Brain structure and function in chronic widespread pain: a UK Biobank study
Poster No:
2529
Submission Type:
Abstract Submission
Authors:
Valeria Oliva1, Aditya Banerjee1, Christine Law1, Dokyoung You1, Lola Falasinnu1, Yiyu Wang1, Gary Glover1, Sean Mackey1, Kenneth Weber2
Institutions:
1Stanford University, Palo Alto, CA, 2Stanford University, PALO ALTO, CA
First Author:
Co-Author(s):
Introduction:
Chronic widespread pain is a debilitating condition affecting up to 10–15% of the global population (1). Brain imaging studies, often with small sample sizes (n<50), have shown both functional and structural changes in the brain that are associated with chronic pain (2). Here we aimed to identify structural and functional brain changes associated with chronic widespread pain in the large, heterogeneous brain imaging dataset available from the UK Biobank (3).
Methods:
We identified 584 participants (377 females, 207 males, age = 60.9 ± 7.5 years) with chronic widespread pain (pain of >3 months duration within four or more body sites) at the initial imaging assessment as well as 584 (377 females, 207 males, age = 61.1 ± 7.3 years) pain-free controls that were matched based on sex, age, body mass index, imaging site, and date of imaging. T1-weighted structural (1 mm ✕ 1 mm ✕ 1 mm) and resting-state functional MRI (TR = 0.735, TE = 39ms, eyes-open, 6-minute duration) brain images were obtained. The UK Biobank's standard image processing pipeline was employed, which included the identification of the gray-white boundary surface and pial surface using FreeSurfer, automatic denoising of the resting-state functional images using FSL's FIX, and spatial normalization of the images (3). Additional preprocessing steps included spatial smoothing (10 mm3 FWHM for the cortical thickness maps and 3 mm3 FWHM for the resting-state functional images) and bandpass temporal filtering of the resting-state functional images (0.01 Hz – 0.10 Hz).
First, we performed mixed-effects between group comparisons in cortical thickness using FreeSurfer with cluster-based familywise error correction (cluster-forming threshold < 1.3, cluster-wise p-threshold < 0.05) (4). Next, to uncover the functional corollaries of the structural differences, we placed 5 mm3 spheres at four brain regions corresponding to the peaks of group differences in cortical thickness, and we extracted the first principal component of the preprocessed fMRI time series from each region for each participant. For each seed, subject-level whole-brain connectivity maps were generated in FSL using the extracted time series. Then we performed mixed-effects between group comparisons in functional connectivity using FSL's flame with cluster-based familywise error correction (cluster-forming threshold < 3.1, cluster-wise p-threshold<0.05) (5). Finally, we extracted the parameter estimates from regions showing significant differences in group-level functional connectivity and assessed whether the strength of functional connectivity was associated with the number of pain sites.
First, we performed mixed-effects between group comparisons in cortical thickness using FreeSurfer with cluster-based familywise error correction (cluster-forming threshold < 1.3, cluster-wise p-threshold < 0.05) (4). Next, to uncover the functional corollaries of the structural differences, we placed 5 mm3 spheres at four brain regions corresponding to the peaks of group differences in cortical thickness, and we extracted the first principal component of the preprocessed fMRI time series from each region for each participant. For each seed, subject-level whole-brain connectivity maps were generated in FSL using the extracted time series. Then we performed mixed-effects between group comparisons in functional connectivity using FSL's flame with cluster-based familywise error correction (cluster-forming threshold < 3.1, cluster-wise p-threshold<0.05) (5). Finally, we extracted the parameter estimates from regions showing significant differences in group-level functional connectivity and assessed whether the strength of functional connectivity was associated with the number of pain sites.
Results:
We identified areas of decreased cortical thickness in the left dorsal anterior cingulate, left precuneus, left inferior parietal, right lateral precentral, bilateral superior parietal, and bilateral occipital cortices in chronic widespread pain (Figure 1). Chronic widespread chronic pain demonstrated decreased functional connectivity between the left dorsal anterior cingulate cortex and the right dlPFC as well as decreased functional connectivity between the precuneus and the right supramarginal gyrus (Figure 2). Neither the right lateral precentral or superior parietal seeds showed significant differences in functional connectivity between the two groups. The strength of functional connectivity was not significantly correlated with the number of pain sites.
·Figure 1. Cortical thickness differences between chronic widespread pain participants (N = 584) and pain-free controls (N = 584, p < 0.05).
·Figure 2. Functional connectivity differences between chronic widespread pain participants (N = 584) and pain-free controls (N = 584, p < 0.001).
Conclusions:
The findings, drawn from a large heterogeneous dataset, reveal altered brain structure and function in chronic widespread pain compared to matched pain-free volunteers. The link between the brain changes and clinical measures remains to be investigated and will require datasets with richer clinical information including pain severity, pain duration, and pain-related disability.
Modeling and Analysis Methods:
fMRI Connectivity and Network Modeling
Univariate Modeling
Neuroanatomy, Physiology, Metabolism and Neurotransmission:
Cortical Anatomy and Brain Mapping 2
Perception, Attention and Motor Behavior:
Perception: Pain and Visceral 1
Keywords:
ADULTS
Cortex
DISORDERS
FUNCTIONAL MRI
Pain
STRUCTURAL MRI
Univariate
1|2Indicates the priority used for review
Provide references using author date format
2. Martucci KT and Mackey SC. Anesthesiology. 2018 Jun;128(6):1241-1254.
3. Alfaro-Almagro F et al. Neuroimage. 2018 Feb 1;166:400-424.
4. Fischl B. Neuroimage. 2012 Aug 15;62(2):774-81.
5. Jenkinson M et al. Neuroimage. 2012 Aug 15;62(2):782-90.