The Value of Pulmonary Bedside Ultrasound System in the Evaluation of Severity and Prognosis of Acute Lung Injury

Abstract

Objective. To evaluate the value of pulmonary bedside ultrasound system in the assessment of severity and prognosis of acute lung injury (ALI). Method. Seventy-two ALI patients in the intensive care unit (ICU) of our hospital from April 2019 to April 2021 were selected as subjects. The changes of lung ultrasound score (LUS) and parameters at D1, D2, and D3 after admission were analyzed (LUS, oxygenation index (PaO2/FiO2), Acute Physiology and Chronic Health Evaluation II (APACHE-II), and Sequential Organ Failure Assessment (SOFA) score). Pearson correlation analysis was used to assess the relationship between LUS and PaO2/FiO2, APACHE-II score, and SOFA score at D1, D2, and D3. Logistic regression analysis was used for influencing factors for the prognosis of ALI patients. Receiver operating characteristic (ROC) curve was used to analyze the predictive value of baseline LUS, PaO2/FiO2, APACHE-II score, and SOFA score for the prognosis of ALI patients. Result. LUSs at D1, D2 and D3 showed an increasing trend with the increase of disease severity ( $P<0.05$ ). From D1 to D3, LUS, PaO2/FiO2, APACHE-II score, and SOFA score showed a downward trend ( $P<0.05$ ). LUS was negatively correlated with PaO2/FiO2 at D1, D2, and D3 but positively correlated with APACHE-II score and SOFA score ( $P<0.05$ ). Logistic regression analysis showed that after controlling for age, PaO2 and PaCO2, an increase in baseline LUS, APACHE-II score, SOFA score, and a decrease in PaO2/FiO2 were independent risk factors for death at 28 d in ALI patients ( $P<0.05$ ). ROC curve showed that LUS, PaO2/FiO2, APACHE-II score, and SOFA score were combined to predict the prognosis of ALI patients with the highest AUC value of 0.920, corresponding sensitivity of 88.89%, and specificity of 95.56%. Conclusion. LUS can evaluate the change of pulmonary ventilation area in ALI patients, further evaluate the severity of the disease, and effectively predict the prognosis of patients.