Research Article: Prediction of intraoperative hypothermia in esophageal cancer radical surgery under general anesthesia: development and validation of a nomogram model
Abstract:
The incidence of unintended intraoperative hypothermia is high in patients undergoing esophageal cancer radical surgery under general anesthesia, which may lead to a series of complications. To effectively prevent intraoperative hypothermia, this study aimed to establish and validate a predictive model for assessing the risk of intraoperative hypothermia in these patients.
This study retrospectively collected clinical data from 601 patients who underwent esophageal cancer radical surgery under general anesthesia at a university hospital between January 1, 2020 and January 31, 2024. Using bootstrap resampling (with 1,000 replications), the data were divided into a modeling cohort of 421 cases and a validation cohort of 180 cases in a 70%:30% ratio. A total of 17 potential risk factors were analyzed, including demographic characteristics, disease status, and surgical factors. The least absolute shrinkage and selection operator (LASSO) regression model was first employed to screen and optimize risk factors. Subsequently, based on the selected important variables, Multivariable logistic regression analysis was used to construct the final intraoperative hypothermia risk prediction model. To evaluate model performance, the C-index was calculated, and calibration curves and clinical decision curves were plotted. The predictive accuracy and clinical applicability of the model were further validated using the validation cohort.
A nomogram for intraoperative hypothermia risk prediction was constructed based on three predictors: body mass index (OR?=?0.039, 95% CI: 0.008–0.148), preoperative body temperature (OR?=?0.130, 95% CI: 0.078–0.213), and intraoperative blood loss (OR?=?0.130, 95% CI: 0.078–0.213). The model showed a C-index of 0.816 (95% confidence interval: 0.779–0.853), indicating good discriminative ability. Decision curve analysis further confirmed that when the threshold probability for intraoperative hypothermia ranged from 6% to 91%, the nomogram provided significant clinical net benefit, suggesting its value in practical applications. Internal validation results supported the robustness of the model, with a C-index of 0.822, further verifying the accuracy and reliability of the nomogram in predicting intraoperative hypothermia risk.
Lower body mass index, lower preoperative body temperature, and greater intraoperative blood loss were identified as independent predictors of intraoperative hypothermia in patients undergoing radical esophagectomy under general anesthesia. In particular, patients with BMI <23.9?kg/m 2 , preoperative body temperature <36.5°C, and estimated intraoperative blood loss ?500?mL were at significantly higher risk. The predictive model constructed based on these key indicators demonstrated good discrimination and calibration, and can serve as an important tool in clinical practice for identifying high-risk patients and guiding individualized temperature management strategies.
Introduction:
Esophageal cancer is the sixth leading cause of cancer-related deaths worldwide, causing approximately 300,000 deaths each year ( 1 , 2 ). Over the past decade, significant progress has been made in the treatment of esophageal cancer. For locally early-stage tumors, endoscopic resection has become a feasible and effective treatment option. For locally advanced but resectable esophageal cancer, the current standard treatment is multimodal therapy, which involves preoperative chemotherapy and radiotherapy followed…
Read more