A novel hyperinflammation clinical risk tool, HI5-NEWS2, predicts mortality following early dexamethasone use in an observational cohort of hospitalised COVID-19 patients

  1. Michael R Ardern-Jones
  2. Hang T.T. Phan
  3. Florina Borca
  4. Matt Stammers
  5. James Batchelor
  6. Isabel C. Reading
  7. Sophie V. Fletcher
  8. Trevor Smith
  9. Andrew S Duncombe

Reviewed by

Read the full article See related articles

Listed in

Log in to save this article

Abstract

Background

The success of early dexamethasone therapy for hospitalised COVID-19 cases in treatment of Sars-CoV-2 infection may predominantly reflect its anti-inflammatory action against a hyperinflammation (HI) response. It is likely that there is substantial heterogeneity in HI responses in COVID-19.

Methods

Blood CRP, ferritin, neutrophil, lymphocyte and platelet counts were scored to assess HI (HI5) and combined with a validated measure of generalised medical deterioration (NEWS2) before day 2. Our primary outcome was 28 day mortality from early treatment with dexamethasone stratified by HI5-NEWS2 status.

Findings

Of 1265 patients, high risk of HI (high HI5-NEWS2) (n=367, 29.0%) conferred a strikingly increased mortality (36.0% vs 7.8%; Age adjusted hazard ratio (aHR) 5.9; 95% CI 3.6-9.8, p<0.001) compared to the low risk group (n= 455, 36.0%). An intermediate risk group (n= 443, 35.0%) also showed significantly higher mortality than the low risk group (17.6% vs 7.8%), aHR 2.2, p=0.005). Early dexamethasone treatment conferred a 50.0% reduction in mortality in the high risk group (36.0% to 18.0%, aHR 0.56, p=0.007). The intermediate risk group showed a trend to reduction in mortality (17.8% to 10.3%, aHR 0.82, p=0.46) which was not observed in the low risk group (7.8% to 9.2%, aHR 1.4, p =0.31).

Interpretation

The HI5-NEWS2 measured at COVID-19 diagnosis, strongly predicts mortality at 28 days. Significant reduction in mortality with early dexamethasone treatment was only observed in the high risk group. Therefore, the HI5-NEWS2 score could be utilised to stratify randomised clinical trials to test whether intensified anti-inflammatory therapy would further benefit high risk patients and whether alternative approaches would benefit low risk groups. Considering its recognised morbidity, we suggest that early dexamethasone should not be routinely prescribed for HI5-NEWS2 low risk individuals with COVID-19 and clinicians should cautiously assess the risk benefit of this intervention.

Funding

No external funding.

Article activity feed

  1. ScreenIT

    SciScore for 10.1101/2021.06.16.21259011: (What is this?)

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

    Table 1: Rigor

    NIH rigor criteria are not applicable to paper type.

    Table 2: Resources

    Software and Algorithms
    SentencesResources
    Data handling and statistical analysis: Structured and semi-structured data was accrued from the trust integration engine using SQL Developer 4.2 queries and then cleaned/transformed using python 3.7 and associated libraries: numpy and pandas.
    python
    suggested: (IPython, RRID:SCR_001658)
    numpy
    suggested: (NumPy, RRID:SCR_008633)
    Analysis was performed using matplotlib, seaborn and scipy.
    matplotlib
    suggested: (MatPlotLib, RRID:SCR_008624)
    scipy
    suggested: (SciPy, RRID:SCR_008058)
    Statistical analysis was undertaken in python 3.7, R (RStudio Version 1.4.1106) and GraphPad, Prism (8.4.3).
    GraphPad
    suggested: (GraphPad Prism, RRID:SCR_002798)
    Prism
    suggested: (PRISM, RRID:SCR_005375)

    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:
    The nature of the rapid reporting during COVID-19 imbues various limitations which are relevant to retrospective analyses such as ours. However, whilst potential pitfalls of bias exist, we have minimised the most important forms of bias as follows: assessment was made of 100% of sequential patients admitted to our institution and the management of COVID-19 remained stable throughout the pandemic. By sampling a single institution, clinical teams and the facilities at the institution were the same for both groups. Furthermore, no differences between dexamethasone and untreated groups were identified including all key potential confounding parameters of ethnicity, sex, diabetes, heart disease, and respiratory disease. Indeed, in our whole cohort analysis the benefit from dexamethasone was small, suggesting that any bias in dexamethasone selection would be unlikely to explain the large subgroup differences associated with high or low risk HI5-NEWS2 status. The original randomised clinical trial from the RECOVERY Collaborative Group (1) showed mortality across all ages for those who received usual care of 25.7% versus 22.9% in those treated with dexamethasone. Our cohort excluded patients ≥85 years and so showed lower overall mortality: usual care was 18.4% (number at risk 612) versus 16.3% in dexamethasone treated groups (number at risk 653) although the trend was similar. Importantly, our results suggest that the benefit first identified by the Recovery trial for dexamethasone t...

    Results from TrialIdentifier: We found the following clinical trial numbers in your paper:

    IdentifierStatusTitle
    NCT04903834Active, not recruitingIdentification of Novel Factors Leading to Activated Macroph…

    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.

    Results from scite Reference Check: We found no unreliable references.

    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 10.1101/2021.06.16.21259011v1 on medRxiv