The crux of the problem - epinephrine continues to improve short-term ROSC with uncertain long-term outcome improvement.
This is a prospective out-of-hospital arrest study from Australia in which epinephrine or saline placebo was given to patients during resuscitation by EMS. And, like many studies before it, it fails to show a meaningful difference between patients receiving epinephrine and patients receiving placebo. Rather, their primary outcome of survival to hospital discharge had 1.9% with placebo and 4.0% with epinephrine - but this result was not statistically significant with a p-value of 0.15.
Of course, what the lack of statistical significance means in this case is that this difference could have occurred by chance 15 times out of 100 times they performed this study - which, while not meeting the gold standard of 5 out of 100, is still a reasonably interesting clinical trend. Like all studies before it, the short-term endpoints met statistical significance, including ROSC of 8.4% for placebo and 23.5% for epinephrine. There are a few confounding differences between groups: more placebo patients had witnessed arrest, although the number with bystander CPR was the same; more placebo patients were endotracheally intubated in the field, which usually confers a survival disadvantage; and more epinephrine patients were ultimately transported to the hospital from the field.
So, there's two ways to look at it: 1) epinephrine works, and we just need to figure out how to salvage more of those ROSC or 2) epinephrine is flogging far too great a number of lost husks back to life that will go on to consume ICU resources and expire regardless.
But, if we're not going to give epinephrine, how do we otherwise look busy during a code? And, what happens downstream to our epinephrine ROSC that fail to leave the hospital or the ER, and can we prevent it?
I am still not sure what the right answer is - like many diseases, cardiac arrest patients are a heterogenous group in which there is almost certainly a subset of patients that benefits from epinephrine, but we don't yet know who that might be.
"Effect of adrenaline on survival in out-of-hospital cardiac arrest: A randomised double-blind placebo-controlled trial."
Thanks to @cliffreid of Resus M.E! for first noting this article.