Patients with temporomandibular disorders and chronic pain of myofascial origin display reduced alpha power density and altered Small-world properties of brain networks

doi:10.21203/rs.3.rs-2496732/v1

Download PDF

Research Article

Patients with temporomandibular disorders and chronic pain of myofascial origin display reduced alpha power density and altered Small-world properties of brain networks

https://doi.org/10.21203/rs.3.rs-2496732/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Chronic pain is one of the most common symptoms of temporomandibular disorders (TMD). Although its pathophysiology is still a challenge, TMD has been associated with changes in central nervous system activity related to pain modulatory capacity. This study was conducted to examine the cortical activity of patients with temporomandibular disorders and chronic pain of myofascial origin using quantitative electroencephalography (qEEG). Individuals with TMD and chronic pain and healthy controls were evaluated using qEEG in four consecutive conditions, all with closed eyes: 1) initial resting condition; 2) non-painful motor imagery task of hand movement; 3) painful motor imagery task of clenching the teeth; 4) final resting condition. Participants with TMD and chronic pain overall presented decreased alpha power density during baseline at rest, and non-painful and painful motor imagery tasks when compared to healthy controls. Furthermore, functional brain connectivity was distinct between groups, with TMD and chronic pain showing lower small-world values for the delta (all conditions), theta (eyes closed, painful and non-painful motor imagery task), and alpha bands (painful motor imagery task), and an increase in the beta band (all conditions). These results suggest that TMD and related chronic pain is associated with maladaptive plasticity in the brain, which may correspond to a reduced ability to modify brain activity during different mental tasks, including painful and non-painful imagery. These changes can be detected by qEEG, a method which may be very important because of its characteristics of good temporal resolution and the possibility to be performed in naturalistic setups.

Temporomandibular disorders

quantitative electroencephalography

motor imagery

brain connectivity

Temporomandibular disorders (TMD) are a set of musculoskeletal and neuromuscular disorders involving the temporomandibular joint, masticatory muscles, and associated structures (Carrara et al. 2010; Greene et al. 2010). TMD affects more than 25% of the world population (Dworkin et al. 1990; Gesch et al. 2004; de Godoi Gonçalves et al. 2010) with prevalence among females being up to four times higher than among males (Ferreira et al. 2016). Chronic pain is one of the most common TMD symptoms (Dahan et al. 2015; Visscher et al. 2016; Tchivileva et al. 2017), and it has been associated with significant changes in the central nervous system neurons (Cruccu et al. 1997). Thus, functional and structural changes in thalamus and somatosensory cortex (Lin 2014), as well as reduced pain modulatory capacity have been reported (Shibukawa et al. 2007). As it occurs with other syndromes, chronic pain in TMD has also been associated with central sensitization, leading to abnormal nociceptive processing and increased perception of pain (Lorduy et al. 2013; Harper et al. 2016).

Quantitative electroencephalography (qEEG) has been used to evaluate pain-related brain functioning in healthy individuals (Nickel et al. 2017) and patients with different chronic pain conditions (Pinheiro et al. 2016). These studies have mainly analyzed absolute and relative power density (Bjørk et al. 2009), as well as peak frequency (Sarnthein et al. 2006; Boord et al. 2008; de Vries et al. 2013; Dinh et al. 2019) during both baseline at rest and imagining a painful task (Sarnthein et al. 2006; Vuckovic et al. 2014). During baseline conditions at rest, patients with neuropathic pain, fibromyalgia, cancer pain, and low back pain (Pinheiro et al. 2016), as well as rheumatoid arthritis (Meneses et al. 2016) displayed increased alpha and theta power densities as compared to healthy individuals. By contrast, studies in patients with abdominal pain secondary to chronic pancreatitis (de Vries et al. 2013), neuropathic pain associated with spinal cord injury (Boord et al. 2008) or with severe forms of neuropathic pain (Sarnthein et al. 2006) have revealed lower-alpha peak frequency (e. g., the discrete frequency with the highest magnitude within the alpha range) than in controls. Also, Dinh et al (2019) showed that patients with chronic pain present increased synchrony in the theta and gamma frequencies in the frontal brain areas. However, although a range of studies have assessed brain functioning in many chronic pain conditions, data from TMD patients with chronic pain are scarce and derived mainly from neuroimaging techniques.

