ABSTRACT: BACKGROUND:Cardiac magnetic resonance techniques permit quantification of the myocardial extracellular volume fraction (ECV), representing a surrogate marker of reactive interstitial fibrosis, and late gadolinium enhancement (LGE), representing replacement fibrosis or scar. ECV and LGE have been independently linked with heart failure (HF) events. In deriving ECV, coronary artery disease type LGE, but not non-coronary artery disease type LGE, has been consistently excluded. We examined the associations between LGE, global ECV derived from myocardial tissue segments free of any detectable scar, and subsequent HF events. METHODS:Mid short-axis T1 maps were divided into 6 cardiac segments, each classified as LGE absent or present. Global ECV was derived from only segments without LGE. ECV was considered elevated if >30%, the upper 95% bounds of a reference group without known cardiac disease (n=28). Patients were divided into 4 groups by presence of elevated ECV and of any LGE. Subsequent HF hospitalization and any death were ascertained. Their relationship with ECV was examined separately and as a composite with Cox proportional hazard models. RESULTS:Of 1604 serial patients with T1 maps, 1255 were eligible after exclusions and followed over a median 26.3 (interquartile range, 15.9-37.5) months. Patients with elevated ECV had increased risk for death (hazard ratio [HR] 2.45 [95% CI, 1.76-3.41]), HF hospitalization (HR, 2.45 [95% CI, 1.77-3.40]), and a combined end point of both outcomes (HR, 2.46 [95% CI, 1.94-3.14]). After adjustments for covariates including LGE, the relationship persisted for death (HR, 1.82 [95% CI, 1.28-2.59]), hospitalization (HR, 1.60 [95% CI, 1.12-2.27]), and combined end points (HR, 1.73 [95% CI, 1.34-2.24]). CONCLUSIONS:ECV measures of diffuse myocardial fibrosis were associated with HF outcomes, despite exclusion of replacement fibrosis segments from their derivation and even among patients without any scar. ECV may have a synergistic role with LGE in HF risk assessment.