Point-of-care lung ultrasound predicts severe disease and death due to COVID-19: a prospective cohort study.

ABSTRACT

ObjectiveThe clinical utility of point-of-care lung ultrasound (LUS) for disease severity triage of hospitalized patients with COVID-19 is unclear. DesignProspective cohort study SettingA large tertiary care center in Maryland, USA between April 2020 to September 2021. PatientsHospitalized adults ([≥]18 years of age) with positive SARS-CoV-2 RT-PCR results. InterventionsNone. Measurements and Main ResultsAll patients were scanned using a standardized protocol including 12 lung zones and followed to determine clinical outcomes until hospital discharge and vital status at 28-days. Ultrasounds were independently reviewed for lung and pleural line artifacts and abnormalities, and the mean Lung Ultrasound Score (ranging from 0 to 3) across lung zones (mLUSS) was determined. The primary outcome was time to ICU-level care, defined as high flow oxygen, noninvasive, or mechanical ventilation, within 28-days of the initial ultrasound. Cox proportional hazards regression models adjusted for age and sex were fit for mLUSS and each ultrasound covariate. A total of 264 participants were enrolled in the study; the median age was 59 years and 114 (43.2) % of participants were female. The median mLUSS was 1 (interquartile range 0.5 to 1.3). Following enrollment, 29 (11.0%) participants went on to require ICU-level care and 14 (5.3%) subsequently died by 28 days. Each increase in mLUSS at enrollment was associated with disease progression to ICU-level care (aHR = 3.63; 95% CI 1.23 to 10.65) and 28-day mortality (aHR = 4.50; 95% CI 1.52 to 13.31). Pleural line abnormalities were independently associated with disease progression to ICU-level care (aHR = 18.86; CI 1.57 to 226.09). ConclusionsParticipants with a mLUSS [≥]1 or pleural line changes on LUS had an increased likelihood of subsequent requirement of high flow oxygen or greater. LUS is a promising tool for assessing risk of COVID-19 progression at the bedside.