Introduction: Although perioperative mortality after radical prostatectomy for prostate cancer is rare, less is known about the extent to which risks posed by patient factors can be modified by care at higher-volume centers. We aimed to assess the incidence and risk factors of mortality within 90 days of radical prostatectomy. Methods: We conducted a retrospective analysis of The National Cancer Database, to identify patients undergoing radical prostatectomy from 2004-2017. We characterized the incidence of death within 90 days of surgery and described clinical, demographic and facility-level factors associated with the outcome, including surgical volume. Mixed effect logistic regression models incorporating a facility random intercept were constructed to examine factors associated with 90-day mortality. Results: We identified 313,071 patients who underwent prostatectomy in the study period. The incidence of 90-day mortality was 0.22%. Mixed effect logistic regression revealed that 90-day mortality was associated with older age (1.47 OR per 5 years, 95% CI 1.39-1.56), higher PSA (>20 mg/mL, OR 2.25, 95% CI 1.88-2.68), receipt of lymph node dissection (OR 1.26, 95% CI 1.05-1.51), high-school degree attainment in patient’s residential zip code by quartile (>93.7% of zip code with degrees, OR 0.70, 95% CI 0.55-0.90, Black race (OR 1.91, 95 %CI 1.57-2.32), higher Charlson-Deyo Morbidity Index (OR 1.50 95% CI 1.25-1.81 for score of 1; OR 2.72 95% CI 2.00-3.72 score of 2; OR 3.72 95% CI 2.36-5.88 for score of 3+) and surgical volume above median (OR 0.80, 95% CI 0.66-0.97). Conclusions: 90-day mortality after radical prostatectomy is rare, but associated with hospital surgical volume and patient- level factors including higher risk disease, higher comorbidity and race. Findings of racial disparity in perioperative mortality suggest a timely need to explore the contributions of patient selection and quality of care. SOURCE OF Funding: This publication was made possible by the Yale School of Medicine Fellowship for Medical Student Research.