Background & Aims

Half of the people who suffer from a whiplash injury will develop chronic neck pain, broadly referred to as chronic whiplash-associated disorders (CWAD). A hallmark of CWAD is fear of neck movements, which contribute to persistent disability and is an important therapeutic target.[1] Fear of movement can be investigated by recalling movement-pain associations through imagination of painful movements. fMRI studies using imagination paradigms to examine fear of movement in people with chronic pain have reported abnormal neural responses in cortical and subcortical areas implicated in sensory, cognitive and affective processing.[2] In this study, we implemented a machine learning-based multivariate pattern analysis (MVPA) in the context of an RCT to 1) identify brain regions encoding fear of movement and, 2) test their response to Pain Neuroscience Education combined with cognition-targeted exercise (PNE+Exercise), a novel therapy that specifically targets fear of movement.[3]

Methods

This fMRI study is a sub-study (N = 63) of a RCT[4] comparing PNE+Exercise to physiotherapy (PT) in people with CWAD. fMRI data was collected at baseline and immediately post-therapy. In addition, 32 pain-free participants completed one fMRI session to serve as control for a baseline comparison. In the scanner, participants were presented with pictures of neck movements and control (neutral) movements and asked to imagine performing the movement. We applied a support vector machine (SVM) to train classifiers of Neck vs Neutral movements in CWAD and pain-free participants separately. Performance accuracy for each classifier was assessed with 10-fold cross-validation using a forced-choice test. We computed the spatial correlation between both classifiers at the whole-brain and parcel levels to examine their differences. Finally, we conducted an univariate analysis to identify treatment differences in brain activation at post-therapy while adjusting for baseline values at each voxel.

Results

The final baseline sample consisted of 60 CWAD and 30 pain-free participants. Forty-five CWAD participants (PNE+Exercise N=20; PT N=25) were retained at post-therapy. We found excellent accuracy for the classifier trained in CWAD participants (97±2.3%) and good accuracy in the classifier trained in pain-free controls (83±6.8%). The two classifiers were moderately spatially correlated at the whole-brain level (r=0.26). More specifically, both classifiers were very similar (r>0.5) in motor and visual regions, but were dissimilar (negative or near-zero correlations) in regions within limbic and attentional networks and key subcortical areas. We retrained the classifiers within these regions to confirm they contained meaningful task-related signal. Performance remained above chance in CWAD participants (92±3.6%), but not in pain-free controls (67±8.6%, p=.1). Thus, we interpret these regions as fear related. PNE+Exercise showed larger reductions than PT in fear-related measures. The univariate analysis revealed more post-therapy activation in multiple brain areas after PNE+Exercise group compared to PT, including in the amygdala, hippocampus and mid-cingulate among others. These regions were among those deemed as fear related in the baseline comparison between classifiers.

Conclusions

People with CWAD distinctly encode neck movements in key regions related to cognitive and affective processing, relative to pain-free controls. PNE+Exercise is superior to conventional physiotherapy in reducing fear of movement, and these effects are reflected in increased activity in brain regions that distinctly encode neck movement in people with CWAD.

References

1.Turk, D.C., et al., Assessing fear in patients with cervical pain: development and validation of the Pictorial Fear of Activity Scale-Cervical (PFActS-C). Pain, 2008. 139(1): p. 55-62.
2.Murillo, C., et al., Neural processing of pain-related distress to neck-specific movements in people with chronic whiplash-associated disorders. Pain, 2023. 164(9): p. 1954-1964.
3.Murillo, C., et al., Reductions in kinesiophobia and distress after pain neuroscience education and exercise lead to favourable outcomes: a secondary mediation analysis of a randomized controlled trial in primary care. Pain, 2023. 164(10): p. 2296-2305.
4.Coppieters, I., et al., A contemporary neuroscience approach compared to biomedically focused education combined with symptom-contingent exercise therapy in people with chronic whiplash associated disorders: a randomized controlled trial protocol. Brazilian Journal of Physical Therapy, 2021. 25(3): p. 356-366.

Presenting Author

Carlos Murillo Ezcurra

Poster Authors

Carlos Murillo Ezcurra

PHD

UNIVERSITY OF COLORADO ANSCHUTZ

Lead Author

E-mail

Inge Timmers

Phd

Tilburg University

Lead Author

E-mail

Marina López-Sola

Phd

University of Barcelona

Lead Author

E-mail

Barbara Cagnie

Phd

Ghent University

Lead Author

E-mail

Mira Meeus

PhD

University of Antwerp

Lead Author

E-mail

Yoni Ashar

PhD

University of Colorado Anschutz Medical Campus

Lead Author

E-mail

Topics

  • Pain Imaging