An Integrative Nutritional Model for Predicting Postoperative Complications in Locally Advanced Gastric Cancer After Radical Surgery and P-HIPEC

Abstract Introduction This study aimed to detect the risk factors of postoperative complications (POCs) and constructed an integrative nutritional model predicting POCs after radical surgery (RS) plus prophylactic hyperthermic intraperitoneal chemotherapy (p-HIPEC). Methods This study included 875 patients receiving RS plus p-HIPEC from January 2019 to March 2024. Least Absolute Shrinkage and Selection Operator regression and logistic regression were used to filter the risk factors affecting POCs, and a model was constructed. Model performance was assessed by the area under the receiver operating characteristic curve (AUC), calibration curves, decision curve analysis, and clinical impact curve. Results Among 875 LAGC patients undergoing RS plus p-HIPEC, 217 (24.8%) developed POCs of Clavien-Dindo grade ≥ II, and 72 (8.2%) experienced severe POCs (grade ≥ III). The predictors included age > 65 (P < 0.001), ASA grade III (P < 0.001), high VSR (P < 0.001), low SMI (P < 0.001), high PNI (P = 0.003), and high SIRI (P < 0.001). These risk factors are shown on the nomogram model and verified. The nomogram showed the AUC of 0.877 (95% confidence interval [CI]: 0.846–0.908) in training set and 0.866 (95% CI: 0.817–0.915) in validation set, demonstrating superior prediction performance and greater clinical application value. Conclusion This study innovatively utilized 6 easily obtained clinical features and inflammatory-nutritional markers to construct an explainable model, providing a practical and reliable tool for predicting POCs in LAGC patients after RS plus p-HIPEC, which can help in making timely personalized clinical decisions for different risk populations.