The association of an implantable cardioverter defibrillator (ICD) with survival in patients with left ventricular assist devices (LVADs) is not well understood. We evaluated all-cause mortality by the presence of an ICD at the time of LVAD implantation, or by ICD implantation after LVAD placement in 191 patients, using Kaplan–Meier survival analyses and Cox models with multivariate adjustment. During the median follow-up of 23 months, 33 of 129 patients (26%) with an ICD and 17 of 62 patients (27%) without an ICD died. Patients had similar all-cause mortality with or without an ICD before LVAD, after censoring for post-LVAD ICD implantation (log-rank p = 0.889). Multivariate models after adjustments revealed no statistically significant survival benefit from an ICD before LVAD (hazard ratio [HR]: 0.65, 95% CI: 0.27–1.57, p = 0.340). Thirty-one of 62 (50%) patients without an ICD before LVAD implantation subsequently received an ICD after LVAD, although these patients did not have significantly better survival when compared with those with no ICD in a time-dependent analysis (HR: 0.70, 95% CI: 0.25–1.95, p = 0.497). Among LVAD patients, neither a previously implanted ICD nor a new ICD implantation after LVAD yielded statistically significant survival benefit. Further studies are warranted to investigate the role of ICD implantation in LVAD patients.