Citations

The advent of high-throughput sequencing technologies has revolutionized cardiovascular genetics, generating vast amounts of data. A major challenge in the genomic era is the efficient and accurate identification of causative genetic variants associated with cardiovascular disease. Artificial intelligence (AI)-driven computational approaches offer a powerful solution by enabling the automation of variant classification, improving consistency and reproducibility, and enhancing predictive accuracy. These methods may guide not only variant classification, variant effect prediction, and clinical prioritization, but also open the door to data-driven precision medicine, enabling more accurate and individualized diagnoses and targeted therapeutic strategies tailored to each patient’s unique genetic and clinical profile. This state-of-the-art review provides a structured overview of AI methodologies applied to cardiovascular genomics, beginning with rule-based expert systems that encode standardized guidelines for consistent variant interpretation. Next, we examine machine learning approaches capable of identifying complex patterns in annotated multimodal clinical and multi-omic datasets. The role of deep learning algorithms is highlighted for their ability to extract features from high-dimensional, unstructured data relevant to cardiovascular disease. In addition, the potential of generative AI is explored, including applications in synthetic data generation, variant impact prediction, and automated summarization of biomedical literature. Despite advances, several challenges remain, including data heterogeneity, the need for explainable AI models to elucidate the decision mechanisms, and the complexity linked to the integration of AI-based tools into clinical workflows. Addressing these issues requires interdisciplinary collaboration among clinicians, geneticists, data scientists, and bioinformaticians to ensure the effective translation of AI-generated insights into clinical practice. This review aims to provide a comprehensive perspective on the evolving role of AI in cardiovascular genomics and its implications for advancing precision medicine.

ABSTRACT Drug‐induced QTc prolongation (diQTP) is a major risk factor for torsades de pointes and sudden cardiac death. This study evaluated whether carriers of the G allele of the NOS1AP rs10494366 T > G variant have greater risk for diQTP when prescribed high‐risk QT‐prolonging medications. A retrospective case‐control analysis was conducted using the Michigan Genomics Initiative (MGI) biobank, linking genotype and electronic health records for 5848 patients of European ancestry prescribed ≥ 1 CredibleMeds high‐risk drug for torsades de pointes between 2001 and 2022. QTc was Bazett‐corrected, and diQTP was defined as change of > 60 ms from the baseline QTc and/or QTc ≥ 500 ms. Associations between rs10494366 and diQTP were tested using unadjusted and propensity‐score‐adjusted logistic regression. The minor‐allele frequency (G) was 0.36, and genotype frequencies were in Hardy–Weinberg equilibrium ( p = 0.07). diQTP occurred in 12.2% of participants. The G allele was not significantly associated with risk of diQTP in either the unadjusted analysis (OR 0.94 [95% CI 0.84–1.05] p = 0.27) or the adjusted analysis (OR 0.94 [95% CI 0.83–1.06] p = 0.30). In this large, real‐world cohort of patients with European ancestry, the NOS1AP rs10494366 variant was not significantly associated with the risk of diQTP. These findings suggest that previously observed associations between this variant and baseline QTc may not extend to all drug‐induced settings.

ABSTRACT Background Genetic testing for cardiomyopathy and arrhythmia (CM/ARRH) provides diagnostic information and informs screening for at-risk relatives. Clinical guidelines recommend genetic testing for these conditions; however, data on how genetic test results influence clinical management recommendations are limited. Here, we determined the frequency of cardiologist-recommended management changes for patients following CM/ARRH genetic testing. Methods This was a retrospective cross-sectional study of patients referred for multigene panel testing between April 2016 and April 2024. Genetically-experienced cardiologists at multicenter academic clinical practices were recruited for participation to complete surveys indicating clinical decision making on patients who had genetic testing. Cases for review were randomly selected to have both positive and non-positive results. Results Among 249 patients (138 positive, 111 non-positive), 136 (54.6%) received clinical management recommendations for their own or their at-risk relatives’ care. Of these, 75 (55.1%) received recommendations for the patient’s own care, most frequently additional diagnostic tests/procedures (n=33). Compared to non-positive results, patients with positive results were more likely to receive recommendations for their own management (66/138, 47.8% vs 9/111, 8.1%; P<0.00001). Patients with positive results in arrhythmogenic cardiomyopathy genes had 263% higher odds of recommended management changes compared to those with TTN (OR=3.63, CI:1.40-9.84, P=0.009). Conclusions The results from CM/ARRH genetic testing on affected patients influenced cardiologists’ medical decision making and management recommendations. Additional research is needed to evaluate how genetic testing for CM/ARRH impact health outcomes.

