Objectives: We performed this study to elucidate whether the heterogeneity of radioactivity uptake in 123I-metaiodobenzylguanidine (MIBG) scanning, the indicator of cardiac sympathetic denervation, could be associated with T-wave alternans (TWA) on Holter ECG or not. Furthermore, we evaluated the prognostic impact of two tests on mortality in patients with non-ischemic dilated cardiomyopathy (DCMP). Methods: We performed 24-hour Holter monitoring and cardiac MIBG imaging in 49 consecutive non-ischemic DCMP patients with lower left ventricular ejection fraction (LVEF) (mean EF = 28짹12%). With baseline characteristics, we investigated MIBG washout (WO) rate, heart-to-mediastinum activity rate (HM), coefficient of variance (CV) of HM activity, and HM dispersion in the early and late image from MIBG scan. Also, maximal TWA using modified moving average analysis on Holter was evaluated. Each parameter from MIBG was analyzed for correlation with maximal TWA. The prognostic impact of all parameters on occurrence of death from ventricular tachyarrhythmia (VT) was tested. Results: With value of maximal TWA, the mean values of WO rate, HM rate, CV of HM activity and its dispersion in the early and late image were listed in table1. All parameters from MIBG were not correlated with maximal TWA after adjustment for age and LVEF. During follow up (mean duration 51.6 짹 15.4 months), death from VT was observed in 19 patients (38.7%) and it seemed to be influenced by HM rate in early and delay image, CV of HM activity in early image in unadjusted analyses. However, multivariate analysis revealed that CV of HM activity in early image was the only prognostic factor [Odd ratio = 14.3 (95% confidence interval 1.56 to 131.2), p=0.019]. Conclusion: This study demonstrated no correlation between TWA on Holter and the parameters from MIBG and MIBG was more potent prognostic tool in survival of patients with non-ischemic DCMP. This finding might be from the property of MIBG scan which is indicator of only sympathetic denervation.