This blog contains dr. Federica Fusina's translations of a selection of ventilab.it (a blog created by dr. Giuseppe Natalini and friends)'s posts. You can click here for more info. Happy reading!

Corticosteroids: a cure for COVID-19 or a cure for ARDS?

July 27, 2020

Original post written on 27th July 2020 by Giuseppe Natalini

Recently, a randomized clinical trial (RCT) known as RECOVERY, on the use of corticosteroid therapy in SARS-CoV-2 disease (COVID-19), was published. The abstract’s conclusions affirm that dexamethasone therapy reduces 28 day mortality in patients “receiving either invasive mechanical ventilation or oxygen alone at randomization but not among those receiving no respiratory support.” (1)

It’s important to maintain the critical approach typical of a scientific attitude even when facing an important RCT like this one, and to integrate the RCT’s results with the knowledge already available to us.

RCT limitations

There is no demonstration that a positive RCT result can give a definite proof of the efficacy of a certain therapy in a specific disease, in spite of the fideistic faith that evidence based medicine has in RCTs.

Recent history teaches us that the results of numerous RCTS on the care of critically ill patients (even those published on the most prestigious medical journals) can be denied after a few years. This has happened, for example, for glycemic control in ICU (2), early-goal directed therapy in septic shock to maintain an ScVO2 >70% (3), human activated C reactive protein in septic shock (4) which was later withdrawn from the market, cisatracurium infusion in initial phases of ARDS (5). Maybe some other examples will cross your mind, too.

There is nothing strange in an RCT being denied. First of all, the so-called evidence based medicine is, in truth, a probability based medicine. This means, to simplify, that every RCT has 1 probability every 20 to show as effective a therapy that, in truth, is not effective. There is an even higher probability that a negative RCT will show as ineffective a treatment that is effective: this happens 1 every 5 times.  These errors, inherent to statistical methods, are called “type I error” (alpha errors) and “type II errors” (beta errors).

Moreover, RCTs results are typically characterized by little external validity, meaning that they may not be applied to real life situations, and cannot be considered as necessarily valid for patients with different characteristics, treated in different contexts and in different times than the ones in which the RCT was developed. (6) 

This doesn’t mean that RCTs are worthless studies. We have to remember that RCT results are important, but cannot be accepted, with an act of faith, as definitive answers on a clinical query. They must be integrated with knowledge we gain from other types of studies. No perfect studies exist, and RCTs make no exception.

Corticosteroid therapy in COVID-19: data from RCTs

Let's go back to the results from the RECOVERY RCT, which was conducted in 176 hospitals in Great Britain. The study included patients with suspected or confirmed SARS-CoV-2 infection. In the end, 15% of the randomized patients did not have a confirmed diagnosis of SARS-CoV-2 infection, which, in presence of non cardiogenic respiratory failure with pulmonary infiltrates or ground glass appearance on chest X-ray (compatible with any type of pneumonia), was only suspected. In these cases, the diagnosis remained “a clinical one based on the opinion of the managing doctor”.
After the exclusion of 2000 patients (the majority of them because they were “not considered suitable for randomization to dexamethasone”), almost 6500 patients were randomized to receive either 6 mg of dexamethasone (oral or intravenous) for up to 10 days or “usual care”, which means they received the treatment normally used in those 176 United Kingdom hospitals.

The study was NOT blinded, so the treating physicians knew who was receiving dexamethasone and who was not: this is an important limitation for any RCT.

Dexamethasone therapy could be started at any time during hospital admission, and in any clinical condition: at randomization, 16% of patients were intubated, 60% were receiving oxygen therapy or non invasive ventilation (these two possibilities were considered as the same thing), and 24% were receiving no respiratory support (not even oxygen).

The global result was a 2,8% reduction in 28 day mortality in patients receiving dexamethasone (22,9% in those receiving dexamethasone versus 25,7% in those not receiving it, Figure 1, red square).

The main reduction in mortality was observed in patients who were receiving invasive mechanical ventilation: in this group, mortality for patients receiving dexamethasone was 29% versus 41% in the “usual care” group.

The efficacy of dexamethasone was a lot less (or absent) in patients with oxygen therapy/non invasive ventilation and in those with no respiratory support: there was no significative difference in mortality when they were taken together (21.7% with dexamethasone versus 22.7% with “usual care”, Figure 1, blue square) .

