Use of corticosteroids in COVID-19 patients vs. acute respiratory distress syndrome of other etiologies: Are there any differences?

The disease resulting from infection by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), widely known as coronavirus disease 2019 (COVID19), has been classified as a global pandemic. It is characterized by respiratory compromise, which may include multisystemic involvement due to endothelial damage as well as activation of multiple thromboinflammatory mechanisms, leading to various serious clinical stages such as multiple organ failure and death due to Acute Respiratory Distress Syndrome (ARDS). In addition, the increased demand for health services, including the need for intensive care units and advanced vital resources such as mechanical ventilation or extracorporeal membrane oxygenation and trained personnel, poses serious challenges in choosing a suitable treatment. In the absence of a standard treatment, the clinical approach to those affected has focused on pharmacological management, particularly the use of antivirals and immunomodulators. According to the phases of the immunological response, the use of glucocorticoids has become widespread given its effect on the control of the inflammatory cascade from its origin (blockade of phospholipase A2) to its wide pharmacological availability and low cost. To date, the results have been controversial. At the beginning of the pandemic, these drugs were not recommended for use in COVID-19 patients. Nonetheless, current preliminary results of a clinical trial in the RECOVERY group show a decrease in mortality in patients receiving dexamethasone, raising the possibility of the systematic use of glucocorticoids for COVID-19 disease. The purpose of this review is to perform a rapid assessment of the evidence for and against the use of Address for Correspondence: Pedro Barrera-López* Clinical Studies and Clinical Epidemiology Division, Fundación Santa Fe de Bogotá, Bogotá, Colombia Email: pbarreramed@gmail.com


INTRODUCTION
The disease resulting from infection by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), widely known as coronavirus disease 2019 (COVID-19), has been classified as a global pandemic. It is characterized by respiratory compromise, which may include multisystemic involvement due to endothelial damage as well as activation of multiple thromboinflammatory mechanisms, 1,2 leading to various serious clinical stages such as multiple organ failure and death due to Acute Respiratory Distress Syndrome (ARDS). 1 In addition, the increased demand for health services, including the need for intensive care units and advanced vital resources such as mechanical ventilation or extracorporeal membrane oxygenation and trained personnel, poses serious challenges in choosing a suitable treatment. 3,4 In the absence of a standard treatment, the clinical approach to those affected has focused on pharmacological management, particularly the use of antivirals and immunomodulators. 5 According to the phases of the immunological response, 6 the use of glucocorticoids has become widespread given its effect on the control of the inflammatory cascade from its origin (blockade of phospholipase A2) to its wide pharmacological availability and low cost. To date, the results have been controversial. At the beginning of the pandemic, these drugs were not recommended for use in COVID-19 patients. 7 Nonetheless, current preliminary results of a clinical trial in the RECOVERY group show a decrease in mortality in patients receiving dexamethasone, 8 raising the possibility of the systematic use of glucocorticoids for COVID-19 disease. 9 The purpose of this review is to perform a rapid assessment of the evidence for and against the use of glucocorticoids to prevent progression to ARDS and related mortality in both COVID-19 and ARDS mediated by other pathologies.

METHODS
A systematic review was carried out in the "PubMed," "EMBASE," and "Cochrane" databases with the following MESH terms: Articles published in the English and Spanish languages were included. For articles related to COVID-19, the deadline was July 31, 2020, and the start date was December 2019. For the meta-analysis related to ARDS from other causes, the start date was 2017, and the search and review time was two months. To assess the effect on COVID-19 patients, only original studies and meta-analyses published to date for other ARDS etiologies were included. Each search yielded 235 and 34 publications, respectively. Studies with samples smaller than ten patients; studies that evaluated the use of glucocorticoids in patients with rheumatological disease, autoimmune disease, and transplant or neoplasm recipients; reviews focused on neonates; lung maturation studies; and one metaanalysis on ARDS and septic shock were excluded. Finally, 14 original articles and 4 meta-analyses were obtained for review ( Figure 1).

RESULTS
The main findings are summarized in the table. In general, a comparative analysis of the findings on the use of glucocorticoids for the treatment of COVID-19 and other causes of ARDS revealed great heterogeneity among treatment types using prednisone, methylprednisolone, dexamethasone, and hydrocor-tisone. Treatment schedules ranged from low doses to pulses of steroids in both groups, and for COVID-19, dexamethasone and methylprednisolone were used more frequently than for ARDS of other causes, where the most used were methylprednisolone and dexamethasone. Only the RECOVERY study specified the application protocol (table 1). Furthermore, it was evidenced that the effectiveness in reducing mortality, recovery from ARDS, or reduction in hospitalization time remained inconclusive, with high heterogeneity in the findings. Two observational studies in China (1326 patients), 10,11 described a greater use of glucocorticoids in seriously ill patients with an increase in mortality; even a meta-analysis by Lu et al. (May, 2020) and the WHO provisional guidelines of January 28, 2020 advised against its use. 12

