Population-based assessment of the surgical management of locally advanced colorectal cancer

Evidence-based guidelines recommend multivisceral resection for patients with locally advanced adherent colorectal cancer because it reduces local recurrence and improves survival. However, this procedure can increase morbidity compared with standard resection and may not be practiced uniformly. We performed a population-based study to examine surgical practice and outcomes among patients with locally advanced adherent colorectal cancer in the United States. Methods: Patients who were 18 years or older and who had surgical resection for nonmetastatic, locally advanced adherent colorectal cancer from January 1, 1988, through December 31, 2002, were identified from the Surveillance, Epidemiology, and End Results (SEER) registry. Logistic regression was used to examine patient, tumor, and geographic factors associated with multivisceral resection. Cumulative early mortality (i.e., at 1 and 6 months after diagnosis) and 5-year survival were obtained from Kaplan-Meier estimates; adjusted risks of death were calculated using Cox proportional hazards models. All statistical tests were two-sided. Results: We identified 8380 patients who underwent surgical resection for locally advanced adherent colorectal cancer, of whom 33.3% were managed with multivisceral resection. Among colon cancer patients, younger age at diagnosis, female sex, SEER region, node negativity, and left-sided tumors were independently associated with having had a multivisceral resection. Among rectal cancer patients, younger age at diagnosis and female sex were positively and statistically significantly associated with multivisceral resection, whereas receipt of neoadjuvant radiation was inversely and statistically significantly associated with multivisceral resection. Compared with standard resection, multivisceral resection was associated with improved overall survival for patients with colon (hazard ratio [HR] =0.89, 95% confidence interval [CI]=0.83 to 0.96) and rectal (HR=0.81,95% CI=0.70 to 0.94) cancer, with no associated increase in early mortality. Conclusions: The majority of patients with locally advanced colorectal cancer did not receive a multivisceral resection. The geographic variation in the application of this procedure in patients with colon cancer suggests that local organizational structures and processes of care may play an important role in patient treatment and, therefore, prognosis.