The extracellular volume fraction (ECV) by T1 mapping can quantify diffuse myocardial fibrosis, and useful as a non-invasive marker for risk stratification for patients with non-ischemic dilated cardiomyopathy (NIDCM). Prolonged QRS interval on electrocardiogram is related to worse clinical outcome for heart failure patients. The purpose of this study was to evaluate the prognostic value of the combination of ECV and QRS duration for NIDCM patients. A total of 60 NIDCM patients (mean age 61 ± 12 years, mean left ventricular ejection fraction 37 ± 10%, mean QRS duration 110 ± 19 ms) were enrolled. Using a 1.5-T MR scanner and 32-channel cardiac coils, the mean ECV value of six myocardial segments at the mid-ventricular level was measured by the modified look-locker inversion recovery method. Adverse events were defined as follows: cardiac death; recurrent hospitalization due to heart failure. Patients were allocated into three groups based on ECV value and QRS duration (group 1: ECV ≦ 0.30 and QRS ≦ 120 ms; group 2: ECV > 0.30 or QRS > 120 ms; group 3: ECV > 0.30 and QRS > 120 ms). During a median follow-up duration of 370 days, 7 of 60 (12%) NIDCM patients experienced adverse events. NIDCM patients with events had longer QRS duration (134 ± 31 ms vs. 106 ± 14 ms, p = 0.01) and higher ECV (0.34 ± 0.07 vs 0.29 ± 0.05, p = 0.026) compared with those without events. On Kaplan–Meier curve analysis, significant difference was found between group 1 and group 3 (p < 0.001, log-rank test). No significant difference was found between group 1 and group 2 (p = 0.053), group 2 and group 3 (p = 0.115). The area under the receiver operating characteristic curve (AUC) for predicting adverse events was 0.778 (95% confidence interval CI 0.612–0.939) for ECV, 0.792 (95% CI 0.539–0.924) for QRS duration, 0.822 (95% CI 0.688–0.966) for combination of ECV and QRS duration. NIDCM patients with high ECV and prolonged QRS duration had significantly worse prognosis compared to those with normal ECV and normal QRS duration. The combination of ECV and QRS duration could be useful as a non-invasive method for better risk stratification for patients with NIDCM.