2026/01/15 by Andrea Baggiano, Gianluca Pontone · 1 voice
Medicine · #Cardiac electrophysiology and arrhythmias #Cardiomyopathy and Myosin Studies #Cardiovascular Effects of Exercise
paper · doi:10.1093/ehjci/jeag015
openalex publication_date 2026/01/15 · openalex created_date 2026/01/21 · openalex updated_date 2026/07/14
This editorial refers to ‘Subendocardium-involved late gadolinium enhancement in non-ischemic dilated cardiomyopathy improves risk stratification of sudden cardiac death’, by X. Jia et al., https://doi.org/10.1093/ehjci/jeaf376. Sudden cardiac death (SCD) remains a major contributor to mortality in patients with non-ischaemic dilated cardiomyopathy (NICM). For decades, primary prevention strategies have relied almost exclusively on left ventricular ejection fraction (LVEF) estimation, with implantable cardioverter defibrillator (ICD) implantation recommended for patients with severely reduced systolic function. However, the limitations of this paradigm are increasingly evident. Many patients fulfilling guideline-based criteria never experience malignant ventricular arrhythmias, while a substantial proportion of SCD events occur in individuals with preserved or only mildly reduced LVEF.1–3 This discordance has driven an ongoing search for more precise, mechanistically grounded approaches to arrhythmic risk stratification. Furthermore, a European Heart Rhythm Association survey by Brociek et al. investigated contemporary clinical practice for primary prevention ICD therapy in NICM. Among 258 European physicians, most reported usually adhering to ESC guideline criteria, yet marked variability in implantation strategies emerged. Beyond LVEF, family history of SCD, cardiac resynchronization therapy indication, unexplained syncope, young age, extent of late gadolinium enhancement (LGE) on cardiac magnetic resonance (CMR), and high-risk genetic profiles strongly influenced decision-making. However, meaningful heterogeneity still persists.4 CMR has progressively assumed a central role in this endeavour. Beyond its established accuracy for ventricular volumes and function, CMR uniquely enables in vivo myocardial tissue characterization, most notably through LGE, a robust marker of replacement fibrosis. Substantial scientific evidence demonstrates that myocardial scar detected by LGE provides powerful and incremental prognostic information for ventricular arrhythmias and SCD in NICM, independent of LVEF.5–7 Nevertheless, incorporation of CMR-derived fibrosis into formal ICD decision-making remains limited, reflecting uncertainty regarding which scar features matter most and how they should be integrated into clinical algorithms. Jia et al.8 address a timely question: whether specific LGE phenotypes, defined by the integration of scar pattern and anatomical location, can improve SCD risk stratification in NICM. Analysing a large cohort of patients undergoing CMR, the authors identify two distinct LGE phenotypes, septal mid-wall LGE and lateral subendocardium-involved LGE, as independent predictors of SCD and surrogate arrhythmic events. Most notably, the coexistence of these two phenotypes conferred a markedly higher arrhythmic risk, enabling the derivation of a pragmatic three-tier risk model that substantially outperformed LVEF alone in discrimination and reclassification. The importance of septal mid-wall LGE as an arrhythmogenic substrate in NICM is well established. Multiple studies have consistently shown that mid-wall fibrosis, particularly when involving the interventricular septum, is associated with an increased risk of ventricular arrhythmias and SCD. Its prognostic relevance has been demonstrated across different populations, making it one of the most reproducible CMR markers of arrhythmic risk in NICM.6,7,9,10 From a pathophysiological perspective, septal mid-wall fibrosis likely reflects diffuse interstitial remodelling and impaired electrical coupling, creating a substrate facilitating re-entrant arrhythmias. In contrast, subendocardial LGE has traditionally been regarded as the hallmark of ischaemic myocardial injury. Its detection in patients classified as having non-ischaemic cardiomyopathy has therefore been viewed with scepticism, often raising concerns regarding occult coronary artery disease or misclassified myocardial infarction with non-obstructive coronary arteries. Jia et al. address this concern directly through careful phenotyping, systematic exclusion of acute coronary syndromes, and assessment of coronary anatomy in the majority of them. Their findings reinforce the concept that subendocardial fibrosis can occur in NICM in the absence of obstructive epicardial coronary disease and that, when present, it carries clinically meaningful prognostic implications. Regardless of the underlying mechanism (chronic microvascular dysfunction, reduced subendocardial perfusion reserve, increased wall stress in a dilated ventricle, etc.), the present study provides evidence that lateral subendocardium-involved LGE is not an incidental finding but a marker of enhanced arrhythmic vulnerability.11–13 The novelty of the work by Jia et al. lies not only in identifying a previously underappreciated LGE phenotype, but also in demonstrating the additive prognostic value of combining scar pattern and location. This integrated approach aligns with a growing body of literature suggesting that arrhythmic risk in NICM is driven less by the global burden of fibrosis than by the spatial organization of scar within the myocardium. A notable example is provided by experience of DERIVATE-NICM Registry, in