Abstract
OBJECTIVE: To investigate high-risk factors for anastomotic fistula after laparoscopic radical resection of rectal cancer and establish a prediction model. METHODS: This is a retrospective cohort study included a total of 306 patients diagnosed with rectal cancer who underwent laparoscopic radical resection in the Second People's Hospital of Foshan from January 2022 to December 2024. The patients were randomly divided into a training set (N = 214) and a validation set (N = 92) in a 7:3 ratio. Models were constructed using univariate logistic regression analysis and multivariate logistic regression analysis based on the training set. Subsequently, the predictive capability of the model was evaluated using calibration curves, receiver operating characteristic (ROC) curves, decision curve analysis (DCA), and validation sets. RESULTS: The training set included 29 patients (13.6%) who developed anastomotic fistulas postoperatively. The study identified five predictive factors: gender (P = 0.032, OR = 2.68, 95% CI = 1.09-6.61), tumor stage (P = 0.008, OR = 3.66, 95% CI = 1.41-9.49), tumor location (P = 0.046, OR = 2.53, 95% CI = 1.02-6.30), surgical duration (P = 0.031, OR = 2.72, 95% CI = 1.10-6.76), and hypoalbuminemia (P = 0.005, OR = 4.28, 95% CI = 1.56-11.72). The AUC of the training set is 0.776 (95% CI = 0.673-0.879). The calibration curve validation showed that the predicted and measured values were in general agreement. DCA highlighted the model's clinical utility. CONCLUSION: The predictive model established in this study provides a tool for clinicians to identify high-risk patients for anastomotic fistula formation following laparoscopic radical rectal cancer surgery at an early stage. This facilitates early identification, detection, intervention and prevention of high-risk anastomotic fistula patients, thereby effectively reducing the risk of anastomotic fistula formation following laparoscopic radical rectal cancer surgery.