Lung ultrasound as diagnostic tool for SARS-CoV-2 infection.

2020 
The aim of this study was to explore the role of lung ultrasound (LUS) in the diagnosis of SARS-CoV-2 infection and to verify its utility in the prediction of lung disease’s severity and outcome. Fifty-three consecutive patients presenting to the Emergency Department of Santa Maria delle Grazie Hospital with high suspicion of SARS-CoV-2 infection underwent diagnostic test for SARS-CoV-2 on samples obtained from nasopharyngeal swab as well as complete proper diagnostic work-up that included clinical evaluation, laboratory tests, blood gas analyses, chest CT and LUS. A semiquantitative analysis of B-lines distribution was performed to calculate the LUS score. Patients were divided into two groups according to the results of both SARS-CoV-2 diagnostic test and other exams (Group A = pneumonia due to SARS-CoV2 infection vs Group B = no SARS-CoV2 infection and another definite diagnosis). LUS showed an excellent accuracy in predicting the diagnosis of SARS-CoV-2 infection (area under the ROC curve of 0.92 with a sensibility of 73% and a specificity of 89% a the cut-off of 12.5). LUS score was more impaired in SARS-CoV-2 patients (18.1 ± 6.0 vs 7.6 ± 5.9, p < 0.00001) and it is significantly negatively correlated with PF ratio values (r = − 0.719, p < 0.0001). An intrahospital mortality rate of 46% was found; patients with adverse outcome had significant higher value of LUS, PF, LDH, and APACHE II score. None of these parameters was predictive of mortality. LUS is a useful tool for the early detection of SARS-CoV-2 infection and for the evaluation of the disease severity, but does not predict mortality. Further studies with repeated evaluations of LUS score are needed to further explore the role of LUS in the assessment of severity in SARS-CoV-2 disease and in the monitoring of the response to treatments.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    18
    References
    11
    Citations
    NaN
    KQI
    []