which the addition of LGE-derived parameters to clinical and functional variables provided a substantial net reclassification improvement for major adverse arrhythmic cardiac events, identifying a sizeable proportion of patients meeting current guideline criteria for primary prevention ICD implantation as being at low arrhythmic risk, while simultaneously refining the identification of truly high-risk individuals most likely to benefit from device therapy.14 Importantly, subsequent analyses from Guaricci et al. from the same registry have further strengthened these observations by demonstrating that segment-specific weighting of mid-wall LGE location significantly enhanced risk discrimination compared with models relying solely on global fibrosis burden, thus underscoring the biological and electrophysiological heterogeneity of fibrotic substrates in non-ischaemic cardiomyopathy.15 Interestingly, differences can be acknowledged between evidence from Jia et al. and DERIVATE-NICM registry experience. The latter, in a larger multicentre and multivendor setting, showed that a pragmatic multiparametric CMR-based score incorporating sex, ventricular remodelling, and segment-weighted mid-wall LGE distribution enables substantial and clinically meaningful reclassification of ICD candidates, while the former extends this concept by identifying a higher-risk fibrotic phenotype characterized by the coexistence of septal mid-wall and lateral subendocardial LGE. These two approaches seem to be not mutually exclusive but rather illuminate different layers of arrhythmogenic substrate complexity. DERIVATE prioritizes robustness, reproducibility, and immediate translational applicability by refining an established mid-wall LGE paradigm, whereas analysis from Jia et al., obtained by a cohort of patients of the same ethnicity from the Fuwai Hospital, introduces a less conventional subendocardial pattern that, if confirmed in multicentre cohorts, may capture a biologically distinct and particularly malignant phenotype. Collectively, these data reinforce a unifying message: arrhythmic risk in non-ischaemic cardiomyopathy is critically determined by scar architecture and location rather than by global systolic impairment alone, and future ICD selection strategies will likely need to integrate both risk scores and refined phenotypic characterization to achieve truly personalized prevention of SCD. Within this evolving concept of differential prognostic weight of LGE distribution, the proposed three-tier risk model represents a logical and clinically appealing step forward. By categorizing patients according to the presence of neither, one, or both high-risk LGE phenotypes, the authors provide a framework that is intuitive, reproducible, and readily applicable in routine practice. From a clinical perspective, these findings have several important implications. First, they reinforce the role of CMR as more than a diagnostic tool, positioning it as a key component of personalized risk assessment in NICM. Second, they support a shift from binary decision-making based on LVEF thresholds towards a more nuanced, phenotype-guided approach to ICD selection. Finally, the reliance on visually assessable LGE features enhances feasibility of implementation across centres with varying levels of technical expertise. Nonetheless, several limitations merit consideration. As with most studies in this field, the analysis is observational from a single-centre cohort, and the number of hard SCD events remains relatively modest. External validation in independent cohorts, including populations with different ethnic, genetic, and aetiological backgrounds, will be essential to confirm generalizability of the proposed model. Moreover, the interaction between LGE phenotypes and other emerging markers of arrhythmic risk, such as parametric mapping, myocardial strain, and pathogenic genetic variants, remains unexplored. Perhaps the most important unanswered question is whether improved risk stratification will translate into improved outcomes. While a wealth of observational data supports the prognostic value of CMR-derived fibrosis, definitive evidence that phenotype-guided ICD implantation improves survival is still lacking. Randomized trials comparing CMR-informed strategies with standard-of-care decision-making represent the logical next step, although their design poses practical and ethical challenges. Until such data become available, CMR-based phenotyping should be viewed as a powerful adjunct to comprehensive clinical assessment rather than a standalone determinant of therapy. In conclusion, Jia et al. provide compelling evidence that refining scar phenotyping beyond the ‘mid-wall paradigm’ can meaningfully enhance arrhythmic risk stratification in NICM. By identifying lateral subendocardium-involved LGE as a clinically relevant and previously under-recognized substrate, and by demonstrating its additive value when combined with septal mid-wall fibrosis, this study reinforces the field towards a more mechanistically informed and personalized approach to SCD prevention. As CMR continues to evolve from a descriptive imaging modality to a decisional tool, phenotype-based models such as this may help bridge the long-standing gap between risk prediction and effective prevention of SCD in non-ischaemic cardiomyopathy. Andrea Baggiano (Conceptualization [lead]; Writing—original draft [lead]), and Gianluca Pontone (Writing—review & editing [supporting]) The data underlying this article will be shared on reasonable request to the corresponding author.