key: cord-1034981-wef2fn3w
authors: Fernandes, Michael; Brábek, Jan
title: COVID-19, corticosteroids and public health: a reappraisal
date: 2021-06-07
journal: Public Health
DOI: 10.1016/j.puhe.2021.05.028
sha: c3650e8be6c2591ab2cbcb7adda44b01f5520794
doc_id: 1034981
cord_uid: wef2fn3w

Objectives To assess whether regulatory guidance on the use of dexamethasone in hospitalised COVID-19 patients is applicable to the larger population of COVID-19 cases. The surge in worldwide demand for dexamethasone suggests that the guidance, although correct, has not emphasised the danger of its wider use. Study design Data from the Randomised Evaluation of COVID-19 Therapy (RECOVERY) trial and the World Health Organisation (WHO) prospective meta-analysis have been deconstructed and analysed. Methods In order to provide context, relevant publications were identified in PubMed using the following keywords: COVID-19, RECOVERY trial, WHO meta-analysis, variants, immunity, public health. Results The WHO guidance ‘Corticosteroids for COVID-19’ was based on their prospective meta-analysis. This meta-analysis was weighted by data from the RECOVERY trial. Conclusions In terms of COVID-19, dexamethasone has value in a narrow indication, namely, in hospitalised patients requiring respiratory support. The media blitz likely resulted in the wider use of dexamethasone in outpatients and as a preventive medication. This is reflected in the surge in worldwide demand for dexamethasone. We ask whether the use of dexamethasone, beyond regulatory indications, may be responsible for the recent increase in mortality? From the public health standpoint, the current guidance for use of dexamethasone in COVID-19 could benefit from clarification and the addition of a cautionary note.

There is no evidence that specific interventions can decrease mortality in acute respiratory distress syndrome (ARDS); therefore, the preliminary results of the Randomised Evaluation of COVID-19 Therapy (RECOVERY) trial, announced in June 2020, were both surprising and welcome. [1] [2] [3] [4] This trial was conducted in hospitalised COVID-19 patients and explored the effect of dexamethasone in the following three severity-based categories: (i) individuals receiving invasive mechanical ventilation (IMV); (ii) individuals receiving oxygen only; and (iii) individuals receiving no oxygen. The organisation and implementation of the RECOVERY trial was phenomenal 5 and upon completion, regulatory and policy action was prompt. In September 2020, the World Health Organisation (WHO), based on results from the RECOVERY trial 2 and its sponsored prospective meta-analysis 6 updated their guidance on the use of corticosteroid drugs in patients with COVID-19. 7 Here, we analyse the RECOVERY trial within the broader context of the natural history of COVID-19 disease and comment on whether the preliminary results are sufficient to formulate global policy. We identify several gaps in the evidence and suggest that policy formulation is deferred until the protocol-specified 180-day follow-up report is published. This would allow for efficacy to be assessed against adverse events in all population categories, especially the elderly, those with relevant co-morbidities and those with a weakened immune system; a 180-day safety report would represent an index of sustained benefit. In this commentary, we do not question the results of these trials, but focus on the interpretation of the analyses and the communication of a consistent message relative to global public health.

Relevant publications were identified in PubMed using the following keywords: COVID-19, RECOVERY trial, WHO meta-analysis, variants, steroids, public health. In order to allow for comparisons between RECOVERY and the WHO meta-analysis, published tables were J o u r n a l P r e -p r o o f deconstructed and analysed. Simple, comprehensive and uniform risk measures [8] [9] [10] were calculated to allow for an understanding of, and comparisons between, the trials.

