Journal Logo

Research Paper

Chronic noncancer pain is not associated with accelerated brain aging as assessed by structural magnetic resonance imaging in patients treated in specialized outpatient clinics

Sörös, Petera,b,*; Bantel, Carstenb,c

Author Information
doi: 10.1097/j.pain.0000000000001756
  • Free
  • C

1. Introduction

Chronic pain is conceptualized as pain lasting beyond regular tissue healing time without the warning and protection function of physiological nociception.92 Chronic pain is a major health care problem affecting approximately 20% of the adult population14 and is associated with changes in cognitive and emotional processing,15 such as impairments of executive functions,9 difficulties with attention,68 and number processing,87,99 as well as an increased risk of depression.14

Although the neurophysiology of nociception is well known, the pathophysiology of chronic pain is far from evident. Acute nociceptive input quickly reshapes neural circuits on the level of the peripheral and central nervous system, such as the C fiber,55 the spinal dorsal horn,10 and the sensorimotor cortex.26,86

In several types of chronic pain, structural alterations of the brain have been demonstrated, starting with the observation that prefrontal and thalamic gray matter was decreased in chronic back pain patients.4 A decrease of gray matter has also been found in patients with chronic tension-type headache,80 migraine,78,95 episodic cluster headache,1 irritable bowel syndrome,27,83 osteoarthritis,44 and chronic complex regional pain syndrome.41 By contrast, other studies found increased gray matter in patients with chronic back pain (basal ganglia and thalamus),79 rheumatoid arthritis (basal ganglia),98 and chronic vulvar pain (hippocampus, parahippocampal gyrus, and basal ganglia).82

In coordinate-based quantitative meta-analyses across various types of chronic pain, several cortical and subcortical areas of decreased and increased gray matter were identified.17,84 In a smaller number of studies, changes in cerebral white matter have also been investigated.52 In patients with chronic musculoskeletal pain, lower fractional anisotropy, indicating impaired integrity of white fiber tracts, has been found in the corpus callosum and the cingulum.57

The functional significance of gray and white matter changes in patients with chronic pain is still unresolved.84 Whether these changes are cause or consequence76 of chronic pain and whether they represent predisposing factors or compensatory neuroplasticity63 is under debate. Moreover, the neuropathological substrates of these changes are not firmly established. Animal research suggests that rapid changes in synaptic spines and glia cells, as well as neuronal loss and, conversely, neurogenesis contribute to macroscopic changes in chronic pain.54

A decrease of total and regional gray matter is one of the hallmarks of brain aging, even in the absence of brain disease.35,88 Previously, 2 publications suggested that patients with chronic pain due to fibromyalgia and temporomandibular disorder experience accelerated or premature brain aging.53,67 Very recently, a study on elderly (>60 years) adults concluded that the brains of participants with chronic pain seemed older compared to participants without pain, based on an established machine learning approach.23 Here, we test the hypothesis that patients with chronic noncancer pain demonstrate accelerated brain aging, compared to healthy participants, using T1-weighted structural MR images and machine learning,18 in a sample of 119 participants aged between 30 and 68 years.

2. Methods

2.1. Participants

This study analyzes structural T1-weighted MR images obtained from 119 participants of the ChroPain187 and ChroPain2 studies. The combined sample comprises 59 participants with chronic noncancer pain and 60 participants free of chronic pain. Patients for ChroPain1 (n = 36) were recruited from the Pain Outpatient Clinic, University Clinic of Anesthesiology, Critical Care, Emergency Medicine, and Pain Management, Klinikum Oldenburg, located in Oldenburg, Germany. Additional patients were found through advertisements in the local daily newspaper. Patients for ChroPain2 (n = 23) were recruited at a specialized rheumatology practice in Oldenburg, Germany (Dr Markus Voglau, https://www.rheuma-oldenburg.de), and through the local ankylosing spondylitis support group. For every patient with chronic pain, a sex- and age-matched (±5 years) pain-free control participant was recruited. Control participants for ChroPain1 (n = 37) and ChroPain2 (n = 23) were identified with the help of the local newspaper, the University of Oldenburg's web page, flyers, and personal communication.

For the definition of chronic pain, we used a purely temporal definition, following the suggestions of the International Association for the Study of Pain Classification of Chronic Pain.93 The International Association for the Study of Pain defines chronic pain as pain that lasts or recurs for >3 months.93 Because we hypothesized that structural and functional changes of the brain will be more pronounced in patients with longer pain duration, we extended the minimum duration of pain to 12 months in our study. Inclusion criterion for the pain group was thus persistent pain on most days of a month for ≥12 months of mild to severe intensity. This was ascertained by reviewing clinical notes and through direct patient interviews.

Exclusion criteria for the chronic pain and the healthy control groups were as follows: neurological disorders (such as dementia, Parkinson disease, stroke, epilepsy, multiple sclerosis, traumatic brain injury, and migraine), psychiatric disorders (such as schizophrenia or major depression), substance abuse, impaired kidney or liver function, and cancer. Since some of the cognitive tests applied in the ChroPain studies depend on cultural background,87 only participants born and raised in Germany were included.

All participants provided written informed consent for participation in one of the studies. A compensation of 10 € per hour was provided. Both studies were approved by the Medical Research Ethics Board, University of Oldenburg, Germany (# 25/2015; 2017-059). ChroPain2 has been preregistered with the German Clinical Trials Register (https://www.drks.de; DRKS00012791).

2.2. Demographic and clinical data

In a structured interview, every participant was asked for the date of birth, sex, handedness, and highest degree of formal education. In addition, previous and present conditions that may lead to an exclusion from the study and complete medication records were inquired. Smoking and alcohol consumption were also recorded.

Chronic pain patients were asked to estimate the average pain intensity that they had perceived during the last 24 hours before the magnetic resonance imaging (MRI) examination by using the 11-point numerical rating scale, with 0 representing “no pain” and 10 “worst pain imaginable.”13 In addition, pain duration in years was noted.

