Association of procedure volume with radical cystectomy outcomes in a nationwide database

Purpose: Studies of national databases Yield important information about expected outcomes after radical cystectomy and factors that influence patient morbidity and mortality. We examined the hospital characteristics associated with outcomes after radical cystectomy in a cohort study using results from a single, high volume academic institution as well as a nationwide data set of academic institutions. Materials and Methods: We obtained data from the University HealthSystem Consortium Clinical Database on 6,728 patients nationwide who underwent radical cystectomy for bladder cancer between 2002 and 2005 as well as on 421 who underwent cystectomy at our institution during this period. Outcomes were compared by hospital characteristics (geographic location, total hospital discharges and procedure volume). The outcome measures analyzed were length of hospital stay, the complication rate and in hospital mortality. Results: The overall complication rate at our institution was 32.07% with an in hospital mortality rate of 0.95% and an average length of stay of 7.05 days. The overall complication rate in the University HealthSystem Consortium data set was 37.16% with an in hospital mortality rate of 1.47% and an average length of stay of 10.98 days. Institutions with higher cystectomy volumes had significantly better outcomes than institutions with lower procedure volumes. The mortality rate at institutions with greater than 50 cystectomies per year was 0.54% compared to 2.70% at institutions with 10 or fewer per year (p <0.0005). Outcomes varied only minimally with total hospital discharges or geographic region. Conclusions: Even among academic medical centers hospitals with a higher volume of cystectomies in 2002 to 2005 were associated with improved outcomes, including decreased mortality, shorter length of stay and lower rehospitalization rates. These data may provide a framework for self-assessment and help establish criteria for performance evaluation.