In this randomised trial of 6425 patients, 2104 received dexamethasone 6 mg once per day for 10 days and 4321 received usual care. The 28-day mortality was calculated for the total study group, as well as subgroups of individuals who required IMV (n = 1007), oxygen only (n = 3883) and in those who did not require respiratory support (n = 1535 The organisation and implementation of the RECOVERY trial was phenomenal 5 and upon completion, regulatory and policy action was prompt. The results were communicated enthusiastically in the media and positioned as a breakthrough: dexamethasone is the first drug shown to save lives. 3 On 2 September 2020, and based on the preliminary report on the RECOVERY trial and related meta-analyses, the WHO endorsed the use of corticosteroids in cases of severe and critical COVID-19. 7 Dexamethasone reduced deaths by one-third in ventilated patients and by one-fifth in patients receiving oxygen only. However, there was a trend of potential harm of dexamethasone use in those patients who did not require respiratory support (the no-oxygen subgroup). Based on these results, one death could be prevented by J o u r n a l P r e -p r o o f dexamethasone treatment of around eight ventilated patients or around 25 patients requiring oxygen alone. 1 Chief investigator Martin Landray, in an interview with Science stated, "It's very, very rare that you announce results at lunchtime, and it becomes policy and practice by tea time, and probably starts to save lives by the weekend." 11 RECOVERY trial: advantages and limitations of a platform design RECOVERY, a platform trial, involved the following two interventions in hospitalised COVID-19 patients: (i) dexamethasone to all patients; and (ii) additional IMV in patients with severe disease. Platform trials that randomise patients with a homogenous and stable disease to a variety of single treatments is a valid and efficient method to explore benefit under uncertainty. 12 However, in an Intensive Care Unit (ICU) setting, the rapid dynamics of disease may require a severe subgroup to be exposed to more than one intervention. Accordingly, implementation of a platform trial in an ICU can evolve into a treatment trial. Here, interpretation of results is problematic on account of interactions between interventions: can outcomes be assigned to a single interventiondexamethasone, IMV or more prudently to the combination? 13 It is impossible to design a trial in human volunteers to assess a possible beneficial effect of dexamethasone in alleviating the adverse effects of IMV. However, Fernando Fonseca dos Reis and colleagues have demonstrated a beneficial effect of pre-treatment with dexamethasone in ventilator-induced lung injury (VILI) in Wistar rats. 14 The objective of a platform trial is to attribute outcomes to distinct and discrete interventions. This is a relevant concern since IMV can be complicated by a cytokine-related, hyperinflammatory lung injury (termed VILI) that is similar to COVID ARDS. In RECOVERY, both interventions relate to the trial endpoint, which is mortality via multiple organ dysfunction syndrome (MODS). 15, 16 Therefore, it is possible that the beneficial effect of dexamethasone in the severe IMV subgroup was related to its dampening impact on the effects on both viral and mechanical ventilation-induced inflammation, rather than the sole inhibition of a COVID-19 specific mechanism. In RECOVERY, dexamethasone did not show beneficial effects in patients who did not require IMV. 2 J o u r n a l P r e -p r o o f

The literature on this topic is both controversial and confusing. In ARDS, the administration of steroids within the first 72 h of mechanical ventilation is directed to dampen the hyperinflammatory response, as evidenced by an increase in ventilator-free days and lower mortality. Despite a rationale for the prolonged use of steroids in COVID-19, 25 the general experience is that they are ineffective in virus-induced ARDS. 26-28 Furthermore, steroids enhance viral replication, 29 delay viral clearance [30] [31] [32] [33] and may increase mortality. 34 For good reason, its use during active infection is generally discouraged. Huan Li and colleagues performed a metaanalysis to determine safety and efficacy of corticosteroids in severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), SARS-CoV and middle east respiratory syndrome coronavirus (MERS-CoV) infections. The use of corticosteroids delayed viral clearance and did not improve survival, but did reduce duration of hospital stay, ICU admission rate and/or use of IMV. 32 

Liu and colleagues at the Shanghai Jiao Tong University School of Medicine, Shanghai, China, analysed the outcome of corticosteroid treatment, mainly methyl prednisolone, in severe COVID-19 patients with ARDS (n = 409) compared with standard care (n = 365). The endpoint was 28-day all-cause mortality. For patients receiving standard care, mortality was 31% (113 of 365 patients) and for those receiving steroids, mortality was 44% (181 of 409 patients). The increase in mortality in patients receiving steroids was 13% (OR 1.77; 95% CI 1.31 to 2.38; p = 0.0002). 33 Patients with moderate-to-severe COVID-19 pneumonia are likely to benefit from moderate-dose corticosteroid treatment when administered relatively late in the disease course. 34 Prior to the RECOVERY trial, clinical evidence did not show any beneficial effects of J o u r n a l P r e -p r o o f corticosteroid treatment for COVID-19 lung disease. 35, 36 In viral pneumonia, there is a tendency for steroids to delay viral clearance and thereby increase residence time, but this is controversial.

