CUA 2018: Predicting Perioperative Complications in Patients Receiving Radical Cystectomy Using Preoperative Computed Tomography-Measured Adipose Tissue Indices

Halifax, Nova Scotia ( Radical cystectomy (RC) remains the standard of care treatment for muscle-invasive bladder cancer (MIBC), but is also an operation with significant morbidity and peri-operative complications. There have been numerous studies looking at predictors of perioperative complications and outcomes; amongst those, body mass index (BMI) has been assessed, and it has had mixed results as a predictor of outcomes. 

Obesity is a true epidemic, particularly in North American cultures. Recent accounts of obesity in the United States are staggering. While obesity in Canada lags behind its southern neighbor, it is not immune to this particular epidemic. Recent studies have suggested body mass index (BMI) may not be the best measure of obesity. As BMI is purely based on height and weight, it does not take into account distribution of that weight – location (central, peripheral) and tissue type (muscle, fat, etc.). 

In this study, visceral adipose tissue (VAT) and subcutaneous adipose tissue (SAT) levels were measured with computed tomography (CT) scans prior to radical cystectomy (RC). The hypothesis was that patients with higher adipose levels (particularly centrally, the site of RC) would have poorer perioperative and survival outcomes. 

VAT and SAT were calculated using a single slice from the pre-operative CT scan at the level of the umbilicus. Image J software, a proprietary mapping tool, was used to calculate VAT and SAT at this level. 

The authors retrospectively reviewed 202 patients undergoing RC at a single-institution between 2000 and 2012. Demographic evaluation of this population revealed that the median age was 70 years (interquartile range [IQR] 78-60), VAT 165 cm2 (IQR 223-114), SAT 233 cm2 (IQR 316-182), LOS 9 days (IQR 12-7), and age-adjusted Charlson Comorbidity Index (CCI) score was 6 (IQR 8-5). Most patients (76%) were male and 59% were smokers. In terms of pathology and treatment, 71% had ≥pT2 disease and 40% received chemotherapy.

In terms of outcomes, there were 32 (16%) 30-day Grade III-V complications. Over a median follow-up of 37 months (IQR 54-27), there were 43 (21%) bladder cancer and 65 (32%) all-cause deaths. 

Adjusting for CCI and smoking status, SAT predicted Grade III-V 30-day complications (OR 1.004; 95% confidence interval [CI] 1.001-1.008). VAT predicted increasing LOS (OR 0.0233; 95% CI 0.0002-0.0463) when adjusted for CCI and gender. 

In terms of survival outcomes, however, neither VAT nor SAT predicted DSS or OS. Ultimately, the short-term perioperative complications did not appear to affect patient survival.

Based on this, use of VAT and SAT may help improve perioperative risk calculators but have limited utility in prognosis discussions. However, as BMI results were mixed in prior literature, this needs to be validated in multiple other datasets first.

Limitations / Discussion Points:
1. Single slice CT – perhaps multi-slice imaging may help improve characterization?

Presented by: Kim, Michael, MD, Queen's University — Kingston, Canada
Co-Authors: Jaimin Bhatt1, Zachary Klaassen1, Bimal Bhindi1, Thomas Hermanns1, Patrick Richard1, John Kachura2, Robert Hamilton1, Neil Fleshner1, Antonio Finelli1, Michael Jewett1, Alexandre Zlotta1, Girish Kulkarni1.
Author Information: 
1. Surgical Oncology, University Health Network, Toronto, ON, Canada
2. Joint Department of Medical Imaging, University Health Network, Toronto, ON, Canada

Written By: Thenappan Chandrasekar, MD, Clinical Fellow, University of Toronto   Twitter: @tchandra_uromd at the 73rd Canadian Urological Association Annual Meeting - June 23 - 26, 2018 - Halifax, Nova Scotia