Study on the construction and verification of intraoperative pressure injury risk prediction model for children undergoing cardiac surgery.

BACKGROUND: This study applied nomogram to develop an intraoperative acquired pressure injury (IAPI) risk prediction model for pediatric cardiac surgery patients, validated its predictive performance, and aims to provide evidence-based guidance for IAPI prevention in pediatric cardiac surgery. METHODS: A retrospective analysis of clinical surgical data from 1,179 pediatric patients undergoing cardiac surgery at Fujian Children's Hospital between August 2022 and November 2024 was conducted to construct a corresponding dataset. Through LASSO analysis and multivariate logistic stepwise regression, we identified high-risk factors for IAPI in pediatric cardiac surgery patients and developed a ROC curve prediction model. The model's fit and predictive performance were evaluated using the Hosmer-Lemeshow test and ROC area under the curve (AUC), with internal validation performed via bootstrap. RESULTS: A total of 1,179 pediatric cardiac surgery patients were included in the study, with 70 cases (5.94%) developing IAPI. LASSO regression analysis identified 10 variables, and subsequent multivariate logistic regression analysis revealed that preoperative hematocrit, prothrombin time, fibrinogen, Braden-Q score, and concurrent respiratory tract infection were significant predictors of IAPI in these pediatric patients (P < 0.05). The Hosmer-Lemeshow test yielded a chi-square value of 11.251 (P = 0.188). Internal validation demonstrated the model's sensitivity at 0.826, specificity at 0.758, and an ROC curve area under the curve (AUC) of 0.833 (0.741-0.925). CONCLUSION: The machine learning and nomogram-based predictive model for IAPI risk in pediatric cardiac surgery demonstrates significant predictive efficacy, providing a scientific basis for operating room nurses to identify high-risk IAPI patients early and implement timely personalized nursing interventions.