Utilization and Perioperative Outcomes of Partial Cystectomy for Urothelial Carcinoma of the Bladder: Lessons from the ACS NSQIP Database

Abstract

IntroductionWe identified predictors of partial cystectomy in the ACS NSQIP® database of more than 400 hospitals across North America. We also reviewed perioperative outcomes. MethodsWe reviewed the records of patients with an ICD-9 diagnosis of urothelial carcinoma of the bladder who were treated with partial or radical cystectomy from 2007 to 2012. The proportion of patients who underwent partial vs radical cystectomy and the proportion who received neoadjuvant chemotherapy were compared across time. We reviewed 30-day morbidity and mortality, and determined risk factors. Logistic regression was used to identify factors predictive of undergoing partial vs radical cystectomy. ResultsA total of 2,393 patients met study inclusion criteria. The ratio of partial to radical cystectomy was low and stable at 0% to 7% (p = 0.36). While patients undergoing radical cystectomy were more likely to receive neoadjuvant chemotherapy in later years (p <0.001), neoadjuvant chemotherapy use before partial cystectomy was consistently low with time (p = 0.68). The 30-day morbidity rate after partial and radical cystectomy was 23.3% and 58.1% (p = 0.001), and the 30-day mortality rate was 1.6% and 2.1%, respectively (p = 0.66). On multivariate regression factors independently associated with partial vs radical cystectomy were cerebrovascular accident history (OR 4.4, p = 0.005), current nonsmoking (OR 0.43, p = 0.032) and lack of trainee participation in the operation (OR 0.28, p <0.001). ConclusionsThe ratio of the number of partial to radical cystectomies performed was stable. Cerebrovascular accident history, nonsmoker status and lack of trainee participation were associated with partial cystectomy. Patients treated with radical cystectomy but not those who underwent partial cystectomy were more likely to receive neoadjuvant chemotherapy in later years. Large detailed registries such as ACS NSQIP have important potential use for evaluating trends in urological practice.