--to call attention to the excellent blog that Dr. David Newman runs over at Mt. Sinai--he calls it "SMART EM" which stands for "Scientific Medicine and Research Translation [in Emergency Medicine]." I'm going to pay a return visit to his blog, even though the issue he discusses is perhaps of peripheral concern to our topic of ethics and the pharmaceutical industry. It does relate to how Americans end up being seriously overtreated without any resulting health benefits. (It's also very discouraging for those of us Anglophiles who had imagined that the Brits had things better figured out.)
Dr. Newman takes aim at a recent study in Lancet:
I need to provide some general background, review what Dr. Newman says, and then add some commentary.
Background: Dr. Newman quickly reviews the reasons why I had decided long ago that even though I am not into tattoos, if I were to have something tattooed on my chest, it would be "No tPA in case of stroke." Stroke care has been taken over by the group I cynically call the "brain attack mafia" who argue that we should go into crisis mode when a stroke patient rolls into the ER just like we do when a heart attack victim arrives. Their problem: We have proven things to do to save lives in acute heart attack, but what do we have to offer for acute stroke? The logic of the disease suggests that dissolving the clot in the brain artery quickly would save brain tissue, function, and ultimately lives. Sadly, as often happens, logic and scientific outcomes fail to match. The clot buster usually recommended, tPA, has been shown in numerous trials either to have no benefit, or to have a very small benefit which you can only find by tying knots in the statistics, while posing risks of major bleeding.
So now there's a new international trial of tPA in acute stroke, IST-3, just published on line in Lancet (subscription required), that Dr. Newman thoughtfully dissects. Their primary endpoint was death or dependence at 6 months. The study data were clear--no difference in either outcome between the tPA and the control group.
But of course the Mafia would not be satisfied with that negative outcome, so the authors then went back and had to play around with the statistics, and they came up with a new, secondary measure that they claim offered a very slight benefit for the tPA group. So, in what Dr. Newman terms as "breathtaking" evidence of going "stark, raving mad," the authors claim that the study was actually a success for tPA. Not only did Lancet aid and abet the insanity by publishing the paper with this obvious lying-with-statistics, but even added an editorial strongly favoring widespread use of tPA based on this supposed "success."
Commentary: Two main things to note about all this. First, is this obviously biased result due to drug company or industry influence? Well, no and yes, apparently. I looked at the Lancet paper for the usual Pharma footprints. The study was funded by what appears to be a consortium of government health research agencies with no obvious industry money. The authors on the other hand have a long list of ties to industry.
What to make of this? My best guess on the Brain Attack Mafia" is that this is at root genuine therapeutic zealotry among a group of neurologists who are tired of seeing stroke dissed as an "oh well" diagnosis that you just have to take care of and cannot attack with all guns blazing. Folks like this may be biased advocates for relatively useless therapies even when no Pharma money is on the table. But when any industry funding shows up, it's naturally this group that is the greatest ally of the industry that wants to push the drug or the test in question.
Now, second main thing to note. As Dr. Newman points out, even if you believe the crazy statistical gymnasics of IST-3, you are still looking at a miniscule benefit from tPA with a big risk of harm. So an obvious question is: Do we have anything better to offer a patient who has just suffered an acute stroke? And the answer is: Yes.
The literature is now full of studies showing over and over what happens when such a patient is admitted to an acute stroke unit, or the care of a dedicated stroke team, or basically a group of nurses, physical therapists, and other staff in the hospital who wear t-shirts that say "Strokes R Us." It hardly matters if there are any docs involved. What has to happen is that the patient gets care from a team that does stroke all day and every day and has a system not to forget all the little details of care. There's no rocket science, it's just painstakling use of known, beneficial aspects of nursing care, physical therapy, prevention of blood clots in the legs, etc. etc.
What hapens? Here's just one observational study, ironically also published in Lancet:
This Italian study followed 11,572 acute stroke patients. Those admitted to the right sort of stroke unit had a 28% long term mortality rate compared to 36% in the controls. Those too disabled to live at home at the end of follow-up were 43% of the controls vs. 35% for the stroke unit. (You don't like observational studies? There are also controlled studies in the literature showing the same results, which have now been widely replicated.)
So we are talking here about absolute differences (not relative) of around 8 percentage points, which is hugely superior to results seen in any study of tPA. So why, you might ask, is the Brain Attack Mafia so insistent that we have to give every eligible patient tPA, and so unconcerned about whether each hospital that cares for stroke patients has a good stroke team or unit?
Dr. Newman has the answer in a term he calls "scienciness." We in medicine love scienciness. We often don't like science, because it shows us that stuff that we devoutly believe in (like busting clots with tPA) doesn't really benefit patients. Stroke units and teams are extremely unsexy and unexciting (good nursing care??? Give me a break) but are solid science. Clot-busting with tPA is scienciness.
The IST-3 Collaborative Group. The benefits and harms of intravenous thrombolysis with recombinant tissue plasminogen activator within 6 h of acute ischaemic stroke (the third international stroke trial [IST-3]): a randomised controlled trial. Lancet, 23 May 2012,