Thread by @nickmmark

It’s been a busy couple of nights in the ICU. Normally I’d be asleep right now, but I’m pumped to talk about #steroids in #COVID19.

A Brief Thread on #steroids in #COVID19.

(Spoiler alert: more compelling RCTs show that steroids work!)
1/

Early in the pandemic it was unclear if steroids would be beneficial or harmful.

There were studies that steroids could reduce mortality in ARDS (DEXA-ARDS), but also studies showing that they could prolong viral shedding in related viruses (MERS).

Clear equipoise for RCTs.
2/

This changed in June, when the #RECOVERY trial (bit.ly/2QKCSGJ) found an impressive mortality decrease in patients treated with #dexamethasone.

This large, well designed study almost immediately changed practice. But a few questions lingered..

https://twitter.com/nickmmark/status/1275187455117848576?s=20

3/

There have now been 7 studies of steroids in COVID including 3 published today in @JAMA_current. These studies & an accompanying meta-analysis paint a more complete picture:
*CAPE-COVID bit.ly/3lIAGO3
*CODEX bit.ly/31TrKO9
*RE-MAP-CAP bit.ly/334NO7V
4/

These studies differ in steroid (#hydrocortisone, #dexamethasone, #methylpred) & dose, population enrolled (ICU vs wards, intubated vs not), & primary endpoints (mortality, MV duration, organ failure, etc)

All were well done methodologically though several were stopped early.
5/

Individually the studies find several benefits to steroids in COVID
*ReMAP: reducing likelihood of organ failure
*CODEX: reducing time on ventilators by 2.2 days
*RECOVERY: reducing mortality by 3%
6/

Even more compelling is the excellent meta-analysis (including not yet published data) also in @JAMA_current.
bit.ly/2YZo4s4

They do a great job summarizing this (increasingly) complex literature & the conclusion is clear: steroids reduce 28 day mortality in COVID.
7/

As you can see, the studies used a range of doses. Let's converting to prednisone equivalents for an🍎to🍏comparison:
Dexamethasone 20mg/day = 133mg (higher)
Methylprednisone 80mg/day = 100mg (higher)
Hydrocortisone 200mg/day = 50mg (lower)
Dexamethasone 6mg/day = 40mg (lower)
8/

Interestingly, the data suggest that the choice of steroid isn’t critical!
* dexamethasone & hydrocortisone appear to have similar benefit (less data on methylprednisolone)
* lower doses of steroids seem to be (at least) as effective as higher doses.
9/

Also, as long as they are on oxygen, most COVID pt subgroups seem to benefit:
*Age, duration of symptoms, & ward vs ICU doesn’t seem to matter
*Patients on supplemental O2 & those on IMV both benefit

Note: #RECOVERY showed a possible increase in mortality in pts not on O2
10/

Oddly, COVID patients on vasopressors seemed to derive *less* benefit from steroids.
This is counter-intuitive: pt who were sicker & on IMV seemed to benefit so I don't think this is just illness severity.
(I'll think more about this post-hoc subgroup after some sleep...)
11/

🥡Points:
*Multiple RCTs show #steroids lower 28-day mortality (& confer other benefits) in people w/ #COVID19
*Choice of steroid (Dexamethasone≈Hydrocortisone) & dose (Low dose≈high dose) less critical
*Steroids are now standard of care in hypoxemic patients with COVID19
12/12