Thread by @buzzhollandermd

A few patients have asked me to look into the research on ivermectin for Covid-19.
You know, that video of congressional testimony and all, from that Pulmonologist.
IVERMECTIN SOUNDS FABULOUS! 1/20

Now, when I see a gray-haired, highly-confident, white dude speaking strongly on a matter outside his own field of expertise, I get this comfortable feeling:
"That is MY demographic!"
But when he goes on about his own credentials;
and calls a re-purposed med "miraculous"... 2/20

I get that skeptical feeling.
But - I'll try to put that aside, and just look at their data.
It's a lot to review.
These folks, the FRONT LINE COVID-19 CRITICAL CARE ALLIANCE, put out a 33 pg document summarizing the research on ivermectin and covid. 3/20

covid19criticalcare.com/wp-content/upl…

The gist is: ivermectin prevents transmission of covid; it helps prevent mild cases from becoming serious; it helps serious cases not become fatal; it even helps prevent long-haul covid. IT DOES EVERYTHING!
No viral therapeutic could realistically do all that.
Strike one. 4/20

Now, as to their data.
There's a lot to wade through.
One sort of "evidence" is observational -
examples of regions that distributed ivermectin and things got better.
Of course, this ignores confounders. It reminds me of this sort of graph: 5/20

https://twitter.com/covidtweets/status/1340152202258767873

Sometimes regions are just going to improve, whether or not you distribute a miracle drug.
One example they give from Paraguay is particularly outstanding, since the curves were *already starting down* before the ivermectin distribution!
Causation is a tricky thing 🤔 6/20

Prophylactic use of ivermectin would seem most promising (whether discussing zinc for colds, oseltamivir for flu, or acyclovir for shingles, treating viruses early before replication has taken off is generally the best mode).

Most enticing study? 7/20

media.marinomed.com/8b/7a/c7/nota-…

In this study from Argentina, HCWs were either given nasal carageenan (I know, weird)& 12 mg oral ivermectin or just the usual PPE.
Results were ASTOUNDING!
58% of the PPE-only group got Covid-19 over the 3 mo trial.
NONE of the 788 carageenan/ivermectin!
NONE!!!
WOW!!!!!! 8/20

How do you get stunning, miraculous results like this???

YOU MAKE THEM UP!
Yes, I said it.
I usually try to be nice.
But those numbers are ridiculous. Putting those HCWs in private bubbles in Anarctica probably wouldn't prevent at least one false positive! 9/20

The other listed studies for prophylactic ivermectin use were positive but more modest.
None took place in the West.
Am I data-ist?
Guilty as charged.
Relying on the validity of studies from structurally research-disadvantaged nations has gotten me in trouble in the past. 10/20

OK, what about mild disease?
This small study out of Bangladesh looked at Ivermectin alone X5d or 1 dose w/ doxycyclineX5d, vs placebo.
ijidonline.com/article/S1201-…
Symptoms not much affected, but viral load dropped w/ ivermectin.
W/o comorbidities listed, hard to analyze. 11/20

How would this approach - using ivermectin early in disease course - hold up in a Western study, though?
WE HAVE ONE! From Spain:
europepmc.org/article/ppr/pp…
Well... how did it go for ivermectin???
Not so well 😔
Primary endpoint was PCR neg in a week;
zero in either group. 12/20

While viral load was lower in the ivermectin group, and that annoying anosmia improved faster, symptoms in general did not improve vs placebo;
and virus cultured out MORE often from the ivermectin group.
Small, low quality study - but no Milagros en España.13/20

How about severe disease? Here we are banking on the purported anti-inflammatory properties of ivermectin.
Yes, ivermectin has A LOT of properties, if the FLCCCA is to be believed.
Again, many studies from S America, Iraq, India, etc.
Most of them positive. 14/20

For example, this small RCT from Iran is a mishmash of disease severities and combined ivermectin w/ doxycycline while "controls" got a cocktail of "standard therapy." Impressive benefit in time-to-recovery & mortality among severely ill patients. 15/20
medrxiv.org/content/10.110…

But not what I would call "good-quality data."
So - how does this play in the West?
Again, a study from Spain.
The title says it all:
journals.plos.org/plosone/articl…
Tiny study. Messy design. But those Miracles went MIA, again... 16/20

We also are blessed with a study from Florida on hospitalized Covid-19 patients, from none other than a founding member of the FLCCCA.
It might be termed, "promising."
Or, "underwhelming."
Not randomized controlled but retrospective observational. 17/20
journal.chestnet.org/article/S0012-…

Mortality was half in the ivermectin group (12.4 vs 25%).
But the lack of randomization is a big issue.
Who got the ivermectin? Were they as sick as the others?
It reminds me of the Henry Ford Health study on HCQ w/ its poorly matched study arms. 18/20
beckershospitalreview.com/pharmacy/recen…

In fact, everything about the Ivermectin Craze reminds me of HCQ:
It helps everything.
It's a miracle.
It's cheap and safe.
Big Pharma doesn't want you to know about it.
Scores of big observational studies, regional data sets, and tiny RCTs support it.
Hmmmm.... 19/20

We all know what happened w/ HCQ.
It flunked every Western RCT thrown at it.
I suspect the same fate awaits ivermectin.
I would be surprised, but not shocked, if it turns out to display some modest benefit in prevention or treatment of early disease.
No miracles, though. 20/20

I should add, before an Ivermectinite puts out a contract on me:
1) I don't think someone literally made up the data showing a 0% rate of mid-pandemic infection vs 50+% placebo - but I do think there must have been a huge error in the data collection or reporting process.

2) It is criminal intellectual negligence for an entire organization of medical professionals to put out a 33 pg paper summarizing studies of mostly poor quality w/o a hint of acknowledgment that these studies were deeply flawed & sometimes ridiculous. That part gets my goat.

3) I didn't even mention publication bias.
No one wants to publish a low-quality, small RCT from Pakistan showing little or no benefit to ivermectin.
But same study *with* benefit might get a look. It's sexier.

No, it's not 1/67M odds that ivermectin is ineffective. Sigh.