Int J Clin Pract. 2021 Jul 26:e14645. doi: 10.1111/ijcp.14645. Online ahead of print.
BACKGROUND: Evidence-based recommendations on efficacy and safety of corticosteroids in acute respiratory distress syndrome (ARDS) remains a therapeutic challenge. Findings from several systematic reviews and meta-analyses are inconsistent. We aimed to assess the published meta-analyses through a systematic review approach and provide further insight into the current uncertainty.
METHODOLOGY: We followed the Preferred Reporting Items for Systematic Review (PRISMA) guidelines to establish the patients, intervention, control and outcome (PICO) for reviewing published meta-analyses. Data sources such as PubMed/MEDLINE, SCOPUS, Cochrane, and Google Scholar from inception to February 2021 were accessed. Prevention of ARDS, mortality, ventilator days, ICU stay and safety in terms of occurrence of adverse effects were the patient-related outcomes. The review also assessed meta-analysis design related outcomes which includes the quality of meta-analysis, factors contributing to the risk of bias, extent and sources of heterogeneity, publication bias and robustness of findings. AMSTAR-2 checklist assessed the quality of published meta-analyses.
RESULTS: A total of 18 meta-analyses were reviewed comprising a total of 38 primary studies and 3760 patients. 14 studies were in ARDS, 3 in community-acquired pneumonia, and 1 in critical care. The overall quality of meta-analyses was observed to be critically low to high. A non-significant risk of publication bias and non-significant level of heterogeneity was observed in the reviewed meta-analysis. Corticosteroid was significantly effective in preventing ARDS among CAP patients. Effect of corticosteroids on mortality observed to be still inconsistent, whereas significant improvement was observed with ICU and ventilator outcomes, compared to the control group. We observed a significant reduction of mortality in RCTs (RR: 0.78; 95%CI: 0.61 to 0.99) and duration of mechanical ventilation (MD: -4.75; 95%CI:-7.63 to -1.88); significant increase in ventilator-free days (MD: 6.03; 95%CI: 3.59 to 8.47) and ICU-free days (MD: 8.04; 95%CI: 2.70 to 13.38) in ARDS patients treated with corticosteroids compared to the control group.
CONCLUSION: The quality of included studies ranged from critically low to high demonstrating inconsistency in risk of bias. While older studies found no significant effect, recent meta-analyses of RCTs found a significant mortality reduction in the corticosteroid group with considerable levels of heterogeneity. The updated meta-analysis by our team found a significant reduction in mortality in the pooled estimation of RCTs, but not in cohort studies. Corticosteroid therapy was effective in terms of ICU and ventilator outcomes with minimal safety concerns. Future meta-analyses should be well executed with specific research questions and well performed with minimal risk of bias to produce good quality evidence.