Rethinking Anastomotic Leak Rates Among Patients Undergoing Ileorectal Anastomosis

Results from a large cohort, retrospective analysis indicate that anastomotic leaks may not occur as frequently as previously reported


Total abdominal colectomy (TAC) with ileorectal anastomosis (IRA) is an operation that can be applied to a number of different conditions including inflammatory bowel disease, synchronous colorectal cancer, patients with constipation and a variety of other diseases. A recent presentation at the American Society of Colon and Rectal Surgeons Annual Scientific Meeting explored the outcomes of ileorectal anastomosis across different patient populations and found that anastomotic leaks (AL) may occur less frequently than previously reported. 

Advertising Policy

Cleveland Clinic is a non-profit academic medical center. Advertising on our site helps support our mission. We do not endorse non-Cleveland Clinic products or services Policy

“The overall aim of the study was to look at the outcomes of the ileorectal anastomosis across the different patient populations,” explains Stefan Holubar, MD, MS, Section Chief of the Inflammatory Bowel Disease Section and Director of Research for the Department of Colon & Rectal Surgery at the Digestive Disease & Surgery Institute and corresponding author on the study. “Most of the literature reports a maximum of just a couple of hundred patients, so we wanted to take more of a 10,000-foot view approach to the ileorectal anastomosis outcomes in a variety of patient populations.”

Dr. Holubar notes that the impetus for this study came from a paper published 15 years ago that indicated anastomotic leaks can occur after 30 days. But even though that study had a small number of ileorectal patients (n = 30), they had a very high 23% leak rate (n = 7), which caused them to modify their anastomotic techniques.

“A lot of people feel that ileorectal anastomosis — given the size discrepancy between the more feeble lumen and the more robust muscular layers of the small bowel and rectum — that these patients are prone to have a high ileus rate,” explains Dr. Holubar. “We had these questions of, ‘Is the leak rate really that high? And is the ileus rate really that high?’”

Study design

The group performed a retrospective analysis of patients undergoing TAC-IRA or end-ileostomy takedown (EI) with IRA after previous TAC with or without diverting loop ileostomies (DLI) between 1980-2021. Patients were identified from a prospectively maintained institutional database. Short-term (30-day) surgical outcomes were collected using the institutional database, but redo IRA cases were excluded. The study defined AL after IRA as an anastomotic fluid collection seen on cross-sectional imaging with signs of sepsis or radiographic evidence of contrast extravasation.

Advertising Policy

Of the 823 patients included in the study, 767 (93%) underwent TAC-IRA, and 56 (7%) underwent EI-IRA. Among the indications for surgery, constipation (32%), multifocal colon cancer (27%), hereditary syndrome (20%) and inflammatory bowel disease (IBD) (15%) were most common. DLI was performed in 223 (27%) patients, but it was more common among patients with constipation or IBD. The overall AL rate was 3%; 1% and 4% in those with and without DLI, respectively, and DLI was found to be protective against AL (OR 0.28, 95% CI 0.08-0.94, p = 0.04).

Patients with DLI were found to have a higher overall postoperative complication rate (51% vs 36%, p<0.001). Complications included superficial wound infections, urinary tract infections, dehydration, blood transfusions and portomesenteric venous thromboses (all p < 0.04). Stoma-related complications occurred in 4% of DLI patients.

Given the size of the cohort, the group was able to do a multivariate analysis for overall postoperative morbidity and identified a few factors that seemed to increase the risk of the operation. These factors included coagulopathy (OR = 4.30), smoking 1.44), depression (OR = 1.56), DLI (OR = 1.50) having a large BMI (OR = 1.03 kg/m2), and prolonged surgery (OR = 1.02/10 min).  

“The biggest risk factor was actually a coagulopathy, both in terms of patients who have bleeding or are at risk for both bleeding complications and venous thromboembolism complications,” explains Dr. Holubar. “What we also found, which is not that surprising if you think about it, is that the diverting loop ileostomy increased the overall morbidity. That’s likely due to dehydration and the fact that these patients are sicker going into the operation, which goes back to why the surgeon decided to divert them in the first place.”

Advertising Policy


“This was the largest series reported to date by several factors,” says Dr. Holubar. “What we found in terms of ileus was that it wasn’t as high as we might think. The overall ileus rate was decently high, but it wasn’t the extraordinarily high numbers that we had seen in previous studies — it was about 17 to 18%, which is a bit higher in general. We also found that low-risk colorectal surgery patients have at least a 10% risk of ileus, and certainly we observe rates between 10 and 50% all the time. So that confirmed that there’s some increased risk for ileus, but it wasn’t outrageously high, so to speak.”

The group found that both IRA with and without DLI were associated with acceptable anastomotic leak rates. However, they noted that diversion is associated with a lower leak rate but higher rates of complications. The findings reiterate the importance of careful patient selection and consideration of diversion when necessary to minimize the risk of postoperative anastomotic leak.

“I think the findings from our study refute the findings from the original study from 15 years ago that suggested that the ileorectal is associated with a very high leak rate,” says Dr. Holubar. “I think that’s because we place a strong focus on proper patient selection with liberal use of diverting loop ileostomy for the sicker patient presentations.”