Scholarlink

Background <p>Intraoperative hypothermia is common during video-assisted thoracoscopic surgery (VATS) because of prolonged exposure of the pleural cavity and the thermoregulatory effects of anesthesia. Despite its clinical relevance, predictive tools tailored to VATS remain limited. This study aimed to develop and validate a practical model to identify patients at increased risk of intraoperative hypothermia.</p> Methods <p>This retrospective study included 651 patients who underwent VATS at a tertiary hospital in Beijing (January 2023–August 2025). Thirty-three perioperative variables were assessed. Predictors were selected using least absolute shrinkage and selection operator (LASSO) regression, and multivariable logistic regression was performed to build the model. Restricted cubic spline (RCS) analysis was applied to explore potential nonlinear associations between continuous predictors and intraoperative hypothermia. A nomogram was developed and internally validated by 1000 bootstrap resamples. Model performance was evaluated in terms of discrimination, calibration, and clinical utility using receiver operating characteristic (ROC) analysis, calibration plots, the Hosmer–Lemeshow test, and decision curve analysis (DCA).</p> Results <p>Intraoperative hypothermia occurred in 407 patients (62.52%). Independent risk factors included older age, longer anesthesia duration, intraoperative hypotension, bradycardia, and elevated high-density lipoprotein (HDL), whereas higher body mass index (BMI) and forced intraoperative warming were protective. The nomogram showed good discrimination with an area under the ROC curve of 0.890 (95% CI, 0.865–0.914), sensitivity of 0.717, and specificity of 0.658. Calibration demonstrated close agreement between predicted and observed outcomes (mean absolute error = 0.008), and the Hosmer–Lemeshow test indicated adequate fit (χ<sup>2</sup> = 5.951, P = 0.653). DCA confirmed favorable clinical utility across a wide range of threshold probabilities.</p> Conclusions <p>The proposed nomogram demonstrated strong predictive accuracy and clinical value for intraoperative hypothermia in VATS patients. The model identifies patients at residual risk of hypothermia despite standard perioperative care and may support optimized temperature surveillance and warming resource use.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Development and validation of a predictive model for intraoperative hypothermia in patients undergoing video-assisted thoracoscopic surgery: a retrospective study

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年2月11日

查看全文
  • Ruirong Chen,
  • Yingjie Du,
  • Xiongpeng He,
  • Min Liu,
  • Wenjia Shen,
  • Guyan Wang

摘要

Background

Intraoperative hypothermia is common during video-assisted thoracoscopic surgery (VATS) because of prolonged exposure of the pleural cavity and the thermoregulatory effects of anesthesia. Despite its clinical relevance, predictive tools tailored to VATS remain limited. This study aimed to develop and validate a practical model to identify patients at increased risk of intraoperative hypothermia.

Methods

This retrospective study included 651 patients who underwent VATS at a tertiary hospital in Beijing (January 2023–August 2025). Thirty-three perioperative variables were assessed. Predictors were selected using least absolute shrinkage and selection operator (LASSO) regression, and multivariable logistic regression was performed to build the model. Restricted cubic spline (RCS) analysis was applied to explore potential nonlinear associations between continuous predictors and intraoperative hypothermia. A nomogram was developed and internally validated by 1000 bootstrap resamples. Model performance was evaluated in terms of discrimination, calibration, and clinical utility using receiver operating characteristic (ROC) analysis, calibration plots, the Hosmer–Lemeshow test, and decision curve analysis (DCA).

Results

Intraoperative hypothermia occurred in 407 patients (62.52%). Independent risk factors included older age, longer anesthesia duration, intraoperative hypotension, bradycardia, and elevated high-density lipoprotein (HDL), whereas higher body mass index (BMI) and forced intraoperative warming were protective. The nomogram showed good discrimination with an area under the ROC curve of 0.890 (95% CI, 0.865–0.914), sensitivity of 0.717, and specificity of 0.658. Calibration demonstrated close agreement between predicted and observed outcomes (mean absolute error = 0.008), and the Hosmer–Lemeshow test indicated adequate fit (χ2 = 5.951, P = 0.653). DCA confirmed favorable clinical utility across a wide range of threshold probabilities.

Conclusions

The proposed nomogram demonstrated strong predictive accuracy and clinical value for intraoperative hypothermia in VATS patients. The model identifies patients at residual risk of hypothermia despite standard perioperative care and may support optimized temperature surveillance and warming resource use.