We've Replaced The Patient's Blood With PolyHeme. Let's See If They Notice.

April 29, 2008 8:54 AM

The blood substitute PolyHeme has been previously discussed on MetaFilter, but new evidence shows that PolyHeme actually raises the chances of death by nearly 30%. PolyHeme was notable mostly for the reaction to its clinical trials, which, controversially, did not require patient consent.
posted by scrump (19 comments total)

The solution is simple, just have "Please don't administer the PolyHeme™ blood substitute to me if I happen to be hurt. I'll wait for real blood. Thanks!" tattooed on your chest. Like that guy from Memento.
posted by Blazecock Pileon at 9:01 AM on April 29, 2008

Plus there's the disturbing Ray Bradbury stories that can't have helped.
posted by GuyZero at 9:09 AM on April 29, 2008

The experiment made sense in the field because paramedics don't carry blood, but once people arrived at the ED they should have been switched over. Of course now that we know it increases the chance of death by 30% the trial should stop altogether.
posted by brevator at 9:52 AM on April 29, 2008

PolyHeme sounds just like the sort of thing that turns people into vampires or zombies in grade Z horror movies. Maybe the dead were lucky.
posted by Halloween Jack at 10:24 AM on April 29, 2008

scrump, thanks for posting this. I had heard that there was an associated risk of heart attacks, and now we know about the increased death rate. To be clear, it wasn't a manufacturer or sponsor who published this data. Scientists from the NIH (who have been involved with research on other blood substitutes) retrieved data from 16 different trials (which involved 5 different products, not just PolyHeme) and published it in JAMA. They couldn't get all the data because sponsor's wouldn't release it to them.

It is, as usual, the FDA that we should be questioning here. The FDA had unpublished data showing the increased risk of MI with PolyHeme, but let studies continue. When you consider the questionable ethics that were associated with these trials on top of everything else, it makes the IRB's job harder, and really does the public a disservice. The authors of the paper actually call for Congress to make FDA applications open to the public. Maybe I'm biting the hand that feeds me, but the FDA is clearly overburdened with the amount of work they have to do, and maybe opening things up to public scrutiny would help. But at the same time, I understand why sponsors wouldn't want that public scrutiny.

The bottom line is that the lack of published results and the questionable consent practices made it so that *nobody* was informed about these products: not potential participants, their families, nor Institutional Review Boards, whose job it is to determine the ratio of risk to potential benefits. I mean, my job is clincal trials, and I read the newspaper articles about PolyHeme (the so-called "community awareness" practices) and I never really heard or understood what the risks were.
posted by sarahnade at 10:59 AM on April 29, 2008 [4 favorites]

The experiment made sense...

Made sense? MADE SENSE? Are you f@$%ing insane?

How about we start a knife in the chest trial. "Oh of course now that we know a knife in the chest increases the chance of death by 90% the trial should stopped altogether."

Jeez!
posted by mikhail at 11:15 AM on April 29, 2008

Mikhall, did you read the articles? Or my comment?

From the last link:

Among other requirements, patients must be in a life-threatening situation when informed consent is not feasible, and available treatments are unproven or unsatisfactory.

During the study, emergency medical technicians randomly administer either saline solution or PolyHeme to trauma patients in hemorrhagic shock in the field and during ambulance transport. Once in the hospital, those receiving PolyHeme continue to do so unless they or their legal representatives formally withdraw. The control group receives standard blood and saline as required.

The field portion of the trial satisfies the regulations because blood is generally not carried in ambulances..."But the head-to-head comparison of blood and PolyHeme in the in-hospital portion of the trial is not justifiable according to the regulation," "The waiver of consent can only happen when available treatments are unproven or unsatisfactory, and blood is not an unsatisfactory treatment for blood loss."

As I said, the trial made sense in the field, but the patients should have been switched to blood once in the hospital.
posted by brevator at 11:41 AM on April 29, 2008 [1 favorite]

An even simpler solution: all of us who can donate blood should.
posted by Carol Anne at 1:14 PM on April 29, 2008

From New Scientist: real blood transfusions may not be so good for you either.

In fact, most experts now agree that the risk posed by the transfused blood itself is far greater than that of a blood-borne infection. “Probably 40 to 60 per cent of blood transfusions are not good for the patients,” says Bruce Spiess, a cardiac anaesthesiologist at Virginia Commonwealth University in Richmond.

posted by moonmilk at 2:51 PM on April 29, 2008

The proposal for this research crossed my desk, a while back. There really is a huge need for these products. The largest problem with testing these products is that we don't have much of a test population. It does make sense to test it when blood is needed but not available, but I had no idea it was possible to 'get around' informed consent that way.
posted by zennie at 3:18 PM on April 29, 2008

moonmilk, that's in relation to acute coronary syndromes. The authors of that JAMA study were studying the impact and necessity of blood transfusions when given to stable patients during routine heart surgeries.

For critical care/ER, there is absolutely zero doubt that transfusions are necessary.
posted by junesix at 4:12 PM on April 29, 2008

Thanks for the clarification, junesix!
posted by moonmilk at 4:51 PM on April 29, 2008

That's no clarification at all. As a trauma anesthesiologist, I can tell you that a large proportion of transfusions, whether in the operating room, in the ICU, or in the ER, are unnecessary or even harmful. We have limited data to tell us what our "transfusion threshold" should be, i.e., how low should we let the hematocrit drop for a particular patient. Some patients can survive having their hematocrit drop to 10% or less (10% of their blood volume is made up of red cells, rather than the normal 35-50%) if it happens gradually.

