ASCO 2017: Predictors of urinary diversion choice in patients with bladder cancer in integrated care settings

Chicago, IL ( Radical cystectomy (RC) is an integral component of management for patients with muscle-invasive bladder cancer (MIBC), and to date, remains the standard of care for patients that can undergo it. However, it is a major operation with significant morbidity, in part due to the urinary diversion component. Urine diversion is required following RC, but the type of urinary diversion is often determined by expected QOL post-RC, anatomic considerations, renal function, oncologic considerations (urethral disease), prior radiation treatments, and surgeon and patient preference. While ileal conduit urinary diversion (incontinent urinary stoma) [IC] remains the most common, an orthotopic neobladder (NB) or continent catheterizable urinary diversion (continent pouch [CP]) are acceptable alternatives.

In this study, the authors utilize their access to an integrated community health care system (Kaiser Permanente), where financial incentive is mitigated, to explore factors regarding the decision on urinary diversion type outside of an academic setting.

They identified 992 patients older than 21 who underwent RC and urinary diversion in a 5-year period (~200 per year). Of these, consistent with many other studies, 80% had IC diversion. In general, older patients, female patients, higher stage disease patients, and patients with more comorbidities underwent IC diversion rather than continent diversion (NB/CP).

Independent of these factors, the surgeon also accounted for some of the variability – as each surgeon had different training and exposure, their comfort level with continent diversion likely varied. Indeed, of the 169 surgeons, 99 performed only IC, while 70 performed all types of UD (and generally performed more per year).

A predictive model with patient factors alone showed good fit (AUC = .93, Hosmer-Lemeshow test p = .22), but including specific surgeon factors (annual cystectomy volume, specialty training, clinical tenure) did not improve the model fit.

While none of the results seen are surprising, they seem to correlate pretty well with academic practices. Importantly, patient factors drive most of the decision regarding diversion type.

Presented By: Marilyn L. Kwan

Co-Authors: Michael C Leo, Valerie S. Lee, Kim N. Danforth, Sheila Weinmann, Maureen Cecelia O'Keeffe-Rosetti, David Yi, Christopher S. Wendel, Joanna E. Bulkley, Mark C. Hornbrook, Robert S. Krouse, Scott Michael Gilbert, Carmit K. McMullen

Institution(s): Division of Research, Kaiser Permanente Northern California, Oakland, CA; Center for Health Research, Kaiser Permanente Northwest, Portland, OR; Kaiser Permanente, Pasadena, CA; Kaiser Permanente Southern California Department of Research and Evaluation, Pasadena, CA; Southern Arizona Veterans Affairs Health Care System, Tucson, AZ; The Center for Health Research, Kaiser Permanente Northwest, Portland, OR; University of Pennsylvania, Philadelphia, PA; H. Lee Moffitt Cancer Canter and Research Institute, Tampa, FL

Written By: Thenappan Chandrasekar, MD, Clinical Fellow, University of Toronto, Princess Margaret Cancer Centre
Twitter: @tchandra_uromd

at the 2017 ASCO Annual Meeting - June 2 - 6, 2017 – Chicago, Illinois, USA