Objectives

To assess the effects of corticosteroid therapy for patients with severe coronavirus disease 2019 (COVID-19).

Methods

We comprehensively searched articles published in the Cochrane Library, PubMed, Embase, China Biology Medicine (CBM), China National Knowledge Infrastructure (CNKI), Wanfang, and VIP databases from January 1, 2019, to March 20, 2021.

Results

A total of 6771 patients from eight prospective studies were included in our meta-analysis. The results showed that corticosteroid therapy was associated with lower mortality in severe COVID-19 (OR = 0.70, 95% CI = 0.54–0.92, P = 0.009; I² = 54.5%). Since the proportion of the RECOVERY (Randomized Evaluation of COVID-19 Therapy) trial included in the meta-analysis was as high as 71.88%, we removed it and recalculated the pooled OR. The results of the remaining seven studies still suggested such a survival benefit (OR = 0.65, 95% CI = 0.44–0.96, P = 0.030; I² = 59.8%). Furthermore, subgroup analysis suggested that the pooled OR of three studies using corticosteroids in the early stages of treatment was much lower (OR = 0.37, 95% CI = 0.25–0.57, P < 0.001; I² = 47.8%). However, after excluding the RECOVERY trial, the pooled OR of the remaining four studies with unspecific administration timing of corticosteroid therapy no longer supported this result (OR = 0.90, 95% CI = 0.69–1.17, P = 0.415; I² = 0.0%).

Conclusions

In this meta-analysis, evidence based on seven randomized controlled trials and one prospective cohort study indicates that corticosteroid therapy was associated with a reduction in the mortality of severe COVID-19, especially when administered at an earlier time.