A recent systematic review of the literature has shown that TMD with chronic pain is associated with brain changes in the default mode network, trigemino-thalamocortical top-down antinociceptive, motor, lateral medial pain systems, as revealed by functional magnetic resonance imaging (fMRI) (Yin et al. 2020). By contrast, EEG and magnetoencephalographic (MEG) studies have the ability to reveal maladaptive changes in brain functioning with higher temporal resolution, thus adding information to a broader understanding of brain functioning. Among the few studies that examined patients with chronic pain due to TMD, Alonso et al. (2010) have shown that patients with TMD/CP exhibited changes in the perception and processing of facial innocuous stimuli, with a persistent cortical activation after stimulation (Alonso et al. 2010). Of utmost importance was that changes in brain processing occurred in the anterior cingulate, amygdala, and even the primary auditory cortex when vibrotactile stimuli were applied not only to the face but to the index finger (Nebel et al. 2010). Recently, Pietro et al. (2018) have also shown that participants with neuropathic orofacial pain presented increased EEG power density in the range from 4Hz to 25Hz and that this enhancement was positively correlated to pain intensity (Pietro et al. 2018). However, Baroni et al. (2020) could not replicate these differences in EEG power density when comparing TMD with chronic pain patients and healthy controls during a Pressure Pain Threshold (PPT) task.

As it seems that brain processing in TMD with chronic pain could be altered during conditions that replicate natural daily life events such as being touched in the skin, one would argue if other similar conditions such as contracting muscles and/or moving the jaw/temporomandibular joint would also be associated with abnormal activity in the brain. In this sense, although cortical dynamics in response to movement-evoked pain seem to be altered in low back pain patients, little is known about the effects of motor tasks in other chronic pain conditions (Corbett et al. 2018). A recent study using a jaw force task showed that pain elicited by the task is able to attenuate responses in the alpha and beta EEG bands (Wang et al. 2020). Nevertheless, the use of overt pain-evoking motor tasks may provoke high levels of discomfort in participants, and other experimental tasks should be used to explore the effects of motor activity in pain perception. In this sense, it has been repeatedly shown that motor imagery can activate similar brain regions to those involved in movement execution (Case et al. 2015) and, therefore, it could hypothetically alter brain activity associated with pain regulation. In this sense, it has been shown that subjective ratings and N200 amplitudes of event-related potentials elicited by painful stimuli can be increased by asking subjects to imagine a lesion on the forearm (Fardo et al. 2015). Motor imagery may worsen pain and edema in patients with pain (Moseley et al. 2008) by activating the pro-nociceptive network, but probably its discomfort is less than movement experience, although conserving the possibility to show brain dynamics during body actions. As EEG analysis may discriminate between painful versus nonpainful MI (Shamsi et al. 2020), this strategy may be useful to understand brain dynamics during movement-provoked pain in TMD with chronic pain, which is relevant as jaw movements are highly important to daily life.

In addition to analyzing brain oscillations and event-related potentials, EEG enables further understanding of brain dynamics by calculating functional connectivity, as well as small-world properties of brain networks in healthy and different painful conditions (Bassett and Bullmore 2006; Nebel et al. 2010; Kuner and Flor 2016; Nickel et al. 2017; Dinh et al. 2019; Rossini et al. 2019). This small-world architecture of the brain is defined as a large dynamic and functional network characterized by many local connections and few long-range connections, and can be studied using methodologies such as graph theory (Watts and Strogatz 1998). Previous fMRI studies have found that small-world parameters are significantly altered in pain conditions such as low back pain (Liu et al. 2018), and post-herpetic neuralgia (Zhang et al. 2014), showing that chronic pain have unstable and inefficient topology of brain functional networks (reduced functional connectivity in cingulate, somatosensory and frontal cortices, as well as reduced nodal efficiency), when compared with healthy controls. Nevertheless, although some studies have shown altered connectivity in patients with chronic pain associated with TMD (He et al. 2018; Mills et al. 2021; Festa et al. 2021), nothing is known about the small-world properties of brain networks assessed with EEG in those patients.

In summary, the analysis of the current knowledge about brain functioning assessed through EEG in TMD with chronic pain shows potential limitations that should be elucidated. Thus, it should be replicated if patients with chronic pain due to TMD display a significant enhancement of EEG power density as compared with healthy controls. Furthermore, the impact of movement-evoked pain should be further examined in these patients by using a more ecologically-based paradigm, such as motor imagery. In the present study, we propose to investigate these aspects in a specific group of TMD with chronic pain subjects, those with myofascial pain; hence, excluding other sources of pain (articular, neuropathic nature) due to TMD, and increasing sample homogeneity. Moreover, the present study will extend our knowledge about brain alterations associated with chronic pain in this population by assessing small-world properties of brain networks using EEG at rest and during painful and non-painful motor imagery tasks. We predict that patients with chronic pain of myofascial origin due to TMD, in comparison to healthy controls, will show (1) a general enhancement of EEG power density at rest; (2) increased power density of low frequencies (delta and theta), together with decreased power density of higher frequencies (alpha and beta) at rest; (3) attenuation of alpha and theta power densities during the painful motor imagery task; (4) altered small-world properties of brain networks.

