Clinical correlation of lung ultrasound profiles in patients with COVID-19 infection

  1. Chitrakshi Nagpal
  2. Sanchit Kumar
  3. Naveet Wig
  4. Arvind Kumar
  5. Praful Pandey
  6. Manraj Singh
  7. Kapil Dev Soni
  8. Richa Aggarwal
  9. Anjan Trikha

Reviewed by

Read the full article

Listed in

Log in to save this article

Abstract

Background

Lung ultrasound is a popular point of care test that correlates well with computed tomography for lung pathologies. While previous studies have shown its ability to detect COVID-19 related lung pathology, we aimed to evaluate the utility of lung ultrasound in the triage and prognostication of COVID-19 patients by determining its ability to predict clinical severity and outcomes.

Methods

This was a prospective, cross-sectional, observational, single centre study done at JPNATC and AIIMS, New Delhi, India. Consenting eligible patients aged 18 years or more were included if hospitalised with microbiologically confirmed COVID-19 and classified as mild, moderate (respiratory rate >24/min OR SpO2<94% on room air) and severe COVID-19 (respiratory rate >30/min OR SpO2<90% on room air) at the time of enrolment. The lungs were systematically assessed with ultrasound after division into 14 zones (4 anteriorly, 4 axillary and 6 posteriorly). Clinical and laboratory parameters including arterial blood gas analysis at the time of evaluation were recorded. Patients were followed till death or discharge. The primary objective was to determine the correlation between clinical severity and lung ultrasound profiles (no. of A, B and C profiles, and the total number of areas involved). Secondary objectives included assessment of the correlation between lung ultrasound profiles and clinical outcomes and development of a statistical model incorporating ultrasound and clinical parameters to allow prediction of COVID-19 related severity and outcomes.

Findings

Between October 1, 2020, and January 31,2021, patients were screened for inclusion and total n=60 patients were evaluated and included in the final analysis. The most common abnormality seen were B lines, seen in at least one zone in n=53 (88.33%) of cases. A median of 9 (IQR: 5-12) zones of the 14 assessed had a B-profile. The total number of abnormal areas (zones with a B or C profile) correlated significantly with the PaO2/FiO2 ratio (ρ= −0.7232, p<0.0001) and SpO2/FiO2 ratio (ρ= −0.6866, p<0.0001), and differed significantly between mild and moderate vs severe cases (p=0.0026 mild vs moderate, p<0.0001 mild vs severe, p=0.0175 moderate vs severe). The total number of B lines were predictors of mortality (p=0.0188, OR 1.03, 95% CI 1.003-1.060). Statistical models that incorporated total number of B-lines, CRP and anticoagulation use could predict mortality (p=0.0124, pseudo R2=0.1740) with an AUC= 0.7682 (95% CI=0.6176-0.9188), and the total number of involved areas and LDH levels could distinguish severe disease from mild/moderate disease (p<0.0001, Pseudo R2=0.3822), AUC = 0.8743 (95% CI=0.7752-0.9733). A simplified cut off of ≥6 involved areas (of the 14 assessed) was 100% sensitive and 52% specific for differentiating severe disease from mild and moderate ones.

Interpretation

In patients with COVID-19, increasing involvement of the lungs as assessed by ultrasonography correlates significantly with clinical severity and outcomes. These findings may be utilized in future prospective studies to validate the use of lung ultrasound to triage and prognosticate patients with COVID-19 infection.

Author Approval

All authors have seen and approved the manuscript

Competing interests

There are no potential competing interests

Data availability Statement

All data referred to in the manuscript shall be provided when asked for.

Disclaimers

The authors have nothing to disclose

Funding statement

No funding source was involved.

Article activity feed

  1. ScreenIT

    SciScore for 10.1101/2021.04.05.21254935: (What is this?)

    Please note, not all rigor criteria are appropriate for all manuscripts.

    Table 1: Rigor

    Institutional Review Board StatementIRB: The study was reviewed and cleared by the Institute Ethics Committee (IEC).
    Consent: Informed consent was obtained from the participants/next of kin prior to enrolment.
    Randomizationnot detected.
    BlindingProfiles were defined for individual zones as per existing literature.6,7 The investigators were blind to any other imaging performed on the patients.
    Power Analysisnot detected.
    Sex as a biological variablenot detected.

    Table 2: Resources

    Software and Algorithms
    SentencesResources
    Analysis was performed using STATA v12.0.
    STATA
    suggested: (Stata, RRID:SCR_012763)

    Results from OddPub: We did not detect open data. We also did not detect open code. Researchers are encouraged to share open data when possible (see Nature blog).

    Results from LimitationRecognizer: We detected the following sentences addressing limitations in the study:
    Our study is, however, not without limitations. Our study did not include follow-up imaging of the enrolled patients. It was a single centre study with a limited sample size and the inter and intra-observer reliability was not assessed. Scans were performed at variable days after symptom onset, at different stages of the disease. Use of the curvilinear probe rather than a linear probe could have compromised detailed assessment of pleural line abnormalities. Therefore, larger prospective studies using lung ultrasound at different points of patient contact are needed to provide further details on its diagnostic and prognostic performance in different clinical settings in the management of COVID-19 patients.

    Results from TrialIdentifier: No clinical trial numbers were referenced.

    Results from Barzooka: We did not find any issues relating to the usage of bar graphs.

    Results from JetFighter: We did not find any issues relating to colormaps.

    Results from rtransparent:
    • Thank you for including a conflict of interest statement. Authors are encouraged to include this statement when submitting to a journal.
    • Thank you for including a funding statement. Authors are encouraged to include this statement when submitting to a journal.
    • No protocol registration statement was detected.

    About SciScore

    SciScore is an automated tool that is designed to assist expert reviewers by finding and presenting formulaic information scattered throughout a paper in a standard, easy to digest format. SciScore checks for the presence and correctness of RRIDs (research resource identifiers), and for rigor criteria such as sex and investigator blinding. For details on the theoretical underpinning of rigor criteria and the tools shown here, including references cited, please follow this link.

  2. Version published to 10.1101/2021.04.05.21254935v1 on medRxiv