Sage Journals: Discover world-class research

Abstract

Introduction:

Post-traumatic headaches (PTH) are common following mild traumatic brain injury (mTBI). There is evidence of altered central pain processing in adult PTH; however, little is known about how children with PTH process pain. The anterior cingulate cortex (ACC) plays a critical role in descending central pain modulation. In this study, we explored whether the functional connectivity (FC) of the ACC is altered in children with PTH.

Methods:

In this case-control study, we investigated resting-state FC of 5 ACC seeds (caudal, dorsal, rostral, perigenual, and subgenual) in children with PTH (n = 73) and without PTH (n = 29) following mTBI, and healthy controls (n = 27). Post-concussion symptoms were assessed using the Post-Concussion Symptom Inventory and the Child Health Questionnaire. Resting-state functional Magnetic Resonance Imaging (fMRI) data were used to generate maps of ACC FC. Group-level comparisons were performed within a target mask comprised of pain-related regions using FSL Randomise.

Results:

We found decreased FC between the right perigenual ACC and the left cerebellum, and increased FC between the right subgenual ACC and the left dorsolateral prefrontal cortex in children with PTH compared to healthy controls. The ACC FC in children without PTH following mTBI did not differ from the group with PTH or healthy controls. FC between rostral and perigenual ACC seeds and the cerebellum was increased in children with PTH with pre-injury headaches compared to those with PTH without pre-injury headaches. There was a positive relationship between PTH severity and rostral ACC FC with the bilateral thalamus, right hippocampus and periaqueductal gray.

Conclusions:

Central pain processing is altered in children with PTH. Pre-existing headaches help to drive this process.

Trial registration:

The PlayGame Trial was registered in ClinicalTrials.gov database (ClinicalTrials.gov Identifier: NCT01874847).

Keywords

anterior cingulate cortex functional connectivity functional magnetic resonance imaging mild traumatic brain injury pain post-traumatic headache

Introduction

Traumatic brain injury (TBI) occurs in about 1.7 million individuals in the United States annually, 700,000 of whom are children and adolescents (0–19 years old).¹ Although approximately 90% of childhood TBIs are mild (mTBI),² half of these children will continue to experience persistent post-concussion symptoms (PPCS) one month following injury.^3
–5 Post-traumatic headache (PTH) is one of the most common PPCS symptoms, and is still problematic in ^{3,43

–58}% of children one month after mTBI.^3,4

The anterior cingulate cortex (ACC) is a critical region in the central pain processing networks and is involved in many pain-related functions, i.e. the sensory, cognitive, and affective aspects of pain processing, and pain modulation. It modulates connectivity within higher cortical (e.g., the dorsolateral prefrontal cortex, DLPFC; and medial prefrontal cortex, mPFC) and subcortical regions (e.g., periaqueductal grey matter (PAG)).^6

–10 This connectivity may be disrupted following adult and juvenile mTBI i.e. altered ACC and default mode network (DMN) connectivity.^11

–15

There is a significant knowledge gap about pain processing in PTH, especially in children.^16,17 Recent research suggests that functional connectivity within central pain-related regions relate to PPCS and may be particularly important in PTH.^13,18 Caudal-dorsal ACC connectivity with the secondary somatosensory cortex and inferior parietal lobe is significantly increased in adults with chronic PTH,¹³ and early FC changes within pain networks (e.g., decreased PAG-DMN FC two days post-injury) significantly predict the persistence of chronic PTH.¹⁸

In childhood, brain injury occurs during a period of rapid neuroanatomical and physiological change (i.e., ongoing myelination, synaptic pruning and functional integration).^19,20 Not only do developing brains respond differently to injury compared to adult brains,²¹ but childhood headaches may also respond differently to treatment.^22,23 Given the critical role of the ACC in central pain processing, we examined the FC of the ACC with other pain-related brain regions in children with PTH following mTBI, and compared them to those without PTH, and healthy controls. As pre-injury headaches are a risk factor for PTH,²⁴ we also investigated how ACC FC differed in children with and without a prior history of headache. We hypothesized that children with PTH would have lower ACC FC with the DLPFC and PAG, and higher ACC FC with the insula. We also hypothesized that there would be a significant effect of prior headache history on ACC FC.

Methods

Participants in this case-control study were recruited from the PlayGame Trial (NCT01874847); a clinical trial investigating the effect of melatonin treatment on PPCS.²⁵ The study was conducted between 2014 and 2018 according to the Good Clinical Practice guidelines. Ethical approval was obtained from the Calgary Human Research Ethics Board (13-0372) Written informed consent and assent were provided by parents/guardians and participants, respectively. Neuroimaging and assessments for this study occurred at enrolment and before commencing medication. At enrolment, a comprehensive clinical assessment was performed by a physician with expertise in mTBI and data were collected on demographic and injury characteristics, post-concussive complaints including PTH, current headache characteristics, previous headache disorder, past medical history, and family history. PTH severity was verbally reported between 0 (no pain) to 10 (worst pain ever).

Participants

Participants (aged 8–18 years) were eligible if they had a medically diagnosed mTBI accordant with the American Academy of Neurology criteria.²⁶ Neuroimaging was not used to define mTBI, however, no intracranial abnormality was identified in the 10% of mTBI participants who had acute neuroimaging. Exclusion criteria included a loss of consciousness of greater than 30 min at time of injury, a Glasgow Coma Scale (GCS) score less than 13, injury due to assault, a previous mTBI within the last three months, neuroactive medications, any significant medical disorder, or contraindication to MRI.

Criteria for PTH were a current complaint of headache (either new-onset since mTBI or increased headache) at clinical interview, and an increase in the headache score on the Post-Concussion Symptom Inventory (PCSI) compared to pre-injury score (evaluated retrospectively at enrolment). Participants were further categorized as having a pre-injury history of headache (+history) or not (-history) at clinical interview. Pre-injury headaches were classified using the International Classification of Headache Disorders, second Edition (ICHD-II), and/or previous headache diagnosis by a neurologist or headache specialist.

A control group of healthy children of similar age and sex without a history of mTBI were recruited mainly from friends and siblings of participants, and underwent identical evaluations. We excluded children who reported having frequent headaches or headache disorders from the control group. We also recruited additional healthy controls from a separate study (International Collaboration on Neuroinflammation in Traumatic Brain Injury study (ICON-TBI, REB 16-2057)) with the same standardized clinical interview, neuropsychological testing (except the PCSI), and identical neuroimaging parameters.

Outcome assessments

Post-concussion symptoms

The PCSI was used to investigate post-concussion symptoms after mTBI.^27
–29 It consists of 26 symptoms in physical, cognitive, emotional and sleep domains and has moderate-to-high test-retest reliability and good convergent validity.^29,30 Using this, participants were asked to rate their headaches ‘over the last few days’. Pre-injury headaches were rated using the same questionnaire (baseline or pre-injury report). Headaches were rated between 0 (not a problem), to 6 (severe problem). The PCSI symptom total score ranges from 0 to 156.

Child Health Questionnaire (CHQ)

The Bodily Pain/Discomfort score from the Child Health Questionnaire (Child report, CHQ87). The CHQ is a validated measure of children’s general health,^31,32 and has been used to measure bodily pain in individuals with TBI and chronic pain conditions.^33

–38 The CHQ87 Bodily Pain/Discomfort score rates severity (“none” to “very severe”) and frequency (“none of the time” to “every/almost every day”) of bodily pain/discomfort in the past four weeks using a a 6-point scale. Scores are standardized and range from 0 (most severe) to 100 (none). Health-related Quality of Life (HRQOL) was reported based on other CHQ87 domains such as Global Health, Physical Functioning, Social Limitations due to emotional, and behavioral problems.^32,39,40

Neuroimaging data acquisition

Functional magnetic resonance imaging (fMRI) was conducted using a GE 3 Tesla 750w MRI with a 32-channel head coil located at the Alberta Children’s Hospital. To record blood oxygen level dependent (BOLD) responses during resting-state, children were asked to lie in the scanner with their eyes open and look at a fixation cross. Resting-state fMRI was recorded using a gradient echo, echo planar imaging sequence for 5 min and 10 s (TE = 30 ms, TR = 2000 ms, flip angle = 90, field of view (FOV) = 23 cm, matrix 64 × 64, 150 slices, slice thickness = 3.6, 150 volumes, voxel size = 3.6 × 3.6 × 3.6 mm). Structural T₁-weighted anatomical images were used for registration (3D Ax FSPGR BRAVO, flip angle = 10, TI = 600, FOV = 24 cm, matrix 300 × 300, slice thickness = 0.8 mm, 230 slices, voxel size = 0.8 × 0.8 × 0.8 mm).

