. 44% P =.003; multivariate odds ratio = 2.96, 95% confidence interval (CI): 1.42-6.40, P =.005] and inferior fQRS itself was a significant predictor of mortality (P =.03). There was no significant association between fQRS and the need for invasive mechanical ventilation. A total of 112 patients underwent echocardiography. There was a greater incidence of right ventricular (RV) dilatation in the fQRS group (16% vs. 2% respectively, P =.02) and pulmonary hypertension (33% vs. 14% respectively, P =.03) based on echocardiographic criteria.

CONCLUSION

Our study demonstrates that fQRS is significantly associated with RV dilation, pulmonary hypertension, and mortality in critically unwell COVID-19 patients.

Evaluation and Management of Sudden Death Risk in Repaired Tetralogy of Fallot

Although substantial progress has been made to prevent sudden cardiac death in repaired tetralogy of Fallot patients, ventricular arrhythmia and sudden death continue to be major causes of morbidity and mortality in these patients. Greater survival in contemporary cohorts has been attributed to enhanced surgical techniques, more effective management of heart failure, and increased efforts in risk stratification and management of ventricular arrhythmias. More recently, our understanding of predictive risk factors has evolved into personalized risk prediction tools that rely on comprehensive demographic, imaging, functional, and electrophysiological data. However, the universal applicability of these different scoring systems is limited due to differences between study cohorts, types of anatomic repair, imaging modalities, and disease complexity. Noninvasive risk stratification is critical to identify those who may derive benefit from catheter ablation or cardioverter defibrillator implantation for primary prevention. Ultimately, assessment and risk stratification by a multidisciplinary team is crucial to analyze the various complex factors for every individual patient and discuss further options with patients and their families.