Citations

Abstract Aims Left ventricular hypertrophy (LVH) is a strong predictor of cardiovascular disease. We previously compared supervised machine learning techniques to classify cardiac magnetic resonance (CMR)-derived LVH using electrocardiogram (ECG) and clinical variables in 37 534 UK Biobank participants, obtaining an area under the receiving operating curve (AUROC) of 0.85, but with limited specificity and requiring external validation. In this study, we develop a deep learning (DL) model to improve classification with external evaluation in the Study of Health in Pomerania (SHIP). Methods and results We analysed 12-lead ECGs of 48 835 participants from the UK Biobank imaging study. The dataset was split into a training set (70%), validation set (15%), and test set (15%) for performance evaluation. The model architecture was a fully convolutional network, for which the input was the participants’ median ECG and clinical variables and the predicted indexed left ventricular mass (iLVM) as the output. A subsequent logistic regression model was used to recalibrate iLVM predictions. In UK Biobank, 717 (1.5%) participants had CMR-derived LVH and the AUROC for the DL model was 0.97. The ECG components most predictive of LVH were the QRS complex and ventricular rate. The DL model outperformed our supervised algorithms, previous DL modelling efforts and clinical ECG benchmarks. There was modest generalizability of the DL model to 1423 participants in SHIP (AUROC 0.78), with differences in clinical profile, ECG acquisition, and CMR labelling as important factors. Conclusion Our findings support the feasibility of scalable DL-based screening tools for the prediction of LVH from the ECG, whilst highlighting the need for model development using larger datasets with greater diversity to ensure generalizability.

Objective: Four hypertension-mediated left ventricular hypertrophy (LVH) phenotypes have been reported using cardiac magnetic resonance (CMR): normal LV, LV remodelling, eccentric and concentric LVH, with varying prognostic implications. The electrocardiogram (ECG) is routinely used to detect LVH; however, its capacity to differentiate between LVH phenotypes is unknown. This study aimed to classify hypertension-mediated LVH from the ECG using machine learning and test for associations of ECG-predicted phenotypes with incident cardiovascular outcomes. Methods: ECG biomarkers were extracted from the 12-lead ECG of 20 439 hypertensive patients in UK Biobank (UKB). Classification models integrating ECG and clinical variables were built using logistic regression, support vector machine (SVM), and random forest. The models were trained in 80% of the participants, and the remaining 20% formed the test set. External validation was sought in 877 hypertensive participants from the Study of Health in Pomerania (SHIP). In the UKB test set, we tested for associations between ECG-predicted LVH phenotypes and incident major adverse cardiovascular events (MACE) and heart failure. Results: Among UKB participants 19 408 had normal LV, 758 LV remodelling, 181 eccentric and 92 concentric LVH. Classification performance of the three models was comparable in UKB. SVM (accuracy 0.79, sensitivity 0.59, specificity 0.87, AUC 0.69) was taken forward for external validation with similar results in SHIP. There was superior prediction of eccentric LVH in both cohorts. In the UKB test set, ECG-predicted eccentric LVH was associated with heart failure (hazard ratio 3.42, 95% CI 1.06–9.86). Conclusion: ECG-based ML classifiers represent a potentially accessible screening strategy for the early detection of hypertension-mediated LVH phenotypes.

