Late Gadolinium Enhancement Predicts SCD Risk in HCM

Quick Takes

  • The 2022 ESC hypertrophic cardiomyopathy (HCM) sudden cardiac death (SCD) risk model had better model discrimination compared to 2011 ACC/AHA and 2014 ESC SCD risk scores.
  • While model discrimination of the 2022 ESC HCM risk score was similar to the 2020 ACC/AHA model, it had a higher specificity.
  • A threshold of late gadolinium enhancement extent ≥5% on cardiac MRI provided further risk stratification in addition to model predicted risk.

Study Questions:

How does the 2022 European Society of Cardiology (ESC) risk model for sudden cardiac death (SCD) in hypertrophic cardiomyopathy (HCM) perform compared with risk models from previous guidelines? And does late gadolinium enhancement (LGE) further improve SCD risk stratification?


This was a retrospective study of adults with HCM who underwent cardiac magnetic resonance imaging (cMRI) at a single center in China from 2012–2013. Extent of LGE was defined as percent of enhanced myocardium to left ventricular (LV) mass. SCD risk as assessed by the 2022 ESC model was compared to 2014 ESC, 2011 American College of Cardiology/American Heart Association (ACC/AHA), and 2020 ACC/AHA guidelines. The primary outcome of interest was aborted SCD or SCD.


A total of 828 patients were studied, including 69% males with a mean age of 49 years. Over a mean follow-up of 7.4 years, 6% of patients experienced SCD/aborted SCD. Patients with SCD had a larger left atrial size, LV mass index, and LVEF with a higher prevalence of LGE (93% vs. 59%). Compared with the 2014 ESC model, the 2022 ESC model showed increased area under curve (AUC; 0.76 vs. 0.63) with a better sensitivity (76% vs. 43%). The AUC was similar for 2022 ESC and 2020 ACC/AHA models. However, the 2022 ESC model showed better specificity than the 2020 ACC/AHA models. Extent of LGE for SCD prediction showed the highest AUC and c-statistic at a threshold of 5% compared with other cut-offs after multivariable adjustment. In patients with Class of Recommendation II or III for an implantable cardioverter-defibrillator (ICD) according to the ESC 2022 model, LGE ≥5% and <15% showed worse prognosis than patients with LGE <5%.


In a cohort of 778 Chinese patients with HCM, the 2022 ESC-SCD risk model performed better than the 2014 ESC risk model. Presence of LGE on cMRI further enabled risk stratification based on the 2022 ESC risk model.


Risk prediction for SCD in HCM patients remains a clinical challenge. Various models have been introduced since 2011, with recent updates, further complicating the decision-making process for clinicians on which model to employ. In this study, the 2022 ESC model outperformed other existing models. It had much better discriminative statistics than the 2011 ACC/AHA and 2014 ESC models. It also had a higher specificity than the 2020 ACC/AHA model with otherwise similar model discrimination.

While LGE on cMRI has been identified as a significant risk factor for SCD in HCM, how this interacts with the risk prediction model remains unclear. Furthermore, to date, the threshold for significant LGE on cMRI in HCM has been placed at >15% of the LV mass. However, in this study, a threshold of ≥5% LGE had the best AUC and sensitivity. Therefore, in patients stratified as Class of Recommendation II/III for an ICD per the ESC 2022 model, LGE on cMRI at a threshold of 5% further enabled additional risk stratification. Important limitations of this study include lack of generalizability, as it was performed in a Chinese cohort.

Clinical Topics: Heart Failure and Cardiomyopathies, Noninvasive Imaging, Magnetic Resonance Imaging

Keywords: Cardiomyopathy, Hypertrophic, Magnetic Resonance Imaging

< Back to Listings