2.3. Verbal intelligence

A standardized German vocabulary test (Wortschatztest, WST) was used to assess verbal intelligence.81 Participants were required to identify, in each of the 42 rows, an existing German word within 5 nonwords. The number of correct choices was transformed into IQ scores according to the test manual.81 WST results are highly correlated with level of education and general intelligence.77

2.4. Depression

The short German version (Allgemeine Depressionsskala, ADS-K)45 of the Center for Epidemiological Studies Depression Scale74 was used to quantify depressive symptoms. The scale consists of 15 items assessing depressive symptoms during the preceding week. Each item is answered on a 4-point Likert scale: “never or rarely” (<1 day), “sometimes” (1-2 days), “often” (3-4 days), and “always” (5-7 days of the week). The total score ranges from 0 to 45. A score of ≥ 18 supports the diagnosis of a clinically relevant depression.45

2.5. Magnetic resonance imaging

All MR images were acquired on a research-only Siemens MAGNETOM Prisma (Siemens, Erlangen, Germany) whole-body scanner at 3 T with a 64-channel head/neck coil. The scanner is located at the Neuroimaging Unit, School of Medicine and Health Sciences, University of Oldenburg, Germany. A 3-dimensional T1-weighted MPRAGE sequence was used.12 Imaging parameters for ChroPain1 were: TR: 2000 ms, TE: 2.4 ms, voxel dimensions: 0.7 × 0.7 mm2, slice thickness: 0.9 mm, 208 axial slices. Imaging parameters for ChroPain2 were: TR: 2000 ms, TE: 2 ms, voxel dimensions: 0.75 × 0.75 mm2, slice thickness: 0.75 mm, 320 axial slices. Both sequences used in-plane acceleration (GRAPPA) with a PAD factor of 2.59 Siemens prescan normalization filter was used in both sequences for on-line compensation of regional signal inhomogeneities.50

2.6. Brain age prediction

MR images were first converted from Siemens DICOM format to uncompressed NIfTI (.nii) format using dcm2niix (https://github.com/rordenlab/dcm2niix). Structural MR images were then analyzed with the software brainageR (version 1.0, 09 August 2018; https://github.com/james-cole/brainageR).20,21 For additional methodological details, see the Supplementary Information in Cole et al. 2017 and 2018.20,21

brainageR uses SPM12 (University College London, London, United Kingdom; https://www.fil.ion.ucl.ac.uk/spm/software/spm12/) for image preprocessing. Using SPM12's Segment function,7 images were bias-field corrected and segmented into gray matter, white matter, and cerebrospinal fluid. Gray matter and white matter images were nonlinearly registered to a custom template, derived from the training data set (see below), using SPM12's DARTEL toolbox.5 Using FSL's slicesdir script (https://fsl.fmrib.ox.ac.uk/fsl/fslwiki/Miscvis), default slices in sagittal, coronal, and axial directions were created for the segmented and normalized brain images. With these images, visual quality control was performed to ensure accurate image segmentation (Fig. 1).

F1
Figure 1.:
Original T1-weighted MR images (A and D), preprocessed gray matter (B and E), and preprocessed white matter images (C and F) of a 30-year-old (upper row) and a 66-year-old pain-free control participant (lower row). Original images are in individual space, and preprocessed images were normalized to a brain template. Preprocessed gray and white matter images were then vectorized and concatenated for brain age prediction.

The individual segmented and normalized images were loaded into R (https://www.r-project.org)73 and vectorized. Gray and white matter vectors were combined. Brain age was then predicted with a Gaussian processes regression75 using R's kernlab (kernel-based machine learning lab) package (version 0.9-27; https://cran.r-project.org/web/packages/kernlab/index.html)49 and the previously learned regression model.

The brainageR model was trained on 2001 healthy individuals from several publicly available data sets. The following studies contributed more than 200 T1-weighted images each: the Information Extracted from Medical Images (IXI) database (https://brain-development.org/ixi-dataset), the International Consortium for Brain Mapping study (https://ida.loni.usc.edu),66 and the Open Access Series of Imaging Studies (OASIS-1; http://www.oasis-brains.org).64 For details about all studies used for the training data set, see Supplementary Information in Cole et al. 2017 and 2018.20,21 For further statistical analysis, brain age delta was calculated (predicted brain age minus chronological age).

2.7. Statistical analyses

To check for normality, a Shapiro–Wilk test was performed. To check for equality of variances, a Levene test was done. To compare the mean values of brain age delta in both groups, a traditional independent-samples Welch t test was performed30; statistical significance was set at P < 0.05. We also performed a subgroup analysis with patients and controls aged 60 years and older using Welch t test. In addition, a Bayesian independent-samples t test was calculated to compare the mean values of brain age delta in both groups using the default Cauchy prior centered on zero and with r = 0.707.61 To compare further characteristics of the chronic pain and the pain-free control group (Table 1), the Welch t test (for continuous data) and Pearson χ2 test (for categorical data) were performed.

T1
Table 1:
Demographic and clinical characteristics of participants with and without chronic pain.

To study the association between chronic pain and brain age delta, controlling for the influences of chronological age and sex, multiple linear regression analysis was calculated with group (chronic pain patients vs pain-free controls), age, and sex as regressors (Table 2). To study the association between pain duration and brain age delta within the group of chronic pain patients, controlling for pain intensity, age, and sex, another multiple linear regression analysis was calculated with pain duration, pain intensity, age, and sex as regressors (Table 3). In both analyses, bootstrapped confidence intervals for the regression coefficients were calculated (10,000 replications, α = 0.05). A regression coefficient is regarded to be statistically significant if the confidence interval does not include zero.

T2
Table 2:
Summaries of multiple linear regression analysis and relative weight analysis for chronic pain predicting brain age delta, adjusting for age and sex, in all participants (n = 119).
T3
Table 3:
Summaries of multiple linear regression analysis and relative weight analysis for pain duration predicting brain age delta, adjusting for pain intensity, age, and sex, in patients with chronic pain (n = 59).

