Share this article on:

fMRI Reveals Unique Neurologic ‘Signature’ of Physical Pain

Samson, Kurt

doi: 10.1097/01.NT.0000431660.24759.94
Back to Top | Article Outline




Based on functional MRI, investigators reported finding a distinct neurologic signature for thermal pain as distinguished from social pain. Experts say the signature could potentially establish a means for objectively quantifying pain independent of a patient's self-reported level of intensity.

Functional magnetic brain imaging of individuals subjected to graded levels of heat revealed a distinct pain “signature” across multiple brain regions.

In four studies described in an April 11 paper in the New England Journal of Medicine, investigators reported finding a distinct neurologic profile for thermal pain based on levels of significant activity — excitation and inhibition — in the bilateral dorsal posterior insula, the secondary somatosensory cortex, the anterior insula, the ventrolateral and medial thalamus, the hypothalamus, and the dorsal anterior cingulate cortex.

“The signature across multiple brain systems was specific to physical pain and independent of other pain, including emotional pain which has a similar but different signature,” said lead author Tor D. Wager, PhD, professor of psychology and neuroscience and director of the Cognitive and Affective Control Laboratory at the University of Colorado, Boulder.

The findings may lead one day to a means for objectively quantifying pain independent of a patient's self-reported level of intensity, offer new clues into how the brain generates and responds to different types of pain, and could help in developing methods of using brain scans to objectively measure anxiety, depression, anger, and other emotional states, Dr. Wager told Neurology Today in a telephone interview.

Back to Top | Article Outline


In the study — which was funded by the National Institute on Drug Abuse (NIDA), the National Institute of Mental Health, and the National Science Foundation — Dr. Wager and researchers at Johns Hopkins University, New York University, and the University of Michigan-Ann Arbor, examined functional MRI (fMRI) images of 114 brains taken when subjects were exposed to different levels of heat ranging from warm to very hot.

In the first of four experiments, participants were asked to self-report pain for four different intensities of heat, ranging from “innocuous” warmth — defined as 1 on a 1-9 visual analogue scale (VAS) — to three progressively hotter temperatures. Each trial consisted of a warning cue and anticipation period, stimulation, and a pain-recall and rating period, with periods of rest before and after recall. The fMRI signature showed 94 percent or greater sensitivity and specificity in discriminating painful heat from warmth, as well as pain anticipation and recall (95% confidence interval [CI], 89 to 98).

In the second experiment, participants were asked to rate pain intensity across six temperatures; ratings were coded from 0 to 99 for nonpainful events and from 100 to 200 for painful events. Both sensitivity and specificity of 93 percent were observed in the discrimination of pain versus no pain (95% CI, 84 to 100).

In the third experiment, investigators sought to assess specificity relative to social pain. Individuals who had recently experienced a romantic breakup were shown a picture of their former partner and their signature was compared to that of their thermal pain fMRI reading. They were also shown an image of a close friend, and the researchers compared “rejector” [for the romantic breakup] and “friend” versus responses to thermal temperatives.

The neurologic signature response was substantially stronger for physical pain, however, than for any of the other conditions (warmth, rejecter, or friend). It discriminated between physical pain and social pain with 85 percent sensitivity (95% CI, 76 to 94) and 73 percent specificity (95% CI, 61 to 84).

Finally, participants received two intravenous infusions of remifentanil, a short-acting opioid, during fMRI scanning in two series of trials. In the open-infusion series, participants knew they received remifentanil, and in the hidden-infusion series, they were told that no drug was delivered, even though it had been administered. During infusion the signature response was reduced in parallel with increases in the drug effect-site concentration. At the maximum drug concentration, remifentanil was associated with a reduction of 53 percent in the signature response, with no differences across open and hidden administration.

The scientists were surprised that the signature was specific to physical pain because past studies have shown that emotional pain can look very similar to physical pain in terms of the brain activity it produces. Yet the signature was absent when they showed pictures of former partners to “heartbroken” subjects who had recently lost that relationship.

“Right now, there's no clinically acceptable way to measure pain and other emotions other than to ask a person how they feel,” said Dr. Wager. “We found a pattern across multiple systems in the brain that is diagnostic of how much pain people feel in response to painful heat.” The team expected that if a pain “signature” could be found, it would likely be unique to each individual, but instead found that it was transferable across different subjects and different pain types, he told Neurology Today.

