The Extent of LGE-Defined Fibrosis Predicts Ventricular Arrhythmia Severity: Insights from a Preclinical Model of Chronic Infarction

Abstract

Abnormal propagation of cardiac electrical impulses in hearts with fibrosis developed post-infarction often lead to rapid ventricular tachycardia (VT) and sudden cardiac death, a major cause of mortality. Certain values of ejection fraction (e.g. EF < 35%) are used clinically to refer scar-related VT patients for ICD implantation; however, these values are not an indication of VT severity (i.e., how fast is the heart rate in VT). Our aim here is to use a preclinical model of chronic fibrosis to determine whether the extent of fibrosis defined by MRI correlates better than EF with the heart rate during VT (a measure known as VT cycle length). Specifically, n = 10 pigs with prior infarct underwent MR imaging (i.e., cine scans to calculate EF, and high resolution late gadolinium enhancement (LGE) to measure the extent of fibrosis, followed by an X-ray guided VT inducibility study to determine VT cycle length. The total infarct size in LGE images was given by the extent of dense scar plus that of gray zone, GZ (a mixture of viable muscle and collagen fibers, located at the scar periphery), as defined by two clinically accepted segmentation thresholding methods: 5SD (standard deviation) and FWHM (full-width at half maximum), respectively. Overall, LGE-defined scar/GZ corresponded well to infarct heterogeneities observed in collagen-sensitive histological stains. Our quantitative results showed that the amount of LGE-defined fibrosis (relative to the left ventricular volume) correlated well with VTCL (R ~ 0.78), suggesting that it could be a potential clinical predictor of dangerous VT.