As regressors may be correlated, a relative weight analysis47 was performed to supplement multiple linear regression. Relative weight analysis decomposes the total variance predicted in a regression model (R2) into weights that reflect the proportional contribution of the included regressors.90 To test for statistical significance, bootstrapped confidence intervals for relative weights were calculated (10,000 replications, α = 0.05). A weight is considered to be statistically significant if this confidence interval does not contain zero.91

For statistical testing (except relative weight analysis), JASP version 0.10.2 (https://jasp-stats.org) was used.60 Relative weight analysis was performed using RWA-WEB (https://relativeimportance.davidson.edu/multipleregression.html), a web-based interface for a program written in R.73

Individual brain age delta data (https://osf.io/2xd7h/) and the results of statistical analysis (https://osf.io/dqpb5/) have been published at the Open Science Framework.

3. Results

3.1. Demographic, clinical, and neuropsychological data

Demographic, clinical, and neuropsychological characteristics of the pain-free control group and the chronic pain group are presented in Table 1. Both groups were carefully matched for chronological age and sex. Formal education was significantly different between both groups, with chronic pain patients having lower levels of education. Moreover, verbal IQ was on average significantly lower in the pain group. Depression scores on the German version of the Center for Epidemiological Studies Depression were significantly higher in the pain group. Smoking habits were not significantly different between chronic pain patients and pain-free controls. Regular alcohol consumption (at least one alcoholic drink on most days of the week) was significantly more common in chronic pain patients (80%) than in pain-free controls (67%). All participants were Caucasians born and raised in Germany.

All patients with chronic pain were seen in a specialized pain clinic or by a specialist in rheumatology on a regular basis. In the chronic pain group, mean pain duration was 16 ± 11 years. On average, chronic pain patients reported a moderate level of pain (5 ± 2 on the numerical rating scale) during the 24 hours preceding the MRI. In the ChroPain1 study, the primary pain diagnoses were degenerative spinal disease (n = 22), degenerative joint disease (n = 5), abdominal pain (n = 2), and fibromyalgia (n = 7). In the ChroPain2 study, the primary pain diagnoses were rheumatoid arthritis (n = 11), ankylosing spondylitis (n = 9), psoriasis arthritis (n = 2), and CREST syndrome (n = 1). Patients of the ChroPain1 study, primarily recruited through a specialized pain clinic, received mainly opioid and nonopioid analgesics and antidepressants. Patients of the ChroPain2 study, suffering from rheumatic disease, took primarily disease-modifying antirheumatic drugs and biologics and, to a lesser extent, nonsteroidal anti-rheumatic drugs or steroids.

3.2. Association between chronological age and brain age

Figure 2A illustrates the relationship between the chronological age of each participant and the brain age predicted by brainageR (red: chronic pain patients, blue: pain-free controls).

F2
Figure 2.:
(A) Relationship between chronological age and predicted brain age. Blue circles represent pain-free control participants (n = 60), and red circles represent patients with chronic pain (n = 59). (B) Brain age delta (predicted brain age minus chronological age) for pain-free controls (blue) and patients with chronic pain (red). The error bars symbolize mean ± SD.

3.3. Comparison of brain age delta between groups

Figure 2B shows the distribution of brain age delta in the pain-free control and chronic pain groups. The results of the Shapiro–Wilk test did not suggest deviation from normality in neither group (W = 0.983, P = 0.543 for the control group; W = 0.967, P = 0.104 for the chronic pain group). The Levene test suggested equality of variances (F = 2.227, P = 0.138). Group mean values of brain age delta were not significantly different (Welch independent-samples t test; t (111.1) = 0.333, P = 0.74, Cohen's d = 0.061).

A Bayesian independent-samples t test indicated moderate evidence in favor of the null hypothesis (BF01 = 4.875; Fig. 3).

F3
Figure 3.:
(A) Bayesian independent samples t test for the effect size δ. The dashed line illustrates the prior distribution (default Cauchy prior centered on zero, r = 0.707), the solid line shows the posterior distribution. The 2 gray dots indicate the prior and posterior density at the test value. The probability wheel on top visualizes the evidence that the data provide for the null hypothesis (H0: effect sizes are equal) and the alternative hypothesis (auburn, H1: effect sizes are different). The median and the 95% central credible interval of the posterior distribution are shown in the top right corner. (B) The Bayes factor robustness plot. The plot indicates the Bayes factor BF01 (in favor of the null hypothesis) for the default prior (r = 0.707), a wide prior (r = 1), and an ultrawide prior (r = 1.414). All priors suggest moderate evidence for the null hypothesis, which is relatively stable across a wide range of prior distributions. Plots taken from JASP; the reporting of the results follows van Doorn et al.96

In a separate analysis that included only participants aged 60 years and above, mean brain age delta was −6.2 ± 7.9 years in pain-free participants (n = 16) and −8.7 ± 5.7 years in chronic pain patients (n = 14; Welch t test, P = 0.327, Cohen's d = 0.362).

3.4. Association between brain age delta and chronic pain and pain duration

A multiple linear regression was calculated for group predicting brain age delta, while adjusting for age and sex (n = 119, Table 2). In the traditional regression and in the relative weight analysis, the association between group and brain age delta was not significant.

In the group of chronic pain patients (n = 59), another multiple linear regression was calculated for pain duration predicting brain age delta, adjusting for pain intensity, age, and sex (Table 3). In the traditional regression and in the relative weight analysis, the association between pain duration and brain age delta was not significant.

4. Discussion

This study of 59 chronic noncancer pain patients and 60 healthy, age- and sex-matched controls with a mean age of 53 years provided no evidence for the hypothesis that chronic pain is associated with accelerated brain aging. In fact, a Bayesian independent-samples t test provided moderate evidence for the null hypothesis (ie, group mean values were equal).

