Table 1.

Details for included studies

Author year Country	No of patients Steroid/control	Age Steroid/control, years	Inclusion criteria of ARDS	Initiation timing of steroid from ARDS onset Early: < 7 days Late: ≥ 7 days	Regimen of steroids	Comparison	Low tidal ventilation strategy	Main etiology of ARDS	Planned primary outcome
Bernard [7] 1987 USA	50/49	Mean ± SD 55 ± 14.1/56 ± 14	(i) Arterial blood gas oxygen partial pressure ≤ 70 mmHg or arterial blood oxygen partial pressure/alveolar oxygen partial pressure ≤ 0.3 with oxygen inhalation of 40% or more (ii) Bilateral noncardiogenic pulmonary edema (iii) A pulmonary artery wedge pressure ≤ 18 mmHg in the presence of a pulmonary artery catheter	Early	Methylprednisolone Day1: 30 mg/kg 4 times daily	Placebo (Manitol)	Study published before 1999 No data	Not detail	No
Meduria,b [8] 1998 USA	16/8	Mean ± SD 47 ± 15.6/51 ± 18.7	AECC criteria	Late	Methylprednisolone Loading: 2 mg/kg Day1-14: 0.5 Day15–21: 0.25 Day22–28: 0.125 Day29–30: 0.0625 Day31–32: 0.0313 mg/kg 4 times daily	Placebo (0.9% serine)	Study published before 1999 No data	Sepsis 63% (15/24)	No
Steinberg 2006 [9] USA	89/91	Mean ± SD 49.0 ± 19.0/49.2 ± 16.5	AECC criteria	Late	Methylprednisolone Loading: 2 mg/kg Day1–14: 0.5 mg 4 times daily Day15–21: 0.5 mg 2 times daily Tapering off over 4 days	Placebo (5% dextrose)	Yes	Pneumonia 42% (76/180) Sepsis 22% (40/180)	Mortality at day 60
Meduri [10] 2007 USA	63/28	Mean ± SD 50.1 ± 15.3/53.2 ± 15.3	AECC criteria	Early	Methylprednisolone Loading: 1 mg/kg Day1–14: 1 Day15–21: 0.5 Day22–25: 0.25 Day26–28: 0.125 mg/kg/day by continuous infusions	Placebo (0.9% serine)	Yes	Pneumonia 42% (38/91) Sepsis 16% (15/91)	1-point reduction in lung injury score
Liu [30] 2012 China	12/14	Mean ± SD 69.8 ± 14.9/55.9 ± 15.3	AECC criteria	Early	Hydrocortisone Day1–7: 100 mg 3 times daily	Placebo (0.9% serine)	Not described	Pneumonia 42.3% (11/26)	No
Tongyoo [11] 2016 Thailand	98/99	Mean ± SD 64.5 ± 17.3/64.3 ± 16.0	AECC/Berlin criteria (Included acute lung injury)	Early	Hydrocortisone Day1–7: 50 mg 6 times daily	Placebo (Not detail)	Yes	Pneumonia 51% (100/197)	No
Tomazini [12] 2020 Brazil	151/148	Mean ± SD 60.1 ± 15.8/62.7 ± 13.1	Berlin criteria as moderate-to-severe ARDS	Early	Dexamethasone Day1–5: 20 Day6–10: 10 mg once daily	No placebo	Yes	Pneumonia due to SARS CoV-2	Ventilator-free days to day 28
Villar [13] 2020a Spain	139/138	Mean ± SD 56 ± 14/58 ± 15	AECC/Berlin criteria as P/F ratio 200 or less	Early	Dexamethasone Day1–5: 20 Day6–10: 10 mg once daily	No placebo	Yes	Pneumonia 53% (147/277) Sepsis 24% (67/277)	Ventilator-free days to day 28
Villar c [31] 2020b Spain	7/12	Median (IQR) 62 (48–68)/60 (52–69)	Berlin criteria as moderate-to-severe ARDS	Early	Dexamethasone Day1–5: 20 Day6–10: 10 mg once daily	No placebo	Yes	Pneumonia due to SARS CoV-2	Mortality at day 60

ARDS acute respiratory distress syndrome, AECC American European Consensus Conference, IQR interquartile range, SD standard deviation, SARS CoV-2 severe acute respiratory syndrome coronavirus 2, ICU intensive care unit

^aMeduri [8] included two articles

^bWe got information by author contact in Meduri [8]. Meduri [8] REPLY: Data on ventilator-free days to 28 days; Intervention (n = 16); Mean ± SD; 12.4 ± 8.1 days, Control (n = 8); 4.0 ± 6.1 days, which referred to all patents randomized before [n = 14] or after [n = 2] day 14 of ARDS

^cWe got information by author contact in Villar [31]. Villar [31] REPLY: In our analysis of 19 patients, we did not analyze ventilator-free days. All patients were mechanically ventilated at the time of randomization. Mortality at 60 days after randomization was as follows: Dexamethasone group: two deaths from 7 enrolled patients; Control group: two deaths from 12 enrolled patients. Mortality was blinded for all groups. Only the study coordinator knew the number of deaths at 28 and 60 days. Regarding infections—Dexamethasone group: three patients out of seven developed pneumonia during the ICU stay. Control group: 7 patients out of 12 developed pneumonia during ICU stay