PURPOSE: We evaluated the clinical outcomes and risk factors for progression in a large cohort of patients with lymph node metastases following en bloc radical cystectomy and bilateral pelvic lymphadenectomy. MATERIALS AND METHODS: From July 1971 through December 1997, 1,054 patients underwent radical cystectomy and bilateral pelvic-iliac lymphadenectomy for high grade, invasive transitional cell carcinoma of the bladder. Of these patients 244 (23%) with a median age of 66 years (range 36 to 90) had pathological lymph node metastases. Overall 139 of the 244 patients (57%) received some form of chemotherapy. At a median followup of greater than 10 years (range 0 to 28) outcomes data were analyzed in univariate analysis according to tumor grade, carcinoma in situ, primary bladder tumor stage, pathological subgroups, total number of lymph nodes removed and involved with tumor, and lymph node density (total number of positive lymph nodes/total number removed). In addition, the form of urinary diversion and the administration of chemotherapy were also evaluated. Multivariate analysis was then performed to analyze these variables independently. RESULTS: The incidence of positive lymph nodes increased with higher p stage and pathological subgroups. Of 669 patients 75 (11%) with organ confined primary tumors and 169 of 385 (44%) with extravesical tumor extension had involved lymph nodes. The median number of lymph nodes removed in the 244 lymph node positive cases was 30 (range 1 to 96), while the median number of positive lymph nodes was 2 (range 1 to 63). Overall recurrence-free survival at 5 and 10 years for the 244 patients with lymph node positive disease was 35% and 34%, respectively. Patients with lymph node positive disease and an organ confined primary bladder tumor had significantly improved 10-year recurrence-free survival compared with those with extravesical tumor extension (44% vs 30%, p = 0.003). The total number of lymph nodes removed at surgery was also prognostic. Patients with 15 or less lymph nodes removed had 25% 10-year recurrence-free survival compared with 36% when greater than 15 lymph nodes were removed. Recurrence-free survival at 10 years for patients with 8 or less positive lymph nodes was significantly higher than in those with greater than 8 positive lymph nodes (40% vs 10%, p <0.001). The novel concept of lymph node density was also a significant prognostic factor. Patients with a lymph node density of 20% or less had 43% 10-year recurrence-free survival compared with only 17% survival at 10 years when lymph node density was greater than 20% (p <0.001). On multivariate analysis the total number of lymph nodes involved, pathological subgroups of the primary bladder tumor, lymph node density and adjuvant chemotherapy remained significant and independent risk factors for recurrence-free and overall survival. CONCLUSIONS: Patients with lymph node tumor involvement following radical cystectomy may be stratified into high risk groups based on the primary bladder tumor, pathological subgroup, number of lymph nodes removed and total number of lymph nodes involved. Lymph node density, which is a novel prognostic indicator, may better stratify lymph node positive cases because this concept collectively accounts for the total number of positive lymph nodes (tumor burden) and the total number of lymph nodes removed (extent of lymphadenectomy). Future staging systems and the application of adjuvant therapies in clinical trials should consider applying lymph node density to help standardize this high risk group of patients following radical cystectomy.