Pain 101: Using Brain Imaging to Understand Pain

Brain imaging is proving to be a valuable tool to learn about what causes pain, and how to treat it. Image credit: Jens Maus/Wikimedia Commons.

Editor’s Note: The third North American Pain School (NAPS) took place June 24-28, 2018, in Montebello, Quebec, Canada. This educational initiative brought together leading experts in pain research and management to provide 30 trainees—part of the up-and-coming generation of pain researchers—with scientific education, professional development and networking experiences. 

Six of the trainees were also selected to provide first-hand reporting from the event, including summaries of talks presented at the meeting. Here, Chulmin Cho, who is doing post-PhD research at the University of Toronto Mississauga, Canada, summarizes a talk delivered by Irene Tracey, a professor and pain researcher at the University of Oxford in the UK. Tracey discussed how brain imaging techniques are helping pain researchers learn more about the causes of chronic pain.

The late Patrick D. Wall, the pioneering “father” of pain research, once said: “Pain is not just a sensation but, like hunger and thirst, is an awareness of an action plan to be rid of it.” Unlike hunger and thirst, however, pain, when it becomes persistent, is not easy to get rid of.

To better understand chronic pain—and to leave it behind—it’s important to learn more about the neurobiological mechanisms of pain perception. That is, what is it that happens in the nervous system to create an experience of pain? At the 2018 North American Pain School, Irene Tracey, a professor and pain brain imager at the University of Oxford in the UK, discussed how brain imaging has been used as a tool to study pain perception mechanisms, and how pain researchers are employing this technique now to learn more about the changes in the nervous system that give rise to pain.

Pain is subjective
Pain perception has multiple dimensions. The sensory-discriminative aspect of pain refers to the physical sensation of pain and how intense it is. The affective-motivational component of pain concerns the emotions that come into play during the experience of pain. And the cognitive-evaluative feature of pain refers to a person’s appraisal of pain based on previous experiences and knowledge.

Therefore, pain is highly subjective. There isn’t a one-to-one correlation between the strength of a stimulus that could cause pain, such as excessive heat or chemicals, and our perception of pain. This explains why, for the same level of heat, for instance, people react and rate the pain they feel differently.

Why use brain imaging to study pain?
The subjective nature of pain has made it difficult for scientists to fully understand it. Still, investigators developed research tools to advance knowledge. These tools include self-report measures, where people rate their own pain; physiological measures such as heart rate and breathing rate, which can serve as indicators of changes in the nervous system that can accompany pain; and recordings of the brain’s electrical activity, as well as stimulation of various brain regions, to better understand the role of the brain in creating the experience of pain.

All of these tools have limitations, however, so the pain field has looked for additional ways to investigate pain perception. One of these is brain imaging. Over the past few decades, brain imaging studies have revealed specific patterns of brain activity that correlate with various aspects of pain perception. Currently, advanced brain imaging techniques are allowing researchers to study the brain’s structure, the connections between different brain regions, and how these are altered during chronic pain. Brain imaging also makes it possible to learn more about the brain chemicals and changes in nerve cell activity that contribute to pain.

The first time that scientists used imaging to examine the brain response to pain was in 1991. New techniques, including magnetic resonance imaging (MRI) and positron emission tomography (PET), identified multiple brain regions that respond to painful heat. The brain regions that are active, represented by “blobs” on the images, include the anterior cingulate cortex (ACC) and the somatosensory cortex. The ACC is an area towards the front of the brain that carries out higher cognitive functions such as attention and decision-making. The somatosensory cortex is at the top of the brain and receives all the sensory information coming in from the body. However, the images had poor resolution, because of limitations in the imaging technology, which prevented further studies at the time.

Looking at different aspects of pain perception
As brain imaging techniques advanced, Tracey and others began to dissect different aspects of pain perception. For example, in a study published in 1999 in the journal Science, she and her colleagues used a type of MRI, called functional MRI (fMRI), to see if what happens in the brain during anticipation of pain differed from what takes place during the actual experience of pain.

Here, healthy study volunteers learned to associate colored lights with warm stimulation applied to the back of the left hand. The researchers saw different patterns of brain activity during anticipation of pain, compared to the actual experience of pain, in three brain regions. These included the medial frontal lobe of the brain (which contains the ACC), the anterior insular cortex (a part of the “island” of cortex deeply embedded in the cerebral cortex that is involved in subjective emotional processes), and the cerebellum (a structure in the back of the brain that regulates movement).

Differences in brain activity are also seen when the source of pain differs—for instance, when pain results from cold temperatures, a burn, or a cut. For example, in a paper published in 2002 in the journal Nature Neuroscience, an fMRI study was able to pin down the brain regions implicated in prickle pain evoked by intense cold, including the ACC, the somatosensory cortex and the insular cortex.