Statistical analyses

Demographic characteristics were compared between groups using SPSS⁴¹ (IBM Corp. Released 2016. IBM SPSS Statistics for Macintosh, Version 24.0. Armonk, NY: IBM Corp). Normality was assessed using Shapiro-Wilk test. Group differences in age and motion outliers were analysed using the Kruskal-Wallis test (KW).

The effect of PTH across groups (controls, PTH or NoPTH) and pre-injury headache history on four HRQOL domains (CHQ 87) were analysed using linear regression controlling for age and sex. To account for multiple comparison among these four predictors, we used Bonferroni correction and corrected p-value was set as p < 0.0125.

Neuroimaging analyses

Preprocessing

The FMRIB Software Library (FSL) version 6.0.0 was used for fMRI pre-processing and analyses (www://fsl.fmrib.ox.ac.uk).⁴² Pre-processing included skull removal for T1 images using the Brain Extraction Tool (BET),⁴³ motion correction using the six parameter affine transformation using MCFLIRT,⁴⁴ slice-time correction, spatial smoothing (Gaussian kernel of 6 mm) at full-width half-max, linear registration of fMRI data to the structural images using six DOF and normalization of the data by linear registration to the MNI152 (2 mm) standard template using 12 DOF.⁴⁵ To mitigate the effect of head motion on the data, the FSL Motion Outliers was used (https://fsl.fmrib.ox.ac.uk/fsl/fslwiki/FSLMotionOutliers) to identify the volumes affected by significant head motion^46,47 with a frame-wise displacement (FD) threshold greater than 0.3 (mm). Scans with more than 25% outlier time-points were excluded (i.e., less than 113 out of 150 volumes remaining after excluding the volumes with motion). A nuisance model was generated using FEAT in FSL to account for the effects of motion and other nuisance factors including global signal (GS), white matter (WM) and cerebrospinal fluid (CSF) timeseries. Global signal regression (GSR) was used to mitigate the effects of non-neuronal noise factors such as respiration, cardiac impulses, motion, vasodilation, and scanner-related artifacts.^48

–54 As a complementary analysis, we further analysed the ACC FC maps without GSR (Supplementary Figures 2 to 4).^55
–57 WM and CSF masks were generated using the FSL FAST segmentation tool.⁵⁸ Masks were registered to the functional and standard space, thresholded at 0.4 for CSF and 0.66 for WM and binarized. Average time-series of WM, CSF, and GS were added as explanatory variables (EVs), alongside motion outliers (volumes), motion parameters and their derivatives as additional confound EVs. The FILM-prewhitening was activated in the analysis to account for the autocorrelation among the nuisance factors and to enhance the model’s efficacy.⁵⁹ The cleaned data was normalized via z-scoring and subsequently centered on 100, using the fslmaths command.

First-level analyses

Region-of-interest (ROI) analyses focused on the ACC. Using the coordinates reported by Margulies et al., Kelly et al. and others.^60

–63 we created five functionally differentiable seeds of the ACC: caudal, dorsal, rostral, perigenual, and subgenual. These ACC seeds have been found to have unique short-term and long-term connectivity with various pain-related regions.^60
–62,64 These seeds were generated for the right (R) hemisphere only as previous studies have shown that connectivity patterns of the ACC are similar for the right and left (L) hemispheres,⁶⁰ and pain processing is more lateralized toward the right hemisphere.⁶⁵ A sphere was created with a radius of 3.5 (mm) around the reported coordinates (see Figure 1). ACC seeds were then transformed into each participant’s native space and the mean time-series for each seed was extracted. For the first-level analysis, five separate models were generated within the FEAT model in FSL, by adding each seed time-series as an explanatory variable.

Figure 1.

ACC seeds and target pain mask. Upper left panel: Five ACC seeds: (1) caudal ACC (cACC: pink), (2) dorsal ACC (dACC: red), (3) rostral ACC (rACC: light blue), (4) perigenual ACC (pACC: dark blue), (5) subgenual ACC (sACC: green). Lower left panel: target ROIs combined in the target pain mask. MNI-152 (2 mm) was used as a standard atlas; Right diagram shows summary of the analysis steps.

Pain mask: A pain mask was created using brain regions implicated in normal or maladaptive pain processing: bilateral primary and secondary somatosensory cortex, OFC, DLPFC, mPFC, anterior and posterior cingulate cortices, insula, basal ganglia, nucleus accumbens, thalamus, amygdala, cerebellum, PAG and rostroventromedial medulla (RVM) (Table 1).^66
–68 Fourteen bilateral ROIs were created using the Harvard-Oxford cortical and subcortical atlases (https://fsl.fmrib.ox.ac.uk/fsl/fslwiki/Atlases).^42,69,70 The cerebellum was selected from the MNI structural atlas. The thresholds for all regions were set as 25%, and the masks were binarized. In addition, two subcortical regions located in the brainstem, the PAG and RVM, were included in the target mask. A 7 mm sphere was drawn covering PAG and RVM seeds (PAG: X = −2, Y = −28, Z = −8 and RVM: X = 4, Y = −34, Z = −50).⁷¹ All 17 pain-related regions were merged into a single pain mask.

Table 1.

ACC seeds and target ROIs.

ROIs		Region	Center coordinate or Atlas name	Source
ACC Seeds	1	Caudal ACC	Coordinate-based mask (X = 5, Y = −10, Z = 47)	From Kelly et al. (2009); 3.5 mm radius
	2	Dorsal ACC	Coordinate-based mask (X = 5, Y = 14, Z = 42)
	3	Rostral ACC	Coordinate-based mask (X = 5, Y = 34, Z = 28)
	4	Perigenual ACC	Coordinate-based mask (X = 5, Y = 47, Z = 11)
	5	Subgenual ACC	Coordinate-based mask (X = 5, Y = 25, Z = −10)
Pain Mask	1	OFC	Frontal Orbital Cortex	Harvard-Oxford cortical atlas
	2	DLPFC	Middle Frontal Gyrus
	3	ACC	Cingulate Gyrus, anterior division
	4	PCC	Cingulate Gyrus, posterior division
	5	MPFC	Frontal medial cortex
	6	Subcallosal cortex	Subcallosal Cortex
	7	SSC1	Postcentral Gyrus
	8	SSC2	Parietal Operculum Cortex
	9	Insula	Insular Cortex
	10	Basal ganglia	R & L Caudate, R & L Putamen, R & L Pallidum	Harvard-Oxford subcortical atlas
	11	Nucleus accumbens	R and L Accumbens
	12	Hippocampus	R and L Hippocampus
	13	Thalamus	R & L Thalamus
	14	Amygdala	R & L Amygdala
	15	Cerebellum	Cerebellum	MNI structural atlas
	16	PAG	Coordinate-based mask (X = −2, Y = −28, Z = −8)	From Kong et al. (2010) (104)
	17	RVM	Coordinate-based mask (X = 4, Y = −34, Z = −50)	From Kong et al. (2010) (104)

Note: ACC and target pain mask were created with FSL 5.0.8 version using Harvard-oxford cortical and sub-cortical atlases and MNI structural atlas provided in FSL software (https://fsl.fmrib.ox.ac.uk/fsl/fslwiki/Atlases). Abbreviations: ACC, anterior cingulate cortex; DLPFC, dorsolateral prefrontal cortex; MPFC, medial prefrontal cortex; PAG, periaqueductal gray; PCC, posterior cingulate cortex; OFC, orbitofrontal cortex; RVM, rostroventromedial medulla; SSC1, primary somatosensory cortex; SSC2, secondary somatosensory cortex.

Group-level analyses

The z-stat outputs from first-level analyses were transformed to standard space. Randomise was used to analyse group results using the threshold-free cluster enhancement (TFCE) statistic with 5000 random permutations (voxel-wise).⁷² A nonparametric permutation test was used to reduce the level of false positives.^72
–74 The TFCE statistic provides a higher probability of finding true signal and better signal-to-noise ratio in the absence of a priori threshold for cluster identification.⁷⁵ Multiple comparison correction in Randomise is based on family-wise error rate (FWE) correction⁷⁶ and results of TFCE statistics in randomise are presented as ‘tstat’ with the p-values generated as ‘1-FWE corrected p.’⁷² We report on clusters of voxels with p-values less than p < 0.05 and their peak voxel values (Supplementary Table 1).

