Emanuele Bobbio

The manuscript, _“Beyond the Heart and Lungs: Extrapulmonary and Treatment-Related Emergencies in Critically Ill Patients with Sarcoidosis,”_ addresses a clinically relevant topic. The author has made a considerable effort to summarize the available...

Artificial intelligence (AI) is transforming cardiovascular medicine by extracting clinically relevant patterns from complex digital health data, extending beyond traditional statistical approaches. However, the translation of AI from proof-of-concept studies to routine cardiovascular care remains constrained by methodological, ethical, and regulatory challenges. Key barriers include data quality and representativeness, model generalizability, calibration and clinical utility, interpretability, and the risk of bias amplification. To address this gap, this review synthesizes current evidence on AI applications in cardiology, demystifies foundational machine learning concepts for a clinical audience, and outlines a practical framework for best practices in model development, evaluation, and responsible clinical deployment.

Background: Mutations in the Four-and-a-Half LIM Domains 1 ( FHL1 ) gene are increasingly recognized as a rare cause of inherited cardiomyopathies, often associated with skeletal myopathy and adverse cardiac outcomes. The phenotypic spectrum and clinical implications of FHL1 variants remain poorly defined. Objective: To systematically review published cases of FHL1 -related cardiomyopathy and characterize the clinical, genetic, and pathological features. Methods: We conducted a systematic literature search in PubMed and EMBASE up to July 2025 using predefined criteria to identify studies reporting clinical cases of patients with FHL1 mutations and cardiac involvement. Data on genotype, phenotype, cardiac and neuromuscular features, and clinical outcomes were extracted and synthesized. Results: Twenty-two studies were included, comprising 114 patients with pathogenic or likely pathogenic FHL1 mutations. Most patients were male (69%), with a median age of onset of 18 (IQR 10–26) years. Cardiac involvement consisted in left ventricular hypertrophy (56%), followed by arrhythmias (51%), and conduction abnormalities (8%). The incidence of sudden cardiac death was 7%, and heart transplantation was reported in 5% of patients. Skeletal muscle involvement was present in 75%, ranging from mild contractures to more severe myopathic phenotypes with functional impairment. Creatine kinase levels were variably elevated. Truncating variants were reported in several severe cardiac presentations in young males, while isolated cardiac disease occurred with selected variants. Conclusions: FHL1 -related cardiomyopathy is a rare but important diagnosis. Genetic testing should be considered in patients with cardiac hypertrophy and neuromuscular features. Further research is needed to define prognostic markers and guide management.

BACKGROUND: Sudden arrhythmic death syndrome (SADS) refers to sudden cardiac death with structurally normal hearts at autopsy, most frequently attributed to inherited arrhythmia syndromes or concealed cardiomyopathies. Postmortem genetic testing may help identify underlying genetic causes. We aimed to investigate the yield of postmortem genetic testing in SADS cases by determining the prevalence of pathogenic or likely pathogenic variants in channelopathy- and cardiomyopathy-associated genes in autopsy-negative SADS victims. METHODS: This systematic review and meta-analysis followed Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines and was registered in PROSPERO (REGISTRATION: URL: https://www.crd.york.ac.uk/PROSPERO/ ; Unique identifier: CRD420251067244). PubMed and Embase were searched on June 4, 2025, for observational studies including individuals aged 1 to 50 years with SADS and negative or nonspecific findings at autopsy. Eligible studies reported postmortem genetic testing for channelopathy and cardiomyopathy genes. Pathogenic or likely pathogenic variant classification followed American College of Medical Genetics and Genomics criteria and ClinGen gene-disease associations. Pooled prevalence was estimated using random-effects models. RESULTS: A total of 45 studies involving 2498 SADS cases were included. Among 1697 SADS victims tested for both channelopathy and cardiomyopathy genes (33 studies), the pooled prevalence of pathogenic or likely pathogenic variants was 11.1% (95% CI, 4.1%–26.6%, I ² =50.7%). Testing for cardiomyopathy genes (33 studies, 1697 cases) and for channelopathy genes (42 studies, 2354 cases) yielded a prevalence of 7.0% (95% CI, 1.9%–22.9%, I ² =51.9%) and 6.3% (95% CI, 2.0%–18.4%, I ² =49.8%), respectively. The most frequently involved genes encoded sarcomeric proteins and ion channels, with TTN , MYBPC3 , MYH7 , KCNH2 , and SCN5A among the most commonly affected. CONCLUSIONS: Postmortem genetic testing identifies pathogenic or likely pathogenic variants in a significant subset of SADS cases, supporting its utility in postmortem evaluation.