Abstract Aims In light of recent advances in imaging techniques, molecular understanding and therapeutic options in hypertrophic cardiomyopathy (HCM), we performed a systematic review of current guidelines for the diagnosis and management of HCM in order to identify consensus and discrepant areas in the clinical practice guidelines. Methods and results We systematically reviewed the English language guidelines and recommendations for the management of HCM in adults. MEDLINE and EMBASE databases were searched for guidelines published in the last 10 years. Following a systematic search, three guidelines on the diagnosis and management of HCM were identified, all of which were robustly developed (AGREE rigour of development score ≥50%). These guidelines were authored by the major European (European Society of Cardiology; 2023), American (American Heart Association /American College of Cardiology/American Medical Society for Sports Medicine /Heart Rhythm Society/Pediatric and Congenital Electrophysiology Society/Society for Cardiovascular Magnetic Resonance; 2024), and Japanese [Japanese Circulation Society (JCS)/Japanese Heart Failure Society (JHFS); 2018] cardiovascular societies. There was broad consensus on echocardiographic recommendations, the medical and invasive management of HCM, the application of genetic testing and family screening, and exercise and reproductive recommendations in HCM. There were areas of variability in the definition and diagnostic criteria for HCM, cardiovascular magnetic resonance imaging recommendations, and assessment of sudden cardiac death (SCD) risk and prevention strategies. Due to the JCS/JHFS guidelines being older, there are no recommendations on the use of cardiac myosin ATPase inhibitors. Conclusion Contemporary guidelines for HCM achieve consensus across a broad range of criteria and recommendations concerning diagnosis and management. However, variations in the approach towards risk assessment for SCD exist between the guidelines. There are also more subtle differences concerning diagnostic criteria and the utility of late gadolinium enhancement for risk stratification, which will likely evolve as the evidence-base broadens.

Background: Mitral valve prolapse (MVP) is a common valvular abnormality. Observational studies demonstrate an association between MVP and increased risk of sudden cardiac death from ventricular arrhythmias, coined arrhythmogenic MVP. A genetic component has been postulated but remains under-investigated. Hypothesis: Predicted deleterious variants in Mendelian cardiomyopathy and arrhythmia genes are more frequent in arrhythmogenic MVP (aMVP) compared to non-arrhythmogenic MVP. Aims: To comprehensively phenotype and genotype a single-center aMVP cohort and compare to non-aMVP from two large population-based cohorts and investigate their clinical outcomes. Methods: The aMVP cohort included 46 prospectively enrolled patients from a regional referral center for inherited cardiovascular disease and had undergone multi-modality imaging and clinical gene panel testing. The non-aMVP comparator cohorts included 453 participants from the UK Biobank (UKBB), a prospective cohort of 500,000 UK residents, and 1,418 participants from the National Institutes of Health (NIH) All of Us (AoU), a prospective cohort of 450,000 USA participants. Gene variants were annotated in ANNOVAR. Participants with and without pathogenic/likely pathogenic (P/LP) variants reported on ClinVar with ≥2-star level of evidence were considered genotype positive (G+) and negative (G-) respectively. Results: The demographics for aMVP (age 61.2 ± 12.8 years, 56.5% female) and non-aMVP (UKBB: age 59.8 ± 7.3 years, 50.7% female and AoU: age 65.3 ± 13.4 years, 76.2% female) were comparable. G+ participants were more prevalent in the aMVP cohort 10/46 (21.7%) versus non-aMVP in the UKBB 8/453 (1.8%) and in the AoU 21/1418 (1.5%); p<0.001. The most frequently implicated gene in aMVP was TTN . Of 46 in aMVP cohort, 33 (71%) underwent CMR. G+ participants had a higher frequency of late gadolinium enhancement [6/6 (100%) vs. 14/27 (51.8%), p<0.001] and wall motion abnormalities [3/6 (50%) vs. 1/27 (3.7%), p<0.001] compared to G- participants. The aMVP cohort had significantly more implantable cardioverter defibrillator (ICD) implants compared to non-aMVP cohort [5/46 (10.9%) vs. 1/453 in UKBB (0.2%), p<0.001]. G+ aMVP cohort showed a trend for developing heart failure and ICD implants compared to G- aMVP . Conclusion: P/LP variants in cardiomyopathy and arrhythmia genes are more prevalent in aMVP participants compared to non-aMVP participants. This could be a hitherto unrecognized effect modifier predicting aMVP .

