Artificial intelligence-based assessment of pre-operative body composition is associated with early complications after radical cystectomy.

Abstract

We aimed to use a validated artificial intelligence(AI) algorithm to extract muscle and adipose areas from CT images before radical cystectomy(RCx), and then correlate these measures with 90-day post-RCx complications. A tertiary referral center's cystectomy registry was queried for patients who underwent RCx between 2009-2017 for bladder cancer. 843 RCx patients with CT imaging within 90 days preceding surgery were included, to allow for extraction of body composition parameters by AI. We assessed complications within 90 days of surgery including: wound, infectious, major complications, re-admission, and death. Multivariable logistic regressions associated pre-RCx measures to post-RCx complications. Increasing subcutaneous adipose tissue(SAT) was associated with more wound complications while patients with increasing visceral adipose tissue(VAT) had greater odds of infectious related complications. After adjusting for patient characteristics, every 10 cm2 increases in fat mass index(FMI) were associated with more infectious (OR 1.04, p=0.002) and wound (OR 1.06, p<0.001) complications. On multivariable analysis, higher pre-operative skeletal muscle index(SMI) was associated with lower odds of major complications (OR 0.75 for every 10 cm2, p=0.008), while higher intramuscular adipose(IMA) was associated with higher odds of major complications (OR 1.93, p=0.008). Automated AI body composition measurements pre-operatively are associated with post-RCx complications. These measurements, in addition to patient (ECOG performance status, smoking status) and surgical (robotic approach, continent diversion) characteristics can then be utilized to individualize patient counseling and facilitate triage of nutritional and rehabilitation efforts.