Abstract
Background: The effectiveness of corticosteroids in acute respiratory distress syndrome (ARDS) and COVID-19 still remains uncertain. Since ARDS is due to a hyperinflammatory response to a direct injury, we decided to perform a meta-analysis and an evaluation of robustness of randomised clinical trials (RCTs) investigating the impact of corticosteroids on mortality in ARDS in both COVID-19 and non-COVID-19 patients. We conducted a systematic search of the literature from inception up to 30 October 2020, using the MEDLINE database and the PubMed interface. We evaluated the fragility index (FI) of the included RCTs using a two-by-two contingency table and the p-value produced by the Fisher exact test; the fragility quotient (FQ) was calculated by dividing the FI score by the total sample size of the trial. Results: Thirteen RCTs were included in the analysis; five of them were conducted in COVID-19 ARDS, including 7692 patients, while 8 RCTS were performed in non-COVID ARDS with 1091 patients evaluated. Three out of eight RCTs in ARDS had a FI > 0 while 2 RCTs out of five in COVID-19 had FI > 0. The median of FI for ARDS was 0.625 (0.47) while the median of FQ was 0.03 (0.014). The median of FI for COVID-19 was 6 (2) while the median of FQ was 0.059 (0.055). In this systematic review, we found that FI and FQ of RCTs evaluating the use of corticosteroids in ARDS and COVID-19 were low.
Highlights
The effectiveness of corticosteroids in acute respiratory distress syndrome (ARDS) and COVID-19 still remains uncertain
We focused on randomised clinical trials (RCTs) regarding two major fields of critical care research such as ARDS and COVID-19 and applied this new statistical methodology to investigate the role of corticosteroids in these two clinical settings
Corticosteroids in COVID-19 ARDS were evaluated in 5 RCTs including 7692 patients [14,15,16,17,18], while 8 RCTs with 1091 patients evaluated the use of steroids in non-COVID19 ARDS [19,20,21,22,23,24,25,26]
Summary
The use of p value < 0.05 was introduced as an arbitrary threshold to declare the statistical significance [1]. Finding a p value < 0.05 implies that the null hypothesis (i.e., no difference in outcome between groups), should be rejected [2]. The FI represents the number of patients responsible for the statistical significance of a trial finding and it is an intuitive measure of the robustness of the RCTs [5]. We focused on RCTs regarding two major fields of critical care research such as ARDS and COVID-19 and applied this new statistical methodology to investigate the role of corticosteroids in these two clinical settings. Several RCTs showed a beneficial effect of corticosteroids on short-term mortality and a reduction in the need for mechanical ventilation in COVID19 ARDS but data are too sparse to draw any conclusions [9]. We decided to perform a meta-analysis and an evaluation of robustness of RCTs investigating the impact of corticosteroids on mortality in ARDS in both COVID-19 and non-COVID-19 patients
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