COVID-19 is a progressive disease that primarily affects the lungs. About 85% of COVID-19 cases are asymptomatic, and it is estimated that 15% require hospitalisation and a smaller fraction need IMV. It is not possible to predict possible progression or lack thereof in individual COVID-19 cases. In those with serious progressive disease, hospitalisation is indicated and management is predicated on the need for oxygen or IMV. Although subgroups facilitate analysis, they are not distinct or stable. It should be noted that progression of the disease is a continuum and ranges from 'no oxygen required' to 'oxygen only' and 'IMV'. Type I and III interferons, the body's first line of antiviral defence, are cytokines that are secreted by host cells in response to viral infection and which block virus replication at several levels. [46] [47] In COVID-19, this response may be dampened by the early administration of glucocorticoids. [48] [49] [50] This, in part, may explain the role of a weakened and uncoordinated immune system in both the recent surge in mortality and the generation of variants. 51 

The RECOVERY trial demonstrated that dexamethasone decreased 28-day mortality in about one-third of hospitalised patients receiving IMV. Dexamethasone is about 25 times more potent than hydrocortisone. Steroids accelerate viral replication, delay viral clearance and predispose individuals to nosocomial infection. For good reason, its use during active infection is generally discouraged. Accordingly, a careful distinction should be made between early intervention in J o u r n a l P r e -p r o o f progressive disease and mass prevention, especially with an agent with a known safety liability.

Dexamethasone is risky in mild cases. 61 Unfortunately, these concerns have not received attention. Both the statement of the chief investigator, Peter Horby, "this treatment can be given to pretty much anyone" 3 and the guidance offered to primary care physicians to consider dexamethasone for home treatment 62 do not appear appropriate. This is important since the preliminary report on the RECOVERY trial, and the media blitz that followed, may have prompted a worldwide surge in demand for dexamethasone for outpatient use. [63] [64] [65] [66] Following the RECOVERY announcement, US drug suppliers struggled to keep up with the demand for dexamethasone. Group drug purchaser VIZIENT, which supplies medicines to about half of the hospitals in the US, saw a 610% increase in requests for dexamethasone. 67 It is unlikely that the narrow clinical indication (i.e. use limited to ICU patients on respiratory support) was the cause for this surge in demand.

According to Ralph Baric at the University of North Carolina in Chapel Hill, US, "in COVID-19 disease early administration of steroids can cause more harm than good because they may dampen the immune response before it has the virus at bay. The best time to start dexamethasone is when patients first need respiratory support." (The Economist, Technology Quarterly, March 27, 2021). Shane Crotty cautions that if steroids are prescribed too early "you could really shoot oneself in the foot because this might be somebody whose adaptive immune response is just getting going." 43

In September 2020, the WHO issued the guidance entitled 'Corticosteroids for COVID-19'. This guidance was prompted by the RECOVERY trial and supported by a WHO sponsored prospective meta-analysis. 6, 7 Their two recommendations were to use systemic corticosteroids in patients with severe and critical COVID-19, and to avoid corticosteroids in patients with nonsevere COVID-19.

The prospective meta-analysis pooled data from seven randomised clinical trials that evaluated the efficacy of corticosteroids in 1703 critically ill patients with COVID-19. Patients were assigned to steroids (dexamethasone, hydrocortisone or methylprednisolone) [n = 678] or to J o u r n a l P r e -p r o o f usual care or placebo (n = 1025). The primary outcome was 28-day all-cause mortality. There were 222 deaths among the 678 patients randomised to corticosteroids and 425 deaths among the 1025 patients randomised to usual care or placebo. This corresponds to an absolute mortality risk of 33% for patients receiving corticosteroids compared with 41% for patients receiving usual care or placebo (OR 0.7; 95% CI 0.56 to 0.84; p = 0.0003). The WHO meta-analysis relative to the RECOVERY data is deconstructed in table 2.