DISCUSSION
Despite the heterogeneity in the findings described above, an understanding of the pathophysiology and evolution of COVID-19 disease, mainly the syndrome of innate release of cytokines mediated by the immune system (macrophages and proinflammatory granulocytes) through the production of TNF-a and IL-6 was achieved 6 ; in addition to recognition of the phases of SARS-CoV-2 infection, which initiates a response mainly mediated by "natural killer" cells and cytotoxic T lymphocytes to the secretion of type I interferon by infected cells. 5 Subsequently, a series of deleterious mechanisms are activated, leading to lung tissue infiltration and damage progressing to ARDS, and endothelial impairment. With the foregoing, the use of corticosteroids alone or concomitantly with  other immunomodulatory therapies for the management of severe forms of COVID-19 has been reconsidered and recommended, to reduce mortality in patients who have already developed lung damage. 13 The most recent systematic reviews and metaanalyses (Table 1), besides corroborating previous findings, demonstrate a decrease in the duration of mechanical ventilation compared with the usual management, without evidence of a significant increase in side effects. Only lower clearance of the virus was observed, but low certainty in the findings remained, given the heterogeneity of the studies. In other ARDS etiologies, marginal benefits of drug therapies have been shown over time. In the four meta-analyses that condensed 17 clinical trials and more than 1,500 patients, the results are equally heterogeneous; cases of reduction in mortality or duration of mechanical ventilation are specific cases that do not allow recommendations to be made for systematic application. In one of the meta-analyses, there is even evidence of a decrease in mortality at day 28 for all patients, 14 a finding that so far is the main argument for the use of dexamethasone in COVID-19 8 ; however, this finding it is not demonstrated in other settings. The findings of side effects of steroids in ARDS are similar, mainly the risk of superinfection, which is not shown to be persistently increased in all patients. Within the pathophysiology of COVID-19 disease, the progression to severe forms does not depend only on the infectious agent but also on the interaction with the host's immune system, in which the human leukocyte antigen (HLA) plays a central role. HLA represents one of the most polymorphic systems that participate in the immune response, and its various polymorphisms have been associated with worse outcomes both in COVID-19 6 and in other viral infections, such as Influenza AH1N1. 15 Therefore, we consider that glucocorticoid therapy for patients with COVID-19 should not be based solely on pulmonary involvement. Hence, the indication should be individualized according to the patient's immune response and immunophenotype, particularly in those with a proinflammatory state that can decline over the course of the disease, based on laboratory markers that have been associated with a worse outcome and a deleterious immune response, such as IL-6 and tumor necrosis factor alpha.

CONCLUSION
It is hypothesized that the demonstrated benefits of steroids in modulating the inflammatory response, slowing the progression to ARDS and associated mortality, both for COVID-19 and for ARDS due to other causes, will be enhanced if the type of immune response can be identified. Administration of a treatment aimed at the specific pathophysiological mechanism will also allow reduction of the possible adverse results associated with the use of these drugs. We understand that the limited access and costs associated with the collection and processing of immunological profiles or HLA subtypes compared with the benefits of the application of glucocorticoids limit the possibility of carrying out a targeted treatment according to the immunophenotype, but we believe that this is as area of potential research and development to optimize individualized therapies.

LIMITATIONS
The main limitations of the study were related to time, phrases and languages that could be analyzed, which restricted the criteria for article inclusion in the metaanalysis, because there are many other references of great importance in languages such as Italian and Mandarin. Likewise, the time interval covered by the review is an important limitation because the production time in COVID-19 is in constant growth, and it is not possible to keep reviews like this one completely up-todate. It is important to note that the objective was always to carry out a systematic review of the literature with a subjective comparison and analysis, and the application of a statistical analysis (meta-analysis) was not within the objectives and design of the study.

ADDENDUM
After the search window for inclusion of articles in the systematic review, a meta-analysis published on September 2, 2020 was found that evaluated the usefulness of steroids to reduce mortality in COVID-19. This included seven studies with 1703 patients, of which 678 patients received dexamethasone, hydrocortisone, or methylprednisolone, and 1025 patients received a placebo. The conclusion is that patients who receive glucocorticoids systematically presented a reduction in mortality at day 28, compared with the placebo. 16 In addition, on February 25, 2021, the RECOVERY group updated its results where the use of dexa- Use of corticosteroids in COVID-19 patients vs. acute respiratory distress syndrome of other etiologies Barrera-López et al. Use of corticosteroids in COVID-19 patients vs. acute respiratory distress syndrome of other etiologies Barrera-López et al.
methasone was associated with a lower mortality on day 28 among patients who required the use of invasive mechanical ventilation or supplemental oxygen, which was not the case for patients with noninvasive mechanical ventilation. 17 These new results continue to reinforce our recommendation that the benefits of steroids can be enhanced by applying targeted therapy to specific immunophenotypes.