Physiological aging is characterized by an inevitable—but also highly variable—decline in cognitive, motor, and sensory functions and also by changes of gray and white matter structure.71 In cross-sectional studies of healthy aging, voxel-based morphometry6 and surface-based morphometry25 have demonstrated a decrease of cortical gray matter density89 and thickness,34 predominantly in prefrontal and temporal areas. In addition, changes of white matter fiber tracts, as assessed by diffusion tensor imaging, have been investigated across age span. In a longitudinal study on participants aged between 50 and 90 years at baseline and with a follow-up after 2 years, a decline of white matter integrity was seen throughout the brain.8

In recent years, several approaches have been developed to determine the degree of physiological or pathological aging of individual adults by predicting individual brain age.19–21,39,56,58,97 These studies have used individual structural MR images of the brain, large MRI training data sets of healthy individuals, and machine learning methods to compare individual brains with the training data (for reviews, see Refs. 18,38). The results of these studies have been shown to be accurate and reliable.18 Accelerated brain aging has been demonstrated in patients with Alzheimer's disease,37 after traumatic brain injury,19 in HIV-positive individuals,22 and in psychiatric disease, such as schizophrenia, major depression, and borderline personality disorder.51 These methods are able to successfully predict conversion from mild cognitive impairment to Alzheimer's disease.40 Moreover, in a group of 669 elderly participants with a mean age of 73 years (the Lothian Birth Cohort 193628), increased brain age delta was associated with decreased physical fitness (weaker grip strength, poorer lung function, and slower walking speed), lower fluid intelligence, and increased mortality risk.21

Of note, brain age delta results derived by these methods are age dependent; brain age of younger participants is overestimated and brain age of older participants is underestimated.56 This inherent bias is driven by a regression toward the age mean of the training data set because the error in the regression model is not orthogonal to age.56,85 This explains why mean brain age delta was approximately −5 years in both groups of our study, where most participants were older than 40 years. Because we recruited a thoroughly age-matched control group for our chronic pain patients, this underestimation of individual brain age does not weaken our conclusion that individual brain age is similar in chronic pain patients and in pain-free controls.

Very recently, Cruz-Almeida and coworkers published a comparison of brain age delta between 33 elderly individuals with chronic pain (mean age: 70.6 ± 5.5 years) and 14 individuals without chronic pain (mean age: 71.5 ± 7.3 years). Although the mean predicted brain age was smaller than the chronological brain age in both groups (P = 0.592), participants with chronic pain had a larger brain age delta after adjustment for chronological age, sex, and exercise levels (analysis of covariance, F [1, 41] = 4.9; P = 0.033).23 There are major differences between Cruz-Almeida et al.'s study and our study. Importantly, our study is characterized by a considerably larger sample size (119 vs 47 participants), a younger age (the majority of participants are between 30 and 60 years of age), and a longer pain duration (15.9 ± 11 vs 6.3 ± 8.8 years). Comparing only participants aged 60 years or older in our sample did not result in a significant difference of brain age delta between groups and did not provide evidence for the notion that chronic pain accelerates brain aging in seniors only. Of note, Cruz-Almeida et al. investigated community dwelling individuals representative of the aging population, who were not necessarily treated for chronic pain. When comparing brain age delta between participants who received treatment for pain (including self-remedies) during the preceding 3 months and those who had no treatment, brain age delta was significantly smaller in the treatment group, ie, the brains of participants seeking treatment seemed younger.23 In Cruz-Almeida et al.'s study, brain age delta of participants without chronic pain (−4.0 ± 1.9 years) and participants with chronic pain seeking treatment (−3.9 ± 1.5 years) was very similar. These results closely resemble the findings in our study, where all patients with chronic pain received specialized care.

Our results have important implications for the pathogenesis of structural alterations of the brain and, ultimately, cognitive deficits in patients with chronic pain. Our results suggest that chronic pain does not induce widespread neural and glial degeneration, presumably the leading cause of age-related structural brain changes.69 By contrast, our results indirectly support recent alternative models of regional, network-specific structural and functional brain alterations in chronic pain.17,33 These models also suggest that the frequently observed cognitive deficits in chronic pain are the direct consequence of persistent nociceptive input, mediated by the aforementioned network-specific structural and functional changes, rather than the result of generalized accelerated aging of the brain.

Unfortunately, neuroimaging of the human brain, at least at 3 T, does not allow to identify the molecular or cellular mechanisms of chronic pain. Growing evidence predominantly from animal studies, however, suggests that inflammation of the central and peripheral nervous system contributes to the development of chronic pain and that novel anti-inflammatory therapies may improve chronic pain.16,46,100 Basic research for instance has shown that neuropathic pain is characterized by an ongoing neuroinflammation and is associated with the release of proinflammatory cytokines, such as interleukin 1β.3,29 Interestingly, healthy aging is also commonly associated with inflammation in the central nervous system.31 Several studies have reported elevated levels of inflammatory cytokines in the brains of aged rodents.43,62,101 Because aging seems to be associated with systemic augmentation of inflammation as well,36 it has been hypothesized that circulating inflammatory mediators prime microglia in the aging brain.32,65,70,72 These neuroinflammatory findings in chronic pain and in aging may suggest that both conditions share pathogenetic mechanisms on the molecular and cellular level.

Our study has important limitations. Chronic pain patients included here had heterogeneous diagnoses and were taken from 2 related, but separate studies. In particular, the effect of rheumatic diseases, generalized autoimmune disorders, on the brain is not well understood.48,98 Moreover, although we were able to match for age and sex, chronic pain patients and pain-free controls differed in terms of education, verbal IQ, depressive symptoms, and alcohol consumption. These differences reflect lower levels of education11 and higher incidence of depression14,24,94 of individuals with chronic pain observed in the general population.

Our study sample was also relatively young with a mean age of 53 years. We cannot rule out the possibility that pain-related changes of the brain mainly develop in higher age, eg, because of a decrease in brain plasticity and an impaired capacity to adapt to disease-specific structural changes. We performed a subgroup analysis with 16 pain-free participants and 14 chronic pain patients aged 60 years and older, and could not find significant differences in brain age delta. Due to the small sample sizes and the heterogeneity of both groups, this analysis has to be interpreted with caution.

Because all our patients received specialized medical care, we were unable to compare our results with chronic pain patients who do not seek treatment. In future studies, the relationship between chronic pain, brain morphology, medical treatment, and motivation for treatment should be investigated in greater detail. In addition, future studies should relate the extent of physical activity and exercise, which may relieve pain and improve pain-related functional impairments,2,42 to changes of brain morphology.

Conflict of interest statement

The authors have no conflicts of interest to declare.

