Résumé
The main objective was to evaluate the precision of the Modified Lung Ultrasound (MLUS score) for predicting the need for respiratory support in the first 48 hours in patients with dyspnea due to the coronavirus disease 2019 (COVID-19). The secondary objectives were 1) to compare the MLUS and National Early Warning Score 2 (NEWS2), as well as the combination of both scores, as predictors of severity according to the World Health Organization (WHO) Ordinal Scale for Clinical Improvement; and 2) to compare severity assessed by ultrasound scoring to severity assessed by lung computed tomography (CT).
Multicenter prospective observational cohort study conducted from March 30 to April 30, 2020, in 2 university hospitals. Adult patients with dyspnea due to COVID-19 were included. An initial lung ultrasound was performed, and the results of MLUS, NEW2, and lung CT evaluations were recorded. Patients were classified by risk according to the WHO scale at 48 hours, as follows: low risk (score less than 5) or high risk (score of 5 or more).
A total of 100 patients were included: 35 (35%) were classified as low risk and 65 (65%) as high risk. The correlation between the MLUS and WHO assessments was positive and very high (Spearman rank correlation = 0.832; P .001). The area under the receiver operating characteristic curves of the MLUS, NEW2 and combined ultrasound scores, in relation to prediction of risk classification were, respectively, 0.96 (0.93-0.99), 0.89 (0.82-0.95) and 0.98 (0.96-1.0). The MLUS and lung CT assessments were correlated.
An early lung ultrasound score can predict clinical severity in patients with dyspnea due to COVID-19.