Skip to main content

Table 2 Summary of key studies on corticosteroids for non-Covid-19 and Covid-19-related ARDS

From: Promises and challenges of personalized medicine to guide ARDS therapy

Studies

Date

Design

Participants

Interventions

Primary Outcome

Heterogeneity across subgroups

Comments

Non-COVID-19 ARDS studies

Steinberg et al. [94]

2006

Country: USA

Centers: 25

Placebo-controlled, double-blind

2 parallel groups

N = 180 (400 planned)

Adults

Persistent ARDS for 7 days at least and 28 days at most

Methylprednisolone bolus of 2 mg/kg followed by 0.5 mg/kg every 6 h for 14 days, 0.5 mg/kg every 12 h for 7 days, and then tapering of the dose

Comparator: placebo

60-day mortality rates 29.2 (95% CI 20.8–39.4) versus 28.6 (20.8–38.6), P = 1.00

Methylprednisolone was associated with + 27% absolute difference in mortality rate in patients randomized after 14 days from onset of ARDS

And + 26% in patients with low levels of procollagen peptide III in the bronchoalveolar lavage fluid

Sample size was changed from 400 to 180 owing to external information on baseline risk of death and low recruitment rate

Methylprednisolone was associated with significantly more mechanical ventilation free days, and more risk of acquired muscle weakness

Meduri et al. [95]

2007

Country: USA

Centers: 5

Placebo-controlled, double-blind

2 parallel groups

Sequential analyses

N = 91 (400 planed)

Adults

ARDS within 72 h of onset

Methylprednisolone bolus of 1 mg/kg followed by 1 mg/kg/24 h continuous infusion for 14 days, 0.5 mg/kg/24 for 7 days, 0.25 mg/kg/day for 4 days, 0.125 mg/kg/day for 3 days

Comparator: placebo

2:1 scheme

The RR for extubation of improvement in Lung Injury Score by 1 or more point at study day 7 was 1.96 (1.16–3.30) in favor of methylprednisolone

Interaction of responses to treatment and adrenal status by Synacthen test could not be performed owing to small sample size

36% of patients in the placebo group did received open label methylprednisolone

Tongyoo et al. [96]

2016

Country: Thailand

Centers 1

Placebo-controlled, double-blind

2 parallel groups

N = 206 (194 planned)

Adults

Septic shock and ARDS

Hydrocortisone, 200 mg/day in 4 bolus of 50 mg for 7 days

Comparator: placebo

The RR of dying at 28-day mortality was 0.82 (0.50–1.34) in favor of hydrocortisone

There was no interaction between response to treatment and subgroups based on age or severity of illness

Effects of corticosteroids were consistent across all secondary outcomes without increased in the risk of adverse reactions except for the risk of hyperglycemia

DEXA-ARDS [97]

2020

Country: Spain

Centers: 17

Open-label

2 parallel groups

N = 277 (314 planned)

Adults

Moderate to severe ARDS

Dexamethasone 20 mg bolus daily to day 5, followed by 10 mg daily to day 10

The mean number of ventilator-free days was 4.8 days [95% CI 2.57 to 7.03] higher in dexamethasone group than in controls

At 60-day the between-group difference in mortality was − 15.3%; − 25.9 to − 4.9 in favor of dexamethasone

–

Trial stopped prematurely for low recruitment rate

Benefits from corticosteroids were consistent across secondary outcomes

COVID-19 studies

RECOVERY [47]

2021

Country: UK Platform trial

Multicenter, randomized

Open-label

N = 6425

Adults

Suspected or confirmed

COVID-19

Hospitalized

Dexamethasone

6 mg/d orally or intravenously

For 5 to 10 days

Control: usual care

The age-adjusted rate ratio for 28-day mortality was 0.83; 95% CI 0.75–0.93

On invasive mechanical ventilation, the RR was, 0.64; 95% CI 0.51–0.81)

On oxygen without invasive mechanical ventilation, the RR was 0.82; 95% CI 0.72–0.94

No respiratory support at randomization, the RR was 1.19; 95% CI 0.92–1.55

Trial stopped prematurely for efficacy

CoDEX [98]