Figure 1


Control group: “usual care”

A question arises after reading this RCT: dexamethasone was more effective than what other treatment, exactly, in patients receiving invasive mechanical ventilation? The answer is that dexamethasone was more effective than “usual care”. Still, we have to ask: what was considered as “usual care” for the 1007 invasively mechanically ventilated patients in the 176 U.K. hospitals that participated in the RECOVERY trail (5-6 patients per hospital)? What were the criteria for intubation? Which ventilatory settings (tidal volume, driving pressure, PEEP, Plateau pressure) were applied? How often was prone positioning used? Did patients receive muscle relaxants? And so on.

It is well known that, even in the era of protective ventilation, 35% of ARDS patients receive an excessive tidal volume  (> 8 m/kg ideal body weight)  and 80% receive a tidal volume greater than 6 ml/kg ideal body weight (7). We also know that a high tidal volume, inappropriate PEEP levels or a stress index >1 cause an increase in inflammatory cytokines. (8,9) Could corticosteroid therapy be effective on this possible source of inflammation?

We know that prone positioning, which is protective for pulmonary stress (10–12), is often not used in patients with severe ARDS (13). How often was it used in RECOVERY patients?

As we saw at the beginning of our post, RCTs results are highly dependent on “usual  care”. This means that they depend on the clinical context in which the results were obtained, and by the combination with other treatments. The possibility to generalize any RCT findings to our clinical practice depends on these elements . If we do not know how RECOVERY patients were ventilated, we cannot know how much the results can be valid for us.

Are corticosteroids effective in COVID-19 or in ARDS? 

If corticosteroid therapy had really been effective in COVID-19, it should have reduced mortality in all patients groups, and not only in invasively mechanically ventilated patients. Why did it work well only in intubated patients and not so well in the others?

Those of us who have seen COVID-19 patients who need invasive mechanical ventilation know well that, at this stage, these patients have a bilateral pneumonia. That is to say, they have an ARDS (see 24/06/2012 post for ARDS diagnosis).

It seems logic to infer that RECOVERY results support the efficacy of corticosteroid therapy in ARDS. The result is not groundbreaking, since corticosteroid therapy was already recommended by the European Society of Intensive Care Medicine and the Society of Critical Care Medicine 2017 joint guidelines (14) and confirmed by a recent RCT in ARDS patients (15)

The value of RECOVERY is to dissolve any uncertainty of the use of steroids in COVID-19 patients when they develop ARDS. Corticosteroid therapy was initially discouraged in COVID-19 for fear of a reduced viral clearance (16). RECOVERY results confirm this fear in patients without a severe respiratory failure (15-20% mortality), but show that, when ARDS develops, benefits outweigh risks. Steroid use is therefore justified in ARDS secondary to COVID-19.


The results of the RECOVERY RCT show that steroid therapy is not effective in COVID-19 (if it was, it would have reduced mortality for every disease severity).

This study confirms that corticosteroid therapy is effective in ARDS, even when due to COVID-19.

The efficacy of dexamethasone is COVID-19 ARDS is valid when patients are treated according to RECOVERY’S “usual care”(intubation indications, ventilatory volumes and pressures, associated drugs…), which are not available to us (at least for now). These results might not be reproduced in hospitals with a “usual Care” from the participating UK hospitals.

As usual, a smile for all ventilab friends, and happy holidays! This year, we really deserve them



  1. The RECOVERY Collaborative Group. Dexamethasone in Hospitalized Patients with Covid-19 — Preliminary Report. N Engl J Med2020;NEJMoa2021436.doi:10.1056/NEJMoa2021436.

  2. van den Berghe, G, Wouters P, Weekers F, Verwaest C, Bruyninckx F, Schetz M, Vlasselaers D, Ferdinande P, Lauwers P, Bouillon R. 110801 Intensive Insulin Therapy in Critically Ill Patients. N Engl J Med 2001;345:1359–1367.

  3. Rivers E, Nguyen B, Havstad S, Ressler J, Muzzin A, Knoblich B, Peterson E, Tomlanovich M. Early Goal-Directed Therapy in the Treatment of Severe Sepsis and Septic Shock. N Engl J Med 2001;345:1368–1377.

  4. Bernard GR, Dhainaut J-F, Helterbrand JD. Efficacy and Safety of Recombinant Human Activated Protein C for Severe Sepsis. N Engl J Med 2001;344:699–709.

  5. Papazian L, Forel J, Gacouin A, Penot-Ragon C, Gilles P, Loundou A, Jaber S, Arnal J, Perez D, Seghboyan J, Constantin J, Courant P, Lefrant J, Claude G, Prat G, Morange S, Roch A. Neuromuscular Blockers in Early Acute Respiratory Distress Syndrome. N Engl J Med 2010;363:1107–1116.