Acknowledgments

A previous version of this article has been published through the biorxiv preprint server (https://doi.org/10.1101/627935).

The authors thank Dr James Cole, King's College London, London, United Kingdom, for his help with setting up the brainageR software and interpreting the results. The authors also thank Katharina Koch and Melanie Spindler for their active support of the ChroPain1 and ChroPain2 studies, including recruitment, data acquisition, and data analysis. The authors are grateful for the help of Dr Markus Voglau and his team and of Mara Schier and the Selbsthilfegruppe Morbus Bechterew Oldenburg during recruitment for the ChroPain2 study. The authors also thank Katharina Grote and Gülsen Yanc for assisting with MRI data acquisition. The authors appreciate the very constructive comments by two anonymous reviewers and by Kathryn Beard on the manuscript. This work was supported by the Neuroimaging Unit, University of Oldenburg, funded by grants from the German Research Foundation (DFG; 3T MRI INST 184/152-1 FUGG and MEG INST 184/148-1 FUGG). The ChroPain1 study was also funded by an intramural grant from the School of Medicine and Health Sciences, University of Oldenburg, to C. Bantel (Forschungspool, 2015-1).

References

[1]. Absinta M, Rocca MA, Colombo B, Falini A, Comi G, Filippi M. Selective decreased grey matter volume of the pain-matrix network in cluster headache. Cephalalgia 2012;32:109–15.
[2]. Ambrose KR, Golightly YM. Physical exercise as non-pharmacological treatment of chronic pain: why and when. Best Pract Res Clin Rheumatol 2015;29:120–30.
[3]. Apkarian AV, Lavarello S, Randolf A, Berra HH, Chialvo DR, Besedovsky HO, del Rey A. Expression of IL-1beta in supraspinal brain regions in rats with neuropathic pain. Neurosci Lett 2006;407:176–81.
[4]. Apkarian AV, Sosa Y, Sonty S, Levy RM, Harden RN, Parrish TB, Gitelman DR. Chronic back pain is associated with decreased prefrontal and thalamic gray matter density. J Neurosci 2004;24:10410–15.
[5]. Ashburner J. A fast diffeomorphic image registration algorithm. Neuroimage 2007;38:95–113.
[6]. Ashburner J, Friston KJ. Voxel-based morphometry—the methods. Neuroimage 2000;11:805–21.
[7]. Ashburner J, Friston KJ. Unified segmentation. Neuroimage 2005;26:839–51.
[8]. Barrick TR, Charlton RA, Clark CA, Markus HS. White matter structural decline in normal ageing: a prospective longitudinal study using tract-based spatial statistics. Neuroimage 2010;51:565–77.
[9]. Berryman C, Stanton TR, Bowering KJ, Tabor A, McFarlane A, Moseley GL. Do people with chronic pain have impaired executive function? A meta-analytical review. Clin Psychol Rev 2014;34:563–79.
[10]. Bird GC, Han JS, Fu Y, Adwanikar H, Willis WD, Neugebauer V. Pain-related synaptic plasticity in spinal dorsal horn neurons: role of CGRP. Mol Pain 2006;2:31.
[11]. Blyth FM, March LM, Brnabic AJ, Jorm LR, Williamson M, Cousins MJ. Chronic pain in Australia: a prevalence study. PAIN 2001;89:127–34.
[12]. Brant-Zawadzki M, Gillan GD, Nitz WR. A three-dimensional, T1-weighted, gradient-echo sequence-initial experience in the brain. Radiology 1992;182:769–75.
[13]. Breivik H, Borchgrevink PC, Allen SM, Rosseland LA, Romundstad L, Hals EK, Kvarstein G, Stubhaug A. Assessment of pain. Br J Anaesth 2008;101:17–24.
[14]. Breivik H, Collett B, Ventafridda V, Cohen R, Gallacher D. Survey of chronic pain in Europe: prevalence, impact on daily life, and treatment. Eur J Pain 2006;10:287–333.
[15]. Bushnell MC, Ceko M, Low LA. Cognitive and emotional control of pain and its disruption in chronic pain. Nat Rev Neurosci 2013;14:502–11.
[16]. Caraci F, Merlo S, Drago F, Caruso G, Parenti C, Sortino MA. Rescue of noradrenergic system as a novel pharmacological strategy in the treatment of chronic pain: focus on microglia activation. Front Pharmacol 2019;10:1024.
[17]. Cauda F, Palermo S, Costa T, Torta R, Duca S, Vercelli U, Geminiani G, Torta DM. Gray matter alterations in chronic pain: a network-oriented meta-analytic approach. Neuroimage Clin 2014;4:676–86.
[18]. Cole JH, Franke K. Predicting age using neuroimaging: innovative brain ageing biomarkers. Trends Neurosci 2017;40:681–90.
[19]. Cole JH, Leech R, Sharp DJ; Alzheimer's Disease Neuroimaging Initiative. Prediction of brain age suggests accelerated atrophy after traumatic brain injury. Ann Neurol 2015;77:571–81.
[20]. Cole JH, Poudel RPK, Tsagkrasoulis D, Caan MWA, Steves C, Spector TD, Montana G. Predicting brain age with deep learning from raw imaging data results in a reliable and heritable biomarker. Neuroimage 2017;163:115–24.
[21]. Cole JH, Ritchie SJ, Bastin ME, Valdés Hernández MC, Muñoz Maniega S, Royle N, Corley J, Pattie A, Harris SE, Zhang Q, Wray NR, Redmond P, Marioni RE, Starr JM, Cox SR, Wardlaw JM, Sharp DJ, Deary IJ. Brain age predicts mortality. Mol Psychiatry 2018;23:1385–92.
[22]. Cole JH, Underwood J, Caan MW, De Francesco D, van Zoest RA, Leech R, Wit FW, Portegies P, Geurtsen GJ, Schmand BA, Schim van der Loeff MF, Franceschi C, Sabin CA, Majoie CB, Winston A, Reiss P, Sharp DJ, COBRA C. Increased brain-predicted aging in treated HIV disease. Neurology 2017;88:1349–57.
