Implantable cardioverter-defibrillator (ICD) shock is associated with mortality. There are, however, conflicting data whether ventricular arrhythmia (VA) leading to ICD shock or shock itself increases mortality. In addition, paced electrical delay is related to both VA substrate and cardiac resynchronization therapy (CRT) response. Little is known about the significance of VA substrate as a prognostic marker of CRT. The aim of this study is to assess the relationship between VA substrate and prognosis such as cardiac death and/or heart failure (HF) hospitalization. We investigated consecutive 330 CRT patients with the follow-up data one year after CRT implantation. VA was defined as lasted ≥30s or treated by anti-tachycardia pacing (ATP) or shock. CRT responder was defined as more than 15% reduction of left ventricular end-systolic volume. Primary endpoint was composite endpoint of cardiac death and/or HF hospitalization. Forty-three patients had VA within one year after CRT implantation. During median follow-up of 2.8 years (interquartile range 1.5-5.0 years), the patients with VA within one year had higher risk of primary endpoint (p<0.01). VA >200bpm was associated with increased risk of primary endpoint (VA >200bpm vs VA ≤200bpm, hazard ratio [HR] 2.34; 95% confidence interval [CI] 1.08-5.08, p=0.03), though shock was not associated with primary endpoint (shock vs anti-tachycardia pacing, HR 1.66; 95%CI 0.76-3.61, p=0.20). The patients with VA within one year had lower prevalence of CRT responder (19 [44%] vs 184 [64%], p=0.01) and longer time of left-ventricular pacing to right ventricular sensing (174±23ms vs 143±36ms, p<0.01) than the patients without. VA occurrence within one year was related to paced electrical delay and poor response to CRT. VA substrate could be associated with poor prognosis among the CRT patients.