Abstract:Objective To investigate the value of bedside ultrasound in the early diagnosis and dynamic assessment of ICU ventilator-associated pneumonia (VAP). Methods A total of 56 patients with clinically suspected ventilator-associated pneumonia underwent lungSultrasound (LUS) and chest CT and laboratory examinations. The clinical value of LUS in diagnosing VAP was analyzed. ROC (receiver operating characteristic) was used to analyze pulmonary ultrasound score, white blood cell count (WBC), procalcitonin (PCT), high-sensitivity C-reactive protein (hs-CRP), and clinical pulmonary infection scores(CPIS) at days 1, 3, 7, and 10 after treatment. Results Compared with the clinical follow-up diagnostic criteria, the sensitivity and specificity of bedside LUS in diagnosing VAP were 93% and 76.9%, respectively. The sensitivity and specificity of chest CT in diagnosing VAP were 95.3% and 84.6%, respectively. There was no significant difference in the sensitivity and specificity of chi-square test between LUS and CT for diagnosis of VAP (p>0.05). At the 1st, 3rd, 7th and 10th d after treatment, the pulmonary ultrasound score and WBC, PCT, hs-CRP infection index and CPIS score of VAP patients decreased with the prolongation of treatment time. Among them, the LUS score and inflammatory index were consist in increasing, but had a slow decrease. ROC analysis showed that the area under curve (AUC) of the LUS scoring at 1, 3, 7, and 10d after treatment was 0.918, 0.920, 0.900, and 0.900,respectively(P<0.05), which was consistent with CPIS; The 95% CI of the bedside LUS score at the 1<sub><sup>st</sup></sub> day (0.823 to 0.998) and the 95% CI of CPIS at the 3<sub><sup>rd</sup></sub> d (0.806 to 1) were the best value. Conclusions Pulmonary ultrasound for diagnosing ventilator-associated pneumonia has high accuracy, can dynamically observe lesion changes in real time. It is a non-invasive, reproducible and reliable method for early diagnosis and dynamic assessment of ICU ventilator-associated pneumonia, which is worthy of clinical application.