Article Text

Download PDFPDF

Correspondence on ‘Preliminary predictive criteria for COVID-19 cytokine storm’
Free
  1. Desiree Tampe1,
  2. Martin S Winkler2,
  3. Peter Korsten1,
  4. Samy Hakroush3,
  5. Onnen Moerer2,
  6. Björn Tampe1
  1. 1Department of Nephrology and Rheumatology, University Medical Center Göttingen, Göttingen, Germany
  2. 2Department of Anesthesiology, Emergency and Intensive Care Medicine, University Medical Center Göttingen, Göttingen, Germany
  3. 3Institute of Pathology, University Medical Center Göttingen, Göttingen, Germany
  1. Correspondence to Dr Björn Tampe, Department of Nephrology and Rheumatology, University Medical Center Göttingen, 37075 Göttingen, Germany; bjoern.tampe{at}med.uni-goettingen.de

Statistics from Altmetric.com

We read with great interest the recent article by Caricchio et al reporting preliminary predictive criteria for COVID-19 cytokine storm (COVID-CS).1 Early risk stratification for disease course and mortality is important especially in critically ill patients with severe COVID-19 to guide physicians in their evaluation to define patients at risk not surviving COVID-19 who may benefit from specific interventions.2 To develop a predictive model, the authors used univariate logistic regressions to identify variables and clusters associated with COVID-CS, resulting in defined optimal cut-off values.1 The model identified patients at risk for COVID-CS, associated with longer hospitalisation and increased mortality. While ferritin and C-reactive protein (CRP) did not add predictive power, these parameters were included in the final criteria per expert preference for clinical reassurance of ongoing systemic inflammation. Therefore, we here describe the performance of the COVID-CS predictive model for the requirement of intensive care unit (ICU) supportive care and mortality in a single-centre cohort of 20 critically ill patients with confirmed SARS-CoV-2 infection. In addition, we analysed the impact of including ferritin and CRP measurements for the usability of the COVID-CS model to predict disease severity and mortality in severe COVID-19.

Assessed at ICU admission, 7 of 20 (35%) patients fulfilled the criteria of the COVID-CS predictive model including ferritin and CRP. Risk classification using the COVID-CS predictive model assessed at ICU admission was associated with prolonged ICU length of stay before patients could be relocated to a non-ICU medical ward (HR 3.3, 95% CI 1.0–10.4; p=0.0428; figure 1A). In the group of patients who met the COVID-CS criteria, we observed 3 of 7 (42.9%) patients not surviving COVID-19 as compared with the group not fulfilling the COVID-CS criteria at ICU admission with 2 of 13 (15.4%) patients (HR 2.3, 95% CI 0.38–13.7; p=0.3698; figure 1B,C). Interestingly, excluding ferritin and C-reactive protein (CRP) measurements further improved the prediction of ICU stay of length (HR 4.0, 95% CI 1.3–12.2; p=0.0163) and mortality (HR 9.1, 95% CI 1.6–53.2; p=0.0141; figure 1D–F). In summary, we here confirm the accuracy of the model established by Caricchio et al to predict mortality in a single-centre cohort of severe COVID-19 patients requiring ICU supportive care. Additionally, the COVID-CS predictive model for disease severity and mortality assessed at ICU admission is superior when levels of ferritin and CRP are excluded.

Figure 1

(A) ICU stay of length in patients with severe COVID-19 grouped for COVID-CS. (B,C) Mortality and probability of survival analysis grouped for COVID-CS. (D) ICU stay of length in patients with severe COVID-19 grouped for COVID-CS without ferritin and CRP measurements. (E,F) Mortality and probability of survival analysis grouped for COVID-CS without ferritin and CRP measurements. COVID-CS, COVID-19 cytokine storm; CRP, C-reactive protein; ICU, intensive care unit.

The final model reported by Caricchio et al classified patients at risk for COVID-CS based on (1) documented COVID-19, (2) ferritin >250 ng/mL and CRP >4.6 mg/dL and (3) one feature from each cluster: cluster I (low albumin, low lymphocytes, high neutrophils), cluster II (elevated alanine aminotransferase, aspartate aminotransferase, D-dimers, lactate dehydrogenase, troponin I) and cluster 3 (low anion gap, high chloride, high potassium, high blood urea nitrogen:creatinine ratio).1 We here confirm that risk classification for mortality using the COVID-CS predictive model assessed at ICU admission also identifies patients at risk not surviving severe COVID-19 requiring ICU supportive care. Furthermore, the COVID-CS predictive model for disease severity and mortality when assessed at ICU admission is superior by exclusion levels of ferritin and CRP. This is in line with the study by Caricchio et al reporting that ferritin and CRP did not add predictive power.1 In conclusion, we here confirm that risk stratification for COVID-CS is also predictive in an independent cohort of critically ill patients with severe COVID-19, highlighting the relevance of hyperinflammation and tissue damage associated with disease severity and COVID-19 mortality as reported previously.3–6

References

Footnotes

  • Twitter @pekor002

  • Contributors DT and BT conceived the letter, collected and analysed data, and cowrote the first draft. MSW, PK, SH, PK and OM participated in the construction and editing of the manuscript.

  • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

  • Competing interests None declared.

  • Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.

  • Patient consent for publication Not required.

  • Ethics approval This study was approved by the institutional review board (IRB) of the University Medical Centre Göttingen, Germany (reference number 25/4/19Ü) and informed written consent was obtained. All data were anonymized to comply with the provisions of personal data protection legislation.

  • Provenance and peer review Not commissioned; internally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Linked Articles