2.1 Study design and sample

In this cross-sectional study, 18 female subjects with TMD chronic pain of myofascial origin and 18 age-matched female healthy controls were recruited. Two TMD with chronic pain subjects and two HC were excluded after the EEG artifact rejection protocol. Thus, we had a final sample of 16 TMD with chronic pain subjects (mean age = 40.19 ±10.68 years) and 16 healthy controls (mean age = 32.94 ±11.68 years). Data collection occurred between January 2014 and March 2016 at the Functional Electrostimulation Laboratory of the Federal University of Bahia, Brazil. The study was carried out in accordance with the principles of the Declaration of Helsinki and approved by the Research Ethics Committee of Maternity Climério de Oliveira, Salvador-Bahia, Brazil (# 1234340).

A multidisciplinary team of dentists and speech-language therapists with more than 10 years of clinical experience in orofacial pain established the diagnosis of our participants. The Fonseca Anamnestic Questionnaire (Berni et al. 2015) was used during the screening of symptomatic participants. To confirm the diagnosis, we used the Research Diagnostic Criteria for Temporomandibular Disorders (RDC/TMD) questionnaire axis I. Only those with myofascial TMD were included in the study. Psychosocial aspects were assessed with RDC/TMD axis II, culturally adapted for the Brazilian population (Lucena et al. 2006). The healthy group was also assessed through the Fonseca Anamnestic Questionnaire to screen and confirm the absence of TMD symptoms.

As inclusion criteria, subjects with TMD should also present daily or near-daily pain of myofascial origin for at least six months. Pain in the last six months should have an intensity equal to or greater than 3 in the Visual Numerical Scale (VNS), ranging from 0 to 10. Subjects with inflammatory connective tissue disease (e.g. ankylosing spondylitis, rheumatoid arthritis, and psoriatic arthritis), fibromyalgia, neurological disorders, dental pain, ongoing orthognathic surgery, or using dental splints were excluded. All subjects signed and received a copy of the informed consent form.

2.2 Evaluation procedure

All subjects were assessed in the same environment, by the same evaluator (author CHI), who is a physiotherapist with 20 years of professional experience and knowledge in TMD. She was trained for electroencephalographic data acquisition by a clinical neurophysiologist and an EEG technician.

Clinical Assessment

For the initial evaluation, the Brazilian version of the RDC/TMD questionnaire (axis I and II) was used. Axis I is used for confirming the diagnosis and to classify the type of TMD, which may be myogenic, arthrogenic, or mixed. Axis II was used to assess the quality of life and to obtain sociodemographic data (Lucena et al. 2006).

Clinical assessment questionnaires were used immediately before the electroencephalographic evaluation. For both groups, we used an anxiety/depression assessment scale. The Hospital Anxiety and Depression Scale (HAD) is composed of 14 items for the screening of anxiety and depression symptoms (Bjelland et al. 2002). The total score of the two subscales is 21, with a cutoff point of 8 for anxiety and 9 for depression (Castro et al. 2006).

Pain quality was assessed through a culturally adapted McGill pain questionnaire for the Brazilian population (Pimenta et al. 1996; Santos et al. 2006). This questionnaire evaluates pain experience in a subjective and multidimensional way, providing qualitative measures of clinical pain. The McGill pain questionnaire is composed of four categories of pain descriptors: sensory-discriminative, affective-motivational, cognitive-evaluative, and miscellaneous (Melzack 1975). The maximum score of "number of pain descriptors" for this Brazilian version is 20. The "total pain level", defined as the sum of pain intensity values, has a maximum score of 68 (Santos et al. 2006).