ACC FC differences between groups

To investigate differences in average FC between the ACC seeds and pain-related regions between participant groups (PTH: n = 65, NoPTH: n = 22, controls: n = 21), general linear model (GLM) design matrices were created for each seed and regressed against the z-stat FC maps. The GLM included three EVs for group intercepts together with age, sex, and number of volumes with motion outliers. Another EV was added to adjust for any variability related to differences in control groups (PlayGame Trial vs. ICON-TBI). Positive and negative group effects were also generated as within-group averages. A one-way ANOVA was used to compare the three groups, and t-contrasts compared pairs of groups. To examine ACC FC in participants with new-onset PTH (n = 35) compared to participants without PTH with no headache history (n = 16) and controls (n = 21), participants with a headache history were excluded (reference Figure 1 consort diagram). Differences in the ACC FC maps between groups (PTH, NoPTH, controls) were analysed using a GLM as above.

Effect of pre-injury headache on ACC FC

To investigate the effect of pre-injury headache history, we compared ACC FC between PTH participants with and without prior headaches (PTH + history: n = 30; PTH − history: n = 35). This was analysed using a GLM with two EVs (PTH − history and PTH + history) controlling for sex, age and motion outliers. T-tests were used for between-group contrasts. Exploratory analyses were performed to examine the specific effect of pre-injury headache history on ACC FC by comparing participants without PTH with a headache history (n = 6) to participants with PTH with a headache history. Results of these analyses are included in the Supplementary Figure 1.

Influence of PTH severity and bodily pain

FC ACC and pain-related regions was investigated in the PTH group using GLM with a covariate for PTH severity for each ACC seed. We entered one EV for the group intercept with PTH together with PTH severity as a covariate of interest, while controlling for sex, age and motion outliers. Positive and negative linear slopes of PTH severity were computed. Similarly, methods were used to investigate the effect of CHQ bodily pain score.

Results

One hundred and twenty-nine participants were eligible. Nineteen participants did not meet eligibility criteria (see Figure 2 for study details). Twenty-one scans were removed due to excessive motion. One hundred and eight scans were included in the final analyses: PTH (n = 65; mean age = 14.6 SD = 2.3 years; 40% male), No PTH (n = 22; mean age = 14.1 SD = 2.5 years; 60% male) and controls (n = 21; mean age = 14.5 SD = 2.8 years; 43% male). Groups did not differ significantly in age or sex distribution (see Table 2). Injury characteristics are presented in Table 3. Participants were between 22 and 54 days post-injury (mean = 37.4 days, SD = 5.9). The most common mechanism of injury was sport-related. Headache was the most common acute symptom.

Figure 2.

Details of the study participants.

Table 2.

Demographics of participants in the control, PTH with a headache history, PTH with no headache history, without PTH with no headache history and without PTH with a headache history groups.

	Control (n = 21)	PTH (n = 65)		NoPTH (n = 22)		Statistics
	Control (n = 21)	+ history (n = 30)	– history (n = 35)	+ history (n = 6)	– history (n = 16)	χ²(4)	P-value
Mean age, y (SD)	14.5 (2.7)	14.7 (1.7)	14.3 (2.7)	14.3 (1.8)	13.3 (2.7)	1.0	0.910
Sex, male (%)	9 (43)	11 (37)	15 (43)	5 (83)	8 (50)	4.67	0.323
Handedness, R (%)	19 (91)	28 (93)	30 (86)	4 (67)	14 (88)

+ Pre-injury headache history.

– No pre-injury headache history.

NoPTH, No post-traumatic headache; PTH, post-traumatic headache; SD, Standard deviation; y, year.

Table 3.

Injury characteristics in the PTH and NoPTH groups.

	PTH (n = 65)		NoPTH (n = 22)
	+ history (n = 30)	– history (n = 35)	+ history (n = 6)	– history (n = 16)
Days since injury, median (range)	35.5 (22–54)	38 (31–49)	36 (28–45)	39 (28–48)
GCS, mean (SD)	14.5 (0.5)	14.29 (0.4)	14.83 (0.4)	14.5 (0.6)
LOC, n (%)	4 (13)	0	1 (17)	5 (31)
LOC duration, n (%)
<5 s	4 (13)		1 (16.7)	3 (19)
Evidence of direct blow to the head, n (%)	30 (100)	35 (100)	5 (83)	15 (94)
Mechanism of injury, n (%)
Sport	21 (70)	27 (77)	3 (50)	14 (88)
Fall	2 (8)	2 (6)	3 (50)	2 (12)
MVA	2 (7)	3 (9)	0	0
Acute symptoms, n (%)
Headache	28 (93)	30 (86)	6 (100)	11 (69)
Appeared dazed or stunned	21 (70)	23 (66)	2 (33)	11 (69)
Confusion	18 (60)	22 (63)	3 (50)	6 (37)
Answered questions slowly	12 (40)	13 (37)	1 (17)	4 (25)
Repeated questions	3 (10)	6 (17)	0	1 (6)
Forgetful (recent information)	10 (33)	9 (26)	1 (17)	4 (25)
Nausea/vomiting	16 (53)	16 (46)	4 (67)	10 (62)
Dizziness	21 (70)	27 (77)	5 (83)	12 (75)
Double/blurry vision	8 (27)	14 (40)	4 (67)	6 (37)
Impact on school performance, n (%)
No impact	6 (20)	8 (23)	5 (83)	10 (62)
Mild impact	16 (53)	18 (51)	1 (17)	3 (19)
Moderate impact	7 (23)	9 (26)	0	2 (12)
Impact on social functioning, n (%)	20 (67)	17 (49)	1 (17)	7 (44)
Previous mTBI, n (%)	15 (50)	10 (29)	1 (17)	5 (31)
Number of previous mTBI injuries, mean (SD, range)	1 (1.4, 0–5)	0.5 (0.9, 0–3)	0.3 (0.8, 0–2)	0.5 (1.2, 0–5)

+ Pre-injury headache history.

– No pre-injury headache history.

PTH, post-traumatic headache; GCS, Glasgow coma scale; MVA, motor vehicle accident; LOC, loss of consciousness; NoPTH, mTBI without PTH.

The clinical details and headache characteristics are reported in Tables 4 and 5. Migraine was the most common pre-injury headache disorder (90%). PTHs were often diffuse or occurred in multiple locations and were usually described as a pressure-like pain. Commonly associated symptoms were photophobia, phonophobia, nausea, and exacerbation with exercise.

Table 4.

Headache history and PTH characteristics in the control, PTH and NoPTH groups.

	Control (n = 21)	PTH (n = 65)		NoPTH (n = 22)
	Control (n = 21)	+ history (n = 30)	– history (n = 35)	+ history (n = 6)	– history (n = 16)
Previous headache type, n (%)
Migraine with aura		9 (30)	—	0	—
Migraine without aura		18 (60)	—	4 (67)	—
Other migraine		1 (3)	—	1 (17)	—
Tension-type		2 (7)	—	0	—
Chronic daily headache		1 (3)	—	0	—
Other headache disorders		2 (7)	—	1 (17)	—
Family history of headache disorders, n (%)
Mother	11 (52)	17 (57)	20 (57)	5 (83)	9 (56)
Father	1 (5)	10 (33)	8 (23)	0	2 (12)
Siblings	4 (19)	4 (13)	10 (29)	0	6 (37)
PTH Location, n (%)
Frontal		0	4 (11)
Temporal		3 (10)	4 (11)
Orbital		0	1 (3)
Vertex		1 (3)	—
Multiple regions		20 (67)	20 (57)
PTH quality, n (%)
Pressure (or squeezing)		15 (50)	19 (54)
Pulsatile (or Throbbing/pounding)		4 (13)	1 (3)
Sharp (or stabbing)		—	2 (6)
Other (dull, achy, et.)		—	1 (3)
Mix		7 (23)	3 (9)
PTH laterality, n (%)
Unilateral		3 (10)	10 (28)
Bilateral		4 (13)	8 (23)
PTH severity* (1 to 10), mean (SD)		5.4 (2)	4.7 (2)
Duration, mean hours, (SD)		9.1 (9)	7.5 (8)
PTH associated symptoms, n (%)
Nausea		18 (60)	13 (37)
Vomiting		1 (3)	3 (9)
Aura		5 (17)	1 (3)
Photophobia		26 (87)	24 (69)
Phonophobia		20 (67)	21 (60)
Osmophobia		0	3 (9)
Disabling		5 (17)	4 (11)
Worse with exercise		16 (53)	15 (43)
Orthostatic features		4 (13)	2 (6)
History of motion sickness		4 (13)	6 (17)

+ Pre-injury headache history.

