The present study cohort included a total of 4048 patients (61.6% males) with a mean age of 60, and a mean BMI of 27.81 kg/m2. The cohort consisted of 1166, 1836, and 1046 patients with tumors in the lower, middle, and upper third of the rectum, respectively. Patients with low rectal cancer had lower mean age (58.90 vs 60.23 vs 61.31, p < 0.001), lower mean BMI (27.49 vs 27.76 vs 28.23 kg/m2, p = 0.026), and lower rate of dyspnea (26.6% vs 30.6% vs 34.2%, p < 0.001) compared to patients with middle and upper rectal cancers. Patients with upper rectal cancer had lower rates of preoperative chemotherapy (34.9% vs 54.4% vs 56.0%, p < 0.001) and radiation (30.2% vs 51.0% vs 50.9%, p < 0.001) as compared to middle and lower cohorts (Table 1,2).

Within the entire cohort, 2,804 patients (69.3%) had surgery with an ileostomy. Patients with tumors in the upper third of the rectum had a much lower rate of having a concurrent ileostomy compared to middle and lower cohorts (49.9% vs 74.5% vs 78.5%, p < 0.001). When comparing intraoperative characteristics of the three patient cohorts, patients with lower rectal cancer had longer operative time (319.08 vs 293.13 vs 269.41 min, p < 0.001), lower rate of open surgery (24.0% vs 30.1% vs 30.7%, p < 0.001), and a higher rate of robotic surgery (23.4% vs 18.4% vs 12.7%, p < 0.001) compared to middle and upper rectal cancer cohorts (Table 2).

Organ space SSI and AL were the postoperative outcomes. For the entire cohort, the organ space SSI and AL rates were 6.7% and 4.3%, respectively. Additionally, to assess the incidence rate of organ space SSI and AL with different severity, we focused on the rates of severe AL and patients with organ space SSI who also had reoperation. For the entire cohort, the overall rate of having organ space SSI with reoperation was 2.1% and severe AL rate was 3.3%.

There was no statistically significant difference in rates of organ space SSI (6.3% vs 7.4% vs 6.0%; p = 0.274) or AL (3.9% vs 4.6% vs 4.4%; p = 0.709) among low, middle, and upper rectal cancer cohorts. There was also no statistically significant difference among the three cohorts in organ space SSI with reoperation (2.0% vs 2.0% vs 2.6%; p = 0.533) and rates of severe AL (2.6% vs 3.4% vs 3.8%; p = 0.172).

A sub-analysis comparing postoperative outcomes in patients with and without an ileostomy was performed. In the overall cohort, patients who had an ileostomy had longer mean operative time (303.51 vs 274.11 min, p < 0.001) and lower rate of having intraoperative transfusion (4.4% vs 6.7%, p = 0.002). Patients with diverting ileostomy were more likely to undergo minimally invasive surgery (laparoscopic: 32.3% vs 26.3%; robotic: 20.0% vs 14.8%, p < 0.001), while patients without an ileostomy were more likely to have open surgery (32.7% vs 26.6%, p < 0.001). There were no statistically significant differences between patients with and without an ileostomy in rates of organ space SSI (7.0% vs 6.1%, p = 0.302), and AL (4.7% vs 3.5%, p = 0.064). There were also no statistically significant differences between patients with and without an ileostomy in rates of severe cases of AL (3.6% vs 2.7%, p = 0.177) or organ space SSI with reoperation (2.4% vs 1.5%, p = 0.069) (Table 3).

The rates of organ space SSI and AL in patients with and without an ileostomy were compared in each cohort separately (upper, middle, and lower rectal cancers). There were no statistically significant differences between patients with and without an ileostomy with low rectal cancer in rates of organ space SSI (6.0% vs 7.2%, p = 0.501), AL (4.4% vs 2.4%, p = 0.153), organ space SSI with reoperation (2.1% vs 1.6%, p = 0.626) or severe AL (3.0% vs 1.2%, p = 0.285). There were no statistically significant differences between patients with and without an ileostomy in the mid rectal cancer cohort in organ space SSI (7.5% vs 7.2%, p = 0.88), AL (5.0% vs 3.4%, p = 0.162), severe AL (3.7% vs 2.6%, p = 0.376), or organ space SSI with reoperation (2.3% vs 1.3%, p = 0.189). There were also not statistically significant differences between patients with and without an ileostomy in the upper rectal cancer cohort in AL (4.8% vs 4.0%, p = 0.538), severe AL (4.2% vs 3.4%, p = 0.806), or organ space SSI with reoperation (3.4% vs 1.7%, p = 0.078). Surprisingly, patients with an ileostomy in the upper rectal cancer cohort had significantly higher rate of organ space SSI (7.5% vs 4.6%, p = 0.049) (Table 4).

As for the entire cohort, on multivariate analysis controlling for confounding variables of preoperative patient characteristics (age, sex, BMI, etc.), comorbidities (dyspnea, diabetes, etc.), preoperative chemotherapy and radiation therapy, intraoperative characteristics (operation time, operational approach, blood transfusion, etc.), adding an ileostomy did not decrease the odds of organ space SSI, AL, severe AL, or organ space SSI with reoperation (Supplement Table 1). When running a similar multivariate analysis for patients with upper, middle and lower rectal cancers, adding an ileostomy did not decrease the odds of space organ SSI, AL, organ space SSI with reoperation, or severe AL (Supplement Table 2a,b,c).