Drug-induced QT prolongation (diQTP) can lead to rare, but potentially fatal adverse effects of many medications, yet individual susceptibility varies due to both clinical and genetic factors. SDHAF3 p.F53L (rs62624461) is a genetic variant that plays a role in mitochondrial function and was previously associated with diQTP in one small prior study. Therefore, the objective of this retrospective pharmacogenetic study was to evaluate whether SDHAF3 p.F53L is associated with clinically significant diQTP. Data were obtained from the Michigan Genomics Initiative, which links genotype information with electronic health records at Michigan Medicine. Adult patients with available genotype and electrocardiogram data who received at least one high-risk QT-prolonging drug between 2001 and 2022 were included. The analysis was limited to 5848 patients of European ancestry, of whom 320 (5.5%) were carriers of the SDHAF3 p.F53L variant. QT prolongation was defined as a QTc greater than or equal to 500 ms or an increase greater than 60 ms from baseline during high-risk QT-prolonging drug prescription. Logistic regression under a dominant genetic model assessed associations between variant carrier status and diQTP, with and without propensity score adjustment. Baseline demographics and comorbidities were similar between carriers and noncarriers. No significant association was observed between SDHAF3 p.F53L and diQTP [unadjusted odds ratio (OR) = 1.06, 95% confidence interval (CI) = 0.76–1.48, P = 0.742; adjusted OR = 1.05, 95% CI = 0.74–1.51, P = 0.773]. These results suggest that SDHAF3 p.F53L does not meaningfully influence diQTP risk in a large, real-world clinical cohort. Future studies should examine this variant across diverse populations.

ABSTRACT Background and Purpose White matter hyperintensities (WMH) are a core neuroimaging marker of cerebral small vessel disease (CSVD). Sleep apnoea (SA) is a recognized vascular risk factor, but its associations with regional WMH burden, short‐interval WMH change and cognitive performance in population‐based cohorts remain incompletely defined. We examined these associations in the UK Biobank imaging cohort. Methods We conducted cross‐sectional analyses with an exploratory longitudinal subset in propensity score–matched cohorts excluding major neurologic and cerebrovascular disease before imaging. SA was identified from hospital records and self‐reports prior to brain magnetic resonance imaging (MRI). Head‐size–normalized WMH volumes were analysed on the log1p scale using multivariable linear regression with robust standard errors. Total WMH was the primary outcome; periventricular WMH (PWMH), deep WMH (DWMH) and cognitive measures were secondary outcomes. Mediation analyses evaluated WMH as a mediator of SA–cognition associations. Results In cross‐sectional analyses ( N = 12,890; mean age, 67.4 years; 27.2% women), SA was associated with greater WMH burden. Total WMH volume was 10.7% higher among participants with SA (95% CI, 5.0%–16.6%; p < 0.001). PWMH and DWMH volumes were 10.0% (95% CI, 4.7%–15.7%; q < 0.001) and 15.7% (95% CI, 7.1%–25.1%; q < 0.001) higher, respectively. SA was associated with modestly worse cognitive performance, with effect estimates attenuated after adjustment for WMH burden and cardiometabolic conditions. Mediation analyses indicated that WMH burden, particularly PWMH, mediated associations between SA and cognitive performance. In the longitudinal subset ( N = 698; median follow‐up, 2.26 years), no significant between‐group differences were observed in annualized WMH change. Conclusions In this large population‐based imaging cohort, SA was associated with higher WMH burden, a key neuroimaging marker of CSVD. Cognitive differences were modest, but WMH burden, particularly PWMH, consistently mediated associations between SA and cognitive performance. Longer term studies are needed to clarify temporal relationships and potential treatment effects.

