Thirty-Day Morbidity after Simultaneous Resection of Colorectal Cancer and Colorectal Liver Metastasis: American College of Surgeons NSQIP Analysis

Abstract

Approximately 20% of patients with colorectal cancer (CRC) present with synchronous liver metastases (CRLM). The decision to resect simultaneously or sequentially remains controversial. The primary aim of this study was to determine whether simultaneous resection of CRC and CRLM is associated with increased complications compared to isolated resection. Prospective data from the American College of Surgeons (ACS) NSQIP, including the ACS NSQIP procedure-specific colectomy and hepatectomy modules from 2014 to 2017, were reviewed in a retrospective cohort study. Primary study outcome was combined 30-day complication rates; secondary outcomes included colectomy and hepatectomy-specific complication. Multivariable logistic regression was performed to control for confounding factors associated with postoperative complication. A total of 23,643 patients underwent colectomy, 7,462 hepatectomy, and 592 simultaneous resection for CRC and CLRM. Overall morbidity was higher among patients treated with simultaneous resection (29.9%) compared with either isolated colorectal (22.2%) or hepatic resection (17.1%; p < 0.001). Additionally, postoperative ileus (36.4% vs 19.1%) and anastomotic failure (7.9% vs 3.8%) were more common after simultaneous resection compared with colorectal resection (p < 0.05). Similarly, rates of bile leak (8.3% vs 6.2%, p=0.195) and post-hepatectomy liver failure (8.7% vs 3.8%, p < 0.001) were higher after simultaneous resection compared with isolated hepatectomy. By multivariable logistic regression, simultaneous resection was associated with increased overall complication compared with isolated colon (odds ratio 1.64 [95% CI 1.36 to 1.96]) or liver resection (odds ratio 2.11 [95% CI 1.75 to 2.55]), as well as increased procedure-specific complication. Although simultaneous resection offers definitive resection for patients with synchronous CRC and CRLM, it is associated with significantly increased 30-day overall and procedure-specific postoperative morbidity.