Hacker News new | ask | show | jobs
by timr 1545 days ago
No, it isn't. There's a saying in medical research: "mice lie, and monkeys exaggerate". This is because most findings in mice don't translate to humans.

Representing an experimental result as directly relevant to humans when it wasn't done in humans is misleading.
2 comments
Barrin92 1545 days ago
> as directly relevant to humans when it wasn't done in humans is misleading

The authors point out that the same observation was made in human brain tissue of covid patients as well, it is literally in the paper.

"We had the opportunity to examine human cortex and subcortical white matter samples from a cohort of nine individuals[...]"

link
timr 1545 days ago
The results shown in Figure 3 are essentially unrelated to the rest of the paper. It's an experiment that they've included to claim some relevance to humans, and head off the criticism that this is a paper about mice.

At best, you can argue that they show a different inflammatory marker in a group of people who were already unwell. Here's the paper describing the people from whom the 9 samples were chosen:

https://www.nejm.org/doi/full/10.1056/NEJMc2033369

> We performed conventional histopathological examination of the brains of 18 patients. Fourteen patients had chronic illnesses, including diabetes and hypertension, and 11 had been found dead or had died suddenly and unexpectedly. Of the 16 patients with available medical histories, 1 had delirium, 5 had mild respiratory symptoms, 4 had acute respiratory distress syndrome, 2 had pulmonary embolism, and the symptoms were not known in 3

I can't tell which of these patients were used in this study, but...there's a lot going on with these samples. One of the patients was a meth addict, and another was a heroin addict, yet another was an alcoholic, and two others had recurrent seizures from prior head injuries!

https://www.nejm.org/doi/suppl/10.1056/NEJMc2033369/suppl_fi...

link
trumpablehump 1544 days ago
I wasn't talking about the results and relevance of this particular paper but I was replying to a cheap, offhand generalisation.

No doubt that one needs to be cautious about animal tests but they do work and are highly effective in many cases, especially for finding indications of possible effects on humans. Or is that not true?

link