Precision medicine, which among other aspects includes an individual's genomic data in diagnosis and management, has become the standard‐of‐care for Mendelian cardiovascular disease (CVD). However, early identification and management of asymptomatic patients with potentially lethal and manageable Mendelian CVD through screening, which is the promise of precision health, remains an unsolved challenge. The reduced costs of genomic sequencing have enabled the creation of biobanks containing in‐depth genetic and health information, which have facilitated the understanding of genetic variation, penetrance, and expressivity, moving us closer to the genotype‐first screening of asymptomatic individuals for Mendelian CVD. This approach could transform health care by diagnostic refinement and facilitating prevention or therapeutic interventions. Yet, potential benefits must be weighed against the potential risks, which include evolving variant pathogenicity assertion or identification of variants with low disease penetrance; costly, stressful, and inappropriate diagnostic evaluations; negative psychological impact; disqualification for employment or of competitive sports; and denial of insurance. Furthermore, the natural history of Mendelian CVD is often unpredictable, making identification of those who will benefit from preventive measures a priority. Currently, there is insufficient evidence that population‐based genetic screening for Mendelian CVD can reduce adverse outcomes at a reasonable cost to an extent that outweighs the harms of true‐positive and false‐positive results. Besides technical, clinical, and financial burdens, ethical and legal aspects pose unprecedented challenges. This review highlights key developments in the field of genotype‐first approaches to Mendelian CVD and summarizes challenges with potential solutions that can pave the way for implementing this approach for clinical care.

ImportanceThe population prevalence of cardiac transthyretin amyloidosis (ATTR) caused by pathogenic variation in the TTR gene (vATTR) is unknown.ObjectiveTo estimate the population prevalence of disease-causing TTR variants and evaluate associated phenotypes and outcomes.Design, Setting, and ParticipantsThis population-based cohort study analyzed UK Biobank (UKB) participants with whole-exome sequencing, electrocardiogram, and cardiovascular magnetic resonance data. Participants were enrolled from 2006 to 2010, with a median follow-up of 12 (IQR, 11-13) years (cutoff date for the analysis, March 12, 2024). Sixty-two candidate TTR variants were extracted based on rarity (minor allele frequency ≤0.0001) and/or previously described associations with amyloidosis if more frequent.ExposureCarrier status for TTR variants.Main Outcomes and MeasuresAssociations of TTR carrier status with vATTR prevalence and cardiovascular imaging and electrocardiogram traits were explored using descriptive statistics. Associations between TTR carrier status and atrial fibrillation, conduction disease, heart failure, and all-cause mortality were evaluated using adjusted Cox proportional hazards models. Genotypic and diagnostic concordance was examined using International Statistical Classification of Diseases, Tenth Revision codes from the hospital record.ResultsThe overall cohort included 469 789 UKB participants (mean [SD] age, 56.5 [8.1] years; 54.2% female and 45.8% male). A likely pathogenic/pathogenic (LP/P) TTR variant was detected in 473 (0.1%) participants, with Val142Ile being the most prevalent (367 [77.6%]); 91 individuals (0.02%) were carriers of a variant of unknown significance . The overall prevalence of LP/P variants was 0.02% (105 of 444 243) in participants with European ancestry and 4.3% (321 of 7533) in participants with African ancestry. The LP/P variants were associated with higher left ventricular mass indexed to body surface area (β = 4.66; 95% CI, 1.87-7.44), and Val142Ile was associated with a longer PR interval (β = 18.34; 95% CI, 5.41-31.27). The LP/P carrier status was associated with a higher risk of heart failure (hazard ratio [HR], 2.68; 95% CI, 1.75-4.12) and conduction disease (HR, 1.88; 95% CI, 1.25-2.83). Higher all-cause mortality risk was observed for non-Val142Ile LP/P variants (HR, 1.98; 95% CI, 1.06-3.67). Thirteen participants (2.8%) with LP/P variants had diagnostic codes compatible with cardiac or neurologic amyloidosis. Variants of unknown significance were not associated with outcomes.Conclusions and RelevanceThis study found that approximately 1 in 1000 UKB participants were LP/P TTR variant carriers, exceeding previously reported prevalence. The findings emphasize the need for clinical vigilance in identifying individuals at risk of developing vATTR and associated poor outcomes.

Abstract Aims Left ventricular hypertrophy (LVH) is an established, independent predictor of cardiovascular disease. Indices derived from the electrocardiogram (ECG) have been used to infer the presence of LVH with limited sensitivity. This study aimed to classify LVH defined by cardiovascular magnetic resonance (CMR) imaging using the 12-lead ECG for cost-effective patient stratification. Methods and results We extracted ECG biomarkers with a known physiological association with LVH from the 12-lead ECG of 37 534 participants in the UK Biobank imaging study. Classification models integrating ECG biomarkers and clinical variables were built using logistic regression, support vector machine (SVM) and random forest (RF). The dataset was split into 80% training and 20% test sets for performance evaluation. Ten-fold cross validation was applied with further validation testing performed by separating data based on UK Biobank imaging centres. QRS amplitude and blood pressure (P &lt; 0.001) were the features most strongly associated with LVH. Classification with logistic regression had an accuracy of 81% [sensitivity 70%, specificity 81%, Area under the receiver operator curve (AUC) 0.86], SVM 81% accuracy (sensitivity 72%, specificity 81%, AUC 0.85) and RF 72% accuracy (sensitivity 74%, specificity 72%, AUC 0.83). ECG biomarkers enhanced model performance of all classifiers, compared to using clinical variables alone. Validation testing by UK Biobank imaging centres demonstrated robustness of our models. Conclusion A combination of ECG biomarkers and clinical variables were able to predict LVH defined by CMR. Our findings provide support for the ECG as an inexpensive screening tool to risk stratify patients with LVH as a prelude to advanced imaging.

Background: There is a paucity of data regarding the phenotype of dilated cardiomyopathy (DCM) gene variants in the general population. We aimed to determine the frequency and penetrance of DCM-associated putative pathogenic gene variants in a general adult population, with a focus on the expression of clinical and subclinical phenotype, including structural, functional, and arrhythmic disease features. Methods: UK Biobank participants who had undergone whole exome sequencing, ECG, and cardiovascular magnetic resonance imaging were selected for study. Three variant-calling strategies (1 primary and 2 secondary) were used to identify participants with putative pathogenic variants in 44 DCM genes. The observed phenotype was graded DCM (clinical or cardiovascular magnetic resonance diagnosis); early DCM features, including arrhythmia or conduction disease, isolated ventricular dilation, and hypokinetic nondilated cardiomyopathy; or phenotype-negative. Results: Among 18 665 individuals included in the study, 1463 (7.8%) possessed ≥1 putative pathogenic variant in 44 DCM genes by the main variant calling strategy. A clinical diagnosis of DCM was present in 0.34% and early DCM features in 5.7% of individuals with putative pathogenic variants. ECG and cardiovascular magnetic resonance analysis revealed evidence of subclinical DCM in an additional 1.6% and early DCM features in an additional 15.9% of individuals with putative pathogenic variants. Arrhythmias or conduction disease (15.2%) were the most common early DCM features, followed by hypokinetic nondilated cardiomyopathy (4%). The combined clinical/subclinical penetrance was ≤30% with all 3 variant filtering strategies. Clinical DCM was slightly more prevalent among participants with putative pathogenic variants in definitive/strong evidence genes as compared with those with variants in moderate/limited evidence genes. Conclusions: In the UK Biobank, ≈1 of 6 of adults with putative pathogenic variants in DCM genes exhibited early DCM features potentially associated with DCM genotype, most commonly manifesting with arrhythmias in the absence of substantial ventricular dilation or dysfunction.

Arrhythmogenic cardiomyopathy (ACM) is an inherited cardiomyopathy characterised by ventricular arrhythmia and an increased risk of sudden cardiac death (SCD). Numerous genetic determinants and phenotypic manifestations have been discovered in ACM, posing a significant clinical challenge. Further to this, wider evaluation of family members has revealed incomplete penetrance and variable expressivity in ACM, suggesting a complex genotype-phenotype relationship. This review details the genetic basis of ACM with specific genotype-phenotype associations, providing the reader with a nuanced perspective of this condition; whilst also proposing a future roadmap to delivering precision medicine-based management in ACM.