LAUSR

OBJECTIVES To better understand the time-course in which major complications occur after radical cystectomy and to describe associations with complications at 30 and 90 days. METHODS A database of radical cystectomy cases was queried for pre-operative, perioperative, and post-operative data. Follow-up extended to 90 days post-surgery and included major complications (Clavien III-V). Early (30-day) and late (90-day) complication rates were compared via McNemar's test, and patient characteristics were compared across complication time groups by one-way ANOVA or Fisher's exact tests. Multinomial logistic regression was used to explore associations between patient characteristics and complication timing. RESULTS Of 969 patients undergoing radical cystectomy, 210/969 (21.7%) experienced a complication within 90 days. The rate of major complication significantly differed at 30 and 90 days (14.4% (CI: 12.4% - 16.9%) vs. 21.7% (CI: 19.2% - 24.4%) respectively, p=<0.0001). COPD (p=0.03), CCI (p=0.02), and Indiana pouch diversion (p=0.002) were significant predictors of early complication. Diabetes was the strongest predictor for late complication (OR: 2.42; p = 0.01). Diabetes was also a significant predictor for late genitourinary complications (OR 3.39; p=0.01), and smoking history was a significant predictor for late infectious complications (OR 3.61; p=0.01). CONCLUSIONS We identified a significant number of complications occurring after 30 days post-cystectomy, including the majority of deaths and genitourinary complications. These findings suggest that assessment of complications exclusively at 30 days would fail to capture a large proportion of major complications and deaths. Understanding the time-course of complications post-cystectomy will serve to better inform design of future outcome studies.