Anastomotic leak remains one of the most dreaded complications after restorative rectal surgery. It is widely accepted that proximal diversion with an ileostomy or colostomy reduces the clinical significance of postoperative anastomotic leaks.

No absolute indications for proximal diversion currently exist, but the most commonly accepted risk factors for postoperative leak are as follows:[1],[2]

• Comorbid conditions of the patients (e.g., cardiac disease, diabetes mellitus)
• Level of the anastomosis (the lower the level, the higher the risk)
• Male sex
• Prior pelvic radiotherapy
• Smoking history
• Technical mishaps with the anastomosis (e.g., incomplete stapler donut/ring, positive air leak test, need for reinforcing sutures, multiple stapler firings during a minimally invasive proctectomy)

A randomized trial of rectal cancer patients undergoing anastomoses < 8cm from the anal verge demonstrated a leak rate of 5.8% in the stoma group, versus a leak rate of 16.3% in the no stoma group (p=0.04).[3] In this study, male gender and lack of stoma usage were associated with anastomotic leakage when they employed multivariate analysis. Of note, use of radiochemotherapy, anastomotic height and body mass index were not found to be associated with the development of anastomotic leak.

The use of drains for low pelvic anastomoses is highly controversial. A recent systematic review on the use of prophylactic pelvic drainage after low anterior resection suggests no obvious benefit in regard to reduction in leak rate or overall complications. However, the study did note a reduction in the mortality rate of the drained group.[4] It is notable that the data quality in this area is overall relatively poor quality, and a well-designed multicenter trial with uniform inclusion criteria for patients undergoing low anterior resection to better study this question would be useful.