[23]. Cruz-Almeida Y, Fillingim RB, Riley JL, Woods AJ, Porges E, Cohen R, Cole J. Chronic pain is associated with a brain aging biomarker in community-dwelling older adults. PAIN 2019;160:1119–30.
[24]. Currie SR, Wang J. Chronic back pain and major depression in the general Canadian population. PAIN 2004;107:54–60.
[25]. Dale AM, Fischl B, Sereno MI. Cortical surface-based analysis. I. Segmentation and surface reconstruction. Neuroimage 1999;9:179–94.
[26]. Dancey E, Murphy B, Srbely J, Yielder P. The effect of experimental pain on motor training performance and sensorimotor integration. Exp Brain Res 2014;232:2879–89.
[27]. Davis KD, Pope G, Chen J, Kwan CL, Crawley AP, Diamant NE. Cortical thinning in IBS: implications for homeostatic, attention, and pain processing. Neurology 2008;70:153–4.
[28]. Deary IJ, Gow AJ, Taylor MD, Corley J, Brett C, Wilson V, Campbell H, Whalley LJ, Visscher PM, Porteous DJ, Starr JM. The Lothian Birth Cohort 1936: a study to examine influences on cognitive ageing from age 11 to age 70 and beyond. BMC Geriatr 2007;7:28.
[29]. del Rey A, Apkarian AV, Martina M, Besedovsky HO. Chronic neuropathic pain-like behavior and brain-borne IL-1β. Ann N Y Acad Sci 2012;1262:101–7.
[30]. Delacre M, Lakens D, Leys C. Why psychologists should by default use Welch's t-test instead of Student's t-test. Int Rev Social Psychol 2017;30:92–101.
[31]. Di Benedetto S, Müller L, Wenger E, Düzel S, Pawelec G. Contribution of neuroinflammation and immunity to brain aging and the mitigating effects of physical and cognitive interventions. Neurosci Biobehav Rev 2017;75:114–28.
[32]. Dilger RN, Johnson RW. Aging, microglial cell priming, and the discordant central inflammatory response to signals from the peripheral immune system. J Leukoc Biol 2008;84:932–9.
[33]. Farmer MA, Baliki MN, Apkarian AV. A dynamic network perspective of chronic pain. Neurosci Lett 2012;520:197–203.
[34]. Fjell AM, Westlye LT, Amlien I, Espeseth T, Reinvang I, Raz N, Agartz I, Salat DH, Greve DN, Fischl B, Dale AM, Walhovd KB. High consistency of regional cortical thinning in aging across multiple samples. Cereb Cortex 2009;19:2001–12.
[35]. Fjell AM, Walhovd KB. Structural brain changes in aging: courses, causes and cognitive consequences. Rev Neurosciences 2010;21:187–222.
[36]. Franceschi C. Inflammaging as a major characteristic of old people: can it be prevented or cured? Nutr Rev 2007;65:S173–6.
[37]. Franke K, Gaser C. Longitudinal changes in individual BrainAGE in healthy aging, mild cognitive impairment, and Alzheimer's disease. GeroPsych 2012;25:235–45.
[38]. Franke K, Gaser C. Ten years of BrainAGE as a neuroimaging biomarker of brain aging: what insights have we gained? Front Neurol 2019;10:789.
[39]. Franke K, Ziegler G, Klöppel S, Gaser C; Alzheimer's Disease Neuroimaging Initiative. Estimating the age of healthy subjects from T1-weighted MRI scans using kernel methods: exploring the influence of various parameters. Neuroimage 2010;50:883–92.
[40]. Gaser C, Franke K, Klöppel S, Koutsouleris N, Sauer H; Alzheimer's Disease Neuroimaging Initiative. BrainAGE in mild cognitive impaired patients: predicting the conversion to alzheimer's disease. PLoS One 2013;8:e67346.
[41]. Geha PY, Baliki MN, Harden RN, Bauer WR, Parrish TB, Apkarian AV. The brain in chronic CRPS pain: abnormal gray-white matter interactions in emotional and autonomic regions. Neuron 2008;60:570–81.
[42]. Geneen LJ, Moore RA, Clarke C, Martin D, Colvin LA, Smith BH. Physical activity and exercise for chronic pain in adults: an overview of Cochrane Reviews. Cochrane Database Syst Rev 2017;4:CD011279.
[43]. Godbout JP, Chen J, Abraham J, Richwine AF, Berg BM, Kelley KW, Johnson RW. Exaggerated neuroinflammation and sickness behavior in aged mice following activation of the peripheral innate immune system. FASEB J 2005;19:1329–31.
[44]. Gwilym SE, Filippini N, Douaud G, Carr AJ, Tracey I. Thalamic atrophy associated with painful osteoarthritis of the hip is reversible after arthroplasty: a longitudinal voxel-based morphometric study. Arthritis Rheum 2010;62:2930–40.
[45]. Hautzinger M, Bailer M, Hofmeister D, Keller F. Allgemeine depressionsskala. Göttingen: Hogrefe, 2012.
[46]. Huh Y, Ji RR, Chen G. Neuroinflammation, bone marrow stem cells, and chronic pain. Front Immunol 2017;8:1014.
[47]. Johnson JW. A heuristic method for estimating the relative weight of predictor variables in multiple regression. Multivariate Behav Res 2000;35:1–19.
[48]. Jones AK, Huneke NT, Lloyd DM, Brown CA, Watson A. Role of functional brain imaging in understanding rheumatic pain. Curr Rheumatol Rep 2012;14:557–67.
[49]. Karatzoglou A, Smola A, Hornik K, Zeileis A. Kernlab—an S4 package for kernel methods in R. J Stat Softw 2004;11:1–20.
[50]. Kaza E, Klose U, Lotze M. Comparison of a 32-channel with a 12-channel head coil: are there relevant improvements for functional imaging. J Magn Reson Imaging 2011;34:173–83.
