The efficacy of corticosteroids in acute respiratory distress syndrome (ARDS) has been a subject of controversy for decades ^[1][2][55,56]. In animal models of ARDS, corticosteroids decreased the expression of pro-inflammatory mediators in lung tissue, including TNF-a, IL-1a, IL-1b, IL-6 and IL-12 p40, and reduces lung injury through the reduction of oxygen radicals produced by neutrophils ^[3][4][57,58]. Beyond their anti-inflammatory effects during the acute phase of inflammation, corticosteroids also contributed to the resolution of inflammation, trough reprogramming effects on macrophages. Corticosteroids have been administered during two distinct phases of ARDS, during the early stage of ARDS when inflammation is expected to be most important and during late phase of ARDS, when lung fibrosis predominates. The biological and pathological characteristics of these two entities differ greatly, explaining the observed conflicting results in the effects of corticosteroids in these two distinct conditions ^[5][59].

The early phase of ARDS is characterized by major alveolar inflammation. Thus, corticosteroids, potent anti-inflammatory agents, are theoretically expected to be relevant treatment for ARDS. In practice, clinical trials found variably favorable, neutral or harmful effects of corticosteroids in ARDS.

In an ancillary analysis of a RCT focused on septic shock, Annane et al., found that 7-day treatment with low dose of steroids was significantly associated with better outcomes in septic shock associated with early septic ARDS in non-responders to short cosyntropin stimulation test ^[6][47]. In a trial of 24 ARDS patients, early corticosteroid treatment (methylprednisolone 2 mg/kg/d followed by progressive dose tapering over 32 days] was associated with a significant reduction in lung injury score (LIS) (p < 0.003 at 5 days) ^[7][60]. Similar findings were observed in a larger cohort (LIS 69.8% in placebo group vs. 35.7% in corticosteroids group; p = 0.02), with methylprednisolone 1 mg/kg/d (progressively tapered off over 28 days) ^[8][61]. In an Egyptian study, early administration of hydrocortisone in septic ARDS was associated with improved oxygenation parameters and LIS without achieving a survival benefit on day 28 ^[9][62]. Trial of short course of high dose corticosteroids (vs. placebo] found no evidence for improved 45-day mortality in adults with ARDS (60% vs. 63% p = 0.74) ^[10][63]. More recently, Villar et al., found that in ARDS, dexamethasone (20 mg IV daily between day 1 to 5, then 10 mg daily between day 6 to 10) compared to placebo, increased the number of ventilator-free days (between-group difference 4.8 days [95% CI 2.57 to 7.03]; p < 0.0001), and reduced mortality at day 60 (between-group difference −15.3% [−25.9 to −4.9]; p = 0.0047) ^[11][64].

Late-stage ARDS is characterized histologically by ongoing inflammation with fibroproliferation, presence of hyaline membranes, and persistent diffuse alveolar damage, leading to prolonged mechanical ventilation and a higher risk of death ^[12][21]. Meduri et al., reported in 9 ARDS patients with pulmonary fibrosis, that high dose of methylprednisolone may improve the LIS ^[13][65]. Wajanaponsan et al., found that low dose methylprednisolone administered >14 days after onset of ARDS was associated with increased mortality rates at 60 and 180 days ^[14][66]. The largest multicenter placebo-controlled trial, found no evidence for difference in 60-day mortality with corticosteroids initiated for late-stage ARDS (36% vs. 27% p = 0.26) ^[15][67].

Not all corticosteroids exhibit the same biological properties. The dose and type of corticosteroid may yield variable effects on patients’ outcomes. In a trial of 304 patients with sepsis, high doses of methylprednisolone led to numerically more patients with ARDS in corticosteroids vs. placebo (32% vs. 25% p = 0.1), fewer reversions of ARDS (31% vs. 61% p = 0.015), and a higher 14-day mortality (52% vs. 22% p = 0.04) ^[16][68]. In another ARDS trial, high doses of methylprednisolone [30 mg/kg every 6 h for 1 day] did not reduce mortality (p = 0.74) or reverse ARDS (p = 0.77) ^[10][63]. In another trial in patients with ARDS and critical illness related corticosteroids insufficiency, hydrocortisone administered 3 times a day (1 mg/kg/d) for seven days increased survival rates and reduced shock rate (5/12 vs. 10/14, p < 0.05), with no significant effect on 28-day mortality ^[17][69].

