|Endpoint||Year||n||Methods||Events (n)||Results||Ref. #|
|Inducible VT on EPS||2005||48||LGE (>2 SD)||18||Unadjusted logistic regression (model with infarct performed better than EF); More VT with large infarct mass, p < 0.0005||(18)|
|2007||47||LGE: FWHM peri-infarct||20||Adjusted regression model for infarct, EF, and EDV; Higher peri-infarct size associated with inducible VT, p = 0.015||(19)|
|Ventricular arrhythmia or ICD therapy||2011||55||LGE: SD&FWHM Total scar & peri-infarct||14||Receiver-operating curve; Total scar: AUC 0.75, p < 0.01 Peri-infarct: p = NS||(20)|
|2009||91||LGE: FWHM Total scar & peri-infarct||18†||CPH model, adjusted for EF and infarct size; Peri-infarct: HR: 1.49/10 g [1.01–2.20], p = 0.04; Infarct core: p = NS||(21)|
|Composite endpoint: ICD therapy or SCD||2012||59||LGE: SD&FWHM Total scar & peri-infarct||10||CPH model, adjusted for indication of ICD insertion; Total scar: HR: 3.7 [1.1–12.1], p = 0.03, NPV 86% Peri-infarct: p = NS||(16)|
|⁎2012||137||LGE (% LV mass)||30||CPH model adjusted for EF, scar size and NYHA; Scar size (>5% LV mass): HR: 4.8 [1.6–13.7], p = 0.004||(22)|
|2012||137||LGE: FWHM peri-infarct||27||CPH model adjusted for diuretic use, infarct, and hs-CRP; Peri-infarct: HR: 4.6 [1.4–15] (3rd tertile vs.1st), p = 0.01||(23)|
AUC = area under the curve; CMR = cardiac magnetic resonance; EDV = end-diastolic volume; EPS = electrophysiology study; FWHM = full width half maximum method; HR = hazard ratio; hs-CRP = high-sensitivity C-reactive protein; LGE = late gadolinium enhancement; LV = left ventricle; NYHA = New York Heart Association functional class; Ref = reference; VT = ventricular tachycardia.
↵⁎ Study included patients with ischemic cardiomyopathy (ICM) and dilated cardiomyopathy (DCM) (no mention what percentage was each category) and variable range of ejection fraction (EF). Implantable cardiac defibrillator (ICD) therapy or sudden cardiac death (SCD) was the secondary endpoint; all-cause death or ICD therapy was the primary endpoint (39 events) with similar results on the Cox proportional hazard (CPH) model.
↵† Of the 18 events, 12 were for adenosine triphosphate therapy, whereas the rest were for ICD shock.