  6. Frieden TR. Evidence for Health Decision Making — Beyond Randomized, Controlled Trials. In: Drazen JM, Harrington DP, McMurray JJV, Ware JH, Woodcock J, editors. N Engl J Med 2017;377:465–475.

  7. Bellani G, Laffey JG, Pham T, Fan E, Brochard L, Esteban A, Gattinoni L, van Haren F, Larsson A, McAuley DF, Ranieri M, Rubenfeld G, Thompson BT, Wrigge H, Slutsky AS, Pesenti A, for the LUNG SAFE Investigators and the ESICM Trials Group. Epidemiology, Patterns of Care, and Mortality for Patients With Acute Respiratory Distress Syndrome in Intensive Care Units in 50 Countries. JAMA 2016;315:788.

  8. Ranieri VM, Suter PM, Tortorella C, Tullio RD, Dayer JM, Brienza A, Bruno F, Slutsky AS. Effect of Mechanical Ventilation on Inflammatory Mediators in Patients with Acute Respiratory Distress Syndrome: A Randomized Controlled Trial: JAMA 2000;44:11–12.

  9. Terragni PP, Filippini C, Slutsky AS, Birocco A, Tenaglia T, Grasso S, Stripoli T, Pasero D, Urbino R, Fanelli V, Faggiano C, Mascia L, Ranieri VM. Accuracy of Plateau Pressure and Stress Index to Identify Injurious Ventilation in Patients with Acute Respiratory Distress Syndrome: Anesthesiology 2013;119:880–889.

  10. Mentzelopoulos SD. Prone position reduces lung stress and strain in severe acute respiratory distress syndrome. Eur Respir J 2005;25:534–544.

  11. Cornejo RA, Díaz JC, Tobar EA, Bruhn AR, Ramos CA, González RA, Repetto CA, Romero CM, Gálvez LR, Llanos O, Arellano DH, Neira WR, Díaz GA, Zamorano AJ, Pereira GL. Effects of Prone Positioning on Lung Protection in Patients with Acute Respiratory Distress Syndrome. Am J Respir Crit Care Med 2013;188:440–448.

  12. Galiatsou E, Kostanti E, Svarna E, Kitsakos A, Koulouras V, Efremidis SC, Nakos G. Prone Position Augments Recruitment and Prevents Alveolar Overinflation in Acute Lung Injury. Am J Respir Crit Care Med 2006;174:187–197.

  13. Guérin C, Gurjar M, Bellani G, Garcia-Olivares P, Roca O, Meertens JH, Maia PA, Becher T, Peterson J, Larsson A, Gurjar M, Hajjej Z, Kovari F, Assiri AH, Mainas E, Hasan MS, Morocho-Tutillo DR, Baboi L, Chrétien JM, François G, Ayzac L, Chen L, Brochard L, Mercat A, for the investigators of the APRONET Study Group, the REVA Network, the Réseau recherche de la Société Française d’Anesthésie-Réanimation (SFAR-recherche) and the ESICM Trials Group. A prospective international observational prevalence study on prone positioning of ARDS patients: the APRONET (ARDS Prone Position Network) study. Intensive Care Med 2018;44:22–37.

  14. Annane D, Pastores SM, Rochwerg B, Arlt W, Balk RA, Beishuizen A, Briegel J, Carcillo J, Christ-Crain M, Cooper MS, Marik PE, Meduri GU, Olsen KM, Rodgers SC, Russell JA. Guidelines for the Diagnosis and Management of Critical Illness-Related Corticosteroid Insufficiency (CIRCI) in Critically Ill Patients (Part I): Society of Critical Care Medicine (SCCM) and European Society of Intensive Care Medicine (ESICM) 2017. Crit Care Med 2017;45:11.

  15. Villar J, Ferrando C, Martínez D, Ambrós A, Muñoz T, Soler JA, Aguilar G, Alba F, González-Higueras E, Conesa LA, Martín-Rodríguez C, Díaz-Domínguez FJ, Serna-Grande P, Rivas R, Ferreres J, Belda J, Capilla L, Tallet A, Añón JM, Fernández RL, González-Martín JM, Aguilar G, Alba F, Álvarez J, Ambrós A, Añón JM, Asensio MJ, Belda J, Blanco J, et al. Dexamethasone treatment for the acute respiratory distress syndrome: a multicentre, randomised controlled trial. Lancet Respir Med2020;S2213260019304175.doi:10.1016/S2213-2600(19)30417-5.

  16. World Health Organization. Clinical management of severe acute respiratory infection (SARI) when COVID-19 disease is suspected. Interim guidance 13 March 2020. 2020;at <https://www.who.int/docs/default-source/coronaviruse/clinical-management-of-novel-cov.pd&gt;.

Read more ...