Postoperative morbidity and mortality for malignant colon obstruction: the American College of Surgeon calculator reliability

Abstract

BackgroundThe American College of surgical risk calculator (SRC) score has never been validated specifically for surgery in emergency. The objective was to evaluate the reliability of this calculator in patients with malignant colon obstruction. Materials and methodsWe retrospectively have analyzed the morbidity and mortality observed in operated patients. Risk factors for postoperative morbidity and mortality were analyzed by logistic regression model. We have compared the morbidity and mortality estimated by the SRC score with that observed using the Brier Score (BS). A BS of 0 indicated perfect prediction, whereas a BS of 1 indicated the poorest prediction. ResultsSixty-nine patients aged 75 y (41-93) have been operated on emergency from November 2001 to August 2015. The tumor was localized in the sigmoid in 33 cases (48%), in the splenic flexure in nine cases (13%), and in the right colon in 17 cases (25%). The surgical procedures were as follows: right colectomy with anastomosis (29%), diverting proximal iliac colostomy (23%), and subtotal colectomy with anastomosis (19%). The SRC score indicated a good predictivity for mortality (9.8% predicted versus 8.7% observed, BS = 0.058), for morbidity (33.4% versus 40.6%, BS = 0.209), and for serious morbidity (25.5% versus 17.4%, BS = 0.131). In multivariate analysis, SRC was an independent risk factor for mortality (P = 0.030 odds ratio [OR] = 1.07 [1.01-1.15]) and morbidity (P = 0.001 OR = 1.16 [1.08-1.27]). ConclusionsSRC score is a reliable tool for assessing the morbidity and mortality of obstructive colon cancer and could help with adapting the surgical gesture to the risks predicted.