RESUMEN
AIM: The morbidity of anastomotic dehiscence may be mitigated by a defunctioning stoma, but it is unclear if it is required for most low rectal anastomoses. Preoperative risk factors leading to anastomotic complications and the indications for faecal diversion have yet to be clearly defined. METHOD: Using the American College of Surgeons-National Surgical Quality Improvement Project (ACS-NSQIP) participant-use file, patients were identified who underwent low anterior resection with anastomosis for cancer at the 211 participating hospitals in 2005-08. RESULTS: A total of 1791 patients underwent low anterior resection. Patients were subdivided into two groups based on the level of the anastomosis. Of these 1266 patients had a low pelvic anastomosis (LPA) and 525 a coloanal anastomosis (CAA). In the LPA group, 606 patients had a stoma and 660 had no stoma. There were no differences in wound complications, sepsis or septic shock. Patients who had a stoma were more likely to have postoperative acute renal failure (1.7 vs 0.5%, P = 0.0485, OR 3.674). In the CAA group, 352 had a stoma and 173 had no stoma. In patients without faecal diversion, there was a significantly greater incidence of sepsis (8.7 vs 3.7%, P = 0.022, OR 2.47), septic shock (3.5 vs 0.57%, P = 0.018, OR 6.29) and need for reoperation (11 vs 1.7%, P = 0.0001, OR 7.11). Hospital length of stay was significantly longer with CAA and no stoma. On multivariate analysis, not having a stoma with a CAA was a risk factor for serious postoperative morbidity. CONCLUSION: While a defunctioning stoma with a coloanal anastomosis seems to protect from postoperative sepsis, septic shock and need for reoperation, it is likely that it is overused in rectal cancer surgery.