– No pre-injury headache history.

* PTH severity was reported as a score between 0 to 10 in the clinical interview.

NoPTH, No post-traumatic headache; PTH, post-traumatic headache; SD, Standard deviation

Table 5.

PCSI scores and bodily pain in the control, PTH and NoPTH groups.

	Control (n = 21)	PTH (n = 65)		NoPTH (n = 22)
	Control (n = 21)	+history (n = 30)	–history (n = 35)	+ history (n = 6)	–history (n = 16)
PCSI score, median (n, range)
Pre-injury headache score	0 (13, 0–3)	0 (30, 0–3)	0 (35, 0–1)	0 (6, 0)	0 (16, 0–1)
Pre-injury total score	3 (13, 0–13)	5 (30, 0–46)	3 (35, 0–32)	0 (6, 0–5)	0.5 (16, 0–24)
Post-injury headache score	—	3 (30, 1–6)	3 (35, 1–6)	0 (6, 0–2)	0 (16, 0–2)
Post-injury total score	—	36 (30, 1–127)	30 (35, 3–122)	5 (0–15)	4 (16, 0–36)
Neck pain, n (%)	—	11 (37)	13 (37)	0	2 (12)
CHQ, median (n, range)
Global Health	85 (14, 60–100)	85 (30, 30–100)	85 (35, 6–100)	100 (5, 85–100)	100 (16, 60–100)
Physical Functioning	98.1 (14, 85.2–100)	85.2 (30, 18.5–100)	77.8 (35, 18.5–100)	100 (6, 96.3–100)	98.1 (16, 66.7–100)
Role/social limitations-emotional	100 (14, 66.7–100)	77.8 (30.0–100)	77.7 (35, 0–100)	100 (6, 88.9–100)	100 (16, 33.3–100)
Role/social limitations-behavioral	100 (14, 66.7–100)	100 (30, 0–100)	88.9 (35,0–100)	100 (6, 77.8–100)	100 (16, 77.8–100)
Bodily pain/discomfort	80 (14, 60–100)	40 (30, 0–80)	40 (35, 0–80)	75 (6, 60–100)	75 (16, 40–100)

+ Pre-injury headache history.

– No pre-injury headache history.

CHQ, child health questionnaire-self report; NoPTH, No post-traumatic headache; PTH, post-traumatic headache.

The regression models across groups (PTH, NoPTH, and controls), pre-injury headache history, age and sex significantly predicted domains of the HRQOL including Physical functioning (F _{(5, 95)} = 6.066, p < 0.0005, R ² = 0.242), Role social limitations/emotional domain (F _{(5, 95)} = 6.294, p < 0.0005, R ² = 0.249), Role social limitations/behavioral domain (F _{(5, 95)} = 2.584, p = 0.031, R ² = 0.120), and Bodily pain and discomfort (F _{(5, 95)} = 13.521, p < 0.0005, R ² = 0.416). Overall, children with PTH reported significantly worse HRQOL compared to those without PTH following mTBI on all four HRQOL domains. Also, compared to healthy controls, the PTH group reported worse HRQOL in the Physical functioning, Role social limitations/emotional and Bodily pain. There was no significant effect of headache history on HRQOL domains, see Table 6.

Table 6.

Linear regression analyses exploring the effect of PTH and pre-injury headache history on HRQOL domains controlling for age and sex.

	Physical functioning			Role social limitations/emotional			Role social limitations/behavioral			Bodily pain and discomfort
	Beta	T	p	Beta	t	p	Beta	t	p	Beta	t	p
Group assignments
NoPTH vs. controls	–0.074	–0.572	0.568	–0.075	–0.583	0.561	0.016	0.115	0.909	–0.085	–0.754	0.453
PTH vs. control	–0.557	–4.145	0.000*	–0.435	–3.247	0.002*	–0.333	–2.299	0.024	–0.677	–5.739	0.000*
PTH vs. NoPTH	–0.471	–4.318	0.000*	–0.348	–3.198	0.002*	–0.352	–2.989	0.004*	–0.578	–6.031	0.000*
Pre-injury headache history (yes vs. no)	0.132	1.390	0.168	–0.026	–0.275	0.784	0.035	0.342	0.733	–0.079	–0.941	0.349
age	–0.050	–0.556	0.579	–0.078	–0.868	0.387	–0.015	–0.160	0.874	0.020	0.248	0.805
Sex (male vs. female)	0.027	0.300	0.765	0.241	2.679	0.009*	0.041	0.422	0.674	–0.022	–0.280	0.780

vs., versus. Beta (standardized coefficients) are reported here. *Significance level was set as p < 0.0125 Bonferroni corrected.

ACC FC differences in PTH

As demonstrated in Figure 3, right perigenual ACC FC with the left cerebellum was increased in controls compared to the PTH group. It is important to note that based on within-group averages, FC between right pACC and left cerebellum was negative; therefore, there was more negative FC between pACC and cerebellum in the PTH group compared to the healthy controls. Moreover, FC between right subgenual ACC FC and the left DLPFC was increased in the PTH group when compared to controls. Within-group averages showed that there was a significant negative FC between the subgenual ACC and DLPFC in controls, which was absent in the PTH group. There were no statistically significant differences between the NoPTH group and the PTH or control groups.

Figure 3.
Significant differences in perigenual and subgenual ACC seeds among three groups: controls, PTH, NoPTH (Cluster threshold < 0.05). (A) The left column shows significant positive (yellow-red) and negative (light blue-blue) within-group averages for the R pACC FC with pain-related regions in controls, NoPTH and PTH groups. (B) The bottom left panel shows higher R pACC FC with the Left cerebellum in controls compared to PTH (voxels in cluster: 5, Max: L lobule VIIIb, tstat = 4.35, x = −18, y = −54, z = −60, p = 0.047). (C) The right column shows significant positive (yellow-red) and negative (light blue-blue) within-group averages for the R sACC with pain-related regions in controls, NoPTH and PTH. (D) The bottom right figure shows relatively higher R sACC FC with the left DLPFC in PTH compared to controls (voxels in cluster: 21, Max: L MFG, tstat = 4.83, x = −46, y = 12, z = 46, p = 0.037). MNI-152 standard atlas is used for presentation. ACC, anterior cingulate cortex; FC, functional connectivity; NoPTH, without post-traumatic headache; R, right; pACC, perigenual ACC; PTH, post-traumatic headache; sACC, subgenual ACC.

Influence of pre-injury headache and PTH severity

After excluding participants with pre-injury headaches, there were no significant group (PTH-history; NoPTH-history; controls) differences in ACC FC, suggesting a significant influence of pre-injury headache. Participants with PTH with a prior headache history had significantly higher ACC FC (rACC and pACC) and the cerebellum compared to the PTH without prior headache group, see Figure 4. Within-group ACC maps demonstrated that there was a less pronounced negative FC between rostral and perigenual ACC seeds and the cerebellum in participants with PTH+headache history.

Figure 4.
Significant differences in the perigenual and rostral ACC seeds FC patterns between PTH without headache history and PTH with a headache history groups are demonstrated (Cluster threshold < 0.05). (A) The left column shows significant positive (yellow-red) and negative (light blue-blue) within-group averages for the R pACC with pain-related regions in PTH-history and PTH+history. (B) The bottom left figure shows higher between-group differences R pACC FC with bilateral cerebellum in PTH+history compared to PTH-history (voxels in cluster: 722, Max: L deep cerebellum nuclei (dentate nucleus), tstat = 3.8, x = −22, y = −52, z = −34, p = 0.016). (C) The right column shows significant positive (yellow-red) and negative (light blue-blue) within-group averages for the R rACC with pain-related regions in PTH-history and PTH+history. (D) The bottom right figure shows higher R rACC FC with right cerebellum in PTH+history compared to PTH-history (voxels in cluster: 101, Max: R cerebellum VI b, tstat = 5.12, x = 40, y = −54, z = −24, p = 0.018). MNI-152 standard atlas is used for presentation. ACC, anterior cingulate cortex; FC, functional connectivity; pACC, perigenual ACC; PTH, post-traumatic headache; PTH + history, post-traumatic headache with a headache history; PTH-history, post-traumatic headache with no headache history; R, right; rACC, rostral ACC.

