Analysis of Risk Factors and Development of A Risk Prediction Model for Hypothermia in ICU Patients Undergoing Continuous Renal Replacement Therapy

JI Mingmin; KANG Dandan; ZHANG Xuan; HUANG Xihua; CHEN Xufeng

doi:10.3639/j.issn.2095-3097.2025.12.002

Abstract：Objective To analyze the risk factors for hypothermia in ICU patients undergoing continuous renal replacement therapy and to construct a risk prediction model.Methods A total of 152 ICU patients who received continuous renal replacement therapy at Jiangsu Provincial People's Hospital from March 2023 to March 2025 were selected and divided into a hypothermia group(45 cases)and a non-hypothermia group(107 cases)based on the occurrence of hypothermia.General clinical data were compared between the two groups.Multivariate Logistic stepwise regression analysis was used to identify influencing factors for hypothermia in ICU patients undergoing continuous renal replacement therapy.A nomogram model was constructed and validated.The predictive value of the nomogram model for hypothermia in these patients was assessed using ROC curve analysis.Results The hypothermia group had a higher proportion of patients aged≥60 years,more frequent use of continuous veno-venous hemodialysis,longer treatment duration(≥24 h),higher rates of mechanical ventilation and shock,higher Acute Physiology and Chronic Health Evaluation Ⅱ(APACHE-Ⅱ)scores,and higher blood lactate levels than the non-hypothermia group,while body mass index and baseline body temperature were lower in the hypothermia group(P<0.05).Multivariate Logistic stepwise regression analysis showed that age≥60 years,treatment duration≥24 h,and APACHE-Ⅱscore≥25.70 were risk factors for hypothermia in ICU patients undergoing continuous renal replacement therapy(OR=2.069,95%CI:1.468-2.915,P<0.01;OR=2.732,95%CI:1.778-4.196,P<0.01;OR=2.340,95%CI:1.600-3.422,P<0.01),while body mass index≥22.47 kg/m² and baseline body temperature≥37.20 ℃ were protective factors(OR=0.517,95%CI:0.376-0.712,P<0.01;OR=0.402,95%CI:0.271-0.596,P<0.01).The consistency index(C-index)of the nomogram model was 0.892,and the calibration curve closely aligned with the ideal curve.The area under the ROC curve for predicting hypothermia using the nomogram model was 0.902,with a sensitivity of 0.917,specificity of 0.784,and Youden index of 0.701.Conclusion The influencing factors for hypothermia in ICU patients undergoing continuous renal replacement therapy include age,body mass index,treatment duration,APACHE-Ⅱ score,and baseline body temperature.The constructed nomogram model has high predictive value for hypothermia in this patient population.