Log in to MyACC
CMR for Arrhythmic Risk Stratification in Nonischemic Cardiomyopathy
Quick Takes
- In this observational study involving patients with nonischemic cardiomyopathy, myocardial fibrosis as demonstrated on CMR (late gadolinium enhancement imaging) was more strongly associated with arrhythmic events than LVEF <35%.
Study Questions:
Does the presence of myocardial fibrosis (MF) on cardiovascular magnetic resonance (CMR) predict ventricular arrhythmias better than left ventricular ejection fraction (LVEF)?
Methods:
This observational study included patients with nonischemic cardiomyopathy (NICM) from two British tertiary referral hospitals. Subjects underwent contrast-enhanced CMR between 2009 and 2017. Aside from coronary artery disease, notable exclusion criteria were uncontrolled hypertension, primary valve disease, infiltrative cardiomyopathy, cardiac sarcoidosis, inadequate CMR image quality, and incomplete clinical follow-up. CMR images were assessed visually for MF on late gadolinium enhancement (LGE) imaging, and if it was present, total fibrosis (TF) and gray zone fibrosis (GZF) were quantified using signal thresholding methods. The primary composite arrhythmic endpoint comprised sudden cardiac death, ventricular fibrillation, and sustained ventricular tachycardia.
Results:
The derivation cohort included 866 patients (64.8% male), followed for a median of 7.6 years. The validation cohort included 848 patients (63.7% male), followed for a median of 6.8 years. As compared with the validation cohort, the derivation cohort had more patients with hypertension and diabetes, and a lower proportion of patients with MF on visual assessment (33.7% vs. 56.6%; p < 0.001). In the derivation cohort, 52 patients (6.0%) met the primary endpoint.
Based on univariate analyses in the derivation cohort, MF by visual assessment was associated with the primary endpoint (hazard ratio [HR], 5.83; 95% confidence interval [CI], 3.15-10.8). The association between LVEF <35% and the primary endpoint was less robust (HR, 1.91; 95% CI, 1.11-3.29). Quantitative assessments of TF and GZF were the most consistent predictors of the primary endpoint (C-statistic 0.73 for both). Quantification of total fibrosis allowed for creation of three risk categories, with the following annual event rates: no MF, 0.34%; low TF mass, 1.38% (HR, 4.03); and high TF mass, 3.15% (HR, 9.17). In decision curve analyses, MF by visual assessment and quantitative measures was superior to LVEF in predicting the primary endpoint.
Conclusions:
In the setting of NICM, MF on CMR is associated with arrhythmic events, and larger fibrosis burden correlates with increased risk.
Perspective:
This study adds to the extensive body of literature illustrating the role of LGE in event prognostication in various cardiomyopathies. A notable limitation is that genetic testing was not consistently performed in this cohort, so the proportion of patients with heritable cardiomyopathies is unclear. While the authors argue that MF predicts ventricular arrhythmia risk better than LVEF <35%, they did not specifically test other LVEF cutoffs. Advanced tissue characterization techniques including T1 mapping were not used, and further research will be needed to determine how they should be incorporated into arrhythmic risk stratification in clinical practice.
Clinical Topics: Arrhythmias and Clinical EP, Heart Failure and Cardiomyopathies, Noninvasive Imaging, Implantable Devices, SCD/Ventricular Arrhythmias, Atrial Fibrillation/Supraventricular Arrhythmias, Magnetic Resonance Imaging
Keywords: Arrhythmias, Cardiac, Cardiomyopathies, Fibrosis, Magnetic Resonance Imaging
< Back to Listings