PTH severity positively correlated with higher rACC FC with central pain regions (i.e., bilateral thalamus, right hippocampus and PAG) as demonstrated in Figure 5. No significant relationship was found between bodily pain and any ACC FC seeds to pain-related regions in participants with PTH.

Figure 5.
The significant relationship between PTH severity and R rACC FC in the PTH group (Cluster threshold < 0.05). This figure shows positive relationship between PTH severity and rACC FC with the bilateral thalamus, R hippocampus and PAG in participants with PTH (voxels in cluster: 371, Max: right thalamus, tstat = 6.13, x = 18, y = −26, z = −2, p = 0.005). MNI-152 standard atlas is used for presentation. ACC, anterior cingulate cortex; FC, functional connectivity; PTH, post-traumatic headache; R, right; rACC, rostral ACC.

Discussion

We investigated differences in ACC FC with other pain-related brain regions in children and adolescents with PTH following mTBI. Our findings supported our hypothesis that there would be significant differences in ACC FC in children with PTH following mTBI when compared to healthy controls. We also found that PTH was significantly associated with HRQOL. Although a history of pre-injury headache had no effect on HRQOL, it was significantly associated with ACC FC.

Altered ACC FC and central pain processing

The ventral portions of the ACC (rostral, perigenual and subgenual) play an important role in modulating pain while performing cognitive demanding tasks. This is mainly through the attentional modulation of pain, i.e. pain distraction.^77
–79 We found evidence of increased negative perigenual ACC FC with the cerebellum in children with PTH (compared to controls). As increased ventral ACC FC to the cerebellum has previously been associated with attention problems,⁸⁰ the increased pACC-cerebellum FC seen in our study could suggest dysregulated attentional modulation of pain.

Higher cortical areas, such as DLPFC, also modulate descending pain inhibitory mechanisms via other cortical and subcortical pain-related regions (e.g. ACC, insula and PAG).^81
–83 We found increased ACC FC between subgenual ACC and contralateral DLPFC in the PTH group (absence of negative FC) when compared to controls. This suggests a dysregulation in top-down inhibitory pain modulation i.e. attenuation of inhibitory DLPFC inputs to the sACC in children with PTH. Consistent with this, a study examining adults with migraine, reported increased FC between left ACC and right DLPFC, supporting alterations in pain modulation processes.⁸⁴

Influence of pre-injury headache and PTH severity

A pre-injury history of frequent headaches is present in 18–47% individuals with PTH.^23,24 We found weaker negative FC between ACC seeds (rostral and perigenual) and the cerebellum in children with PTH who had a prior headache history. Given that the rostral and perigenual ACC play important roles in descending pain inhibition and the modulation of the dorsal horn neurons,^71,85,86 our finding of lower negative ACC-cerebellum FC suggests a downregulation of descending pain inhibitory mechanisms.

Although there is a lack of consensus on the biological meaning/relevance of negative resting-state FC patterns in the literature,^{57,87

–92} animal and human data support its validity; the anti-correlated hemodynamic pattern originating from inhibitory synchronous neural activity.^93

–97 This is relevant to our data in that the cerebellum has major inhibitory functions via inhibitory GABAergic Purkinje cellsy^98,99 The posterior cerebellum is functionally connected (albeit indirectly) to frontal cortex, ACC and DMN regions and fMRI studies have demonstrated its role in pain inhibition and analgesia.^100

–103 The different rACC/pACC-cerebellar FC in PTH without a headache history found in our study further supports the role of the cerebellum as an important inhibitor of the rACC/pACC and central pain networks.

In addition to altered descending pain inhibitory mechanisms, activation of pain facilitatory mechanisms in the brainstem may lead to increased sensitivity of the central pain networks. Regions such as the ACC, hippocampus, locus coeruleus, thalamus, PAG and RVM are critical elements of this mechanism and alterations of FC involving these regions may contribute to pain persistence.^{67,104

–107} We observed higher rACC FC with the thalamus, hippocampus and PAG regions in PTH participants with higher pain burdens. As the interactions between inhibitory and facilitatory pain mechanisms are important in the development of chronic pain,^105,106,108 these findings could suggest increased activation of descending pain facilitatory mechanism in pediatric PTH. This finding may also be related to pain rumination tendencies, where inhibition of unwanted thoughts is impaired due alterations in cortico-thalamo-hippocampal circuits.^109

–112

Our contention of impaired descending pain modulation systems in pediatric PTH is partially supported by previous studies involving adults with PTH.^13,18 Leung et al. found increased ACC FC with the secondary somatosensory regions in adults with chronic PTH compared to healthy controls,¹³ and Niu et al. found that diminished PAG-DMN FC predicted persistent PTH.¹⁸

It is somewhat debatable whether our findings of altered ACC FC are associated with mTBI per se or the persistent PTH symptom itself. Although we found significant differences in ACC FC in PTH, which was driven (at least partially) by the presence of pre-injury headaches, we failed to find definitive evidence that mTBI played a significant role i.e. no differences in ACC FC were found between PTH groups when participants with pre-injury headaches were excluded. This also could have been due to a loss in power.

In our analyses, we used global signal as a nuissance regressor. Similar analyses without GSR are presented in the Supplementary materials for comparison. The usefulness of GSR in resting-state fMRI analysis is highly debated.^{48

–54,56,113} Similar to others,⁸⁰ we found similar general group difference maps of ACC FC with and without GSR; however, the differences without GSR analysis, did not pass the threshold of p < 0.05. Other studies have also reported that GSR increases the likelihood of identifying significant group differences, especially where regions are negatively associated.^114,115 Our findings further support the use of GSR in rs fMRI to remove the effect of the nuisance factors.

The large sample size is a strength of our study and allowed some subgroup analyses. Other strengths include well-characterized clinical phenotypes, and GSR methodology. This study also has several limitations. As this was a pediatric population, we used a relatively short resting-state scan, which could have decreased the signal to noise ratio,¹¹⁶ although others have demonstrated reliable FC with short 5-min scans.¹¹⁷ Quantification of headache burden could have been improved by using a headache diary and specific headache questionnaire. It should be noted that we can only hypothesize about underlying inhibitory/facilitatory process as in FC studies it is not possible to identify the cause of the observed relationships. Finally, although we excluded healthy controls with headaches/headache disorders, about 75% of healthy controls had a family history of headache disorders.

This study has several important clinical implications. Our findings are in keeping with the common clinical scenario in which a head injury seems to be a trigger/stimulus that changes the presentation of pre-existing (and often infrequent) mild headaches to severe, chronic and migraine-like.²³ Simple analgesics seem to offer little or no relief to these patients.^3,118 Further, our findings of altered ACC connectivity in PTH are similar to those seen in animal and human studies of chronic headache disorders.^119

–122 A mTBI, therefore, seems to incite a shift from acute to chronic pain processing and supports the use of treatment and management strategies targeted toward chronic pain mechanisms.²³ Future research identifying the underlying mechanisms for this rapid shift from acute to chronic pain processing is likely to provide some valuable insights into potential therapeutic options for this clinically challenging problem.¹²³ Moreover, our findings offer potential targets for non-invasive brain stimulation to either inhibit facilitatory pain networks (e.g., cerebellum) or increase the activity of central pain attenuating regions (e.g., DLPFC) that could be used as a treatment for the challenging problem of post-traumatic headache.^124

–127

In summary, we found significant alterations in the functional connectivity of the perigenual and subgenual ACC seeds suggesting that PTH are associated with dysregulation in descending central pain modulatory systems. Pre-injury headaches play a significant role in the central modulation of pain processing. Further studies exploring differences in brain connectivity, especially in pain processing regions, during acute pain tasks could help in understanding the neurobiology of PTH.

