关键词: 急性呼吸窘迫综合征, 早期预警评分, 肺损伤预测评分, 肺炎严重指数, 风险预测模型, 护理

Abstract:

Objective To establish a predictive model for ICU medical staff to early predict the risk of occurrence of acute respiratory distress syndrome（ARDS）.Methods A total of 223 patients who were admitted to the ICU between January 2017 and April 2018 and met the criteria were enrolled in this case-control study. They were classified into an ARDS group（n=69） and a non-ARDS group（n=154）. The clinical data between the groups were compared. A risk prediction model was established by binary logistic regression,and the effectiveness of the model was verified by the area under the ROC curve.Results Age（OR=1.136）,lung injury prediction score（OR=2.328）,UK national early warning score（OR=2.830） and pneumonia severity index（OR=1.033） were included in this ARDS early risk prediction model. The model formula was Z=0.128 × age+0.845 × lung injury prediction score +1.040 × national early warning score+0.032 × pneumonia severity index -20.059. The area under the ROC curve of this model was 0.926 with a sensitivity of 0.906,a specificity of 0.812 and the Youden index of 0.718.Conclusion The risk prediction model described here can effectively predict the occurrence of ARDS in ICU patients,and all 4 indicators can be obtained in the early stage of ICU,thus providing references for ICU medical staff to take preventive care interventions for patients with different risk stratifications in the early stage.

Key words: Acute Respiratory Distress Syndrome, Early Warning Score, Lung Injury Prediction Score, Pneumonia Severity Index, Risk Prediction Model, Nursing Care