Construction of a risk prediction model and nursing strategy of hypothermia during hip replacement surgery for elderly patients based on the decision tree

WANG Hui; WANG Haiting; LU Lilan

doi:10.12102/j.issn.2095-8668.2025.17.020

Abstract：Objective: To construct a risk prediction model for intraoperative hypothermia in elderly patients undergoing hip arthroplasty based on decision tree, and to develop corresponding nursing strategies. Methods: A total of 108 elderly patients who underwent hip arthroplasty at Yanzhou University Jianhu Clinical Medical College from October 2021 to October 2024 were selected as the study subjects. The patients were divided into the hypothermia group (n=43) and non-hypothermia group (n=65) according to whether they developed hypothermia during surgery. Logistic regression analysis was used to identify risk factors for intraoperative hypothermia, and the decision tree algorithm was applied to build the prediction model. The receiver operating characteristic (ROC) curve was drawn to evaluate the predictive performance of the model. Meanwhile, targeted nursing strategies were proposed based on the risk factors. Results: There were statistically significant differences between the two groups in terms of preoperative body temperature, anesthesia time, operation time, operating room temperature, intraoperative fluid volume, and intraoperative blood loss (P<0.05). Logistic regression analysis showed that preoperative body temperature ≤36℃, anesthesia time ≥30 minutes, operation time ≥90 minutes, operating room temperature ≤22℃, intraoperative fluid volume ≥1500 mL, and intraoperative blood loss ≥150 mL were independent risk factors for intraoperative hypothermia in elderly patients undergoing hip arthroplasty (P<0.05). In the decision tree model, intraoperative blood loss was the most important predictive factor for intraoperative hypothermia in elderly patients undergoing hip arthroplasty. The area under the ROC curve (AUC) of the decision tree model (0.910) was higher than that of the logistic regression model (0.783). Conclusion: The risk prediction model for intraoperative hypothermia in elderly patients undergoing hip arthroplasty based on the decision tree method can accurately predict the risk of intraoperative hypothermia. Clinical medical staff can develop targeted nursing strategies based on the model's prediction results.