Clinical implications

Altered functional connectivity of the perigenual and subgenual areas of the ACC in children with post-traumatic headache suggests dysregulation in descending pain modulation.

PTH severity is correlated with greater functional connectivity between the rostral ACC with thalamus, hippocampus and periaqueductal gray.

Children with pre-injury headaches may be prone to developing PTH due to downregulation of descending pain modulation (i.e., reduced negative functional connectivity between the rostral/perigenual ACC seeds and the cerebellum).

Supplemental material

Supplemental Material, sj-pdf-1-rep-10.1177_25158163211009477 - Functional connectivity of the anterior cingulate cortex with pain-related regions in children with post-traumatic headache

Supplemental Material, sj-pdf-1-rep-10.1177_25158163211009477 for Functional connectivity of the anterior cingulate cortex with pain-related regions in children with post-traumatic headache by Zahra Ofoghi, Christiane S Rohr, Deborah Dewey, Signe Bray, Keith Owen Yeates, Melanie Noel and Karen M Barlow in Cephalalgia Reports

Footnotes

Acknowledgements

We would like to acknowledge Brenda Turley, Tina Samuel, Aneesh Khetani and Regan King for their contributions. We thank all of the children and their families who participated in this study.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Ethics or institutional review board approval

The PlayGame Trial (REB 13-0372) and ICON study (REB 16-2057) received local ethical approval.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Researchers acknowledge the support of the Canadian Institutes of Health Research funding (CIHR) (grant number: TBI-293375), the CIHR-Era-Net Neuron (JTC 2016, project number: 10016191).

ORCID iD

Deborah Dewey

Supplemental material

Supplemental material for this article is available online.

References

Faul

Wald

, et al. Traumatic brain injury in the United States: emergency department visits, hospitalizations, and deaths 2002-2006. Atlanta (GA): Centers for Disease Control and Prevention,National Center for Injury Prevention and Control, 2010.

Dewan

Mummareddy

Wellons

3rd , et al. Epidemiology of global pediatric traumatic brain injury: qualitative review. World Neurosurg 2016; 91: 497–509.e1.

Barlow

. Postconcussion syndrome. J Child Neurol 2016; 31(1): 57–67.

Barlow

Crawford

Stevenson

, et al. Epidemiology of postconcussion syndrome in pediatric mild traumatic brain injury. Pediatrics 2010; 126(2): e374–e381.

Zemek

Barrowman

Freedman

, et al. Clinical risk score for persistent postconcussion symptoms among children with acute concussion in the ED. JAMA 2016; 315(10): 1014–1025.

Barthas

Sellmeijer

Hugel

, et al. The anterior cingulate cortex is a critical hub for pain-induced depression. Biol Psychiatry 2015; 77(3): 236–245.

Fuchs

Peng

Boyette-Davis

, et al. The anterior cingulate cortex and pain processing. Front Integr Neurosci 2014; 8: 35.

Rainville

Duncan

Price

, et al. Pain affect encoded in human anterior cingulate but not somatosensory cortex. Science 1997; 277(5328): 968–971.

Tang

Ding

, et al. Pavlovian fear memory induced by activation in the anterior cingulate cortex. Mol Pain 2005; 1: 6.

10.

Zhuo

. Long-term potentiation in the anterior cingulate cortex and chronic pain. Philos Trans R Soc B Biol Sci 2014; 369(1633): 20130146.

11.

Stevens

Lovejoy

Kim

, et al. Multiple resting state network functional connectivity abnormalities in mild traumatic brain injury. Brain Imaging Behav 2012; 6(2): 293–318.

12.

Plourde

Rohr

Virani

, et al. Default mode network functional connectivity after multiple concussions in children and adolescents. Arch Clin Neuropsychol 2020; 35(3): 302–311.

13.

Leung

Shukla

Yang

, et al. Diminished supraspinal pain modulation in patients with mild traumatic brain injury. Mol Pain 2016; 12: 1744806916662661.

14.

Leung

Yang

Lim

, et al. Pain-related white matter tract abnormalities in mild traumatic brain injury patients with persistent headache. Mol Pain 2018; 14: 1744806918810297.

15.

Schwedt

TJ.

Structural and functional brain alterations in post-traumatic headache attributed to mild traumatic brain injury: a narrative review. Front Neurol 2019; 10: 615.

16.

Ofoghi

Dewey

Barlow

. A systematic review of structural and functional imaging correlates of headache or pain after mild traumatic brain injury. J Neurotrauma 2020; 37(7): 907–923.

17.

Toledo

Lebel

Becerra

, et al. The young brain and concussion: imaging as a biomarker for diagnosis and prognosis. Neurosci Biobehav Rev. 2012; 36(6): 1510–1531.

18.

Niu

Bai

Sun

, et al. Disruption of periaqueductal grey-default mode network functional connectivity predicts persistent post-traumatic headache in mild traumatic brain injury. J Neurol Neurosurg Psychiatry 2019; 90(3): 326–332.

19.

Luna

Sweeney

. The emergence of collaborative brain function: FMRI studies of the development of response inhibition. Ann N Y Acad Sci 2004; 1021: 296–309.

20.

Karlin

AM.

Concussion in the pediatric and adolescent population: “different population, different concerns”. PM R 2011; 3(10 suppl 2): S369–S379.

21.

Giza

Prins

. Is being plastic fantastic? mechanisms of altered plasticity after developmental traumatic brain injury. Dev Neurosci 2006; 28(4–5): 364–379.

22.

Powers

Coffey

Chamberlin

, et al. Trial of amitriptyline, topiramate, and placebo for pediatric migraine. N Engl J Med 2017; 376(2): 115–124.

23.

Kuczynski

Crawford

Bodell

, et al. Characteristics of post-traumatic headaches in children following mild traumatic brain injury and their response to treatment: a prospective cohort. Dev Med Child Neurol 2013; 55(7): 636–641.

24.

Hoffman

Lucas

Dikmen

, et al. Natural history of headache after traumatic brain injury. J Neurotrauma 2011; 28(9): 1719–1725.

25.

Barlow

Brooks

Esser

, et al. Efficacy of melatonin in children with postconcussive symptoms: a randomized clinical trial. Pediatrics 2020; 145(4): e20192812.

26.

Giza

Kutcher

Ashwal

, et al. Summary of evidence-based guideline update: evaluation and management of concussion in sports: report of the guideline development subcommittee of the American Academy of Neurology. Neurology 2013; 80(24): 2250–2257.

27.

Lovell

Iverson

Collins

, et al. Measurement of symptoms following sports-related concussion: reliability and normative data for the post-concussion scale. Appl Neuropsychol 2006; 13(3): 166–174.

28.

Lovell

Collins

. Neuropsychological assessment of the college football player. J Head Trauma Rehabil 1998; 13(2): 9–26.

29.

Sady

Vaughan

Gioia

. Psychometric characteristics of the postconcussion symptom inventory in children and adolescents. Arch Clin Neuropsychol 2014; 29(4): 348–363.

30.

Gioia

Schneider

Vaughan

, et al.

Which symptom assessments and approaches are uniquely appropriate for paediatric concussion?

Br J Sports Med 2009; 43(Suppl 1): i13–i22.

31.

HealthActCHQ Inc. CHQ: Child Health Questionnaire [Internet]. Healthactchq, Inc. https://www.healthactchq.com/survey/chq?gclid=CIOtkp-pl84CFZRefgodD20MjAnhg (accessed 5 April 2021).

32.

Landgraf

Maunsell

Speechley

, et al. Canadian-French, German and UK versions of the child health questionnaire: methodology and preliminary item scaling results. Qual Life Res 1998; 7(5): 433–445.

33.

Waters

Salmon

Wake

Hesketh

Wright

. The child health questionnaire in Australia: reliability, validity and population means. Aust N Z J Public Health 2000; 24(2): 207–210.

34.

Konijnenberg

Uiterwaal

Kimpen

, et al. Children with unexplained chronic pain: Substantial impairment in everyday life. Arch Dis Child 2005; 90(7): 680–686.

35.

Kwan

Noel

, et al.

A scoping review of pain in children after traumatic brain injury: Is there more than headache?

J Neurotrauma 2018; 35(7): 877–888.

36.

Batailler

Hours

Maza

, et al. Health status recovery at one year in children injured in a road accident: a cohort study. Accid Anal Prev 2014; 71: 267–272.

37.