[51]. Koutsouleris N, Davatzikos C, Borgwardt S, Gaser C, Bottlender R, Frodl T, Falkai P, Riecher-Rössler A, Möller HJ, Reiser M, Pantelis C, Meisenzahl E. Accelerated brain aging in schizophrenia and beyond: a neuroanatomical marker of psychiatric disorders. Schizophr Bull 2014;40:1140–53.
[52]. Kregel J, Meeus M, Malfliet A, Dolphens M, Danneels L, Nijs J, Cagnie B. Structural and functional brain abnormalities in chronic low back pain: a systematic review. Semin Arthritis Rheum 2015;45:229–37.
[53]. Kuchinad A, Schweinhardt P, Seminowicz DA, Wood PB, Chizh BA, Bushnell MC. Accelerated brain gray matter loss in fibromyalgia patients: premature aging of the brain. J Neurosci 2007;27:4004–7.
[54]. Kuner R, Flor H. Structural plasticity and reorganisation in chronic pain. Nat Rev Neurosci 2017;18:20–30.
[55]. LaMotte RH, Lundberg LE, Torebjörk HE. Pain, hyperalgesia and activity in nociceptive C units in humans after intradermal injection of capsaicin. J Physiol 1992;448:749–64.
[56]. Le TT, Kuplicki RT, McKinney BA, Yeh HW, Thompson WK, Paulus MP, Tulsa I. A nonlinear simulation framework supports adjusting for age when analyzing BrainAGE. Front Aging Neurosci 2018;10:317.
[57]. Lieberman G, Shpaner M, Watts R, Andrews T, Filippi CG, Davis M, Naylor MR. White matter involvement in chronic musculoskeletal pain. J Pain 2014;15:1110–19.
[58]. Liem F, Varoquaux G, Kynast J, Beyer F, Kharabian Masouleh S, Huntenburg JM, Lampe L, Rahim M, Abraham A, Craddock RC, Riedel-Heller S, Luck T, Loeffler M, Schroeter ML, Witte AV, Villringer A, Margulies DS. Predicting brain-age from multimodal imaging data captures cognitive impairment. Neuroimage 2017;148:179–88.
[59]. Lindholm TL, Botes L, Engman EL, Frank A, Jonsson T, Svensson L, Julin P. Parallel imaging: is GRAPPA a useful acquisition tool for MR imaging intended for volumetric brain analysis. BMC Med Imaging 2009;9:15.
[60]. Love J, Selker R, Marsman M, Jamil T, Dropmann D, Verhagen AJ, Ly A, Gronau QF, Smira M, Epskamp S, Matzke D, Wild A, Knight P, Rouder JN, Morey RD, Wagenmakers E-J. JASP — graphical statistical software for common statistical designs. Journal of Statistical Software 2019;88:1–17.
[61]. Ly A, Verhagen J, Wagenmakers EJ. Harold Jeffreys's default Bayes factor hypothesis tests: explanation, extension, and application in psychology. J Math Psychol 2016;72:19–32.
[62]. Maher FO, Martin DS, Lynch MA. Increased IL-1beta in cortex of aged rats is accompanied by downregulation of ERK and PI-3 kinase. Neurobiol Aging 2004;25:795–806.
[63]. Mansour AR, Baliki MN, Huang L, Torbey S, Herrmann KM, Schnitzer TJ, Apkarian AV. Brain white matter structural properties predict transition to chronic pain. PAIN 2013;154:2160–8.
[64]. Marcus DS, Wang TH, Parker J, Csernansky JG, Morris JC, Buckner RL. Open Access Series of Imaging Studies (OASIS): cross-sectional MRI data in young, middle aged, nondemented, and demented older adults. J Cogn Neurosci 2007;19:1498–507.
[65]. Matt SM, Johnson RW. Neuro-immune dysfunction during brain aging: new insights in microglial cell regulation. Curr Opin Pharmacol 2016;26:96–101.
[66]. Mazziotta JC, Toga AW, Evans A, Fox P, Lancaster J. A probabilistic atlas of the human brain: theory and rationale for its development. The International Consortium for Brain Mapping (ICBM). Neuroimage 1995;2:89–101.
[67]. Moayedi M, Weissman-Fogel I, Salomons TV, Crawley AP, Goldberg MB, Freeman BV, Tenenbaum HC, Davis KD. Abnormal gray matter aging in chronic pain patients. Brain Res 2012;1456:82–93.
[68]. Moriarty O, McGuire BE, Finn DP. The effect of pain on cognitive function: a review of clinical and preclinical research. Prog Neurobiol 2011;93:385–404.
[69]. Morterá P, Herculano-Houzel S. Age-related neuronal loss in the rat brain starts at the end of adolescence. Front Neuroanat 2012;6:45.
[70]. Norden DM, Godbout JP. Review: microglia of the aged brain: primed to be activated and resistant to regulation. Neuropathol Appl Neurobiol 2013;39:19–34.
[71]. Park DC, Reuter-Lorenz P. The adaptive brain: aging and neurocognitive scaffolding. Annu Rev Psychol 2009;60:173–96.
[72]. Perry VH. Contribution of systemic inflammation to chronic neurodegeneration. Acta Neuropathol 2010;120:277–86.
[73]. R Core Team. R: A language and environment for statistical computing. Vienna: R Foundation for Statistical Computing, 2019.
[74]. Radloff LS. The CES-D scale: a self-report depression scale for research in the general population. Appl Psychol Meas 1977;1:385–401.
[75]. Rasmussen CE, Williams CKI. Gaussian processes for machine learning. Cambridge: MIT Press, 2006.
[76]. Rodriguez-Raecke R, Niemeier A, Ihle K, Ruether W, May A. Brain gray matter decrease in chronic pain is the consequence and not the cause of pain. J Neurosci 2009;29:13746–50.
[77]. Satzger W, Fessmann H, Engel RR. The equivalence of three German vocabulary tests and the German version of the Wechsler Adult Intelligence Scale-Revised (HAWIE-R). Z für Differentielle Diagnostische Psychol 2002;23:159–70.
[78]. Schmidt-Wilcke T, Gänssbauer S, Neuner T, Bogdahn U, May A. Subtle grey matter changes between migraine patients and healthy controls. Cephalalgia 2008;28:1–4.
