ICD Therapy: Can Cardiac Scars Predict If You Need One?
"New research uses cardiac MRI to analyze scar tissue, potentially improving how we determine who benefits most from ICDs."
Sudden cardiac death (SCD) due to ventricular arrhythmias is a major concern for individuals with ischemic cardiomyopathy (ICM) and nonischemic cardiomyopathy (NICM). An implantable cardioverter-defibrillator (ICD) is often prescribed as a preventative measure, delivering life-saving shocks to restore a normal heart rhythm.
However, ICDs aren't without risks and only a fraction of those who receive them ever actually need their shock therapy. This highlights the critical need for better risk stratification methods to accurately identify individuals who will truly benefit from an ICD, avoiding unnecessary procedures and potential complications for those who won't.
Enter cardiac magnetic resonance imaging (MRI) with late gadolinium enhancement (LGE). This advanced imaging technique allows doctors to visualize scar tissue within the heart muscle. The size, location, and characteristics of these scars can provide valuable insights into a person's risk of developing dangerous arrhythmias. This article explores recent research into how LGE border zones, the areas surrounding dense scar tissue, can be used to predict the success of ICD therapy, potentially refining patient selection and improving outcomes.
ICD Use and Risk Assessment
Implantable cardioverter-defibrillator (ICD) therapy is described as a cornerstone intervention for reducing mortality in people at risk of life-threatening ventricular arrhythmias, including ventricular tachycardia and ventricular fibrillation. Guidelines recommend ICDs for primary prevention of sudden cardiac death in patients with reduced left ventricular ejection fraction (LVEF), but relying on LVEF alone may not identify those at greatest risk. A 2026 multicenter substudy of the Dutch Outcome in ICD Therapy (DO-IT) registry included 252 patients with LVEF ≤35% who underwent cardiac MRI before primary-prevention ICD implantation; 52% had nonischemic dilated cardiomyopathy and 38% had cardiac resynchronization therapy. Median follow-up was 4.3 years.
From Nuclear Resonance to Cardiac MRI
The history of cardiovascular magnetic resonance (CMR) can be traced to C. J. Gorter's 1936 manuscript in *Physica*, which described an unsuccessful attempt to find resonance of lithium-7 nuclei in crystalline lithium fluoride. Later developments in nuclear magnetic resonance helped establish the foundations for CMR imaging. CMR has since become an important clinical and research tool, with a history shaped by advances from experimental physics into medical imaging.
Cardiac Scars: A New Way to Predict ICD Success?
Researchers have been investigating the potential of LGE border zones as independent predictors of ventricular arrhythmias. While the presence and extent of LGE have been linked to predicting arrhythmic events, there's no universal agreement on the best method to analyze these border zones. To address this, a recent study compared four different algorithms for assessing LGE border zones and their ability to predict appropriate ICD therapy in ICM and NICM patients.
- Expectation Maximization, Weighted Intensity, and A Priori Information (EWA)
- Weighted Border Zone algorithm (WBZ)
- Modified Full-Width Half-Maximum (mFWHM)
- A 2-3SD threshold-based algorithm (2-3SD)
Scar Characterization and Arrhythmic Risk
A review of cardiac MRI tissue characterization describes late gadolinium enhancement (LGE) as a way to examine arrhythmic substrate within the heart muscle. Methods for quantifying LGE vary; one commonly described approach defines scar core as normalized peak signal intensity above 60% and the border zone as 40–60%. However, a study in nonischemic dilated cardiomyopathy reported that separating LGE into core and border-zone regions had not been shown to predict ICD therapy in that population. The evidence therefore supports careful attention to how scar is measured and interpreted.
Imaging and ICD Approaches
Imaging methods may not be interchangeable when assessing eligibility for ICD therapy, particularly for primary prevention in patients with borderline LVEF values of 30–40%, where LVEF may be virtually the only decision parameter. Device approaches also differ: the subcutaneous ICD (S-ICD) and extravascular ICD (EV-ICD) both avoid intravascular hardware, but use different sensing and detection approaches. The EMBLEM MRI S-ICD provides therapy without wires implanted in the heart, leaving the heart and blood vessels untouched. In the CMR GUIDE trial, ICD implantation did not significantly reduce sudden cardiac death or hemodynamically significant ventricular arrhythmia compared with an implantable loop recorder among patients with LVEF 36–50% and myocardial scar.
What This Means for You
This research suggests that LGE border zone quantification can improve patient selection for primary prevention ICDs, especially in those with ICM. By accurately identifying individuals at low risk of arrhythmias, doctors can potentially avoid unnecessary ICD implantations and their associated complications.
AI and the Next Generation of Cardiac MRI
Artificial intelligence (AI) methods based on machine learning and deep learning are already used in cardiac MRI, and a 2024 review examines their current and future impact. Complicated image prescriptions and long acquisition protocols remain constraints on clinical cardiac MRI. Proposed directions include AI-enhanced MRI, faster acquisition sequences, and applications in earlier disease detection and treatment monitoring. A 2025 review also describes AI models intended to detect true arrhythmias while filtering false alerts, predict ventricular arrhythmias weeks before ICD shocks, and forecast which heart failure patients may respond to cardiac resynchronization therapy.
CMR for Risk Stratification in Dilated Cardiomyopathy
Cardiac magnetic resonance (CMR) can support risk stratification in patients with dilated cardiomyopathy (DCM) referred for ICD assessment by characterizing tissue. The reviewed techniques include late gadolinium enhancement (LGE), parametric mapping, and feature tracking. These methods offer information for evaluating risk in this population, while the source frames CMR as a risk-stratification tool rather than a standalone ICD decision.
While these findings are promising, it's important to note that the study also highlights the impact of the algorithm used for LGE border zone analysis. The predictive value of LGE border zone differed depending on the algorithm, emphasizing the need for standardized methods and further research to optimize LGE assessment.
If you're at risk for sudden cardiac death or have been diagnosed with ICM or NICM, discuss these findings with your cardiologist. Cardiac MRI with LGE could be a valuable tool in determining whether an ICD is the right choice for you. This information, combined with other clinical factors, can help you and your doctor make informed decisions about your heart health and treatment options.