Raat

Landgraf

Bonsel

, et al. Reliability and validity of the child health questionnaire-child form (CHQ-CF87) in a Dutch adolescent population. Qual Life Res 2002; 11: 575–581.

38.

Raat

Mangunkusumo

Landgraf

, et al. Feasibility, reliability, and validity of adolescent health status measurement by the Child Health Questionnaire Child Form (CHQ-CF): internet administration compared with the standard paper version. Qual Life Res 2007; 16(4): 675–685.

39.

Michalos

. Encyclopedia of quality of life and well-being research. Dordrecht: Springer International Publishing; 2014.

40.

Rentz

Matza

Secnik

. Psychometric validation of the child health questionnaire (CHQ) in a sample of children and adolescents with attention-deficit/hyperactivity disorder. Qual Life Res 2005; 14(3): 719–734.

41.

IBM Corp. IBM SPSS statistics for macintosh, Version 24.0. Armonk, NY: IBM Corp; 2016.

42.

Jenkinson

Beckmann

Behrens

TEJ

, et al. FSL. Neuroimage 2012; 62(2): 782–790.

43.

Smith

. Fast robust automated brain extraction. Hum Brain Mapp 2002; 17(3): 143–155.

44.

Jenkinson

Bannister

Brady

, et al. Improved optimization for the robust and accurate linear registration and motion correction of brain images. Neuroimage 2002; 17(2): 825–841.

45.

Woolrich

Ripley

Brady

, et al. Temporal autocorrelation in univariate linear modeling of fMRI data. Neuroimage 2001; 14(6): 1370–1386.

46.

Power

Mitra

Laumann

, et al. Methods to detect, characterize, and remove motion artifact in resting state fMRI. Neuroimage 2014; 84: 320–341.

47.

Power

Barnes

Snyder

, et al. Spurious but systematic correlations in functional connectivity MRI networks arise from subject motion. Neuroimage 2012; 59(3): 2142–2154.

48.

Fox

Snyder

Vincent

, et al. The human brain is intrinsically organized into dynamic, anticorrelated functional networks. Proc Natl Acad Sci USA 2005; 102(27): 9673–9678.

49.

Aguirre

Zarahn

D’Esposito

. Empirical analyses of BOLD fMRI statistics. Neuroimage 1997; 5(3): 199–212.

50.

Aguirre

Zarahn

D’Esposito

. The inferential impact of global signal covariates infunctional neuroimaging analyses. Neuroimage 1998; 8(3): 302–306.

51.

Desjardins

Kiehl

Liddle

. Removal of confounding effects of global signal in functional MRI analyses. Neuroimage 2001; 13(4): 751–758.

52.

Zarahn

Aguirre

D’Esposito

Empirical Analyses of BOLD fMRI Statistics. I. spatially unsmoothed data collected under null-hypothesis conditions. Neuroimage 1997; 5(3): 179–197.

53.

Macey

Kumar

, et al. A method for removal of global effects from fMRI time series. Neuroimage 2004; 22(1): 360–366.

54.

Power

Plitt

Laumann

, et al. Sources and implications of whole-brain fMRI signals in humans. Neuroimage 2017; 146: 609–625.

55.

Chai

Castañón

Ongür

, et al. Anticorrelations in resting state networks without global signal regression. Neuroimage 2012; 59(2): 1420–1428.

56.

Murphy

Fox

. Towards a consensus regarding global signal regression for resting state functional connectivity MRI. Neuroimage 2017; 154: 169–173.

57.

Liu

Nalci

Falahpour

The global signal in fMRI: nuisance or information?

Neuroimage 2017; 150: 213–229.

58.

Zhang

Brady

Smith

. Segmentation of brain MR images through a hidden Markov random field model and the expectation-maximization algorithm. IEEE Trans Med Imaging 2001; 20(1): 45–57.

59.

Bright

Tench

Murphy

. Potential pitfalls when denoising resting state fMRI data using nuisance regression. Neuroimage 2017; 154: 159–168.

60.

Margulies

Kelly

Uddin

, et al. Mapping the functional connectivity of anterior cingulate cortex. Neuroimage 2007; 37(2): 579–588.

61.

Kelly

AMC

Di Martino

Uddin

, et al. Development of anterior cingulate functional connectivity from late childhood to early adulthood. Cereb Cortex 2009; 19(3): 640–657.

62.

Kennis

Rademaker

van Rooij

SJH

, et al. Resting state functional connectivity of the anterior cingulate cortex in veterans with and without post-traumatic stress disorder. Hum Brain Mapp 2015; 36(1): 99–109.

63.

Zhou

Shi

Cui

, et al. Functional connectivity of the caudal anterior cingulate cortex is decreased in autism. PLoS One 2016; 11(3): e0151879.

64.

Allman

Hakeem

Erwin

, et al. The anterior cingulate cortex. The evolution of an interface between emotion and cognition. Ann N Y Acad Sci 2001; 935(1): 107–117.

65.

Symonds

Gordon

Bixby

, et al. Right-lateralized pain processing in the human cortex: an fMRI study. J Neurophysiol 2006; 95(6): 3823–3830.

66.

Apkarian

Baliki

Geha

. Towards a theory of chronic pain. Prog Neurobiol 2009; 87(2): 81–97.

67.

Kucyi

Davis

. The dynamic pain connectome. Trends Neurosci 2015; 38(2): 86–95.

68.

Bingel

Tracey

. Imaging CNS modulation of pain in humans. Physiology 2008; 23: 371–380.

69.

Desikan

Ségonne

Fischl

, et al. An automated labeling system for subdividing the human cerebral cortex on MRI scans into gyral based regions of interest. Neuroimage 2006; 31(3): 968–980.

70.

Frazier

Chiu

Breeze

, et al. Structural brain magnetic resonance imaging of limbic and thalamic volumes in pediatric bipolar disorder. Am J Psychiatry 2005; 162(7): 1256–1265.

71.

Kong

Zyloney

, et al. Intrinsic functional connectivity of the periaqueductal gray, a resting fMRI study. Behav Brain Res 2010; 211(2): 215–219.

72.

Winkler

Ridgway

Webster

, et al. Permutation inference for the general linear model. Neuroimage 2014; 92(100): 381–397.

73.

Chen

Yan

. Reproducibility of R-fMRI metrics on the impact of different strategies for multiple comparison correction and sample sizes. Hum Brain Mapp 2018; 39(1): 300–318.

74.

Nichols

Holmes

. Nonparametric permutation tests for functional neuroimaging: A primer with examples. Hum Brain Mapp 2002; 15(1): 1–25.

75.

Smith

Nicholas

. Threshold-free cluster enhancement: addressing problems of smoothing, threshold dependence and localisation in cluster inference. Neuroimage 2009; 44(1): 83–98.

76.

Holmes

Blair

Watson

JDG

, et al. Nonparametric analysis of statistic images from functional mapping experiments. J Cereb Blood Flow Metab 1996; 16(1): 7–22.

77.

Valet

Sprenger

Boecker

, et al. Distraction modulates connectivity of the cingulo-frontal cortex and the midbrain during pain--an fMRI analysis. Pain 2004; 109(3): 399–408.

78.

Lorenz

Minoshima

Casey

. Keeping pain out of mind: the role of the dorsolateral prefrontal cortex in pain modulation. Brain 2003; 126(5): 1079–1091.

79.

Stevens

Hurley

Taber

. Anterior cingulate cortex: unique role in cognition and emotion. J Neuropsychiatry Clin Neurosci 2011; 23(2): 121–125.

80.

Rohr

Dimond

Schuetze

, et al. Girls’ attentive traits associate with cerebellar to dorsal attention and default mode network connectivity. Neuropsychologia 2019; 127: 84–92.

81.

Seminowicz

Moayedi

. The dorsolateral prefrontal cortex in acute and chronic pain. J Pain. 2017; 18(9): 1027–1035.

82.

Gebhart

. Descending modulation of pain. Neurosci Biobehav Rev 2004; 27(8): 729–737.

83.

Ossipov

Morimura

Porreca

. Descending pain modulation and chronification of pain. Curr Opin Support Palliat Care 2014; 8(2): 143–151.

84.

Yuan

Qin

Liu

, et al. Reduced fractional anisotropy of corpus callosum modulates inter-hemispheric resting state functional connectivity in migraine patients without aura. PLoS One 2012; 7(9): e45476.