Aims: Duchenne (DMD) and Becker (BMD) muscular dystrophies often progress to dilated cardiomyopathy and heart failure with reduced ejection fraction (HFrEF), a major cause of death in affected patients. Although guideline-directed medical therapy (GDMT) improves HFrEF outcomes, its real-world use in dystrophinopathies is poorly defined. We assessed GDMT utilization and optimization in the study population. Methods and results: HOPE-MD (Heart failure treatment OPtimization in patiEnts with Muscular Dystrophy) is a European, multicentre, registry including adults with genetically confirmed DMD or BMD and left ventricular ejection fraction (LVEF) ≤ 40%. GDMT was defined as concurrent use of a renin–angiotensin system inhibitor (RASi), beta-blocker, mineralocorticoid receptor antagonist (MRA), and sodium–glucose cotransporter-2 inhibitor (SGLT2i). Doses were categorized as target (≥100%), intermediate (50–99%), or low (<50%). Among 274 patients (age, 28 years; LVEF, 35%; DMD, 80%), 44 (16%) received GDMT at baseline, with only one quarter at ≥50% of the target dose. Underutilization was greatest for MRAs (n = 140, 51%) and SGLT2i (n = 60, 22%), whereas RASi (n = 246, 90%) and beta-blockers (n = 235, 86%) were frequently prescribed. After 24 months (n = 141), 40% received GDMT, and 22% received ≥50% of the target dose, primarily through MRA and SGLT2i initiation/uptitration. Contraindications accounted for only a small proportion of undertreatment. At a median follow-up of 2.2 (1.7–2.6) years, optimized GDMT was not associated with a statistically significant difference in the rates of death or HF hospitalization compared with low-dose or no GDMT (adjusted HR 0.70, 95% CI 0.28–1.74). Conclusion: In this European, multicentre registry, GDMT use and dose escalation were suboptimal in young patients with DMD/BMD and HFrEF.

Sudden cardiac death (SCD) is a major global health issue, defined as sudden natural death presumed to be of cardiac cause. While in the elderly SCD is commonly associated with coronary artery disease, in the younger population it is linked to inherited cardiomyopathies or channelopathies, even though SCD can remain unexplained even after a comprehensive autopsy in a substantial proportion of cases. In this context, genetic testing has gained importance, supported by the widespread availability of techniques such as next-generation and whole-exome/genome sequencing and their reduced costs. This state-of-the-art review summarizes the genetic bases of sudden cardiac death among cardiomyopathies, channelopathies and in sudden unexplained death presumed to be of arrhythmic cause. Among the structural causes, inherited cardiomyopathies such as hypertrophic, dilated, non-dilated left ventricular, arrhythmogenic right ventricular and restrictive ones represent major substrates for malignant ventricular arrhythmias mostly arising from variants in sarcomeric or desmosomal genes. Channelopathies (long or short QT syndrome, Brugada syndrome and catecholaminergic polymorphic ventricular tachycardia) are caused by variants in genes encoding cardiac ion channels and/or regulatory proteins, which equally predispose to high risk of life-threatening ventricular arrhythmias. In sudden arrhythmic death syndrome, with a structurally normal heart, post-mortem genetic testing (molecular autopsy) can uncover an underlying inherited condition. However, variants of uncertain significance are detected in more than half of the cases, underscoring the need for a multidisciplinary approach. Genetic testing also plays a key role in cascade screening of first-degree relatives. While monogenic variants drive risk in inherited cardiac disorders, emerging evidence suggests that polygenic contributions may modulate SCD susceptibility, highlighting future roles for polygenic risk scores in risk stratification.

