Construction and validation of postoperative hypothermia prediction model for patients undergoing joint replacement surgery

Development and validation of a novel risk assessment model for accurate prediction of intraoperative hypothermia in adult patients undergoing different types of surgery: insights from a multicentre, retrospective cohort study

BACKGROUND

Intraoperative hypothermia is a prevalent complication that may significant clinical and economic burdens. Previous risk assessment models have demonstrated limitations in accurately predicting intraoperative hypothermia, particularly in diverse surgical populations. This study aims to develop and validate a model in adult surgical patients to improve outcomes.

METHODS

This retrospective cohort study utilized data extracted from electronic medical records and anaesthesia information management systems between June 2022 and August 2023. The analysis included information of 3,405 adult surgical patients from three independent centres in China who underwent elective surgical procedures with body temperature monitoring. Intraoperative hypothermia was defined as a core temperature below 36 °C during surgery. The Least Absolute Shrinkage and Selection Operator (LASSO) regression employed to select optimal features and multivariate logistic regression was used to identify independent predictors of intraoperative hypothermia and then built the risk assessment model. We further evaluated the discriminative ability, calibration curves, and clinical utility of the predictive model.

RESULTS

The total incidences of intraoperative hypothermia in adult surgical patients were 42.5%. The predictors in the intraoperative hypothermia model included: age, BMI, baseline HR, baseline temperature, minimally invasive surgery, smoking, previous surgery and serum creatine level. In the training cohort, the model demonstrated strong discriminatory ability, with C-index values of 0.721 (95% CI 0.697-0.744). Internal and external validation further confirmed the model's robustness and generalizability.

CONCLUSION

These findings suggest that our model may help us more accurately identify patients at risk of intraoperative hypothermia.

TRIAL REGISTRATION

China Clinical Trial Registration Center (ChiCTR2300071859), Date registered May/26/2023.

Risk factors for postoperative hypothermia in non-cardiac surgery patients: a systematic review and meta-analysis

BACKGROUND

Postoperative hypothermia seems to be a common problem in surgical patients but is easily ignored. This study aimed to identify risk factors for postoperative hypothermia in non-cardiac surgery patients.

METHODS

We searched databases including PubMed, Embase, Web of Science, Cochrane Library, CINAHL, VIP, Wan Fang, CNKI, and CBM from inception to April 2025. The studies were selected using inclusion and exclusion criteria. Two reviewers screened studies, extracted data, and independently evaluated the risk of bias. The quality of the study was assessed with the Newcastle-Ottawa Scale, and a meta-analysis was carried out with Revman 5.4 software.

RESULTS

A total of 17 studies were included. Age ≥ 60 (odds ratio [OR] = 1.80), BMI < 18.5 kg/m2 (OR = 1.83), ASA III-IV (OR = 1.87), endoscopic surgery (OR = 1.93), intraoperative blood loss ≥ 100ml (OR = 2.35), intravenous fluid ≥ 1000ml (OR = 1.87), blood transfusion (OR = 1.80), duration of anesthesia > 1 h (OR = 1.99) and duration of surgery > 1 h (OR = 2.34) were significant risk factors that contributed to postoperative hypothermia in non-cardiac surgery patients.

CONCLUSION

There are many risk factors for postoperative hypothermia in patients undergoing non-cardiac surgery. The results of this research may improve clinician awareness, risk stratification, and prevention of postoperative hypothermia in non-cardiac surgery patients.

Development and Internal Validation of a Risk Model to Estimate Probability of Intraoperative Hypothermia in Adult Surgical Patients

Intraoperative hypothermia is associated with various perioperative complications and an increased risk of mortality. This study aims to develop and validate a reliable risk model, the Intraoperative Hypothermia Risk Estimating Model (IHREM), for assessing the likelihood of intraoperative hypothermia in adult patients receiving different types of surgery and anesthesia. Data from 1815 surgical patients were collected, with 1521 used to develop the IHREM training set. Univariate logistic regression was utilized to evaluate the parameters included in the study. For the first time, parameters showing non-linear associations with the risk of intraoperative hypothermia were evaluated and then incorporated into a primary model using restricted cubic splines (RCS), based on the result of multivariate logistic regression. The final model was comprised of 12 risk factors, including body mass index (BMI), fasting time, preoperative heart rate, preoperative tympanic temperature, intravenous fluid administration volume, intraoperative irrigation volume, estimated blood loss, duration of anesthesia, surgical position, intraoperative warming, operation room temperature, and humidity. The IHREM model demonstrated satisfactory performance in the training set, exhibiting reliable discrimination, calibration, overall performance, and clinical utility. In the temporal validation set (n = 294), the c-index, calibration intercept and calibration slope, Brier score, and R2 were determined to be 0.763 (95% CI, 0.710-0.819), 0.394 (95% CI, 0.118-0.680), 0.865 (95% CI, 0.638-1.114), 0.204 (95% CI, 0.180-0.229), and 0.236, respectively. Meanwhile, decision curve analysis and clinical impact curve showed that IHREM provides promising clinical value. In addition, RCS analysis indicated that maintaining the operation room temperature above 20°C is sufficient to prevent hypothermia while increasing or sustaining the preoperative core temperature to around 36.7-36.8°C significantly reduces the risk of hypothermia. IHREM holds promise as a valuable tool for identifying adult patients at risk of intraoperative hypothermia under various types of surgery and anesthesia, thereby supporting clinical decision-making.

