Abstract
BACKGROUND: Surgical site infection (SSI) is a common and serious complication following gastric cancer surgery, often linked to patient age, surgery duration, and the surgical approach taken. Accurate prediction and personalized mitigation of SSI risk are crucial for improving surgical outcomes. While prior studies have focused on SSI rates after open and laparoscopic gastric cancer surgeries, it is important to also consider robot-assisted procedures. This study aims to develop a predictive model for SSI after radical gastric cancer surgery, validate it through external testing, and provide a reliable tool for clinical use. METHODS: Data from 763 postoperative gastric cancer patients were analyzed, with 601 in the training set from Gansu Provincial People's Hospital and 162 in the validation set from The First Hospital of Lanzhou University. All available variables were considered as potential predictors, and factors influencing SSI post-surgery were identified using logistic regression. A nomogram model was then created for precise SSI risk prediction. RESULTS: Among the 763 gastric cancer patients, 10.9% experienced postoperative SSI. Significant differences were noted in the American Society of Anesthesiologists (ASA) physical status classification system classification, preoperative albumin levels, surgical approach, and reconstruction techniques between groups. Age, surgery duration, surgical approach, total gastrectomy, and tumor diameter were identified as significant predictors of SSI. The nomogram model showed high predictive accuracy, with concordance index (C-index) values of 0.834 in the training set and 0.798 in the validation set. Calibration plots and decision curve analysis (DCA) further validated the model's performance. CONCLUSIONS: This study identified five key predictors of postoperative SSI in gastric cancer and developed a nomogram model to enhance SSI prediction. These findings have important implications for preventing SSI in gastric cancer surgeries.