85.

Bingel

Lorenz

Schoell

, et al. Mechanisms of placebo analgesia: rACC recruitment of a subcortical antinociceptive network. Pain 2006; 120(1–2): 8–15.

86.

Kong

Jensen

Loiotile

, et al. Functional connectivity of the frontoparietal network predicts cognitive modulation of pain. Pain 2013; 154(3): 459–467.

87.

Murphy

Birn

Handwerker

, et al.

The impact of global signal regression on resting state correlations: Are anti-correlated networks introduced?

Neuroimage 2009; 44(3): 893–905.

88.

Qian

Diez

Ortiz-Terán

, et al. Positive connectivity predicts the dynamic intrinsic topology of the human brain network. Front Syst Neurosci 2018; 12: 38.

89.

Chen

Xie

, et al. Negative functional connectivity and its dependence on the shortest path length of positive network in the resting-state human brain. Brain Connect 2011; 1(3): 195–206.

90.

Goelman

Gordon

Bonne

. Maximizing negative correlations in resting-state functional connectivity MRI by time-lag. PLoS One 2014; 9(11): e111554.

91.

Buzsáki

Geisler

Henze

, et al. Interneuron diversity series: circuit complexity and axon wiring economy of cortical interneurons. Trends Neurosci 2004; 27(4): 186–193.

92.

Parente

Colosimo

. Anticorrelations between active brain regions: an agent-based model simulation study. Neural Plast 2018; 2018: 6815040.

93.

Shih

Chen

Shyu

, et al. A new scenario for negative functional magnetic resonance imaging signals: endogenous neurotransmission. J Neurosci 2009; 29(10): 3036–3044.

94.

Duncan

Wiebking

Northoff

. Associations of regional GABA and glutamate with intrinsic and extrinsic neural activity in humans-a review of multimodal imaging studies. Neurosci Biobehav Rev 2014; 47: 36–52.

95.

Shmuel

Augath

Oeltermann

, et al. Negative functional MRI response correlates with decreases in neuronal activity in monkey visual area V1. Nat Neurosci 2006; 9: 569–577.

96.

Devor

Tian

Nishimura

, et al. Suppressed neuronal activity and concurrent arteriolar vasoconstriction may explain negative blood oxygenation level-dependent signal. J Neurosci 2007; 27(16): 4452–4459.

97.

Shih

Chiang

Shyu

, et al. Endogenous opioid–dopamine neurotransmission underlie negative CBV fMRI signals. Exp Neurol 2012; 234(2): 382–388.

98.

Khaliq

Raman

. Axonal propagation of simple and complex spikes in cerebellar purkinje neurons. J Neurosci 2005; 25(2): 454–463.

99.

Zheng

Raman

. Synaptic inhibition, excitation, and plasticity in neurons of the cerebellar nuclei. Cerebellum 2010; 9(1): 56–66.

100.

Ruscheweyh

Kühnel

Filippopulos

, et al. Altered experimental pain perception after cerebellar infarction. Pain 2014; 155(7): 1303–1312.

101.

Mehnert

May

. Functional and structural alterations in the migraine cerebellum. J Cereb Blood Flow Metab 2019; 39(4): 730–739.

102.

Bernard

Seidler

Hassevoort

, et al. Resting state cortico-cerebellar functional connectivity networks: a comparison of anatomical and self-organizing map approaches. Front Neuroanat 2012; 6: 31.

103.

Holmes

Scheinost

DellaGioia

, et al. Cerebellar and prefrontal cortical alterations in PTSD: structural and functional evidence. Chronic Stress (Thousand Oaks) 2018; 2: 2470547018786390.

104.

Mills

Di Pietro

Alshelh

, et al. Brainstem pain-control circuitry connectivity in chronic neuropathic pain. J Neurosci 2018; 38(2): 465–473.

105.

Harper

Ichesco

Schrepf

, et al. Resting functional connectivity of the periaqueductal gray is associated with normal inhibition and pathological facilitation in conditioned pain modulation. J Pain 2018; 19(6): 635.e1–635.e15.

106.

Staud

. The important role of CNS facilitation and inhibition for chronic pain. Int J Clin Rheumtol 2013; 8(6): 639–646.

107.

Chen

Taniguchi

Chen

, et al. Top-down descending facilitation of spinal sensory excitatory transmission from the anterior cingulate cortex. Nat Commun 2018; 9: 1886.

108.

Zhuo

Descending facilitation: From basic science to the treatment of chronic pain. Mol Pain 2017; 13: 1744806917699212.

109.

Kucyi

Moayedi

Weissman-Fogel

, et al. Enhanced medial prefrontal-default mode network functional connectivity in chronic pain and its association with pain rumination. J Neurosci 2014; 34(11): 3969–3975.

110.

Machino

Kunisato

Matsumoto

, et al. Possible involvement of rumination in gray matter abnormalities in persistent symptoms of major depression: an exploratory magnetic resonance imaging voxel-based morphometry study. J Affect Disord 2014; 168: 229–235.

111.

Anderson

Bunce

Barbas

. Prefrontal–hippocampal pathways underlying inhibitory control over memory. Neurobiol Learn Mem 2016; 134(Pt A): 145–161.

112.

Kühn

Vanderhasselt

De Raedt

, et al. Why ruminators won’t stop: The structural and resting state correlates of rumination and its relation to depression. J Affect Disord 2012; 141(2–3): 352–360.

113.

Falahpour

Nalci

Liu

. The effects of global signal regression on estimates of resting-state blood oxygen-level-dependent functional magnetic resonance imaging and electroencephalogram vigilance correlations. Brain Connect 2018; 8(10): 618–627.

114.

Wang

Zhou

Zhuo

, et al. Altered functional connectivity of the cingulate subregions in schizophrenia. Transl Psychiatry 2015; 5: e575.

115.

Wang

, et al. Abnormal resting-state functional connectivity strength in mild cognitive impairment and its conversion to alzheimer’s disease. Neural Plast 2016; 2016: 4680972.

116.

Murphy

Bodurka

Bandettini

. How long to scan? The relationship between fMRI temporal signal to noise ratio and necessary scan duration. Neuroimage 2007; 34(2): 565–574.

117.

Van Dijk

KRA

Hedden

Venkataraman

, et al. Intrinsic functional connectivity as a tool for human connectomics: theory, properties, and optimization. J Neurophysiol 2010; 103(1): 297–321.

118.

Ditommaso

Hoffman

Lucas

, et al. Medication usage patterns for headache treatment after mild traumatic brain injury. Headache 2014; 54(3): 511–519.

119.

Jia

Chen

Tang

, et al. Atypical functional connectivity between the anterior cingulate cortex and other brain regions in a rat model of recurrent headache. Mol Pain 2019; 15: 1744806919842483.

120.

Yuan

Zhao

Cheng

, et al. Altered structure and resting-state functional connectivity of the basal ganglia in migraine patients without aura. J Pain 2013; 14(8): 836–844.

121.

Schwedt

Schlaggar

Mar

, et al. Atypical resting-state functional connectivity of affective pain regions in chronic migraine. Headache 2013; 53(5): 737–751.

122.

Skorobogatykh

van Hoogstraten

Degan

, et al.

Functional connectivity studies in migraine: what have we learned?

J Headache Pain 2019; 20(1): 108.

123.

Watanabe

Bell

Walker

, et al. Systematic review of interventions for post-traumatic headache. PM R 2012; 4(2): 129–140.

124.

Ashina

Iljazi

Al-Khazali

, et al. Persistent post-traumatic headache attributed to mild traumatic brain injury: deep phenotyping and treatment patterns. Cephalalgia 2020; 40(6): 554–564.

125.

Leung

Shukla

Fallah

, et al. Repetitive transcranial magnetic stimulation in managing mild traumatic brain injury-related headaches. Neuromodulation 2016; 19(2): 133–141.

126.

Feng

Zhang

, et al. Effects of non-invasive brain stimulation on headache intensity and frequency of headache attacks in patients with migraine: a systematic review and meta-analysis. Headache 2019; 59(9): 1436–1447.

127.

Bocci

Ferrucci

Priori

, et al. Cerebellar transcranial direct current stimulation (ctDCS) effect in perception and modulation of pain. In: Neurostimulation and neuromodulation in contemporary therapeutic practice. UK: IntechOpen, 2020.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.73 MB