Cross-Combination Analyses of Random Forest Feature Selection and Decision Tree Model for Predicting Intraoperative Hypothermia in Total Joint Arthroplasty

BACKGROUND

In total joint arthroplasty patients, intraoperative hypothermia (IOH) is associated with perioperative complications and an increased economic burden. Previous models have some limitations and mainly focus on regression modeling. Random forest (RF) algorithms and decision tree modeling are effective for eliminating irrelevant features and making predictions that aid in accelerating modeling and reducing application difficulty.

METHODS

We conducted this prospective observational study using convenience sampling and collected data from 327 total joint arthroplasty patients in a tertiary hospital from March 4, 2023, to September 11, 2023. Of those, 229 patients were assigned to the training and 98 to the testing sets. The Chi-square, Mann-Whitney U, and t-tests were used for baseline analyses. The feature variables selection used the RF algorithms, and the decision tree model was trained on 299 examples and validated on 98. The sensitivity, specificity, recall, F1 score, and area under the curve were used to test the model's performance.

RESULTS

The RF algorithms identified the preheating time, the volume of flushing fluids, the intraoperative infusion volume, the anesthesia time, the surgical time, and the core temperature after intubation as risk factors for IOH. The decision tree was grown to 5 levels with 9 terminal nodes. The overall incidence of IOH was 42.13%. The sensitivity, specificity, recall, F1 score, and area under the curve were 0.651, 0.907, 0.916, 0.761, and 0.810, respectively. The model indicated strong internal consistency and predictive ability.

CONCLUSIONS

The preheating time, the volume of flushing fluids, the intraoperative infusion volume, the anesthesia time, the surgical time, and the core temperature after intubation could accurately predict IOH in total joint arthroplasty patients. By monitoring these factors, the clinical staff could achieve early detection and intervention of IOH in total joint arthroplasty patients.

Development and validation of an intraoperative hypothermia nomograph model for patients undergoing video-assisted thoracoscopic lobectomy: a retrospective study

This study aimed to develop and internally validate a nomogram model for assessing the risk of intraoperative hypothermia in patients undergoing video-assisted thoracoscopic (VATS) lobectomy. This study is a retrospective study. A total of 530 patients who undergoing VATS lobectomy from January 2022 to December 2023 in a tertiary hospital in Wuhan were selected. Patients were divided into hypothermia group (n = 346) and non-hypothermia group (n = 184) according to whether hypothermia occurred during the operation. Lasso regression was used to screen the independent variables. Logistic regression was used to analyze the risk factors of hypothermia during operation, and a nomogram model was established. Bootstrap method was used to internally verify the nomogram model. Receiver operating characteristic (ROC) curve was used to evaluate the discrimination of the model. Calibration curve and Hosmer Lemeshow test were used to evaluate the accuracy of the model. Decision curve analysis (DCA) was used to evaluate the clinical utility of the model. Intraoperative hypothermia occurred in 346 of 530 patients undergoing VATS lobectomy (65.28%). Logistic regression analysis showed that age, serum total bilirubin, inhaled desflurane, anesthesia duration, intraoperative infusion volume, intraoperative blood loss and body mass index were risk factors for intraoperative hypothermia in patients undergoing VATS lobectomy (P < 0.05). The area under ROC curve was 0.757, 95% CI (0.714-0.799). The optimal cutoff value was 0.635, the sensitivity was 0.717, and the specificity was 0.658. These results suggested that the model was well discriminated. Calibration curve has shown that the actual values are generally in agreement with the predicted values. Hosmer-Lemeshow test showed that χ2 = 5.588, P = 0.693, indicating that the model has a good accuracy. The DCA results confirmed that the model had high clinical utility. The nomogram model constructed in this study showed good discrimination, accuracy and clinical utility in predicting patients with intraoperative hypothermia, which can provide reference for medical staff to screen high-risk of intraoperative hypothermia in patients undergoing VATS lobectomy.

Construction of risk prediction model for hypothermia during pancreaticoduodenectomy

Purpose

To investigate the factors influencing hypothermia during pancreaticoduodenectomy and establish and verify a prediction model.

