Cytokine prediction of mortality in COVID19 patients

Mathilda Mandel, Gil Harari, Michael Gurevich, Anat Achiron

Research output: Contribution to journalArticlepeer-review

Abstract

Coronavirus disease 2019 (COVID19) is a life-threatening infection with uncertain progression and outcome. Assessing the severity of the disease for worsening patients is of importance in making decisions related to supportive mechanical ventilation and aggressive treatments. This was a prospective, non-randomized study that included hospitalized patients diagnosed with COVID19. Pro-inflammatory cytokines were assessed during hospitalization, and we calculated a prediction paradigm for 30-day mortality based on the serum levels of interleukin1β (IL1β), interleukin6 (IL6), interleukin8 (IL8), and tumor necrosis factor alpha (TNFα) measured by next-generation ELISA. Data of 71 COVID19 patients, mean age 62 years, SD13.8, 50 males, 21 females, were analyzed. Twelve (16.9%) patients died within 7–39 days of their first COVID19 positive nasopharyngeal test. Levels of IL6 and TNFα were significantly higher in patients that did not survive. IL6 predicted mortality at the cut-off value of 163.4 pg/ml, with a sensitivity of 91.7% and specificity of 57.6%. Our findings demonstrate that IL6 expression is significant for the prediction of 30-day mortality in hospitalized COVID19 patients and, therefore, may assist in treatment decisions.

Original languageEnglish
Article number155190
JournalCytokine
Volume134
DOIs
StatePublished - Oct 2020
Externally publishedYes

Bibliographical note

Publisher Copyright:
© 2020 Elsevier Ltd

Keywords

  • Coronavirus disease 2019
  • Interleukin-1β
  • Interleukin-6
  • Interleukin-8
  • Mortality
  • Pro-inflammatory cytokines
  • Tumor necrosis factor alpha

ASJC Scopus subject areas

  • Immunology and Allergy
  • Immunology
  • Biochemistry
  • Hematology
  • Molecular Biology

Fingerprint

Dive into the research topics of 'Cytokine prediction of mortality in COVID19 patients'. Together they form a unique fingerprint.

Cite this