A billion years ago (or so it seems), the RECOVERY trial demonstrated mortality benefit from dexamethasone 6 mg/day for up to 10 days in hypoxemic COVID patients. This rapidly became a standard treatment. It’s a fundamental therapy that has saved lives.
Nonetheless, questions linger about the optimal steroid dose. 6 mg/day dexamethasone is equivalent to 40 mg/day prednisone, which is a moderate dose of steroid. In comparison, the DEXA-ARDS trial (investigating ARDS of various etiologies) as well as the CODEX trial (investigating intubated COVID patients) revealed benefit from much higher doses of dexamethasone (20 mg/day for five days, followed by 10 mg/day for five days). Taken together, the RECOVERY, DEXA-ARDS, and CODEX trials suggested that an evidence-based dose of dexamethasone in COVID pneumonia might lie somewhere within the 6-20 mg/day range.
Meanwhile, several additional RCTs have detected benefit from the addition of baricitinib or tocilizumab on top of 6 mg/day dexamethasone. Although these studies don’t directly investigate steroid dose, they do reveal that 6 mg/day dexamethasone by itself does not provide an optimal degree of immunomodulation.
Now enter from stage left: The COVID STEROID-2 trial. This is a double-blind, multicenter RCT comparing dexamethasone 6 mg/day versus 12 mg/day in more severe COVID pneumonia (patients on high flow nasal cannula, noninvasive ventilation, or invasive ventilation). The primary endpoint (days alive without life support) favored dexamethasone 12 mg/day, but barely failed to meet statistical significance (p=0.07). Secondary endpoints likewise favored 12 mg/day dexamethasone, but not to a statistically significant degree.
This trial detected a 5.2% absolute reduction in 28-day mortality with higher doses of dexamethasone (p=0.1). In comparison, the original RECOVERY trial of dexamethasone 6 mg versus placebo detected a 2.8% absolute mortality reduction (p<0.001). So the mortality benefit in COVID STEROID-2 was actually larger than the mortality benefit in the RECOVERY trial. However, COVID STEROID-2 contained 971 patients, whereas RECOVERY contained 6,425 patients. This indicates that COVID STEROID-2 probably uncovered a clinically relevant mortality difference, but it is underpowered to unequivocally show this.
Some of the most important data from COVID STEROID-2 reflects the safety profile of higher-dose steroid. There was a lower rate of superinfections among patients treated with 12 mg dexamethasone. There was a trend towards increased gastrointestinal bleeding in patients treated with higher steroid doses, albeit with very low absolute risks (1.8% vs. 1%). Overall, this data is very reassuring. However, stress ulcer prophylaxis might be considered more broadly in these patients (especially given the perils of performing gastrointestinal endoscopy in patients with severe hypoxemia who aren’t intubated).
In isolation, COVID STEROID-2 isn’t strong enough evidence to change practice. However, we must interpret this study in the greater context of evidence (including DEXA-ARDS, CODEX, COVID-BARRIER, and the larger body of evidence regarding steroid use in critical respiratory failure). Specifically, we now have several studies which support the following concepts:
- 6 mg/day dexamethasone alone provides suboptimal immunomodulation for the patients with COVID pneumonia (especially for sicker COVID patients).
- Steroid doses in the range of 12 mg/day dexamethasone for 10 days are relatively safe (a dose equivalent to 80 mg/day prednisone).
Synthesizing all the available evidence and experience, a reasonable conclusion may be that currently 12 mg/day dexamethasone appears superior to 6 mg/day dexamethasone for sicker patients with COVID pneumonia. However, one size may not fit all patients. For some patients, 6 mg/day could make sense (e.g., a brittle diabetic with milder COVID pneumonia). For occasional patients with very severe disease and sky-high inflammatory markers, starting with 20 mg/day dexamethasone for a few days could be a rational approach.
Perhaps the most challenging question is what the optimal dose of steroid is for patients who are receiving additional immunomodulators (e.g., tocilizumab or baricitinib). The combination of these immunomodulators with higher doses of steroid has not yet been investigated. Thus, for most COVID patients who are receiving multiple immunomodulators, sticking to 6 mg/day dexamethasone seems reasonable.
- Dexamethasone 6 mg/day alone is likely suboptimal for most patients with severe COVID pneumonia. Dexamethasone 12 mg/day may provide a reasonable intensity of immunomodulation.
- The optimal steroid dose for patients receiving tocilizumab or baricitinib remains unclear. For now, it’s probably best to stick to a standard dose of 6 mg/day dexamethasone when constructing multi-agent immunomodulatory regimens.
- Higher steroid doses may carry an increased risk of stress ulceration, potentially arguing for broader use of stress ulcer prophylaxis.
Going further:
- Manuscript at JAMA.
- Amazing presentation and discussion hosted by Critical Care Reviews below. If you don't have time to watch the whole video, the editorial by Marion Campbell starting 50:37 is not to be missed.
- Further discussion of steroid dosing & immunomodulation in the IBCC chapter.
Opening photo by Ray Hennessy on Unsplash
- Pulmcrit wee: The cutoff razor - April 15, 2024
- PulmCrit Blogitorial – Use of ECGs for management of (sub)massive PE - March 24, 2024
- PulmCrit Wee: Propofol induced eyelid opening apraxia – the struggle is real - March 20, 2024
Prof Artie Kavanaugh highlights the pitfall of Steroid dosing, based on Eminence based medicine – https://rheumnow.com/blog/icymi-steroid-poker
hello,
This is such an amazing article, am so interested
in such articles to read. Thank you for sharing such an amazing article.
Source: http://www.mumbaicoworking.com/coworking-space-in-andheri-west/
Excellent summary. We have routinely been using higher-dose steroids since last year. Like you stated, clinical precedence has been established a while back. It is nice to have supportive and relevant data for justification.
I would caution drawing any firm conclusions from a direct comparison of the absolute reductions in 28-day mortality between RECOVERY and COVID-STEROID 2. RECOVERY included 1500+ patients on no oxygen at all, and those patients had a numerically higher rate of 28-day mortality in the dexamethasone group. This finding obviously impacts how we treat those patients and impacts the type of patients that we enroll in future steroid trials. The 2.8% absolute reduction in mortality in RECOVERY mentioned above includes those patients on no oxygen. It doesn’t seem appropriate to compare that number to the reduction in mortality in COVID-STEROID… Read more »
Hi Josh Farkas,
Thanks for your valuable information about covid vaccine.
Should BMI and/or weight play a role in dexamethasone dosing as well?