There is a huge need for blood substitutes, especially in the battlefield, and during shortages at home.

Certainly we are beginning to understand the negative effects of transfusion, and gradually lowering our transfusion thresholds as the evidence accumulates.
posted by etherist at 5:36 PM on April 29, 2008

Even if we all donated a pint of blood every week, would an ambulance have the capability to haul around all the blood they would need and to do the work up required to pick the right bag to top off their patient? The results of putting the wrong blood in someone is pretty damn ugly.

Mikhail, right now the "best practice" is to replace the blood you lost with phosphate buffered saline or the like. Once you're lost enough blood you might as well go with the knife in the chest treatment as saline's oxygen transport capability is right about zero and you're already screwed.

Let's be careful out there.
posted by Kid Charlemagne at 5:54 PM on April 29, 2008

I remember being stunned to learn back in 2004 that Duke University's medical center was one of the sites that would be testing Polyheme on local patients without their consent. It's worth revisiting an article from back then, especially this part:

...the FDA approved a no-consent trial in 1998 for a blood substitute, HemAssist, manufactured by Baxter Healthcare. That trial was aborted after almost half the subjects receiving HemAssist died, far more than those receiving blood.

The rest looks even more horrid in retrospect than it did at the time:

The federal consent-waiver rule allows for non-consensual testing only in strictly limited circumstances. In particular, the subjects must be in a life-threatening situation, "available treatments are unproven or unsatisfactory," and the collection of the data is necessary to demonstrate the effectiveness of a given treatment...Northfield, in fact, is calling its trial an "ambulance study."

But the study protocol allows for administration of PolyHeme for up to 12 hours, most of which occurs in the hospital setting after the victim has arrived. There, of course, a well-proven and satisfactory substitute is readily available--whole blood. This aspect of the study has drawn fire from medical ethicists and others who say it's immoral to test PolyHeme on people unless they say it's OK. "It is not until after artificial blood is proven to be equivalent to real blood, using consenting subjects, that it should be used experimentally in non-consenting patients," says Leonard Glantz, a professor of health law at the Boston University School of Medicine...

The use of PolyHeme for 11 or more hours in the hospital isn't the only aspect of the study that flunks the smell test. Those conducting no-consent studies must undertake an extensive process of public disclosure and community consultation. And to minimize the chance of enrolling people who would object to the experimental treatment, they must create an opt-out mechanism...the requirements are vague, and the practical application of these safeguards amounts to very little. To date, Duke has set up a hotline, held four public meetings, spoken to a Rotary Club gathering and set up information tables, twice at Northgate Mall and at the July 4 Bulls game. (Attempts to set up a table at Southpoint mall and speak to the congregations of two black churches in Durham were rebuffed.) In addition, Duke placed notices in the local dailies and a few public service announcements on the radio...

Even more absurd is the idea that the opt-out option has any practical value. As do all the PolyHeme sites, Duke will offer a blue plastic bracelet (Carolina blue, it should be noted) to anyone who chooses not to participate. Objectors can wear the bracelet 24/7, and if they happen to suffer a life-threatening traumatic injury, emergency personnel will recognize the bracelet and keep the PolyHeme out of their veins. Duke is also providing wallet-sized cards, like organ donor cards, and is compiling a do-not-test list that ambulance attendants will allegedly have on hand and allegedly check when trauma victims are picked up--as long as the list doesn't get too long and unwieldy...

It's understandable why Northfield wants to test PolyHeme as broadly and quickly as possible. PolyHeme is the company's only product, and Northfield is locked in a race with two competitors to bring its blood substitute to market first.

And now it turns out this experimental new product that absolutely, positively *needed* to be tested on non-consenting patients so Northfield could get it to market ahead of its competition is, in fact, deadlier than expected.

Well, gosh. Color me surprised.
posted by mediareport at 6:55 PM on April 29, 2008 [2 favorites]

These reports are really dismaying to me. Consent is important, and even when patients can't be fully informed, an attempt at consent should at least be tried. It's true that advances in E/R medicine are stymied by the difficulty of obtaining consent in E/R situations. That's what E/R researchers need to be aware they're signing up for when they plan clinical trials. Waivers of consent need to occur from time to time - retrospective observational trials are a good place, because no one can be harmed - but prospective treatment trials had better not waive consent unless there is absolutely no other way to gain data on the treatment under study.

That wasn't the case here, and it set back artificial blood trials another decade, which is too bad, because I've been wanting to get my hands on some of this stuff for cerebral vasospasm patients for at least that long. Now no IRB is going to want to come near it, ever. One team's sloppy disregard for standard ethical approaches has poisoned the well for everyone who comes after.

Jackasses.
posted by ikkyu2 at 10:36 PM on April 29, 2008 [1 favorite]

This immediately brought Medical Apartheid to mind. I can't praise this book enough. I wish I could say I was shocked by the whole PolyHeme mess, but it sounds like it's just par for the course.
posted by fiercecupcake at 6:21 AM on April 30, 2008

In other news we apparently need to set up a study to chuck people out of planes without a parachute.

That is, if we can muster up the energy.
posted by BrotherCaine at 1:47 AM on May 1, 2008

The issue that strikes me is that because sponsors didn't know about previous failures -- the same process was tried again repeatedly with the same poor results. Healthcare product companies are very hardlined about sharing anything that might give a hint to competitors. I work in an association for professionals in the field and they have a hard time sharing stories to learn from one another because *everything* is considered proprietary. If the results were not considered proprietary and the trials had been required to be reported then subsequent studies would not have reinvented the wheel, so to speak.
posted by Librarygeek at 8:48 AM on May 1, 2008