EEG data recording

The cortical activity was recorded by an EEG cap with 20 gold-plated electrodes, positioned on the scalp according to the 10/20 International System and using a BrainNet device (EMSA, Brazil). We used Nuprep abrasive gel (Weaver and Company, Aurora, CO, USA) to prepare the scalp and electroconductive paste to place EEG electrodes over the scalp. Subjects were seated comfortably in a Faraday cage and were instructed to be relaxed, with eyes closed, while monitored to stay awake. The EEG sampling rate was 200 Hz, with impedance lower than 5 kΩ, reference in Cz, and a notch filter of 60 Hz. The electrodes used were: Fp1, Fp2, F3, F4, F7, F8, Fz, T3, T4, T5, T6, C3, C4, P3, P4, Pz, O1, O2, Oz and a ground electrode, positioned on the left shoulder. Electromyography electrodes were placed on the flexor muscles of the dominant hand and at the masseter muscle of the dominant side to ensure participants were not making actual voluntary movements while attempting to perform imagery tasks.

At the beginning of the experimental session, participants were instructed about the two motor imagery tasks to be used and had a small period of training (described next). The trial period was finished after two or three attempts when the participants declared that the task was clear. Next, the participants had their eyes closed and four conditions were applied sequentially: a) initial rest; b) non-painful motor imagery task of hand movement; c) painful motor imagery of clenching the teeth; d) final rest, after the motor imagery tasks The latter condition was included in an attempt to clarify if possible changes during motor imagery would persist after the end of the motor imagery tasks. For the imagery tasks, subjects were instructed to perform kinesthetic motor imagery, which consisted in performing the imagery of clenching the dominant hand, representing a non-painful motor imagery task, and in the next condition clenching the teeth, representing a painful motor imagery task. Both resting-state conditions had a four minutes duration each, whereas each motor imagery condition lasted two minutes. The time interval between conditions was three minutes. An audio recording of the word ''clenching'' and “relaxing” was played to cue participants to perform the two motor imagery tasks. The “relaxing” cue was played five seconds after the “clenching” and the next “relaxing” cue was emitted after three seconds. In total, participants performed 16 “clenching” trials in each MI condition. See figure 1 for an illustration of the experimental procedure.

Before EEG data collection, subjects were asked to make clenching movements of hands and teeth in order to assure that EMG activity was recorded properly. During data collection, EMG data was evaluated only qualitatively by visual inspection. In the event of an increased signal, data collection was stopped and reinitiated.

2.3 EEG data processing and analysis

EEG data were analyzed using the EEGLab (ver. 13) toolbox in MATLAB environment (ver. 8.3). Data were filtered with a 0.5-45 Hz band-pass filter. To follow the same parameters used by our previous studies (Meneses et al. 2016; Santana et al. 2021), electrodes were re-referenced to the mean of all channels. Continuous recordings were segmented in epochs of 1,280 milliseconds, resulting in 186 epochs for the baseline conditions (nearly four minutes) and 93 epochs for the imagery tasks (nearly two minutes). This epoch size allowed a consistent evaluation of power densities in the frequency range of 1.5–30 Hz. All epochs were corrected by subtracting each data point from the mean amplitude of the first 100 ms. In addition, a semi-automatic protocol was used to remove amplitude artifacts using an upper limit of +750 µV, a lower limit of -750 µV, and ±2 global standard deviations as rejection criteria. In addition, epochs were visually assessed by two of the authors (CHI and FQC) to decide which epochs should be removed. The inter-rater reliability was high (approximately 95%). The individual’s highest rejection rate was approximately 60% (resulting in 77 epochs) during the resting-state conditions, whereas the individual’s highest rejection rate was around 50% (resulting in 29 epochs) during the imagery conditions. On the basis of these data, we decided to use the same number of epochs for all conditions and subjects. Therefore, only 29 epochs were selected for all participants in each of the four conditions.

The absolute power density was calculated for each electrode separately by applying and averaging the Fast Fourier Transform of each epoch. Then, we computed relative PD for each frequency band by dividing the absolute value of each electrode by their values in the total power spectrum. In addition, we computed regions of interest by grouping electrodes and averaging their relative power densities as follows: frontal (Fp1, Fp2, F3, Fz, F4), temporal (T3, T5, T4, T6), central (C3, C4), parietal (P3, Pz, P4) and occipital (O1, Oz, O2). The power densities of the following frequency bands were analyzed: delta [1.5-3.5 Hz], theta [4-7 Hz], alpha [8-12 Hz], and beta [13-30 Hz].

2.4 Statistical analyses

We used the Shapiro-Wilk test to analyze the normality of the data distribution. Differences in sociodemographic and clinical data between groups were assessed by Student's t-test or Mann-Whitney U test (continuous variables) and Fisher's exact test (categorical variables). We used the Bonferroni method to correct for multiple comparisons when analyzing sociodemographic, clinical data, and post hoc analysis for ANOVAs described below.

