Aug 192013

strokeassociation FAST

One of the most difficult tasks required routinely of ER doctors is obtaining informed consent for the use of IV tPA in cases of acute ischemic stroke.┬áIt is a difficult task because it combines two things that are each — standing alone — among the most difficult things ER doctors are expected to do.

First, we have to decide for ourselves whether or not we want to give tPA. This is an extremely difficult clinical decision. There is a substantial risk of causing catastrophic life-threatening hemorrhage and other serious harm; tPA is not a risk-free treatment. The decision to give tPA must usually be made under conditions of severe time pressure. Studies and guidelines suggesting that tPA benefits victims of ischemic stroke emphasize that it must be given within a very short time period after the onset of symptoms, usually three hours. To complicate matters, the scientific evidence supporting the decision to treat acute ischemic stroke with tPA is not obvious or clear. It is perhaps the most controversial current debate in emergency medicine, with passionate supporters and opponents all wielding literature that is inconclusive and incapable of resolving the debate.

Second, given the risks of tPA, modern medical ethics demands that before we push the drug we explain to the patient or her family members the risks and benefits of treatment to the extent that they are able to give their “informed consent.” In the best of circumstances, obtaining truly informed consent is very challenging. But the circumstances surrounding the tPA decision could not be worse for the patient or family member of the patient — the clock is ticking; they have to make the decision quickly. They are being asked to weigh highly incommensurable outcomes against each other, e.g. death from intracranial hemorrhage, or regaining the ability to speak. In many cases a family member asked to make the decision at one remove — not what I want, but what would my family member want were she able to make the decision for herself. And, of course, the honest physician will inform the patient that the scientific evidence for tPA is not ironclad, and that there is a raging debate among doctors about whether it is beneficial, useless, or harmful.

“So, after all that, what do you want to do, Mrs. Jones? You have, let’s see, about 15 minutes to make your decision.”

My own approach to this difficult task has evolved and will evolve further as I accumulate experience and as more scientific evidence surrounding tPA (and surrounding informed consent) is published. But for now, this is how I approach the problem.

1) It is impossible to familiarize the patient or her family members with the entirety of the tPA evidence in the time they have available for making their decision.

It takes many, many hours for emergency physicians and neurologists to digest the tPA literature; so much so that most busy clinicians rely upon trusted intermediaries to digest the literature for them, pro and con. To then expect that you will be able to fully inform a layperson within the allotted time who is already in great distress from seeing their loved one suddenly immobile on one side of their body and unable to speak is fantasy. So don’t try. It won’t work, and you will be wasting precious time.

2) The goal should be to help the patient or her family member to make a decision that they will be least likely to regret should the outcome turn out to be unfavorable.

Regardless about what the decision about tPA turns out to be, there is a significant chance that the outcome will be far less than perfect. The patient may refuse the tPA and end up paralyzed and fed through a PEG tube for the rest of their lives. Or, the patient may choose the tPA and suffer a fatal intracranial hemorrhage. Post hoc, it will be impossible for anyone to say if the outcome would have been “worse” had the patient made the opposite decision. I don’t think that having a more subtle understanding of the NINDS trial, or being told about the difference between the Cochrane metaanalysis of stroke studies and the ACEP guideline’s review of the tPA literature will do much to alleviate regret if, in this one case, the outcome is bad. What will help is if, in hindsight, the patient’s family can say that the decision was consistent with the patient’s values and overall approach to risk.

3) To maximize the chance that this will be the case, the doctor’s discussion with the patient or the family member should emphasize three informational elements: a) the near clinical equipoise of the published studies b) the choice between potentially reversing the disability caused by the stroke with tPA, and the potential for that drug causing possibly fatal bleeding c) your personal enthusiasm for or against tPA, if the patient asks.

There is rarely enough time for a discussion of much more than these three things, and in the end, these are the three things that matter most for the decision.

4) Remember that the patient or her family member is not being asked to take sides in the controversy over the tPA evidence.

They are being asked to share in the decision making about whether tPA should be given here, now, in this particular case. This is not the same thing as taking sides on the issue of whether routine tPA should be the standard of care across all cases of acute ischemic stroke.

5) Translate the decision into one that the patient is familiar with after a lifetime of making their own high-stakes decisions under uncertainty.

In general, this is the choice between “doing something” to perhaps improve the situation, vs avoiding iatrogenic harm.

Most patients that physicians see tend to fall into one of these two camps. The former will actively take all kinds of medications and seek out all kinds of therapies for even the most subtle of symptoms. The latter will wait until the last possible moment to take that daily aspirin for their heart disease or to see that surgeon for their postprandial right-upper-quadrant pain, in order to avoid iatrogenic stomach ulcers or an inadvertent surgical sponge in the abdomen. If physicians are honest, they will admit that much of their own enthusiasm (or lack thereof) for tPA in stroke is guided by their own tendency to either “do something” or to avoid causing harm.

This choice between action under conditions of uncertainty vs avoidance of action under uncertainty is something that we all, regardless of background or education, understand. It’s a choice that’s fairly thoroughly informed by our general values and approach to living. When faced with the problem of water pollution, for example, some of us dream of huge water processing plants and massive pipelines; others of us dream of discontinuing the production of toxic waste. Faced with the possibility of a stock market crash, some of us short a bunch of stock, others of us put all our money in cash under the bed.

I like to ask my patients “are you a “do-something” kind of person, or are you a “don’t screw things up” kind of person? I tell them frankly that they, or their loved one, is in a very bad position and that the uncertainties are huge. I tell them that the outcome could be tragically bad whether we give or don’t give the tPA. I say that I understand that they are being asked to make a difficult decision in a situation with a high degree of uncertainty and a high likelihood of extremely consequential results, either way. I offer to answer any questions that they have, to the best of my knowledge. I will give them my opinion about tPA: not enough evidence yet to say whether it’s helpful, harmful, or inconsequential — a scientifically unproven therapy that deserves further study.

“So after all that, what do you want us to do, Mrs. Jones?”

Ugh. This is why I’m a fan of doing all we can to prevent ischemic stroke.


Some links I like:

ACEP clinical policy on tPA in stroke [PDF]

David Newman’s SMART EM podcast on tPA evidence

Dr. Newman’s blog post on same topic

Links to Cochrane review of thrombolytics in ischemic stroke

BMJ debate about tPA (and steroid) guidelines and conflicts of interest


 Leave a Reply

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>