Lung ultrasound: a new approach for detection of severity of COVID-19 viral pneumonia and associated mortality.

BACKGROUND AND

objectivesLung ultrasound has established itself as a dependable bedside imaging modality, with demonstrated concordance with computed tomography in the assessment of pulmonary abnormalities. Prior studies have also validated its utility in identifying COVID-19-associated lung pathology. In this context, the present study aims to evaluate the correlation between lung ultrasound findings and clinical parameters in order to determine its effectiveness in assessing disease severity and predicting mortality among patients with COVID-19 infection.

methodsThis study enrolled hospitalized patients with microbiologically confirmed SARS-CoV-2 infection, who were clinically stratified into mild (absence of moderate or severe COVID-19 symptoms), moderate (respiratory rate ≥24/min or SpO₂ <94% on room air) and severe disease categories (respiratory rate ≥30/min, SpO₂ <90% on room air, or requirement for non-invasive/invasive mechanical ventilation). For systematic lung ultrasound evaluation, each lung was divided into 7 anatomical zones-comprising 2 anterior, 2 axillary and 3 posterior zones-resulting in a total of 14 assessed zones. The primary objective was to comprehensively evaluate the correlation between lung ultrasound profiles and established clinical parameters, including PF and SF ratios. Secondary objectives were to investigate the association of lung ultrasound findings with clinical disease severity and mortality, while also establishing optimal prognostic cut-off values for mortality prediction.

resultsAmong the total cohort of 60 patients, B-line abnormalities involving at least one lung zone were identified in 88.33% of cases. The total number of lung zones demonstrating ultrasonographic abnormalities (B/B' or C/C' profiles) showed a significant negative correlation with both PF ratio (ρ = - 0.7232, p < 0.0001) and SF ratio (ρ = - 0.6866, p < 0.0001). A simplified threshold of ≥10 abnormal lung zones demonstrated 71.43% sensitivity and 72.00% specificity in distinguishing severe from non-severe illness. The total B-line burden was significantly associated with mortality, with a cut-off of ≥35 B-lines yielding 65.38% sensitivity and 58.82% specificity for mortality prediction.

conclusionA greater extent of pulmonary abnormalities detected on lung ultrasound demonstrates significant correlation with clinical severity and patient outcomes in COVID-19 pneumonia. These findings support the potential role of lung ultrasound in future prospective studies as an effective and safer modality for triaging and prognosticating patients with viral pneumonia other than COVID-19 in any resource-limited setting.