Sunday, July 21, 2013

Pinching Patients - Pain, Consciousness and Practical Neurology

How to Solve the Problem of Consciousness With Pinching

I’ve just completed two weeks with Neuro Critical Care in the Neurological Intensive Care Unit (NeuroICU) taking care of very sick patients. The NeuroICU, like the hospital in general, is not the sort of place you want to end up in. You get to spend time with us if you’ve had a bleed in your brain, or had a massive stroke, or had severe head trauma. There are many things I could say about my experience, but I will focus here on one that is philosophically relevant.

Depending on what service you happen to be on, you report different things. Surgeons are concerned about bowel stasis after abdominal surgery, so are very interested in things like pooping and passing gas (woe unto you if you forget to report to your team whether your patient pooped since last they saw her). Cardiologists care about heart sounds and psychiatrists care about mood. But what about doctors in the NeuroICU?

Many of our patients had sustained brain injury to the point of being unable to speak. How do you assess whether things are getting worse or getting better in the brain, especially when the patient can’t talk? Do we need to perform an operation, do a procedure, or give another drug? First, we check to see if the patient can move despite not being able to talk. “Mr. Johnson, move your fingers. Mr. Johnson, move your eyes.” But sleeping people also wouldn’t follow commands. What’s the difference between a sleeping person, a comatose person, and a dead person? How can you separate these three? How do you know if “anybody’s in there”?

One of the best (and only) tests we have is response to noxious stimulation. That is, we try to hurt them and see how they respond. Every morning, my job was to approach my comatose patient and, to see how she was progressing, pinch her on each of her extremities. Neurologists, I found, are quite adept at causing pain and I was a quick study. A patient might not respond at all (this is a bad sign). A patient might respond, but do so in a “decerebrate” or “decorticate” fashion, that is, their response is a brain stem or spinal cord reflex and not a conscious decision [1]. In other words, some people will respond to a pinch on the inside of the wrist by pulling in toward the pain rather than away from it. Still further, a person who was particularly awake would make some effort to stop the pain. A patient who tries to bat away the pinch is in relatively good shape.

I’m also reading on the philosophy of mind, and I realized that this crude test is actually an experiment demonstrating several things. Minds (whatever they are ontologically) have various capacities and experiences including sensation, thought and intention. I realized that our pinching test is trying to answer these questions. Is the patient experiencing the sensation of pain? As Stanford pain expert Sean Mackey likes to say, “The strain of pain lies mainly in the brain” (i.e. is the “noxious stimulus” of the pinch making it to the brain and consciousness?). Movement, facial grimacing, or increased heart-rate all suggest the sensation of pain is intact. Is the patient thinking about where the pain is? Does the patient intend to free herself from the pain by movement? If she moves away from the pinch, it would suggest both that she has knowledge about where the pain is coming from and intends to stop the pain. Batting at the pinch indicates higher-order knowledge of location and how to stop it.

The “pinch test” has obvious limitations, the most concerning of which is for patients who are “locked in.” Patients who lose the connection between brain and body will “fail” the pinch test, but may still be entirely conscious. Patients we have declared “vegetables” by our crude tools might actually be awake but unable to communicate. The most intriguing work that I know about in trying to improve these tools is Adrian Owen. He put “vegetables” (patients in “minimally conscious states” and “persistent vegetative states”) into fMRI scanner, a machine that can read where blood is flowing in the brain, and asked them to perform various well-characterized tasks. He would say things like, “If you can hear me, imagine swinging a tennis racket.” In their sample of 50 patients, 5 of them could. But ability to follow a simple command, even in the brain, is far from fully conscious. So he tried to go further. Using this strategy, he asked various yes or no questions, and found that one patient was able to respond correctly to personal information. Owen continues his work in this area and is trying to do the same thing with EEG (which would be much cheaper and more scalable).

Pinching is a pretty crude way to answer the Problem of Consciousness. But it is a strategy in line with the core of medicine that I have long loved: it’s about helping patients. Medicine since the time of Hippocrates has been an empirical profession; at our core, our job as doctors is not to discover answers to the deep questions of nature or philosophy. Doctors qua doctors use whatever science or art we can to help patients. Unfortunately in 2013, the best tool we have to assess consciousness is a well-placed pinch. Would it be nice if there were better tests? Sure. But we've got what we've got and there’re sick people to take care of right now. So let’s get busy.

[1] I think it is a fair question to ask how we know for sure that a “decerebrate” response is truly “unconscious.” I’m skeptical that this has been rigorously tested. As with many things in medicine, there is (necessarily) a lot of tradition. Even things as basic as prognosis are notoriously under-studied and even well-trained physicians are exceptionally poor at prognosticating. I read one study of cardiac arrest victims who were given a “poor” or “grave” prognosis: 21% recovered, 54% were supported and died or had poor recovery and 25% had support withdrawn and died. This suggests that on the order of 5% of the total population died because a doctor said they wouldn’t live.

1 comment: