Well, after dismissing the observational trial of remdesivir for #COVID19 last evening. I wake to find a surprise that causes me to relook.
@EricTopol, who is an experienced and thoughtful scientist was impressed> https://twitter.com/EricTopol/status/1248699556591579138?s=20
What's more...
Another major thinker in cardiology @ProfDFrancis did not seem put off by Gilead's announcement that the next trial will recruit more patients and use a 7-pt ordinal endpoint.
He explains here https://twitter.com/ProfDFrancis/status/1248812803760193536
First the remdesivir NEJM. @EricTopol was drawn to the fact that more than have the patients on the drug were extubated and 3 of 4 pts on ECMO stopped receiving it. Notable seems a good word choice.
Those numbers are good if you look at historical controls which have shown more dismal rates of death while intubated in ICU. What's more while the drug did have adverse effects (liver and kidney) these might be expected in ICU patients and might have been seen in controls.
I regret calling the study utterly useless b/c there is a potential signal: a) pts did not seem to do worse, b) at least Gilead and the authors collected and published the data. c) compared to historical controls, there is potential, 4) the authors do call for RCTs
But my worry is that this report of compassionate use will be held out as more than it is. See. https://twitter.com/guyll/status/1248857631206379520
Of course that is not the authors' fault.
The reason I felt this study is not helpful stems mostly from the lack of control arm. Yes, while a youngish cohort of pts receiving the drug did seem to do OK, without an active control arm you cannot know if the drug *caused* the improvement.
...IOW the historically terrible ICU outcomes may not be a reliable comparator. Other problems: incomplete f/u: ≈1/3 did not have an ultimate disposition (some still on vents). There is also 8 of the 61 pts without data -- no accusations of bad stuff but that is 13% of total
Final: This report tells us we should have a proper RCT.
But @ProfDFrancis thread describing no placebo arm and non-blinding of doctors making the clinical decisions is problematic.
You can follow @drjohnm.
Tip: mention @twtextapp on a Twitter thread with the keyword “unroll” to get a link to it.

Latest Threads Unrolled: