Analysis of Prognostic Risk Factors and Preliminary Model Construction of Extracorporeal Membrane Oxygenation in ICU Patients
SU Yanqin
REN Junhui
WANG Fangyuan
Abstract：Objective To analyze the risk factors of poor prognosis after extracorporeal membrane oxygenation(ECMO)therapy in intensive care unit(ICU)patients and construct a logistic regression prediction model for clinical prevention and treatment.Methods A total of 92 patients with ICU in Henan Chest Hospital from October 2021 to October 2023 were retrospectively analyzed.According to the prognosis within 28 days after ECMO treatment,they were divided into 36 patients with good prognosis and 56 patients with poor prognosis.The clinical data of the two groups were compared,and the influencing factors of poor prognosis were analyzed by logistic regression.Model 1 was established based on clinical data,and model 2 was established based on clinical data combined with biochemical indicators,and the predictive efficacy of the two models for poor prognosis was evaluated.Hosmer-Lemeshow was used to test the fitting degree of the model.Results Age,nosocomial pneumonia,hemorrhage,sequential organ faliure assessment(SOFA)score,acute physiology and chronic health evaluation Ⅱ(APACHE Ⅱ)score,blood lactic acid,time from ICU to ECMO and cardiac tropoin T(cTnT)level were independent risk factors for poor prognosis.Lymphocyte count,oxygen consumption,oxygen uptake rate and oxygen delivery were independent protective factors for poor prognosis(P＜0.05).The area under the curve(AUC),sensitivity and specificity of model 1 for predicting poor prognosis were 0.833,73.21％ and 83.33％,respectively.The AUC,sensitivity and specificity of model 2 for predicting poor prognosis were 0.950,91.07％ and 88.89％,respectively.The Hosmer-Lemeshow test showed that model 2 had good calibration ability.Conclusion Age,nosocomial pneumonia,hemorrhage,SOFA score,APACHE Ⅱ score,blood lactic acid,time from ICU to ECMO,and cTnT level are independent risk factors for poor prognosis after ECMO treatment in ICU patients.Lymphocyte count,oxygen consumption,oxygen uptake rate,and oxygen delivery are independent protective factors.The logistic regression prediction model was established according to the above influencing factors,and the model had better predictive efficacy for poor prognosis.Clinical intervention measures could be formulated according to the above model to reduce the occurrence of poor prognosis.
Keywords：severe diseaseICUextracorporeal membrane oxygenationprognosisforecastlogistic regression analysis
Publication Date：2024-10-31
Online Publishing Date：2026-09-12(First online date of this platform, not the publication date of the document)
Pages：6( 3678-3683 )