Abstract
PURPOSE: A reappraisal of the validity of the conclusions of systematic reviews (SRs) and meta-analyses related to corticosteroids use for the treatment of COVID-19. MATERIAL AND METHODS: An overview of SRs (umbrella review). The methodological quality of the SRs was assessed using tha AMSTAR-2 checklist; quality of the evidence was appraised following the GRADE approach. RESULTS: 35 SRs were included in this overview. Data were from 307 overlapping reports, based on 121 individual primary studies (25 randomized clinical trials (RCTs), 96 non-RCTs. In critically ill patients the use of steroids significantly reduced mortality compared to standard of care in 80% of the SRs, more often with moderate/high level of certainty; however, in patients not requiring oxygen supplementation the use of steroids increased the overall mortality in 2/3 of the comparisons. Clinical progression of diseases (need for mechanical ventilation, or for intensive care admission) was more commonly observed among controls compared to steroids recipients (in 9 out of 14 comparisons; certainty of evidence from very-low to moderate). The occurrence of adverse events was similar among steroids recipients and controls. Other outcomes (i.e., viral clearance, length of hospital stay) or issue related to optimal dose and type of steroids were addressed in a minority of SRs, with a high level of uncertainty, so that no definitive conclusions can be drawn. CONCLUSIONS: There is moderate certainty of evidence that corticosteroids reduce mortality and progression of disease in critically ill COVID-19 patients compared to standard of care, without increasing the occurrence of adverse events.