It can be seen that the RECOVERY data was a major contributor to the WHO meta-analysis (weight = 57%). Both hydrocortisone and methyl prednisolone were ineffectual. These trials were incomplete (underpowered) and although seeming to favour the use of steroids, did not We conclude that for patients receiving IMV, dexamethasone demonstrates efficacy and that corticosteroids other than dexamethasone are ineffective in COVID-19. We wait with anticipation for the follow-up report of the RECOVERY trial in order to assess the effect of age, obesity, cardiovascular disease, diabetes and hypertension on the incidence of death. 69, 70 In addition, a 90-day mortality assessment in RECOVERY would confirm sustained efficacy and help further the benefit-risk analysis. Carl Heneghan, director of the Centre for Evidence Based Medicine at the University of Oxford, UK, has suggested that a follow-up beyond 28 days and additional analyses would clarify whether dexamethasone could harm patients in the longer term. 65

The futurefrom re-purposed drugs to purposive science J o u r n a l P r e -p r o o f A recent editorial in The Lancet calls for an increase in research towards a broader range of therapies. 71 In this complex situation, generated by several inter-related mechanisms, it is not possible to assign success to the inhibition of a putative and primary causal process.

Misattribution of outcomes may have the effect of not recognising and funding epidemiology, public health and mechanistic research and development in COVID-19 ARDS. Furthermore, the COVID-19 model is mechanistically relevant to related, multi-causal, common and fatal conditions, such as septic shock. 72 Although morbidity and proximate cause of death is COVID-19, related to defined pulmonary and coagulation complications, it is a systemic disease. 73 Based on pathophysiology, a comprehensive research and development approach would necessitate a broad portfolio.

Unfortunately, the media blitz on steroids has resulted in a de-emphasis of related research in coprimary mechanisms, such as cytokine release, 74-77 the bradykinin-kallikrein system, 78-80 the complement cascade, 81-85 contact activation and coagulation 86, 87 and neutrophil extracellular traps (NETosis). [88] [89] [90] [91] The patterned response of the host reflects parallel and inter-related mechanisms. The initiating event is likely an interaction between the virus and endothelial elements in the blood vessels leading to immunothrombosis. 92, 93 

More than 95% of all trials in sepsis and ARDS fail to demonstrate a positive and reproducible mortality effect. 94 Armand Girbes and Harm-Jan de Grooth at the VU University Medical Center, Amsterdam, The Netherlands, point to the limitations of large trials with mortality endpoints in patients with sepsis and ARDS. 95 When patients with the same syndrome diagnosis do not share the same pathways that lead to death (the attributable risk), any therapy can only lead to small effects. Larger and more 'pragmatic' randomised trials are not the solution because they decrease diagnostic precision, the effect size and the probability of finding a beneficial effect. A logical approach is a focus on mechanistic research into the complexities of critical illness syndromes.

The success of dexamethasone in the treatment of serious COVID-19 patients receiving IMV has been an electrifying advance in therapeutics and we congratulate the RECOVERY investigators and await a follow-up report listing predisposing conditions, such as demographics (especially age and gender), relevant co-morbidities, concomitant medicines, adverse effects and the 90-day mortality data. 96, 97 This information would be of interest to an actionable audience, especially decision-makers in public health. 98 In the management of a serious disease, on a pandemic scale, and in real time, therapeutic 103 Thirty-five years ago, Geoffrey Rose explained that although individuals may not gain directly from population strategies, the beneficial effect of the 'population approach' for the present and for the future is enormous. 104 In the absence of simple and universally applied public health measures, especially vaccination, COVID-19 will remain with us and spreadthis virus knows no borders.

