DESIGN: Systematic review and meta-analysis.
SETTING: Electronic search for randomized controlled trials and observational studies (MEDLINE, EMBASE, CENTRAL).
PARTICIPANTS: Hospitalized adults ≥ 18 years old who were SARS-CoV-2 PCR positive.
INTERVENTIONS: High-dose and low-dose corticosteroids.
MEASUREMENTS AND MAIN RESULTS: A total of twelve studies (n=2759 patients) were included in this review. The pooled analysis demonstrated no significant difference in mortality rate between the high-dose and low-dose corticosteroids groups (n=2632; OR: 1.07 [95%CI 0.67, 1.72], p=0.77, I2=76%, trial sequential analysis=inconclusive). No significant differences were observed in the incidence of intensive care unit (ICU) admission rate (n=1544; OR: 0.77[95%CI 0.43, 1.37], p=0.37, I2= 72%), duration of hospital stay (n=1615; MD: 0.53[95%CI -1.36, 2.41], p=0.58, I2=87%), respiratory support (n=1694; OR: 1.51[95%CI 0.77, 2.96], p=0.23, I2=84%), duration of mechanical ventilation (n=419; MD: -1.44[95%CI -4.27, 1.40], p=0.32, I2=93%), incidence of hyperglycemia (n=516, OR: 0.91[95%CI 0.58, 1.43], p=0.68, I2=0%) and infection rate (n=1485, OR: 0.86[95%CI 0.64, 1.16], p=0.33, I2=29%).
CONCLUSION: The meta-analysis demonstrated high-dose corticosteroids did not reduce mortality rate. However, high-dose corticosteroids did not pose higher risk of hyperglycemia and infection rate for COVID-19 patients. Due to the inconclusive trial sequential analysis, substantial heterogeneity and low level of evidence, future large-scale randomized clinical trials are warranted to improve the certainty of evidence for the use of high-dose compared to low-dose corticosteroids in COVID-19 patients.