Another feature of pain is that it has psychological and social components, and these are also reflected in patterns of brain activity. For instance, a study published in the journal Science in 2004 looked at what happens in the brain during empathic pain, that is, when experiencing pain by observing the painful experience of a loved one. The study found that this affective (emotional) quality of pain depended upon brain activity in the anterior insular cortex and ACC, but not in other regions involved in processing the sensory aspects of pain. Similarly, these affective regions of the brain came into play during non-physical, hypnosis-induced pain where hypnotic suggestions were used to change how unpleasant pain feels.

Tracey and others have also imaged the brains of infants. To their surprise, they have found that the pattern of brain activity in response to a non-damaging pinprick in infants was more or less indistinguishable from that of adults. This suggests that the pain circuitry is developed and working from birth.

Brain imaging + machine learning
Over the past several decades, the resolution of brain imaging has improved significantly. But there are still approximately 5.5 million nerve cells represented by a single “blob” on a brain image. In order to identify and predict specific patterns of brain activity within the same “blob” that correspond to different aspects of pain, scientists began to use other ways of analyzing the data. The goal is to decipher what underlies the “blob” and to determine whether different patterns exist. One of these new methods of analysis involves machine learning; this is a process where a computer can automatically “learn” on its own as more data is fed into it. For Tracey and other pain researchers, this data often consists of brain images from many different pain studies.

A 2015 study published in the journal Science Translational Medicine showed the power of the machine learning approach in pain research. Scientists gathered data from multiple published studies that looked at changes in brain activity, in people with pain, in response to eight different medications known to relieve pain. Using machine learning and fMRI, they were able to identify patterns of brain activity in response to the drugs that correlated with how well the drugs worked. This may allow researchers to predict how well a new pain drug might work, by comparing the pattern of brain activity in response to the new drug to the pattern produced by the previously tested drugs. If the patterns are similar, the new drug may be more likely to be effective, making drug developers more confident in the drug’s prospects to treat pain.

A special role for the insula?
Tracey’s group also published a study in Nature Neuroscience in 2015 that identified a role for an area of the insula, called the dorsal posterior insula, during on-going or slowly varying pain—things that are ordinarily hard to image. To do this, they applied varying degrees of a heat pain stimulus to healthy volunteers who had capsaicin, the component of chili peppers that makes them hot, applied to their legs (to cause ongoing pain) and then recorded changes in brain activity over several hours.

They then looked for a brain region that was always active and whose strength of activity changed based on how intense the pain stimulus was. That turned out to be the dorsal posterior insula. The study was controversial, though, as many researchers interpreted the work as claiming that this particular brain area was the only area needed for pain, though Tracey and her colleagues said that this is not what they had stated.

Regardless, reducing activity in this brain region may relieve pain, and animal studies and patient work are ongoing to further verify the role of the dorsal posterior insula in pain and its potential as a therapeutic target.

“Good Cop” vs. “Bad Cop”
Once the brain receives information about potential harm to the body, it can then either facilitate or inhibit pain by signaling back down to the spinal cord. The ability of the brain to alter pain in this way is called descending pain modulation. During chronic pain, the balance between the “good cop” (inhibiting pain via descending modulation) versus “bad cop” (facilitating pain) goes awry.

For instance, a study published in 2018 in the journal Brain looked at painful diabetic neuropathy (pain resulting from nerve damage caused by diabetes). Here, Tracey’s group, working together with David Bennett at the University of Oxford, used brain imaging to reveal changes in the periaqueductal grey (PAG), which is a brain region important for descending pain modulation. They found that changes in the activity of the PAG were correlated with how much pain people with diabetic neuropathy experienced. This suggests that when the PAG does not function properly, this may favor the role of the “bad cop” in painful diabetic neuropathy.

Pain researchers like Tracey are also studying how to better harness the “good cop” to relieve pain. One example of this is the placebo effect, where a treatment with no active ingredients such as a “sugar pill” relieves pain. Tracey said that placebo hijacks the “good cop” to make people feel better. By studying how placebo does this, scientists can develop ways to enhance the “good cop” and correct the imbalance between pain facilitation and pain inhibition seen during chronic pain.

Over the years, brain imaging has made significant contributions to the understanding of pain. It is no longer a technique that merely gives scientists pretty and colorful images that cannot be explained or interpreted in any meaningful way. Instead, brain imaging is helping pain researchers understand the changes in the nervous system that accompany chronic pain, which will boost efforts to come up with new ways to treat it.

Chulmin Cho is a postdoctoral fellow at the University of Toronto Mississauga, Canada.