Rare cardiovascular diseases are often diagnosed late, not because they are clinically silent, but because their biological signals are fragmented across different diagnostic layers. Genomic variation may be present without clear functional consequences, imaging abnormalities may overlap with common age-related changes, and clinical phenotypes may only fully emerge years after the initial molecular insult (1). In this context, the central challenge is no longer the lack of diagnostic tools, but the difficulty of integrating them into a coherent diagnostic strategy (2).The contributions gathered in this Research Topic address this challenge from complementary angles, collectively arguing that precision in rare cardiovascular disease depends on interrogating the right biological layer at the right moment. Rather than proposing a single "best" technology, these studies show how genome structure, transcript-level function, quantitative imaging phenotypes, and genotype-informed clinical evaluation can converge to reduce diagnostic uncertainty and improve interpretability.One persistent source of uncertainty in congenital heart disease lies in genomic architecture.Despite widespread use of chromosomal microarray analysis and sequencing-based approaches, a substantial proportion of patients, particularly those with complex phenotypes such as heterotaxy, remain without a molecular diagnosis (3). Structural variants, including balanced rearrangements and complex genomic events, represent a plausible contributor to this missing heritability but are incompletely captured by standard methods. Optical genome mapping has emerged as a next-generation cytogenetic approach capable of genome-wide structural variant detection with high resolution and without reliance on amplification (4). Its clinical relevance has been demonstrated in several disease settings, supporting its role as a complementary diagnostic layer rather than a replacement for sequencing.Within the present Research Topic, Min et al. apply optical genome mapping to patients with congenital heart disease and heterotaxy whose prior genetic testing had been inconclusive. By integrating structural variant data with existing genomic information, they identify clinically relevant rearrangements affecting known disease-associated genes, as well as candidate loci that merit further investigation. The broader implication of this work is conceptual: unresolved cases may not reflect "negative genetics," but rather incomplete interrogation of genome structure.Even when sequence variation is identified, diagnostic ambiguity often persists. Variants affecting splicing, transcript stability, or allele-specific expression may remain classified as uncertain when evaluated at the DNA level alone. Transcriptome-informed diagnostics have therefore gained increasing attention, with RNA sequencing shown to improve diagnostic yield and clarify pathogenic mechanisms across Mendelian diseases (5). This functional dimension is particularly relevant for large, multi-exon genes with complex expression patterns., the gene encoding fibrillin-1 and the principal molecular determinant of Marfan syndrome, by developing a targeted whole-blood RNA sequencing assay covering the full coding region of FBN1. Applied to families with suspected Marfan syndrome, the assay not only detects pathogenic variants with high sensitivity but also reveals their transcriptional consequences, including aberrant splicing and RNA polymerase slippage. This functional evidence directly supports diagnostic confidence in a condition where accurate classification has profound clinical implications for aortic surveillance and prophylactic intervention, as emphasized by the revised Ghent nosology (6). More broadly, this work illustrates how RNA-based approaches can act as a bridge between genotype and mechanism, transforming uncertain variant calls into clinically meaningful diagnoses. Molecular diagnosis, however, does not end at variant interpretation. It must ultimately inform risk stratification and monitoring at the organ level. Fabry disease exemplifies this translational challenge. Although its molecular basis is well defined, early cerebrovascular involvement can be difficult to distinguish from common vascular changes using conventional imaging. Prior studies have described characteristic posterior circulation involvement, including basilar artery remodeling, but these findings have rarely been integrated into a unified diagnostic framework (7).brain MRI markers in genetically confirmed Fabry disease. Using high-resolution vessel wall imaging, they demonstrate that perivascular space burden and basilar artery remodeling are coupled in Fabry disease but not in matched controls, suggesting a disease-specific pattern of vascular degeneration. This approach reframes advanced imaging as a quantitative downstream phenotype of molecular pathology positioned between genotype and clinical events, and potentially valuable for early diagnosis and longitudinal monitoring.Finally, molecular diagnostics should not be viewed solely as a means of etiologic classification.Their clinical value also lies in identifying modifiable contributors to disease expression. Pediatric primary cardiomyopathies are rare, genetically heterogeneous, and associated with high morbidity and mortality, prompting major societies to emphasize systematic phenotyping and genetic evaluation as the foundation of care (8). Yet even in genetically defined disease, biological modifiers may influence severity and reversibility.In this context, Jiang et al. explore the relationship between genetic findings and electrolyte and amino acid pathways in children with primary cardiomyopathies. In a genetically characterized cohort, they report improvements in cardiac function among children with dilated cardiomyopathy treated with calcium supplementation. While exploratory, these observations resonate with earlier reports of...

RASopathies are a group of genetic disorders caused by pathogenic variants in the RAS‐mitogen‐activated protein kinase (RAS–MAPK) signaling pathway, often presenting with congenital heart defects, craniofacial dysmorphisms, and developmental delays. To assess the diagnostic yield of genetic testing in patients with suspected RASopathies and to identify predictive clinical “red flags” using machine learning (ML). This retrospective study included patients evaluated at our institution between 2020 and 2023. All patients underwent cardiovascular evaluation, and a subset of individuals with suspected RASopathies underwent genetic assessment including exome sequencing. Thirteen clinical “red flags” were analyzed as predictors of a molecular diagnosis within the subgroup of patients with suspected RASopathies. Diagnostic performance was assessed via sensitivity, specificity, and area under the curve (AUC). A random forest classifier identified the most predictive clinical features. Among 669 patients, 34 were clinically suspected of RASopathy, with a confirmed diagnosis in 24 cases (71%). Noonan syndrome was most frequent (18/24), and PTPN11 was the most commonly mutated gene ( n = 13). Pulmonary valve stenosis (PVS) and facial dysmorphisms were the strongest individual predictors of a positive genetic test result. A threshold of ≥ 2 red flags balanced sensitivity (92%) and accuracy (73.5%). ML analysis identified PVS and facial dysmorphisms as the top predictors. A model including only these features achieved an AUC of 0.86 in the derivation cohort. PVS and facial dysmorphisms are key diagnostic indicators for RASopathies and should prompt early genetic testing. While the ML model showed high performance in this derivation cohort, external validation is needed to confirm its generalizability.