[Factors associated with anastomotic leakage after anterior resection in rectal cancer].

Objective: To investigate the factors associated with the anastomotic leakage after anterior resection in rectal cancer.

Methods: From January 2014 to January 2017 471 patients underwent Dixon procedure for rectal cancer in The Affiliated Hospital of Qingdao University. The data of those patients was collected and reviewed retrospectively. Inclusion criteria included: 1) rectal cancer confirmed by preoperative electron colonoscopy; 2) the standard of total mesorectal excision followed by the surgeon during the surgery; and 3) elective surgery. Exclusion criteria included multi-primary rectal cancer, secondary surgery for tumor recurrence, palliative surgery, Miles procedure, Hartmann procedure, hormone drugs used, presence of rheumatic and immune diseases, and distant metastasis of rectal cancer. The variables, including demograpic characteristics, ASA score, diabetes mellitus, preoperative radiochemotherapy, histopathologic grade, pathological T stage, laparoscopic or open surgery, distance of the tumor from the anal verge ≤5 cm, were analyzed to identify the risk factors for anastomotic leakage.

Results: Of 471 patients, 285 and 186 were men and women, respectively, with a mean age of 61 years (range, 31-92) years. Symptomatic clinically anastomotic leakage occurred in 31 patients (6.6%, 31/471) after Dixon procedure for rectal cancer. On univariate analysis, the occurrence of anastomotic leakage was associated with diabetes (χ=10.972, P=0.001), serum albumin level <35 g/L (χ=9.784, P=0.002), neoadjuvant chemoradiotherapy (χ=6.867, P=0.009), distance ≤5 cm between the tumor and anal edge (χ=5.993, P=0.014), preventive colostomy (χ=5.630, P=0.018), and the use of double-perfusion cannula for abdominal flushing (χ=4.232, P=0.040). Multivariate analysis revealed that diabetes (OR=3.632, 95%CI: 1.620-8.145, P=0.002), neoadjuvant chemoradiotherapy (OR=3.177, 95%CI: 1.283-7.867, P=0.012) and distance ≤5 cm between the tumor and anal edge(OR=2.444, 95%CI: 1.172-5.059, P=0.017) were independent risk factors for anastomotic leakage, while preventive colostomy (OR=0.138, 95%CI: 0.056-0.345, P=0.000) and the use of double-perfusion cannula for abdominal flushing (OR=0.223, 95%CI: 0.086-0.575, P=0.002) were independent protective factors for anastomotic leakage.

Conclusions: For patients with rectal cancer with diabetes, undergoing neoadjuvant chemoradiotherapy, or distance ≤5 cm between the tumor and anal edge, anastomotic leakage after anterior resection of rectal cancer must be paid attention. When necessary, preventive colostomy or use of double-perfusion cannula for abdominal flushing should be considered.