[79]. Schmidt-Wilcke T, Leinisch E, Gänssbauer S, Draganski B, Bogdahn U, Altmeppen J, May A. Affective components and intensity of pain correlate with structural differences in gray matter in chronic back pain patients. PAIN 2006;125:89–97.
[80]. Schmidt-Wilcke T, Leinisch E, Straube A, Kämpfe N, Draganski B, Diener HC, Bogdahn U, May A. Gray matter decrease in patients with chronic tension type headache. Neurology 2005;65:1483–6.
[81]. Schmidt K, Metzler P. Wortschatztest (WST). Weinheim: Beltz Test, 1992.
[82]. Schweinhardt P, Kuchinad A, Pukall CF, Bushnell MC. Increased gray matter density in young women with chronic vulvar pain. PAIN 2008;140:411–19.
[83]. Seminowicz DA, Labus JS, Bueller JA, Tillisch K, Naliboff BD, Bushnell MC, Mayer EA. Regional gray matter density changes in brains of patients with irritable bowel syndrome. Gastroenterology 2010;139:48–57.e2.
[84]. Smallwood RF, Laird AR, Ramage AE, Parkinson AL, Lewis J, Clauw DJ, Williams DA, Schmidt-Wilcke T, Farrell MJ, Eickhoff SB, Robin DA. Structural brain anomalies and chronic pain: a quantitative meta-analysis of gray matter volume. J Pain 2013;14:663–75.
[85]. Smith SM, Vidaurre D, Alfaro-Almagro F, Nichols TE, Miller KL. Estimation of brain age delta from brain imaging. Neuroimage 2019;200:528–39.
[86]. Sörös P, Knecht S, Bantel C, Imai T, Wüsten R, Pantev C, Lütkenhöner B, Bürkle H, Henningsen H. Functional reorganization of the human primary somatosensory cortex after acute pain demonstrated by magnetoencephalography. Neurosci Lett 2001;298:195–8.
[87]. Spindler M, Koch K, Borisov E, Özyurt J, Sörös P, Thiel C, Bantel C. The influence of chronic pain and cognitive function on spatial-numerical processing. Front Behav Neurosci 2018;12:165.
[88]. Sullivan EV, Pfefferbaum A. Neuroradiological characterization of normal adult ageing. Br J Radiol 2007;80:S99–S108.
[89]. Tisserand DJ, van Boxtel MP, Pruessner JC, Hofman P, Evans AC, Jolles J. A voxel-based morphometric study to determine individual differences in gray matter density associated with age and cognitive change over time. Cereb Cortex 2004;14:966–73.
[90]. Tonidandel S, LeBreton JM. RWA web: a free, comprehensive, web-based, and user-friendly tool for relative weight analyses. J Business Psychol 2015;30:207–16.
[91]. Tonidandel S, LeBreton JM, Johnson JW. Determining the statistical significance of relative weights. Psychol Methods 2009;14:387.
[92]. Treede RD, Rief W, Barke A, Aziz Q, Bennett MI, Benoliel R, Cohen M, Evers S, Finnerup NB, First MB, Giamberardino MA, Kaasa S, Kosek E, Lavandʼhomme P, Nicholas M, Perrot S, Scholz J, Schug S, Smith BH, Svensson P, Vlaeyen JW, Wang SJ. A classification of chronic pain for ICD-11. PAIN 2015;156:1003–7.
[93]. Treede RD, Rief W, Barke A, Aziz Q, Bennett MI, Benoliel R, Cohen M, Evers S, Finnerup NB, First MB, Giamberardino MA, Kaasa S, Korwisi B, Kosek E, Lavandʼhomme P, Nicholas M, Perrot S, Scholz J, Schug S, Smith BH, Svensson P, Vlaeyen JWS, Wang SJ. Chronic pain as a symptom or a disease: the IASP classification of chronic pain for the international classification of diseases (ICD-11). PAIN 2019;160:19–27.
[94]. Tsang A, Von Korff M, Lee S, Alonso J, Karam E, Angermeyer MC, Borges GL, Bromet EJ, Demytteneare K, de Girolamo G, de Graaf R, Gureje O, Lepine JP, Haro JM, Levinson D, Oakley Browne MA, Posada-Villa J, Seedat S, Watanabe M. Common chronic pain conditions in developed and developing countries: gender and age differences and comorbidity with depression-anxiety disorders. J Pain 2008;9:883–91.
[95]. Valfrè W, Rainero I, Bergui M, Pinessi L. Voxel-based morphometry reveals gray matter abnormalities in migraine. Headache 2008;48:109–17.
[96]. van Doorn J, van den Bergh D, Bohm U, Dablander F, Derks K, Draws T, Evans NJ, Gronau QF, Hinne M, Kucharský Š. The JASP guidelines for conducting and reporting a Bayesian analysis. doi: 1031234/osfio/yqxfr 2019. PsyArXiv.
    [97]. Varikuti DP, Genon S, Sotiras A, Schwender H, Hoffstaedter F, Patil KR, Jockwitz C, Caspers S, Moebus S, Amunts K, Davatzikos C, Eickhoff SB. Evaluation of non-negative matrix factorization of grey matter in age prediction. Neuroimage 2018;173:394–410.
    [98]. Wartolowska K, Hough MG, Jenkinson M, Andersson J, Wordsworth BP, Tracey I. Structural changes of the brain in rheumatoid arthritis. Arthritis Rheum 2012;64:371–9.
    [99]. Wolrich J, Poots AJ, Kuehler BM, Rice AS, Rahman A, Bantel C. Is number sense impaired in chronic pain patients? Br J Anaesth 2014;113:1024–31.
    [100]. Xanthos DN, Sandkühler J. Neurogenic neuroinflammation: inflammatory CNS reactions in response to neuronal activity. Nat Rev Neurosci 2014;15:43–53.
    [101]. Ye SM, Johnson RW. Increased interleukin-6 expression by microglia from brain of aged mice. J Neuroimmunol 1999;93:139–48.
    Keywords:

    Chronic pain; Musculoskeletal disorders; Rheumatic disease; Magnetic resonance imaging; Brain structure; Gray matter; White matter; Aging

    © 2019 International Association for the Study of Pain