Use of glucocorticoids in patients with acute respiratory distress syndrome: a meta-analysis and trial sequential analysis

• Main Authors: Yazan Zayed, Mahmoud Barbarawi, Esraa Ismail, Varun Samji, Josiane Kerbage, Fatima Rizk, Mohammad Salih, Areeg Bala, Michele Obeid, Smit Deliwala, Sherry Demian, Ibrahim Al-Sanouri, Raju Reddy
• Format: Article
• Language: English
• Published: BMC 2020-06-01
• Series: Journal of Intensive Care
• Subjects: Glucocorticoids; Corticosteroids; Acute respiratory distress syndrome; ARDS; Meta-analysis
• Online Access: http://link.springer.com/article/10.1186/s40560-020-00464-1

Abstract Background Acute respiratory distress syndrome (ARDS) is a common and disabling disease with high rates of mortality and morbidity. The role of steroids in treating ARDS remains controversial. We aim to examine the evidence behind using glucocorticoids in the management of ARDS from the available studies. Methods We performed a literature review of major electronic databases for randomized controlled trials (RCTs) comparing glucocorticoids versus placebo in treating patients with ARDS. Our primary outcome was hospital mortality. Other outcomes included ICU mortality, number of ventilator-free days at day 28, incidence of nosocomial infections, and hyperglycemia. We performed a meta-analysis using a random effects model to calculate risk ratios (RR) and mean difference (MD) with their corresponding 95% confidence intervals (CI). A subsequent trial sequential analysis was performed to examine the strength of evidence and to guard against statistical type I and type II errors for our results. Results Eight RCTs were included in the final analysis totaling of 1091 patients, with a mean age of 57 ± 16, and 56.2% were male. In our pooled analysis, use of glucocorticoids was associated with a significant reduction in hospital mortality (RR 0.79; 95% CI 0.64–0.98; P = 0.03) and ICU mortality (RR 0.64; 95% CI 0.42–0.97; P = 0.04). Furthermore, glucocorticoid use was associated with an increased number of ventilator-free days at day 28 (MD 4.06 days; 95% CI 2.66–5.45; P < 0.01). Regarding adverse events, glucocorticoids use was not associated with an increased risk for nosocomial infections (RR 0.82; 95% CI 0.68–1.00; P = 0.05); however, it was associated with an increased risk of hyperglycemia (RR 1.11; 95% CI 1.01–1.24; P = 0.04). In our trial sequential analysis, the required diversity-adjusted information size (sample size = 2692 patients) was not reached, and the evidence was insufficient from the available RCTs. Conclusion Among patients with ARDS, use of glucocorticoids is associated with a significant reduction in mortality and duration of mechanical ventilation, without increased risk of hospital-acquired infections. However, based on a trial sequential analysis, these findings may be secondary to a false-positive (type I) error. Further studies are needed for a firm conclusion with guarding against possible statistical errors.