Method

The clinical data of patients undergoing pancreaticoduodenectomy in Hunan People's Hospital between January 1, 2022 and October 15, 2022 were analysed. The patients were divided into a hypothermia group (n = 302) and a non-hypothermia group (n = 164) according to whether hypothermia occurred during surgery. A binary logistic regression model was used to analyse the independent risk factors for hypothermia in patients undergoing pancreaticoduodenectomy. A risk prediction model was established, and R software was used to plot a column graph. The predictive value of the model was evaluated using the receiver operating characteristic (ROC) curve.

Results

Among the 466 patients undergoing pancreaticoduodenectomy, 302 (64.81 %) had hypothermia, including 154 men and 148 women, with a median age of 58.6 (38-86) years. The binary logistic regression analysis showed that low body mass index (BMI), room temperature at the time of entry, intraoperative flushing fluid volume and peritoneal flushing fluid temperature were independent risk factors for intraoperative hypothermia in patients undergoing pancreaticoduodenal surgery (P < 0.05). A multivariate logistic regression analysis (backward logistic regression) was used to establish the prediction model. The area under the ROC curve was 0.927, P ≤ 0.001, the sensitivity was 0.921 and the specificity was 0.848, indicating good differentiation by the prediction model.

Conclusion

The nomogram constructed using four independent risk factors: BMI, room temperature at the time of entry, intraoperative peritoneal flushing fluid volume and intraoperative peritoneal flushing fluid temperature, has good predictive efficacy and good clinical application value for predicting intraoperative hypothermia in patients undergoing pancreaticoduodenectomy.

Construction and validation of a risk prediction model for intraoperative hypothermia in elderly patients undergoing total hip arthroplasty

AIMS

To construct and validate an intraoperative hypothermia risk prediction model for elderly patients undergoing total hip arthroplasty (THA).

METHODS

We collected data from 718 patients undergoing THA in a tertiary hospital from January 2021 to December 2022. Of these patients, 512 were assigned to the modeling group from January 2021 to April 2022, and 206 participants were assigned to the validation group from May 2022 to December 2022. A logistic regression analysis was performed to construct the model. The area under the curve (AUC) was used to test the model's predictive ability.

RESULTS

The incidence rate of intraoperative hypothermia was 51.67%. The risk factors entered into the risk prediction model were age, preoperative hemoglobin level, intraoperative blood loss, postoperative hemoglobin level, and postoperative systolic blood pressure. The model was constructed as follows: logit (P) = - 10.118 + 0.174 × age + 1.366 × 1 (preoperative hemoglobin level) + 0.555 × 1 (postoperative hemoglobin level) + 0.009 × 1 (intraoperative blood loss) + 0.066 × 1 (postoperative systolic blood pressure). Using the Hosmer-Lemeshow test, the P value was 0.676 (AUC, 0.867). The Youden index, sensitivity, and specificity were 0.602, 0.790, and 0.812, respectively. The incidence rates of intraoperative hypothermia in the modeling and validation groups were 53.15% and 48.06%, respectively. The correct practical application rate was 89.81%. This model had good application potential.

CONCLUSIONS

This risk prediction model has good predictive value and can accurately predict the occurrence of intraoperative hypothermia in patients who undergo THA, which provides reliable guidance for clinical work and has good clinical application value.

Use of the reliable change index to evaluate the effect of a multicomponent exercise program on physical functions

AIMS

Using the reliable change index (RCI), we aimed to examine the effect of a multicomponent exercise program on the individual level.

METHODS

Overall, 270 adults (mean age, 78 years) completed a multicomponent physical exercise program (strength, aerobic, gait, and balance) for 40 min, 1-2 times per week, continued up to 1 year at a daycare center. Effectiveness was assessed using grip, ankle, knee, and hip strength; Timed Up & Go (TUG); Berg Balance Scale (BBS); gait speed; and 6-min walking distance. These were measured at baseline and every 3 months thereafter. We calculated the RCI using the data between two-time points (baseline and at 3, 6, 9, or 12 months) in each participant and then calculated the mean RCI value across the participants. A paired t-test was also employed to evaluate the effect of the intervention as an average-based statistics.

RESULTS

The highest mean RCI values were on ankle plantar-flexion strength, followed by gait speed, hip abduction strength, BBS, knee extensor strength, 6-min walk distance, grip strength, and finally TUG. Paired t-test also revealed significant improvement with moderate effect sizes for ankle plantar-flexion strength (0.504), gait speed (0.413), hip abduction strength (0.374), BBS (0.334), knee extensor strength (0.264), and 6-min walk distance (0.248). Significant but small effect size was seen on TUG (0.183).

CONCLUSION

The RCI is a convenient method of comparing the effect between different assessments, especially at an individual level. This index can be applied to the use of personal feedback.