2.4.1 EEG relative power density

Repeated measures ANOVAs were used to evaluate differences between “group” (TMD patients with chronic pain vs. healthy controls), “brain region” (frontal vs. central vs. temporal vs. parietal vs. occipital), and “condition” (initial resting-state vs. nonpainful imagery vs. painful imagery vs. final resting-state). Basically, three sets of ANOVAs were performed. The first set was used to test our main hypothesis comparing the two groups during the four experimental conditions. The second set of statistical analyses was applied to test effects of group and experimental conditions separately at the different brain regions. The third set of ANOVAs was specifically performed to test differences between the initial and the final resting-state condition. All ANOVAs were repeated controlling for symptoms of anxiety and/or depression, in case of significant effects due to the main factor “group'' or the interaction “group*condition”. Post-hoc mean comparisons were reported by using corrected t-tests only when the interaction "condition x group" was significant.

Violations of the sphericity assumption were corrected by the Greenhouse-Geisser epsilon. We tested statistical significance at the .05 level. The SPSS 20.0 software package was used for these analyses.

2.4.2 Brain connectivity

The effective or functional connectivity can be assessed using interconnection and clustering parameters over nodes and edges, distinguishing regular from randomized networks generating the global measures (Ioannides 2007) and (Park and Friston 2013). A global measure of the graph interconnectedness can be defined as the average path length L, which is the average number of steps from one node to another, taking the shortest path. A global measure of the graph density can be defined as the clustering coefficient C, which represents the proportion of neighbor nodes that are connected to each other, forming triangles. Ordered graphs exhibit long L and high C, while random ones have short L and low C. There are ordered graphs that show some efficiency in the sense of information diffusion when new long-distance connections arise causing the drop of the average path length while the average cluster coefficient remains practically unchanged. This phenomenon is called ‘small-world’, and here is referring to the idea of connecting any two different EEG channels with a few steps (Bassett and Bullmore 2006). Small-world networks are identified by short L and high C and have been detected in neural contexts being considered an optimal architecture for synchronizing neural activity between brain regions (Smit et al. 2008; Bullmore and Sporns 2009).

We inferred brain connectivity using a nonlinear causality measure called convergent cross-mapping (CCM) (Sugihara et al. 2012) which captures nonstationary coupling effects between brain areas. Given an EEG signal X={x₁,..., x_n}, CCM is based on embedding vectors X_ev(i)=(x_i,x_i+τ, x_i+2τ, x_{i+(m-1) τ}) , i=1,…,n-(m-1) τ, where m is the spatial dimension and τ is the time delay. For each X_ev(i) we calculate w_j , j=1,…,n-1 , a normalized distance-related weight to the other embedding vectors (Mønster et al. 2016). Considering a second EEG signal Y={y₁,..., y_n}, we can estimate predicted information from X, calculating ỹ_i= Σw_jx_j . Considering the prediction vector Ỹ={ỹ₁,..., ỹ_n}, CCM is defined as the correlation between Y and Ỹ. If X and Y are coupled, in the sense of complex systems, CCM shows values close to -1 or 1.

After a narrowband decomposition and considering the connectivity strength between channels, for each subject, condition, and epoch, CCM was evaluated resulting in a 20x20 matrix, where the diagonal values were ignored and the absolute values were taken into account. For each subject and condition, the medians of the CCM values (from pairs of channels) were used as a threshold to turn matrices into a binary representation (adjacency matrices) and to define the graphs. See Figure 2 for an illustration of the process.

Following Humphries et al (2006) and Smit et al (2008), the graph analysis was performed for the adjacency matrices calculating first the average path length (L) and the average clustering coefficient (C), and then the small-world parameter (S), defined as a ratio of standardized C and L values. All these parameters were computed separately for each EEG dataset filtered within the four band frequencies (delta, theta, alpha and beta). To attenuate inter-, and intra-subject variability, for each condition, we considered only the S-values in the interquartile interval, between the 25th and 75th percentiles.

These statistical analyses were carried out by the nonparametric Mann-Whitney U-test for inter-groups comparisons, with a significance level of 0.05. The MATLAB (ver. 8.3) environment was used for these analyses.

All the participants presented at least mild pain during data collection. Several patients were taking only weak analgesics, and they were asked to interrupt their medication for at least 12 hours before data collection.

