High-Dose Corticosteroid Pulse Therapy Increases the Survival Rate in COVID-19 Patients at Risk of Cytokine Storm

Background: COVID-19 is a disease caused by SARS-CoV-2, which can develop into a hyper-inflammatory late phase or “cytokine storm” that poses high risk of death. It has been proposed that the management of this cytokine storm is crucial to increase the survival rate of COVID-19 patients. However, to date there are no diagnostic criteria based on inflammatory markers that could be used to select the COVID-19 patients at risk of developing a cytokine storm or effective interventions that can decreased mortality in these patients. Methods: We carried out a quasi-experimental study to test whether high dose corticosteroid pulse therapy with either dexametaxona or methylprednisolone were associated with reduced mortality in patients at risk of cytokine storm and assess which laboratory markers can be used to pre-select these patients. To control the development and progression of this hyper-inflammtory phase, 64 patients (20.1%) were treated with high dose corticosteroid pulse therapy (HDCPT) by using either dexametaxona or methylprednisolone daily for 2-5 days at doses of at least 30mg or 125mg respectively. To determine diagnostic criteria that can be used to defined patients at risk of cytokine storm, we carried out a 30-day time course analysis of laboratory markers between survivors and non-survivors. A multivariate Cox regression (controlling for co-morbidities and therapies) was carried out to determine whether HDCPT (among other interventions) was associated with decreased mortality in both the full cohort, and in a subgroup containing the patients with the derived criteria. Findings:  Out of 500 patients originally recruited, 381 met the inclusion criteria of SARS-COV-19 detection by PCR or serology (n=272, 71.2%) or high clinical suspicion (n= 46, 16.9 %), defined as having bilateral pulmonary infiltrate or lymphopenia with according clinical profile. HDCPT was the only intervention that showed a significant decrease in mortality (odds ratio = 0.093 [95% CI 0.023 – 0.38]; P = 40 pg/ml and/or two of the following: C-reactive protein >= 100 mg/L, D-dimer >= 1000 ng/ml, ferritin >= 500 ng/ml and lactate dehydrogenase >= 300 U/L. Multivariate Cox regression by pre-selecting the patients using this criteria evidenced also a decreased mortality rate with HDCPT with a lower odds ratio (odds ratio = 0.081 [95% CI 0.011 – 0.56]; P = 0.011). Conclusions: Preventing and/or controlling the development of the cytokine storm in patients at high-inflammatory risk with HDCPT is a widely available therapy to increase survival rate in these patients. We estimate a 14.2% [95% CI 0.792 - 0.975] increase of survival rate. We also suggest some initial clinical variables that can aid with the identification of the patients who will benefit from this intervention. Funding Statement: There were no funding sources in this study. Declaration of Interests: The authors declare no competing interests. Ethics Approval Statement: This study was approved by the Ethics Committee of the Complejo Hospitalario de Jaen (Hospital of Jaen), Spain (0946-N-20). According to the local ethics committee regulations, verbal consent was obtained from all the patients that joined the study.