Utilization of interventional radiology in the postoperative management of patients after surgery for locally advanced and recurrent rectal cancer

: The surgical management of locally advanced primary rectal cancer and locally recurrent rectal cancer requires complex operations frequently resulting in complicated postoperative courses. We sought to evaluate the utilization of interventional radiologic (IR) procedures in the management of postoperative complications. Under Institutional Review Board approval, a prospective database of colorectal cancer patients undergoing resection from July 1999 to January 2010 was analyzed. Data collected included demographics, operative procedure, complications, length of stay, and IR utilization. Fisher's exact tests and logistic regression explored associations with necessitating an IR procedure during the postoperative period. Continuous variables were analyzed using Wilcoxon rank sum tests. One hundred and one patients underwent surgery and 66 received intraoperative electron radiotherapy (IOERT). Primary procedures included pelvic exenteration (n = 35), abdominoperineal resection (n = 25), low anterior resection (n = 23), paraaortic node dissection (n = 7), resection of isolated pelvic/retroperitoneal tumor (n = 7), and colectomy (n = 4). Sixty-two patients required multivisceral resection including partial/total cystectomy (n = 30), small bowel resection (n = 25), oophorectomy (n = 15), vaginectomy (n = 12), hysterectomy (n = 12), hepatectomy (n = 3), and nephrectomy (n = 3). Seventeen partial sacral resections and 47 pelvic sidewall resections were also required. One hundred and thirty-eight complications were identified in 72 patients, 30 of which required a procedural intervention. Twenty-seven IR procedures were performed including drainage of fluid collections (n = 14), nephrostomy tube placement (n = 8), arterial embolization (n = 2), inferior vena cava filter placement (n = 2), and pleural drainage (n = 1). Only three reoperations were required, none related to failure of IR procedures. There were no deaths. Estimated blood loss > 2000 mL (P = 0.002), IOERT (P = 0.03), and incomplete resection (P = 0.02) were found to be associated with postoperative IR utilization. Surgery for locally advanced primary rectal cancer and locally recurrent rectal cancer is associated with significant morbidity but low mortality. IR procedures play a significant role in the postoperative management of these patients and may decrease the need for reoperation.