Objective Analyze the risk factors of pulmonary infection after cardiac surgery and construct a nomogram prediction model, so as to provide reference for early screening of high-risk groups and targeted prevention and control measures. Methods Case data of 1188 patients undergoing cardiac surgery from January 2019 to December 2023 were collected, and the training group and validation group were randomly divided in a ratio of 7:3. Based on Logistic regression to analyze the predictors of pulmonary infection after cardiac surgery, the risk prediction model was established and validated. The model was evaluated by area under the receiver operating characteristic curve (AUC), calibration curve and decision curve analysis. Results Of the 1,188 patients, 148 (12.5%) had pulmonary infections. Independent risk factors for pulmonary infection after cardiac surgery were emergency department admission, history of smoking, chronic obstructive pulmonary disease, postoperative renal insufficiency, reintubation, intraoperative blood transfusion volume and tracheal intubation time. The AUC value was 0.923, specificity and sensitivity were 0.804 and 0.885, respectively; in the validation group, the AUC value was 0.946, specificity and sensitivity were 0.788 and 0.977, respectively, and high model calibration and net benefit. Conclusions The constructed predictive model has certain predictive value and can be used for risk assessment and individualized treatment of pulmonary infection in patients after cardiac surgery. |