Analysis of Prognostic Status and Influencing Factors of Viral Pneumonia during Mechanical Ventilation in Different Decubitus Positions and Con-struction of Prediction Model

LIU Juanjuan; YAO Na; ZHANG Tingting

doi:10.3969/j.issn.1002-3429.2024.07.006

Abstract：Objective To explore the prognosis and influencing factors of patients with viral pneumonia under me-chanical ventilation in different supine positions,and to construct a prediction model,to provide reference for improving the prognosis of patients with viral pneumonia.Methods A total of 363 patients with viral pneumonia under mechanical ventila-tion from January 2020 to May 2023 were selected to analyze the prognostic status of the patients with viral pneumonia under mechanical ventilation in different supine positions.According to 28-day prognosis,they were divided into poor prognosis group and good prognosis group.and the influencing factors of poor prognosis were analyzed.A nomogram prediction model for poor prognosis was constructed,followed by external validation using receiver operating characteristic(ROC)curve,decision curve analysis(DCA),and clinical impact curve(CIC).Results The poor 28-day prognosis rate of patients with viral pneumonia under mechanical ventilation in prone position was 17.19%(33/192),and the good prognosis rate was 82.81%(159/192).The 28-day prognosis rate of patients with virus pneumonia under mechanical ventilation in supine position was 28.65%(49/171),and the good prognosis rate was 71.35%(122/171).Multivariate Logistic regression analysis showed that etiology,body position under mechanical ventilation,mechanical ventilation duration,oxygenation index,Cdyn,compli-cated acute respiratory distress syndrome,LYM and SAA were independent influencing factors for poor prognosis of patients with viral pneumonia under mechanical ventilation(P＜0.01).Based on the independent influencing factors obtained by mul-tivariate Logistic regression analysis,a nomogram model for predicting poor prognosis of patients with viral pneumonia under mechanical ventilation was established.ROC curve analysis showed that the area under the ROC curve(AUC)of the model was 0.903(95%CI:0.868,0.938),DCA curve showed that the model had a good clinical net benefit,and CIC curve showed that the prediction model could effectively distinguish high-risk patients with poor prognosis under mechanical ventila-tion within the threshold probability range.Conclusion The prognosis of patients with viral pneumonia under mechanical ventilation in prone position is better than that of patients in supine position.The poor prognosis of patients with viral pneumo-nia under mechanical ventilation is affected by etiology,body position under mechanical ventilation,mechanical ventilation duration,oxygenation index,dynamic lung compliance,complicated acute respiratory distress syndrome,lymphocyte count and amyloid A.The prediction model constructed based on the above factors has high prediction efficiency and good clinical effectiveness.