Construction and Validation of a Risk Nomogram Model for Inadvertent Intraoperative Hypothermia in Elderly Patients Undergoing Esophagectomy with General Anesthesia

doi:10.12104/j.issn.1674-4748.2025.09.019

Abstract：Objective: To explore the influencing factors of inadvertent intraoperative hypothermia (IIH) in elderly patients undergoing esophagectomy with general anesthesia, and to construct and validate a risk nomogram model based on independent influencing factors. Methods: A total of 114 elderly patients who underwent esophagectomy with general anesthesia from June 2021 to December 2022 were selected as the modeling set, and another 49 patients who underwent the same surgery from January 2023 to June 2023 were selected as the validation set. According to whether IIH occurred, the patients in the modeling set were divided into the hypothermia group (n=60) and the non-hypothermia group (n=54). Univariate analysis and multivariate logistic regression analysis were used to identify independent risk factors, and a predictive model was constructed based on regression analysis. The nomogram was visualized using R 4.2.1 software, and the predictive performance of the risk nomogram was evaluated by receiver operating characteristic (ROC) curves and calibration curves. The predictive performance of the model was externally validated using data from the validation set. Results: Age (OR=1.168), anesthesia duration (OR=1.092), fluid volume (OR=9.199), hypertension (OR=3.995), and diabetes (OR=5.836) were independent risk factors for IIH in elderly patients undergoing esophagectomy with general anesthesia (P<0.05), while body mass index (OR=0.045) and baseline temperature (OR=0.176) were independent protective factors (P<0.05). The risk nomogram constructed based on these seven independent factors had an area under the ROC curve (AUC) of 0.907 [95% CI (0.853, 0.961), P<0.001], with a cutoff score of 0.706, sensitivity of 0.750, and specificity of 0.981, indicating good discrimination ability. The average absolute error (MAE) of the calibration curve was 0.060, and it partially overlapped with the ideal 45° line, indicating minimal systematic error. The external validation results showed good AUC and MAE, suggesting that the model has good external predictive performance. Conclusion: The occurrence of IIH in elderly patients undergoing esophagectomy with general anesthesia is influenced by factors such as age, anesthesia duration, and fluid volume. The risk nomogram model constructed based on these factors has good predictive performance, which can help clinicians screen patients at high risk of IIH and prepare preventive measures in advance to ensure the smooth progress of esophagectomy.