We identify clinicopathological variables predicting overall survival in patients with recurrent bladder urothelial carcinoma after radical cystectomy. We retrospectively collected data on 114 patients treated with radical cystectomy for bladder urothelial carcinoma who subsequently had remote metastasis and/or local recurrence. The Kaplan-Meier method with the log rank test and multivariate Cox regression models were used to address overall survival after recurrence. During followup 99 of the 114 patients died. Median survival in the 114 patients was 11.2 months. One and 3-year overall survival rates were 48.0% and 12.1%, respectively. On multivariate analysis independent predictors of poorer overall survival included less than 1 year to recurrence, symptoms at recurrence, 2 or more metastatic organs at recurrence, high serum C-reactive protein, high lactate dehydrogenase, no post-recurrence platinum based chemotherapy and no metastasectomy. Based on the 4 variables (time to recurrence, symptoms, number of metastatic organs and C-reactive protein), we constructed a risk model predicting post-recurrence overall survival that classified patients into 3 groups with significantly different overall survival (p <0.0001). Our data confirm that recurrent urothelial carcinoma after radical cystectomy is a highly aggressive, lethal disease. Seven clinicopathological factors were identified that predicted post-recurrence overall survival. Our risk model based on the 4 variables could be useful to provide relevant prognostic information to patients and physicians, and better stratify patients in clinical trials.