5. Adverse Events

The administration of corticosteroids may be associated with adverse events. In high-quality trials and meta-analyses in sepsis and in ARDS, indicate the main adverse events associated with corticosteroids may include neuromuscular weakness, gastrointestinal bleeding, hypernatremia and hyperglycemia ^[18][19][52,90]. A meta-analysis of 18 trials including 2826 ARDS patients, found no evidence for increased risk of muscular weakness: RR 0.85 95% CI [0.62 to 1.18] or gastrointestinal bleeding RR 1.20 95% CI [0.43 to 3.34], but increased risk of hyperglycemia RR 1.11 95% CI [1.01 to 1.23] (Table 1).

Author, Reference	Type	Sample Size	Study Population	Treatment	Results
Bernard et al. [10]	RCT, multicenter	99	ARDS as Partial pressure of oxygen ≤ 70 mm Hg on > 40% oxygen, PaO2/PAO2 ratio < 0.3, bilateral lung infiltrates, pulmonary artery wedge pressure ≤ 18 mm Hg	MPS 30 mg/kg IV 6 hourly for 24 h vs. placebo	PEP mortality MPS 30/50 (60%); Pl 31/49 (63.2) OR 0.75 [0.4 to 1.57] p = 0.74
Meduri et al. [7]	RCT multicenter	24	ARDS 1994 7 days of mechanical ventilation with an LIS of 2.5 or greater and less than a 1-point reduction from day 1 of ARDS, and no evidence of untreated infection.	MPS Loading dose of 2 mg/kg; then 2 mg/kg/d from day 1 to day 14, 1 mg/kg/d from day 15 to day 21, 0.5 mg/kg/d from day 22 to day 28, 0.25 mg/kg/d on days 29 and 30, 0.125 mg/kg/d on days 31 and 32. vs. placebo	PEP Lung injury and mortality day 10 MPS 1.7 [0.1]; Pl 3.0 [0.2]; p < 0.001 SEP: Mortality MPS 0/16 (0%); Pl 5/8 (62%) p = 0.002 Mortality in hospital MPS 2/16 (12.5%); Pl 5/8 (62.5%) OR 0.41 [0.06 to 99] p = 0.03
Steinberg et al., ARDSnetwork, [15]	RCT Multicenter	132/180	ARDS 1994 in early and late stage At least 7 days duration ARDS; p/F < 200 Intubated, mechanical ventilation	MPS Loading dose of 2 mg/kg of predicted body weight followed by 0.5 mg/kg 6 hourly for 14 days; 0.5 mg/kg 12 hourly for 7 days; and then tapering of the dose.	In early ARDS (7–13 d) PEP mortality at 60 days MPS (36%); Pl (27%) p = 0.26
Annane et al. [6]	post Hoc RCT	129/300 177 ARDS: 129 non responders, 48 responder	ARDS 1994 bilateral infiltrate on chest radiography, PaO2/FiO2 < 200 mm Hg and Pulmonary occlusion pressure ≤ 18 mm Hg or no clinical evidence of left atrial hypertension	HSHC 50 mg IV 6 hourly and 9-alpha fludrocortisone once a day for 7 days.	PEP: mortality at 28-day In the non-responder subgroup HSHC + FC 33/62 (53%); Pl 50/67 (75%) RR = 0.71; 95% CI [0.54 to 0.94] p = 0.013 OR = 0.35; 95% CI [0.15 to 0.82], p = 0.016). In the responder group No significant result HSHC + FC 16/23 (70%); PL 12/25 (48%) RR = 1.4; 95% CI [0.89 to 2.36] p = 0.130 OR = 2.29; 95% CI [0.49 to 10.64] p = 0.290
Meduri et al. [8]	RCT multicenter
On day 4, regarding medical decision based on predefined criteria, following administration for a total of 8 or 14 days
PEP mortality at day 28		HSHC 25 of 400 patients 6.2%; 95% CI, [3.9 to 8.6] vs. placebo 47 of 395 patients 11.9%; 95% CI, [8.7 to 15.1]		(Absolute difference, −5.6 percentage points; 95% CI, [−9.6 to −1.7];	p = 0.006).