In closing, we believe that dexamethasone is of value in hospitalised COVID-19 patients receiving IMV. At this time, and pending the 180-day follow-up report on RECOVERY, the wide use of steroids for prevention and self-medication is discouraged. Regulators and policy makers in public health need access the detailed trial and follow-up data in order to update initial recommendations. 105 In an environment subject to media overdrive, simple, clear and evidencebacked messages trump (sic) 'U' turns in policies; restraint and caution are required. All things considered, COVID-19 is the prototypic stress test for science and especially public health. 106 J o u r n a l P r e -p r o o f 

Low-cost dexamethasone reduces death by up to one third in hospitalised patients with severe respiratory complications of COVID-19. Press release

Coronavirus breakthrough: dexamethasone is the first drug shown to save lives

Corticosteroids in COVID-19 ARDS: evidence and hope during the pandemic

How we accelerated clinical trials in the age of COVID-19

The WHO Rapid Evidence Appraisal for COVID-19 Therapies (REACT) Working Group

Association between administration of systemic corticosteroids and mortality among critically ill patients with COVID-19. A meta-analysis

Corticosteroids for COVID-19. World Health Organisation

Estimating risk ratios and risk differences: alternatives to Odds Ratios

Should we rely on trials with diseaserather than patient-oriented endpoints?

Proportions and their differences

trial is transforming COVID-19 treatment. Why haven't others delivered more results?

The RECOVERY Platform

Ventilator-induced lung injury and sepsis: two sides of the same coin

Pre-treatment with dexamethasone attenuates experimental ventilator-induced lung injury

Multiple system organ failure: is mechanical ventilation a contributing factor?

Glucocorticosteroids enhance replication of respiratory viruses: effect of adjuvant interferon

Effects of early corticosteroid treatment on plasma SARS-associated coronavirus RNA concentrations in adult patients

Systemic corticosteroid therapy may delay viral clearance in patients with Middle East Respiratory Syndrome coronavirus infection

Impact of corticosteroid therapy on outcomes of persons with SARS-CoV-2, SARS-CoV, or MERS-CoV infection: a systematic review and meta-analysis

Corticosteroid treatment in severe COVID-19 patients with acute respiratory distress syndrome

COVID-19 pneumonia, and acute respiratory distress syndrome

COVID-19: Stay cool toward corticosteroids

Clinical evidence does not support corticosteroid treatment for 2019-nCoV lung injury

Are the steroids a blanket solution for COVID-19? A systematic review and meta-analysis

Corticosteroid treatment of patients with coronavirus disease 2019 (COVID-19)

Adjuvant corticosteroid treatment in adults with influenza A (H7N9) viral pneumonia

The new role of steroids in viral infections

Molecular simulations suggest vitamins, retinoids and steroids as ligands of the free fatty acid pocket of the SARS-CoV-2 Spike protein

Innate immunity during SARS-CoV-2: evasion strategies and activation trigger hypoxia and vascular damage

Adaptive immunity to SARS-CoV-2 and COVID-19

An 'uncoordinated' immune response may explain why COVID-19 strikes some hard, particularly the elderly

Charting a coronavirus infection. The New York Times

Flawed interferon response spurs severe illness

Type I and Type III interferons -induction, signaling, evasion, and application to combat COVID-19

The type I interferon signaling pathway is a target for glucocorticoid inhibition

Glucocorticosteroids enhance replication of respiratory viruses: effect of adjuvant interferon

Glucocorticoids inhibit type I IFN beta signaling and the upregulation of CD73 in human lung

Persistence and evolution of SARS-CoV-2 in an immunocompromised host

Virus variants likely evolved inside people with weak immune systems. The New York Times

Genetic variants of SARS-CoV-2-what do they mean?

SARS-CoV-2 evolution during treatment of chronic infection

Editorial. COVID-19 evolution during the pandemic -implications of new SARS-CoV-2 variants on disease control and public health policies

Escape from neutralizing antibodies by SARS-CoV-2 spike protein variants

The great escape? SARS-CoV-2 variants evading neutralizing responses

Understanding variants of SARS-CoV-2

The coronavirus variants don't seem to be highly variable so far

New SARS-CoV-2 Variants -clinical, public health, and vaccine implications

Breakthrough drug for Covid-19 may be risky for mild cases. The New York Times

A recurrent question from a primary care physician: How should I treat my COVID-19 patients at home?

