Five RCTs (n=518) were included in the review. Sample sizes ranged from 24 to 180 patients. Methodological quality ranged between 11.5 and 14.5. Some trials reported differences in baseline characteristics. Two trials randomised patients after 14 days.
Short-term mortality was significantly reduced in patients who received glucocorticoid treatment (RR=0.76, 95% CI 0.62 to 0.93, p=0.007; five RCTs), but there was evidence of significant statistical heterogeneity (p=0.06). Subgroup analyses by study size showed similar significant reductions in the two small trials, but the reduction in mortality was no longer significant when the three larger trials were combined. Significant reductions in mortality were reported in patients treated before day 14 of ARDS in small trials (p=0.007) and large trials (p=0.02). Significant reductions in mortality were reported in patients who received prolonged prednisolone treatment of greater than one week duration, but randomised before 14 days (p=0.01). There was no evidence of statistical heterogeneity for subgroup analyses.
The number of mechanical ventilation-free days was significantly increased at 28 days in patients who received glucocorticoid treatment (WMD 4.42 days, 95% CI 2.93 to 5.90, p<0.001), but there was evidence of significant statistical heterogeneity (p=0.004). Subgroup analyses indicated a significant increase in the number of mechanical ventilation-free days in trials that investigated treatment (methylprednisolone) of greater than one week duration (WMD 5.59 days, 95% CI 3.49 to 7.68, p<0.001; three trials). There was no evidence of statistical heterogeneity.
Improvements in oxygenation and preventative measures to decrease complications were discussed in the review.