Yes, the Pain Is All in Your Head

This transcript has been edited for clarity. 

Welcome to Impact Factor, your weekly dose of commentary on a new medical study. I’m Dr F. Perry Wilson from the Yale School of Medicine.

I’ve been thinking about Dune a lot lately. I think I might be the only person in the world who prefers the bizarre and grotesque David Lynch movie version to the elegantly crafted Villeneuve oeuvre, including David Lynch himself. We lost a real artist with his passing, and a rewatch of Twin Peaks is very much on my to-do list for this winter.

But back to Dune, because one of the pivotal scenes in the novel and both movie versions is one where young Paul Atreides is tested by the Machiavellian Bene Gesserit. Atreides has to put his hand in a box. What is inside? Pain. Ever increasing pain. He must keep his hand in the box, despite all his instincts telling him to pull it out to prove his fundamental humanity — his ability to exercise control over his own instincts.

Because, as the Reverend Mother points out after the ordeal, his hand is unharmed. The pain is a fabrication — pain by nerve induction, she says. There is no physical damage. It’s all in his mind.

And, of course, that’s true of all pain, isn’t it? It’s not your toe that hurts when you stub it. Signals are sent from your toe, up a nerve to your spinal cord, up another nerve to your thalamus, and then onto the cortex to give it context, emotion, intensity, reality. If that chain is broken, pain simply does not occur. It’s all in your mind.

That’s what makes pain so difficult to treat. It is fundamentally subjective. I’ve had patients with wounds that would make me scream for my mother, yet they sat stoically silent while we worked on them. And I’ve had those who, well, seemed like they were hamming it up a bit.

But right now our best tool to get a window into someone’s level of pain is a tool like this. A pain scale. How much does it hurt?

Given that the answer may lead us to prescribe some ice or some Tylenol or some morphine, well, the answers aren’t always reliable. Or even consistent.

To quantify pain, to put a nonsubjective number on it, we’d need to read someone’s mind. Well, maybe we can.

Pain is on my mind this week, thanks to this study “Predicting Individual Pain Sensitivity Using a Novel Cortical Biomarker Signature,” appearing in JAMA Neurology, from David Seminowicz and colleagues at the University of Western Ontario. Their goal was to use noninvasive brain studies — in this case, electroencephalography (EEG) and transcranial magnetic stimulation (TMS) — to interrogate the level of pain a person is feeling.

To do that, they enrolled 150 healthy participants with no history of chronic pain. Then they needed to hurt them. And they needed to hurt them all in a very standard way. Their model here was to inject a substance called nerve growth factor into the right masseter muscle. That’s the muscle right in front of your ear that does a lot of chewing. I mean, honestly, this is so gom jabbar, right?

When you inject nerve growth factor into a muscle, it hurts. And not just for a moment. It hurts for about 4 weeks; it’s a model of acute pain that turns into chronic pain. I am quite curious how much these 150 volunteers were paid to participate in this study, and more curious to know whether they felt that pay was adequate after the fact.

Before we get to the brain, let’s talk about the effect this injection had on pain levels. The authors used a statistical technique called growth mixture modeling to identify two types of people in their cohort: those for whom this injection caused a lot of pain and those for whom it’s not so bad. Their scores, over time, are shown here.

The yellow group, people who experienced more pain, were in a lot of pain. You can see their scores were around the top, 10 or so, for the first few days of the study before coming back down a few weeks later. The blue group tolerated it better. It still hurt, they reported, but it wasn’t the worst.

So the question at hand is whether you can tell who belongs to which group from the brain alone. And the answer is yes — almost perfectly, in fact.

The authors identified two brain biomarkers (measurements) that could predict quite well which group an individual belonged to. Via EEG, they looked at the sensorimotor peak alpha frequency, a measure of the rate of oscillation of brain waves, and found that slower peak alpha frequency predicted more pain.

TMS might not be as familiar as an EEG, but it’s seeing increasing use. TMS uses a targeted magnetic field to sort of “poke” the brain — give it a little stimulus in a very specific area. You then can see how quickly that stimulus propagates to other areas. In the case of this study, they looked at something called corticomotor excitability — how quickly a signal from the cortex can get to the part of the brain that controls muscles. Slower signaling? More pain.

Combining these two metrics, peak alpha frequency and corticomotor excitability, was all the researchers needed to predict who would belong to which group — the high-pain or the low-pain group. And it worked with about 90% weighted accuracy.

But there’s more to the study than that. The thing that really blew my mind was the effect of including psychological information about the participants.

There’s a standard survey called the “pain catastrophizing scale” that helps to quantify the psychological impact of pain. Certain individuals may tend to ruminate on pain more, or have more anxiety about the idea of pain, or feel helpless to manage pain. 

But here’s the thing: The inclusion of those scores didn’t improve upon the brain measures at all. To predict how much pain someone is going to report, knowing about their psychology was essentially useless in this study, provided you know about the electrical signals in their brain.

This is an amazing way to think about and understand pain, and people’s complaints about pain. It’s not that some people react worse to the same level of pain; it’s that some people’s brains process those signals differently. 

Psychology is a slave to biology, not the other way around.

There is one last twist to the study that I want to leave you with. One of the two metrics assessed — the peak alpha frequency — was measured before the injection into the masseter muscle. Alone, that measurement couldn’t perfectly predict who would have more pain, but it was still pretty good.

That means this study has opened the door to two new ways to think about and deal with pain. First, a way to predict the severity of pain before you experience it. Think of this as a tool to be used before elective surgery, for example, to help physicians and patients prepare for postoperative pain and manage therapy. And second, to quantify the degree of pain a patient may be feeling acutely, in order to better dose pain medication.

There’s a dark version of this, of course. There’s a world where we stop listening to patients’ complaints and just strap something on their head that spits out how much oxycodone they should get. I don’t want that world, and I’m sure you don’t either. But there are cases where this will help — particularly those cases where the cause of pain isn’t obvious (such as fibromyalgia), and doctors might minimize the symptoms when the patient describes them. Tools like this could prove what patients have been telling us all along. And when doctors dismissively say, “It’s all in your head," patients can accurately say, "Indeed it is. Treat it, please."

F. Perry Wilson, MD, MSCE, is an associate professor of medicine and public health and director of Yale’s Clinical and Translational Research Accelerator. His science communication work can be found in the Huffington Post, on NPR, and here on Medscape. He posts at @fperrywilsonand his book, How Medicine Works and When It Doesn’t, is available now.