Creating and validating an anastomotic leakage risk prediction model after laparoscopic low anterior resection for rectal cancer

PURPOSE: Anastomotic leakage (AL) is a serious complication after rectal cancer surgery, and there is still a lack of effective prediction tools. This study aims to provide a basis for the development of individualized AL prevention plans. METHODS: 585 patients who underwent laparoscopic low anterior resection for rectal cancer in Xuzhou Central Hospital from 2019 to 2023 were retrospectively enrolled and randomly divided into a training set (410 cases) and a validation set (175 cases). Predictors were screened by LASSO regression, and a nomogram prediction model based on logistic regression was constructed. The area under the curve (AUC), calibration curve, decision curve and clinical impact curve were used to evaluate the model performance. RESULTS: The incidence of AL was approximately 13 % (76/585). According to LASSO regression, 8 predictors were identified: male gender, larger tumor diameter, a shorter distance between the tumor's lower margin and the anal verge, non-preservation of the left colic artery during surgery, preoperative neoadjuvant therapy, higher levels of carbohydrate antigen 19-9, no preventive stoma, and prolonged operative time. The AUC of the model in the training set and validation set was 0.745 (95 % CI: 0.675-0.814) and 0.733 (95 % CI: 0.606-0.859), respectively, and the calibration and clinical practicality were also favorable. CONCLUSIONS: The prediction model is relatively accurate and can provide a basis for the formulation of individualized AL prevention strategies.