Development and Validation of a Predictive Nomogram for Intraoperative Hypothermia in Elderly Patients Undergoing Da Vinci Robot-Assisted Urological Tumor Resection: A Retrospective Cohort Study

This study aims to construct a Nomogram for intraoperative hypothermia (IH) in elderly patients undergoing robot-assisted urological tumor resection (RAUTR) and to evaluate the effect of the model by internal and external validation. Using convenient sampling to enroll patients in a large hospital from February 2022 to July 2024 as the modeling and validation cohort. Identifying the independent risk factors for IH by univariate and multivariate logistic regression, and developing a Nomogram by the R software. The Nomogram's discrimination and accuracy were tested by receiver operating characteristic (ROC) curves and the Hosmer-Lemeshow (H-L) test, internal validation was performed with 1000 Bootstrap resamples and calibration curves. External evaluation was conducted on a validation cohort using ROC curves and H-L tests. The modeling cohort included 420 patients, with an IH rate of 39.8%. Univariate and multivariate logistic regression showed that baseline temperature (odds ratio [OR] = 0.087), preoperative psychological score (OR = 1.114), body mass index (BMI) (OR = 0.820), and anesthesia time (OR = 1.013) were independent risk factors for IH. The ROC curve of the Nomogram had an area under the curve of 0.844 (95% confidence interval [CI]: 0.807-0.881), a maximum Youden index of 0.563, a best cutoff value of 0.383, a sensitivity of 0.772, and a specificity of 0.791. The H-L test yielded a chi-square value of 10.173 and a p-value of 0.253. Internal validation with 1000 Bootstrap resamples showed a consistency coefficient of 0.844, the calibration curve fits well. A total of 120 patients were included in the validation cohort, including 45 with hypothermia (37.5%). The area under the ROC curve for the prediction of IH in the external validation cohort was 0.854 (95% CI: 0.781-0.927), and the H-L test yielded a chi-square value of 5.207 and a p-value of 0.735. The IH rate is high in elderly patients undergoing RAUTR. Baseline temperature, preoperative psychological score, BMI, and anesthesia time are independent risk factors. And the Nomogram could be used to predict IH.