Dexamethasone demand soars after positive Covid-19 Study

US hospital orders soar for Covid-19 steroid treatment

Covid-19: Demand for dexamethasone surges as RECOVERY trial publishes preprint

Worrying situation regarding the use of dexamethasone for COVID-19

With dexamethasone's sudden COVID-19 blessing

Effect of dexamethasone on days alive and ventilator-free in patients with moderate or severe acute respiratory distress syndrome and COVID-19. The CoDEX Randomized Clinical Trial

Characteristics of and important lessons from the coronavirus disease 2019 (COVID-19) outbreak in China: summary of a report of 72314 cases from the Chinese Center for Disease Control and Prevention

Covid-19: risk factors for severe disease and death

Curing COVID-19

Sepsis and septic shock

A rampage through the body

Cytokine release syndrome in severe COVID-19

Raloxifene and Bazedoxifene could be promising candidates for preventing the COVID-19 related cytokine storm, ARDS and mortality

COVID-19: consider cytokine storm syndromes and immunosuppression

The dynamic changes in cytokine responses in COVID-19: a snapshot of the current state of knowledge

Kinins and Cytokines in COVID-19: a comprehensive pathophysiological approach

A mechanistic model and therapeutic interventions for COVID-19 involving a RAS-mediated bradykinin storm

Kallikrein-kinin blockade in patients with COVID-19 to prevent acute respiratory distress syndrome

Is the COVID-19 thrombotic catastrophe complementconnected?

Rationale for targeting complement in COVID-19

A Review: does complement or the contact system have a role in protection or pathogenesis of COVID-19?

Complement as a target in COVID-19?

COVID-19: Complement, Coagulation, and Collateral Damage

The coagulopathy, endotheliopathy, and vasculitis of COVID-19

The contact activation system as a potential therapeutic target in patients with COVID-19

Neutrophil extracellular traps contribute to immunothrombosis in COVID-19 acute respiratory distress syndrome

NETosis, complement, and coagulation: a triangular relationship

Targeting potential drivers of COVID-19: Neutrophil extracellular traps

A NET-thrombosis axis in COVID-19

Endothelial cell infection and endotheliitis in COVID-19

Endothelial dysfunction and immunothrombosis as key pathogenic mechanisms in COVID-19

Negative trials in critical care: why most research is probably wrong

Time to stop randomized and large pragmatic trials for intensive care medicine syndromes: the case of sepsis and acute respiratory distress syndrome

Predictors of adverse prognosis in COVID-19: a systematic review and meta-analysis

Corticosteroids in severe COVID-19: a critical view of the evidence

Making decisions in a COVID-19 world

Remdesivir in adults with severe COVID-19: a randomised, double-blind, placebo-controlled, multicentre trial

Remdesivir and COVID-19

Clinical trials of disease stages in COVID 19: complicated and often misinterpreted

Toning down the 2019-nCoV media hype-and restoring hope

Commentary: the prevention paradox in lay epidemiology--Rose revisited

Sick Individuals and Sick Populations

Dexamethasone in the management of covid-19

COVID-19: a stress test for trust in science

We thank the reviewers for their guidance.

Not required.

This work has not received any specific grant from funding agencies in the public, commercial or not-for-profit sectors.

None declared. The initial innate immune response is depicted in green while the later adaptive response consists of antibodies (orange) and T cells (blue). In the usual infection, the coordinated response results in a decrease in the viral load (purple). An uncoordinated and delayed immune response results in an increased and sustained viral load. The latter is likely related to a weak T-cell response. The period of severe COVID-19 clinical disease is shaded gray. Note: T cells refer to virus-specific CD4+ and CD8+ T cells and antibodies refer to virus-specific neutralising antibodies.

Highlights  The success of the RECOVERY trial on the beneficial effect of dexamethasone in COVID-19 patients is both surprising and welcome. However, back extrapolation of results in hospitalized patients on mechanical ventilation to unsupervised use in outpatients, and for protection, constitutes a serious public health concern.  This has been amplified by the WHO guidance based on their meta-analysis. In our de-construction of the prospective meta-analysis it is clear that the RECOVERY trial, by virtue of its size, drove the conclusion.  Regulatory guidances have been correct, but the media blitz has trumped medical advice. This is reflected by the global surge in demand for corticosteroids. In this context, corticosteroids have a safety liability. COVID-19 represents a stress test for science and public health. Under uncertainty, the balance between providing useful information and that which fuels inappropriate action, is delicate.