1424: Mapping Central Nervous System Responses To Painful Stimuli In Patients With Axial Spondyloarthritis

Background/Purpose: Chronic pain is one of the common causes of disability in axial spondyloarthritis (axSpA). The mechanism behind this pain is not well understood, hindering the development of safe and efficacious treatments. The aim of this exploratory study is to examine differences in brain responses to painful stimuli between axSpA patients and healthy controls (HC).

Methods: Six axSpA patients and 6 age-and-sex matched HC underwent functional brain magnetic resonance imaging (fMRI) while tactile pressure to the thumbnail was applied: a light pressure (LSTIM) that is perceived but not painful, and a hard pressure (HSTIM) rated as a 1 pain, on a scale of 0-10 (Fig 1). A visual cue appeared 1.4 – 4.2 seconds prior to the stimulus, indicating a light or hard level of the subsequent stimulus. Analysis of fMRI data was exploratory due to small sample size. AxSpA and HC groups were combined to maximize power to detect signal changes associated with the pressure task. Areas with increased signal were used to create regions of interest (ROIs), which were then analyzed for group effects, using repeated measures mixed effects models.

Results: AxSpA patients were 50% females, had a mean age of 40 and mean ASDAS-CRP score of 2.7. The visual cue conditions produced increased signal throughout the brain, including areas in the occipitotemporal pathway, anterior insula, frontal operculum, medial superior frontal gyrus, and parietofrontal (PF) complex. Signal decreases were seen in areas typically associated with the default mode network (DMN). The tactile stimulation produced increased signal throughout the insula bilaterally, somatomotor areas along the central sulcus, fusiform gyrus, middle frontal gyrus, and the area at the PF junction (Fig 2). Decreased activity was seen in occipitotemporal areas, occipitoparietal areas, frontal pole, and precentral sulcus, bilaterally.

When comparing axSpA vs HC responses to LSTIM and HSTIM (within and between groups), there were statistically significant differences in 3 ROIs: the fusiform gyrus, the PF junction, and the frontal pole (Fig 3). At the level of the fusiform gyrus (involved in visual processing and pain anticipation), axSpA patients had a decreased signal over time in response to HSTIM compared to LSTIM, while the opposite was seen in HC. The PF junction, which is part of the salience network (SN), is involved in cognitive functions and pain processing. AxSpA patients had more pronounced response with HSTIM compared to LSTIM at the level of the PF complex compared with no difference in stimulus response in HC. The frontal pole, including the anterior cingulate cortex and medial prefrontal cortex, and part of the DMN, is involved in cognitive functions and modulating chronic pain. In the axSpA group, there was an increased signal over time in response to HSTIM compared to LSTIM; this signal was decreased in HC.

Conclusion: In this proof-of-concept exploratory study, patients with active axSpA had altered brain response to painful stimuli at regions encompassing SN and DMN, which have been previously implicated in chronic pain conditions. Further studies are planned to characterize central nervous system dysregulation and its role in chronic pain in larger cohorts of axSpA patients.