“If extended to the clinic, fMRI could be used to confirm pain in situations in which patients are unable to communicate pain effectively or when self-reports are otherwise suspect,” he said. “However, pain is a complex phenomenon and different people have different responses. This does give us a way to begin parsing out the brain systems driven by physical pain and break different acute types of pain into the various neurologic components affected. This is a beginning, not an end.

“The results do not yet allow us to quantify physical pain, but they lay the foundation for future work that could produce the first objective tests of pain by doctors and hospitals,” continued Dr. Wager. “This is not a pain detector. We can confirm pain in brain activity and discriminate between discomfort caused by other factors, but pain is subjective by definition. Self-reporting is important and must be considered because different people experience pain differently.”



In ongoing research, Dr. Wager and his colleagues have now tested the fMRI signature in mechanical pain and pain from electrical shocks, with similar success, he told Neurology Today in a telephone interview. They will also be studying inferred, chemical, ischemic, and laser pain.

Back to Top | Article Outline


In an accompanying editorial, Assia Jaillard, MD, PhD, of Grenoble University in Grenoble, France, and Allan H. Ropper, MD, of Brigham and Women's Hospital in Boston, said the findings point to the promise of neurimaging in pain research and the need for a means of subjectively measuring pain.

However, they also pointed to limitations mentioned in the papers. For example the studies only considered cutaneous pain, and not pain caused by disease. “They also do not shed light on the issue of chronic pain, one of the most vexing problems in general medicine,” they wrote.

Further, the use of a photograph of an ex-partner in the social pain study was an uncertain stimulus in terms of neural processes that are engaged, they wrote. “Participants in these studies may have experienced many feelings, including social rejection, love, or attachment, which lead to changes in the activity of reward centers in the brain.”

They also noted the low sensitivity of the 1.5-T fMRI system that was used for most testing only yields limited spacial resolution.“[T]his may have led to the misidentification of small deep-brain structures that contributed to the neurologic signature response for pain.”

Nonetheless, the investigators provide “an example of how functional neuroimaging may help clinicians assess clinical symptoms, such as somatic and emotional pain, that were previously thought to be impenetrable,” they wrote. “Being doctors, though, we may ultimately have to acknowledge that ‘pain is pain’ and can be reported only by the patient.”

“I think the major message is that you can do imaging to establish whether a patient is in pain or not in a presumably quantitative, non-subject-dependent manner,” said Costantino Iadecola, MD, professor of neurology and director of the Brain and Mind Research Institute at Weill Cornell Medical College in New York City. “Today, the only way we have of measuring pain is the 1-10 scale of self reporting by patients, which is very subjective, said Dr. Iadecola, who was not involved with the study.

The novelty of having a pain “signature” to help quantify level of pain in a non-subject dependent fashion would be very useful for pain researchers and for evaluating pain in patients who are “locked in” due to disorders like amyotrophic lateral sclerosis or brain stem stroke, where communication is difficult, he said.

However, in the clinical setting there are too many variables that come into play in the subjective experience of pain, so using neuroimaging to any great effect would be very difficult, he commented.

“Even if every patient underwent scanning, we still would not know what to do with their subjective rating of pain levels. Some will be higher than the MRI signature indicates, some lower, and some will be right on it, I think. Subjectivity is a major problem, as is sensitivity to pain changes, especially in patients with chronic pain.”

Unless every pain clinic had a fMRI available, he said the findings would be hard to translate into clinical reality. “This is a very interesting finding from a research standpoint, but in terms of clinical screening, I don't think it's ready for prime time.”

Back to Top | Article Outline


•. Wager TD, Atlas LY, Lindquist MA, et al. An fMRI-based neurologic signature of physical pain. N Engl J Med 368;15:1388–1397.
    •. Borsook D, Becerra L, Hargreaves R. A role for fMRI in optimizing CNS drug development. Nat Rev Drug Discov 2006;5:411–424.
      •. Brodersen KH, Wiech K, Lomakina EI, et al. Decoding the perception of pain from fMRI using multivariate pattern analysis. Neuroimage 2012;63:1162–1170.
        ©2013 American Academy of Neurology