Muscle Invasive Bladder Cancer: Who Should Not Undergo Radical Cystectomy? A Risk Predictive Model

Abstract

Background: The gold standard treatment of muscle invasive bladder cancer (MIBC) is radical cystectomy (RC) for eligible patients. Recently, trimodal therapy has been recommended as an alternative bladder preserving approach. We aim to identify patients to whom trimodal therapy is the optimal approach by constructing risk calculators of morbidity and mortality.

Methods: Using ACS-NSQIP database, we selected the patients having MIBC undergoing RC, making 10642 patients. Our primary outcome was mortality and secondary outcome was morbidity within 30 days of the procedure. Morbidity was assessed using prolonged length of stay (>10 days). We underwent multivariate logistic regression to obtain the best fit model for each outcome on 60% of the sample. Validation of the models was done on 40% of the sample. Model performance was assessed using discrimination and calibration abilities. Risk calculator was constructed using Excel.

Results: Of the full cohort, 199 patients (1.9%) experienced death and 2328 patients (21.9%) experienced morbidity. For the mortality model, the area under the curve was 70% with Hosmer-Lemeshow statistic of 0.64. Variables significant in the model included age, frailty, ASA status and preoperative creatinine. For the morbidity model, the area under the curve was 65% with a Hosmer-Lemeshow statistic of 0.8. Variables significant in the model included age, frailty, albumin, preoperative creatinine, robotic surgery and continent diversion.

Conclusion: We provide statistically significant and clinically relevant models to be used in the clinical setting to provide patients with individualized risks of morbidity and mortality from RC.