Construction and validation of a model for predicting surgical site infection in elderly patients with colorectal cancer based on perioperative multidimensional variables

OBJECTIVE To explore the influencing factors for surgical site infection (SSI) in elderly patients with colorectal cancer (CRC) after surgery, and to construct and validate a nomogram prediction model. METHODS A total of 1 268 elderly patients with CRC admitted to Liaocheng People's Hospital from Jan. 2021 to May 2025 were selected as the study subjects and randomly assigned to a model training set (n=888 cases) and a validation set (n=380 cases) in a 7∶3 ratio. LASSO-logistic regression was used for variable selection, and a nomogram model was constructed. Receiver operating characteristic (ROC) curves and calibration curves were plotted to internally validate the model performance. RESULTS Diabetes mellitus (OR=2.857, 95%CI: 1.695-4.815), ASA class III (OR=2.081, 95%CI: 1.210-3.580), multiple tumors (OR=5.613, 95%CI: 2.745-11.479), rectal tumors (OR=3.086, 95%CI: 1.809-5.265), carbohydrate antigen 19-9 (CA19-9) >39 U/ml (OR=3.516, 95%CI: 2.026-6.103) and operation duration (OR=1.519, 95%CI: 1.179-1.957) were identified as risk factors for SSI in elderly patients with CRC after surgery. CD₄⁺/CD₈⁺ ratio (OR=0.443, 95%CI: 0.241-0.813), serum albumin (OR=0.901, 95%CI: 0.855-0.950) and elective surgery (OR=0.109, 95%CI: 0.032-0.375) were identified as protective factors for SSI in elderly patients with CRC after surgery. Based on the aforementioned indicators, a nomogram model was constructed. The area under the curve (AUC) of the validation set was 0.797 (95%CI: 0.717-0.865), indicating good discriminatory power of the model. The Hosmer-Lemeshow goodness-of-fit test showed that the model had good accuracy and consistency (χ²=6.315, P=0.097). CONCLUSIONS The nomogram prediction model constructed based on LASSO-logistic in this study has good predictive value, which is helpful for early identification of high-risk patients for SSI before surgery and implementation of targeted infection control measures, optimizing perioperative management.