Corticosteroids in COVID-19: one size does not fit all.

To the Editor:

Edalatifard et al. [1], in a single-blind randomised controlled clinical trial, showed that methylprednisolone pulse administration (250 mg per day intravenously for 3 days) to hospitalised, not intubated patients with severe coronavirus disease 2019 (COVID-19), significantly improved survival and duration of recovery compared to standard care; the effect was more prominent in patients receiving noninvasive ventilation and reservoir mask support.

The dose of systemic corticosteroids used is remarkably higher compared with the one used in the RECOVERY trial [2], which was the first to demonstrate the beneficial survival effect of systemic corticosteroids in severely and critically ill COVID-19 patients requiring respiratory support (proportional reduction of mortality of one-fifth and one-third, respectively); the dose used in RECOVERY was 6 mg of dexamethasone administered for 10 days. In line with RECOVERY, the living World Health Organization (WHO) guideline on drugs for COVID-19 provided a strong recommendation for systemic corticosteroids in patients with severe and critical COVID-19, and a weak or conditional recommendation against systemic corticosteroids in patients with non-severe disease, i.e. in those not requiring respiratory support. Although dosing and optimal timing of drug initiation remains uncertain, a relatively low dose, equivalent to 6 mg of dexamethasone, is proposed. The guideline release was supported by a prospective meta-analysis of seven randomised clinical trials by the WHO Rapid Evidence Appraisal for COVID-19 Therapies Working Group, which also confirmed that administration of corticosteroids (in various types and doses) in critically ill patients with COVID-19 was associated with lower 28-day all-cause mortality, compared with usual care or placebo [3]. This analysis was not powered to assess the optimal dose and duration of treatment, leaving this question unanswered.

Even though the living WHO guideline proposes the use of low corticosteroid doses, the current study suggests that the use of pulses of high corticosteroid doses is efficacious and has an acceptable short-term safety profile. There is additional evidence that even higher doses would be worthy of consideration in selected cases of severe COVID-19. For example, a beneficial mortality effect of high-dose dexamethasone has been recently shown in moderate and severe acute respiratory distress syndrome, with no excess of adverse events compared to placebo [4]. Although diffuse alveolar damage has been described in severe COVID-19, there are also reports suggesting that organising pneumonia and acute fibrinous and organising pneumonia (OP/AFOP) represents the predominant pattern in pathological samples of some some COVID-19 patients. A systematic review of COVID-19 pathological findings revealed AFOP features in 26% of patients [5]. Moreover, computed tomography scans of COVID-19 patients often present OP features, including peripheral bilateral ground-glass opacifications with or without consolidation or intralobular lines and atoll signs [6].

OP/AFOP may be the consequence of viral alveolar epithelial injury, which generates immune responses at a later stage of the disease. Notably, dexamethasone is more beneficial in patients treated after the first week of COVID-19 illness, when inflammatory, rather than infectious, mechanisms are predominant. Since rapidly progressive and extensive OP require treatment with high doses of corticosteroids, we suggest that critically ill COVID-19 patients may benefit more from higher systemic corticosteroid doses, especially in the presence of OP/AFOP radiological features.

The use of high systemic corticosteroid doses in clinical practice is not without adverse events and its consequences in virus clearance is unknown, thus we suggest that the decision for higher corticosteroid doses in patients with severe COVID-19 should rely upon the recognition of OP/AFOP features in computed tomography scans by experienced respiratory physicians and radiologists.

This one-page PDF can be shared freely online.

Shareable PDF ERJ-00224-2021.Shareable