The peri-infarct gray zone of myocardial fibrosis is a better predictor of ventricular arrhythmias than dense core fibrosis in patients with previous myocardial infarction

Abstract

Abstract Funding Acknowledgements Type of funding sources: None. Background Current sudden cardiac death (SCD) risk stratification relies heavily on left ventricular ejection fraction (LVEF), but markers to refine risk assessment are needed. Dense core fibrosis (DCF) and peri-infarct "gray zone" of myocardial fibrosis (GZF) on late gadolinium enhancement (LGE) cardiac magnetic resonance (CMR) have been proposed as potential arrhythmogenic substrates. The aim of our study was to determine whether DCF and GZF could predict the occurrence of ventricular arrhythmias in patients with previous myocardial infarction. Methods We performed a single centre retrospective study enrolling consecutive patients with previous myocardial infarction undergoing CMR before implantable cardioverter-defibrillator (ICD) implantation. Areas of LGE were subdivided into "core" DCF and "peri-infarct" GZF zones based on signal intensity (>5 SD, and 2-5 SD above the mean of reference myocardium, respectively). The primary endpoint was a composite of sudden arrhythmic death, appropriate ICD shock, ventricular fibrillation (VF), or sustained ventricular tachycardia (VT) as detected by the device. Results A total of 88 patients (median age 61 years [IQR 54-73], 84% male, median LVEF 30% [IQR 23-36%], 14% secondary prevention) were included. During a median follow-up of 23 months [IQR 9-38], 13 patients reached the primary endpoint (10 appropriate ICD shock, 2 sustained VT or VF, and 1 sudden arrhythmic death). Patients who attained the primary endpoint had similar DCF (30.4g ± 14.7 vs. 28.0g ± 15.3; P = 0.601) but a greater amount of GZF (18.1g ± 9.6 vs. 11.9g ± 6.7; P = 0.005). On univariate analysis, GZF was associated with the composite endpoint (HR: 1.09 per gram; 95%CI: 1.02-1.15; P = 0.006), whereas DCF was not (HR: 1.01 per gram; 95%CI: 0.98-1.05; P = 0.571). After adjustment for LVEF, GZF remained independently associated with the primary endpoint (adjusted HR: 1.06 per gram; 95% CI: 1.01-1.12; P = 0.035). Decision tree analysis identified 11.9g of GZF as the best cut-off to predict life-threatening arrhythmic events. The primary endpoint occurred in 11 out of the 35 patients (31.4%) with GZF ≥11.9g, but in only 2 of the 53 patients (3.8%) with GZF <11.9g – Figure. Conclusions The extent of peri-infarct GZF seems to be a better predictor of ventricular arrhythmias than DCF. This parameter may be useful to identify a subgroup of patients with previous myocardial infarction at increased risk of life-threatening arrhythmic events.