A chest-CT and clinical chemistry based flowchart for rapid COVID-19 triage at emergency departments – a multicenter study

  1. Steef Kurstjens
  2. Eva-Leonne Göttgens
  3. Bob Smit
  4. Bent Postma
  5. Carl Kluge
  6. Armando van der Horst
  7. Eva H.J. Lamboo
  8. Caroline M.M. Janssen – Te Slaa
  9. Robert Herpers
  10. Yvette C.M. Kluiters-de Hingh
  11. Arthur du Mée
  12. Marjan Veuger
  13. Rob van Marum
  14. Peter de Jager
  15. Marc G. L. M. Elisen
  16. Ron Kusters
  17. Martin Schuijt
  18. Marc H. M. Thelen

Reviewed by

Read the full article See related articles

Discuss this preprint

Start a discussion What are Sciety discussions?

Listed in

Log in to save this article

Abstract

Background

Due to the large number of patients with coronavirus disease 19 (COVID-19), rapid diagnosis at the emergency department (ED) is of critical importance. In this study we have developed a flowchart based on two well-known diagnostic methods: the ‘corona-score’ and the ‘CO-RADS’. This flowchart can be used in hospitals that use chest-CT, instead of chest X-ray, for COVID-19 suspected patients at the ED.

Methods

ED patients (n=1904) from the Jeroen Bosch Hospital, Amphia Hospital, HagaHospital, Elisabeth TweeSteden Hospital, Bernhoven Hospital and Slingeland Hospital were included. A laboratory-based ‘corona-score’, without radiology, called the ‘lab-corona-score’ was combined with a chest-CT based radiology scoring system (CO-RADS), to develop a flowchart. The performance was assessed by sensitivity/specificity analyses using the RT-PCR outcome or the physician’s final diagnosis as golden standard.

Results

Out of the 1904 patients, 611 (32.1%) patients tested positive for the SARS-CoV-2 virus. The lab-corona-score alone had an AUC of 0.86, a sensitivity of 87% and a specificity of 88% using cut-off values of 0-2 (negative) and 8-10 (positive). Of 255 patients, from the Amphia and Slingeland Hospitals, a CO-RADS score was determined. The flowchart, which combined the ‘CO-RADS’ with the ‘lab-corona-score’, was developed based on data from Slingeland Hospital (sensitivity 97%, specificity 96%). Hereafter, the performance of the flowchart was validated using an independent dataset from Amphia hospital, and reached a sensitivity of 98% and specificity of 93%. A decision could be made in 79% of the patients, which was correct in 95% of the cases.

Conclusion

This flowchart, based on radiology (CO-RADS) and clinical chemistry parameters (lab-corona-score), results in a rapid and accurate diagnosis of COVID-19 at the ED.

Article activity feed

  1. ScreenIT

    SciScore for 10.1101/2020.10.29.20218743: (What is this?)

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

    Table 1: Rigor

    Institutional Review Board StatementIRB: The execution of this retrospective observational study of patient records was approved by the regional Medical Research Ethics Committee (METC Brabant), which declared that the use of anonymized patient data is not subject to the regulations of the WMO (Dutch Medical Research Involving Human Subjects Act).
    Randomizationnot detected.
    Blindingnot detected.
    Power Analysisnot detected.
    Sex as a biological variablenot detected.

    Table 2: Resources

    Software and Algorithms
    SentencesResources
    Statistics: Data were analyzed using Excel 2010 (
    Excel
    suggested: None
    (Microsoft Corporation, VS) and SPSS (IBM, Version 25.0, VS).
    SPSS
    suggested: (SPSS, RRID:SCR_002865)

    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:
    This study also has a set of limitations. During the development of the flowchart, the medical diagnosis of the physician was used as the golden standard. However, in the validation of the flowchart, the results of the RT-PCR were used as golden standard. Because false-negative RT-PCR results as a result of sampling are not uncommon, it is possible that a number of COVID-19 patients have been incorrectly classified as negative. This could have led to an underestimation of the diagnostic performance of the flow chart. A second limitation is that the use of the flowchart requires the performance of a CT-scan for all suspected patients. If a hospital is unable to perform a CT-scan, they will not be able to use this flowchart. In that case, we would like to recommend the use of the original corona-score, with cut-off values of 0-4 and 11-14.(8) A third limitation might be that the diagnostic accuracy of the flowchart might be affected by geographical location and population characteristics, as this study was performed in Dutch hospitals only.(7) Therefore we recommend that hospitals perform a small local validation of the performance of the flowchart prior to implementation. Lastly, as of now we have not been able to whether the performance of the flowchart would be diminished when a large portion of the patients is infected by season flue (influenza). However, it is unlikely that influenza leads to such drastic changes in clinical chemistry parameters, such as ferritin and LD, a...

    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 10.1101/2020.10.29.20218743 on medRxiv