2020

Country: Brazil

Centers: 51

Open-label

N = 299 (350 planned)

Adults

Moderate to severe

ARDS

Onset < 48 h before randomization Invasive Mechanical ventilation

Probable or confirmed COVID-19

Dexamethasone

Intravenous bolus 20 mg/day for 5 days, then 10 mg/day for 5 days

Control: usual care

The mean ventilator-free days was 6.6 (95% CI 5.0–8.2) in dexamethasone group versus 4.0 (95% CI 2.9–5.4) in controls

The difference was 2.26; 95% CI 0.2–4.38, in favor of dexamethasone

The RR for 28-day mortality was 0.86; 95% CI 0.64–1.15)

There were no evidence for interaction between response to treatment and age, severity of illness, degree of hypoxia, duration of disease prior to randomization, or vasopressor-dependency

Trial was stopped prematurely following external information from the RECOVERY trial

CAPE COVID [99]

2020

Country: France

Centers: 28

Embedded, randomized

Double-blinded

Placebo controlled

N = 256 (290 planned)

Adults

Admitted to ICU or intermediate care unit

Oxygen ≥ 6 L/min

Probable or confirmed COVID-19

Hydrocortisone, continuous infusion for 8 days or 14 days

200 mg/day for 4 days or 7 days; 100 mg/day for 2 or 4 days, and 50 mg/day for 2 or 3 days

Comparator: Placebo

The OR for 21-day mortality was 0.46 95% CI 0.20–1.04

–

Trial was stopped prematurely following external information from the RECOVERY trial

REMAP-CAP [100]

2020

Country: Europe, USA, Canada, Australia, New Zealand, Saudi Arabia Platform trial

Centers, Platform trial

Open-label

Bayesian analyses

N = 403

ICU adults

High-flow nasal oxygen with FIO2 ≥ 0.4 at ≥ 30 L/min, noninvasive or invasive

Ventilatory support, or vasopressors

Probable or confirmed

COVID-19

Hydrocortisone intravenously fixed 7-day course of 50 mg or 100 mg every 6 h)

OR

A shock-dependent course of 50 mg every 6 h when shock was clinically evident)

Comparator: usual care

The median adjusted OR and Bayesian probability of superiority were 1.43 (95% CI 0.91–2.27) and 93% for fixed-dose hydrocortisone, respectively, and 1.22 (95% CI 0.76–1.94) and 80% for shock-dependent hydrocortisone compared with no hydrocortisone

–

Trial was stopped prematurely following external information from the RECOVERY trial

MetCOVID [101]

2020

Country: Brazil

Center, 1

Double-blinded

Placebo

2 parallel groups

N = 416 (416 planned)

ICU adults

suspicion of COVID-19, SpO2 ≤ 94% with room air, required supplementary oxygen, or required IMV

Methylprednisolone intravenously 0.5 mg/kg twice daily for 5 days

Comparator, Placebo

The OR for 28-day mortality was 0.92, 95% CI 0.67–1.28

There was no evidence for interaction between response to treatment and age, level of respiratory support, biomarkers of inflammation

Post hoc analysis suggested survival benefit from MP in patients of > 60 years old whereas younger patients may have increased risk of death with MP

GLUCOCOVID [102]

2021

Country: Spain

Centers, 5

Open-label

2 parallel groups

N = 64 (180 planned)

Adults

Suspicion of COVID-19, disease duration < 7 days

Moderate to severe ARDS

CRP > 15 mg/L

Or IL-6 > 20 pg/mL

Or D-dimer > 800 ng/mL, or ferritin > 1000 mg/dL

Methylprednisolone intravenous 40 mg bid for 3 days, then 20 mg bid for 3 days

Comparator: usual care

The age-adjusted RR for the composite of death, progression to ICU admission, or progression to NIV was 0.68, 95% CI 0.37–1.26, in favor of corticosteroids

There was no evidence for an interaction between treatment response and duration of symptoms prior to randomization

The study was stop prematurely following the release of the RECOVERY trial and the low recruitment rate