Introduction/Background Anastomotic leak (AL) is a critical complication in colorectal surgery, with reported incidences showing significant variability (2.8% to 30%). This variability is probably due to the considerable heterogeneity of the populations studied, each characterized by specific risk factors. This study aimed to assess colorectal-AL incidence and associated risk factors in patients undergoing cytoreductive surgery (CRS) for ovarian(OC), colorectal(CRC), and upper-gastrointestinal(GI) cancer.

Methodology This retrospective analysis included patients undergoing at least rectosigmoid resection and colorectal anastomosis for CRS for OC, GI, CRC at Fondazione Policlinico Universitario A.Gemelli,IRCCS, Rome, between January-2015 and December-2022. Binomial logistic regression examined potential risk factors for AL using a multivariable logistic regression.

Results A total of 1324 patients were included (OC:941, CRC:280, GI:103). 304 patients(23.0%) received any chemo or radiation treatment before rectal resection (21.3% vs. 20.4% vs. 45.6% respectively). The AL rate did not significantly differ among groups (OC:2.7% vs. CRC:3.9% vs. GI:3.9%,p=0.478). OC patients had a significantly lower diverting ostomy rate than GI-cancer and CRC patients (35.6% vs. 44.3% vs. 53.3%, p<0.001). Independent AL risk factors varied: for OC, BMI<18 kg/m2, preoperative albumin<30 mg/dL, anastomosis distance<10cm, pelvic lymphadenectomy, and post-operative anemia (Hb≤8.0 g/dL); for GI-CRS: ECOG≥1 and preoperative albumin<30 mg/dL; for CRC surgery: preoperative chemoradiotherapy and medium-low anastomosis.

Conclusion AL is a relatively rare postoperative complication in a high-volume center with extensive experience in both colorectal and cytoreductive surgery, regardless of extent of carcinomatosis, additional procedures performed, and number of bowel resections. Nevertheless, specific risk factors can be identified within the three groups, indicating inherent differences in the pathologies. Among modifiable factors, malnutrition and performance status emerge as key predictive factors in both ovarian and GI cancer patients, highlighting the importance of preoperative nutritional optimization. Interestingly, diverting ostomy did not prove to be a protective factor against AL in any of the studied groups.

Disclosures No conflicts of interest to disclose.