What if a simple, non-invasive brain scan could tell if someone would develop chronic pain after a medical procedure or not?
Now, a new study published in JAMA Neurology has identified brain biomarkers that accurately predict whether people will experience high and prolonged discomfort after a painful injection (1). The findings bring science closer to finding reliable predictors of chronic pain.
Chronic pain conditions affect more than 50 million Americans (2). Medicine’s inability to help has been a significant contributor to the opioid epidemic. Pain scientists have searched the body and brain for markers that anticipate chronic pain, as early intervention could be key to managing and preventing it.
This search has only been partly successful, said A. Vania Apkarian, a pain researcher at Northwestern University, who was not involved in the current study. Biomarkers derived from functional magnetic resonance imaging (fMRI) have provided some answers. Using the technique, said Apkarian, “one can find multiple parameters that are both associated with the state of chronic pain and are also predictive before the development of chronic pain.” But, the field has struggled to replicate these findings across multiple labs and pain states (3). Furthermore, fMRI is an expensive and restrictive technique.
For me, it was especially painful during eating.
-David Seminowicz, University of Western Ontario
Quick and reliable indicators of chronic pain would transform the field and patients’ lives. Senior author of the new study and a neuroscientist at the University of Western Ontario, David Seminowicz, recruited 150 volunteers for an unusual pain study to test a solution.
The researchers injected a molecule called nerve growth factor into their participants' jaws. The resulting pain lasts for several weeks. “For me, it was especially painful during eating,” said Seminowicz, who puts himself through all the pain interventions he studies. The injections are an ethical way of studying prolonged pain in healthy participants — a necessary step before involving patients.
Seminowicz and his colleagues measured two biomarkers in the study: peak alpha frequency (PAF) and corticomotor excitability (CME). These are brain activity measures recorded by electroencephalography (EEG) and transcranial magnetic stimulation (TMS), respectively. EEG, which records electrical signals through the scalp using a cap, and TMS, which uses magnetic fields to stimulate the brain, are easier and cheaper than functional brain imaging. What dimensions of chronic pain PAF and CME provide insight into is unclear, although the authors suggest that it captures aspects of pain processing in the brain and the subsequent motor response.
After injections, the participants in the study recorded the pain they felt when chewing and yawning every day for a month. Some volunteers reported an initial pain rating of 9 out of 10, whereas others reported just a 3 on the same scale. Average scores returned to 0 by day 30.
Logistic regression analysis based on PAF and CME predicted perfectly whether a patient would go on to have high or low pain over the next month in a small validation set of 100 participants. This prediction was based solely on measures recorded in the five days after injection.
The scientists then applied the prediction model to the remaining 50 participants — the test set. The model classified the participants correctly 88 percent of the time. “It's actually pretty excellent compared to other markers,” said Paul Geha, a neuroscientist at University of Rochester who was not involved with the study.
Seminowicz was excited by this performance but remained clear that results in a clinical population will be the real test. He is leading an ongoing clinical study to record PAF and CME in people diagnosed with the jaw pain condition temporomandibular disorder (TMD). “A good number of people, unfortunately, who experience acute jaw pain do go on to develop chronic pain TMD,” explained Seminowicz.
The two measures are sensitive to chronic pain states, but whether the changes recorded are specific to these states remains unclear. Apkarian said that PAF fluctuates all the time. “If I close my eyes, that peak power will dramatically change, and if I have a cup of coffee, it will change as well,” he explained. Geha said that further work could be useful to understand what PAF and CME record about chronic pain.
Seminowicz reasoned that the measures have already been tested in some other models of chronic pain, pointing to a small study that showed the same connections between PAF and post-surgical pain (4). Seminowicz thinks this area could benefit from new biomarkers. Patients at high risk of developing chronic pain might choose, for example, to turn down elective surgeries. “The procedure you're having is not worth developing chronic pain that you'll have potentially for the rest of your life,” he said.
References
- Chowdhury N.S. et al. Predicting individual pain sensitivity using a novel cortical biomarker signature. JAMA Neurol 82, 237-246 (2025).
- Katzman J.G. and Gallagher R.M. Pain: The silent public health epidemic. J Prim Care Community Health 15, (2024).
- Zhang Z. et al. Brain imaging biomarkers for chronic pain. Front Neurol 12, (2022).
- Millard S.K. et al. Predicting postoperative pain in lung cancer patients using preoperative peak alpha frequency. Br J Anaesth 128, E346-E348 (2022).