PURPOSE The Lynch syndrome (LS) genes, MLH1 , MSH2 , MSH6 , and PMS2 , are considered to be some of the highest-priority genes to be tested for population genomic screening, with the aim to prevent and detect cancers early in those with a higher risk for such diagnoses. The objective of this study was to determine the risk of cancer associated with pathogenic variants in MLH1 , MSH2 , MSH6 , or PMS2 in the general population to build evidence-based guidelines for population genomic screening METHODS This retrospective case-control study uses Helix Research Network data from 230,159 participants across nine US health systems. Variant interpretation was performed following the American College of Medical Genetics and Association for Molecular Pathology guidelines. Clinical diagnoses were identified from electronic health records and claims data for 11 cancer types associated with LS, including colorectal and endometrial cancers. RESULTS Individuals with pathogenic variants in MLH1 , MSH2 , and MSH6 showed substantial risk for LS-related cancers compared with those without pathogenic variants. Conversely, individuals with pathogenic variants in PMS2 had no increase in risk for all 11 LS-associated cancers with hazard ratio (HR) of 1.0 (95% CI, 0.6 to 1.8) compared with those without a pathogenic variant. There was a small increase in risk when restricting the analysis to colorectal cancer with an HR of 3.5 (1.7-7.4). This increase in colorectal cancer risk was predominantly observed after age 60 years, 10 years after US Preventive Services Task Force guidelines recommend starting colonoscopies for the average population. Up to age 60 years, 2.6% of individuals with PMS2 pathogenic variants were diagnosed with colorectal cancer, closer to the general population (0.7%) than those with MLH1 (37.3%), MSH2 (28.9%), or MSH6 (6.7%) pathogenic variants. For 100,000 individuals screened for PMS2 , 854 additional colonoscopies would be performed, which would prevent or detect early 3.2 colorectal cancers. CONCLUSION This analysis across nine health systems highlights the clinical phenotypes of PMS2 pathogenic germline variants in the entire population.

Background: The timing of coronary artery bypass grafting (CABG) in patients with non-ST-elevation myocardial infarction (NSTEMI) may critically impact clinical outcomes, yet the ideal surgical window remains unclear. We aimed to evaluate the association between time-to-CABG and in-hospital outcomes among NSTEMI admissions undergoing CABG. Methods: This retrospective cohort study utilized the US National Inpatient Sample database from 2017 to 2021 to evaluate the association between time-to-CABG and key in-hospital outcomes, including mortality, cardiac arrest, cardiogenic shock, ischemic stroke, acute kidney injury, and blood transfusion rates, while assessing length of stay (LOS) and hospital charges exclusively among survivors. Adult admissions with principal diagnosis of NSTEMI who underwent CABG without prior interhospital transfer were grouped by the calendar day of CABG from admission (0, 1, 2, 3, 4, 5, 6, and ≥7). Multivariate regression models were used to assess the outcomes across these eight groups. Results: Of 142,200 included admissions, same-day CABG (hospital day 0) was associated with the highest rates of in-hospital mortality, cardiac arrest, cardiogenic shock, and ischemic stroke. The adjusted odds of mortality, cardiac arrest, and cardiogenic shock were lower among admissions undergoing CABG between days 2 and 5. Likewise, the adjusted odds of ischemic stroke were lowest when CABG occurred on days 2 to 4. The odds of AKI increased when CABG was delayed beyond day 6, whereas blood transfusion rates showed no statistically significant association with time-to-CABG. Among survivors, LOS and hospital charges progressively increased with delayed CABGs. Conclusions: Among NSTEMI admissions undergoing CABG, CABG performed on hospital days 2 to 4 was associated with lower adjusted odds of several adverse in-hospital outcomes compared with same-day CABG. Further prospective studies are needed to validate these results and guide clinical decision-making.

ABSTRACT Objective Hypertension is a key risk factor for aortic dissection (AD). AD, if left untreated, carries significant mortality rates. Our aim is to analyse trends in mortality due to AD in adults with primary hypertension in the United States (US). Methods We used the CDC WONDER database to extract mortality data for patients with primary hypertension who died due to AD. Age‐adjusted mortality rates (AAMRs) and crude mortality rates (CMRs) were extracted per 100 000 persons. Annual percentage changes (APCs) and average APCs (AAPCs) in AAMRs and CMRs were calculated using Joinpoint regression. Results From 1999 to 2020, a total of 13 128 deaths due to AD were reported in patients with primary hypertension in the US. Males displayed a higher overall AAMR (0.3) than females (0.2) throughout the study period. Slight regional variations were observed, with the West showing the highest overall AAMR (0.4), followed by the Midwest (0.3), and the Northeast and South (0.2). In urban areas, AAMRs were higher than in rural areas until 2008. From 2009 to 2020, AAMRs remained stable in urban areas (0.3) but increased in rural areas from 2010 to 2020 (4.7). The highest state‐level AAMRs were observed in Hawaii, Oregon, and Oklahoma. Conclusion Significant differences were observed in AAPCs compared to AD‐related mortality trends in the general population. Mortality trends revealed an initial decline followed by a gradual rise. Clinicians should focus on high‐risk groups and raise awareness about the disease in these populations.

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.