To analyze the risk factors that affect the occurrence of hypothermia during laparoscopic colorectal cancer surgery, and establish a nomogram warning model, and verify the model.

The clinical data of 275 patients undergoing laparoscopic colorectal cancer surgery in Hainan Branch of Chinese People′s Liberation Army General Hospital from January 2019 to December 2020 were retrospectively selected as the modeling group. The patients were divided into hypothermia group and normal group according to whether hypothermia occurred during the operation, and multivariate Logistic regression analysis was used to screen the risk factors that affect the occurrence of hypothermia during the operation. To use R software to establish a nomogram early warning model, and use Hosmer-Lemeshow (HL), calibration curve, area under the ROC curve and clinical decision curve (DCA) to evaluate the prediction effect of the model. In addition, 60 patients undergoing laparoscopic colorectal cancer surgery in the hospital from January to April 2021 were selected as the verification group for external verification of the model.

The incidence of intraoperative hypothermia in 275 patients undergoing laparoscopic colorectal cancer surgery was 42.91% (118/275). In this study, 4 risk factors including anesthesia time >150 min (OR = 3.939, 95% CI 1.762-8.804), intraoperative blood loss ≥150 ml (OR = 2.053, 95% CI 1.187-3.550), intraoperative fluid supplement ≥1 500 ml (OR = 2.084, 95% CI 1.199-3.624), and total CO₂ dosage≥200 L (OR = 2.707, 95% CI 1.571-4.663) were included to construct an intraoperative hypothermia warning model. Internal verification: the area under the ROC curve was 0.797 (95% CI 0.744-0.851), the HL goodness of fit test of the Logistic regression model with χ² = 6.27, P = 0.617, the predicted occurrence risk value of the calibration curve was basically the same as the actual occurrence risk value. External validation: the area under the ROC curve of the external validation group of the prediction model was 0.745 (95% CI 0.662-0.628). The 95% CI area of the GiViTI calibration curve did not cross the 45°diagonal bisector (P = 0.735), and when the DCA threshold range was 0.01-0.97, the nomogram model was reasonable to predict intraoperative hypothermia and the patient could get high net benefit.

The nomogram warning model established based on the risk factors of hypothermia during laparoscopic colorectal cancer surgery has good discrimination, consistency and clinical practicability, and can provide a certain reference value for intraoperative medical staff to predict the occurrence of hypothermia.

Laparoscopic surgery; Influencing factor; Colorectal cancer; Surgical hypothermia; Nomograms