3.1 Sociodemographic data

Sociodemographic data of both groups is shown in Table 1. TMD patients with chronic pain subjects presented higher scores of anxiety and/or depression [X²(1)=11.221, p=.002]. Thus, we adjusted all ANOVA models for those symptoms, which were significantly different between groups in the clinical evaluation. Clinical pain characteristics of TMD patients with chronic pain are described in Table 2.

Table 1. Comparison of Demographic, Clinical Characteristics of Women with Temporomandibular Disorders and Healthy Controls.
	Total sample(n=32) N(%) or mean (±SD)	TMD (n=16) N(%) or mean (±SD)	Controls (16) N(%) or mean (±SD)	P-value
Demographic Characteristics
Age, in years	36.56(±11.61)	40.19(±10.68)	32.94(±11.68)	.07
Race/color				.70
White	10(31.3)	4(25.00)	6(37.50)
Black, Mixed race, Others^a	22(68.80)	12(75.00)	10(62.50)
Marital Status				.47
Married	17(53.10)	10(62.50)	7(43.80)
Separated/Divorced/ Widow	15(46.90)	6(37.50)	9(56.30)
Clinical Characteristics
Depression/Anxiety (HADS)				.002*
With anxiety and/or depression	11(34.40)	10(62.50)	1(6.30)
Without anxiety and/or depression	21(65.60)	6(37.50)	15(93.80)
Differences between groups were tested using Student’s t-test for continuous variables, and Fisher’s test or Mann Whitney for categorical variables.^*P < .05, ^a Others = total of individuals self-declared ``Yellow/Oriental`` and ``Indian``; SD = Standard Deviation; TMD = Temporomandibular Disorders; HADS = Hospital Anxiety and Depression Scale.

Table 2. Characteristics of the Pain in Women with Temporomandibular Disorders.
Characteristics of pain	Mean (±SD)
VNS ^a	3.69(±3.15)
VNS in the past six months ^b	7.31(±1.53)
Pain duration, in years ᶜ	4.22(±5.81)
Number of pain descriptors ^d	14.13(±3.64)
McGill pain index ^e	32.60(±11.29)
SD=Standard Deviation, VNS = Visual Numerical Scale ^a٫^b٫^c Research Diagnostic Criteria for Temporomandibular Disorders (RDC/TMD) – Axis I ^d٫^e McGill Pain Questionnaire variables

3.2 EEG relative power density

3.2.1 Group differences across experimental conditions

Repeated measures ANOVA using the between-subjects factor “group” (TMD patients with chronic pain vs. healthy controls) and the within-subjects factors “condition” (initial resting vs non-painful imagery vs. painful imagery vs. final resting) and “brain region” (frontal vs. central vs. temporal vs. parietal vs. occipital) were used to analyze overall differences in power density at the four EEG band frequencies: delta, theta, alpha and beta. These analyses revealed significant differences in power density due to the factor “group” at the alpha [F(1,30) = 7.786, P = .009] and the beta frequency band [F(1,30) = 5.359, P = .028]. In addition, a significant interaction "condition x group" was found in power density at the delta band [F(3,360) = 3.175, P = .043, GG epsilon = .634]. No significant differences were observed in power density at the theta band.

To further analyze whether the above findings were due to group differences in mood, additional ANCOVAs were performed on power density of delta, alpha, and beta frequency bands controlling for anxiety and depression. Significant effects due to “group” [F(1,29) = 6.613, P = .016] and "condition x group" [F(3,348) = 3.893, P =.022, GG epsilon = .736] were only observed in power density in the alpha band, confirming that regardless of differences in anxiety/depression, TMD patients with chronic pain had a decreased alpha power density at initial resting condition (mean difference=.166, P=.003), as well as during nonpainful (mean difference=.173, P=.006), and painful motor imagery tasks (mean difference=.125, P=.045) (Table 3). No significant effects were observed at the final resting condition. Regarding the beta band, only significant effects due to the factor “group” were observed [F(1,29) = 5.942, P = .021], indicating higher power density in TMD patients with chronic pain than in healthy controls. Finally, no significant effects were yielded in power density at the delta band when anxiety/depression was controlled for.

Table 3. Relative power density for the interaction between conditions and groups, controlling for symptoms of anxiety and/or depression, in women with TMD/CP and HC.
Frequency band and Condition	TMD/CP Participants (n=16) Mean (±SD)	HC Participants (n=16) Mean (±SD)	F (3,348)	P-value
Delta			2,558	.080
REST_1	1.40(.16)	1.37(.16)		.55
MI_1	1.41(.16)	1.37(.16)		.48
MI_2	1.41(.17)	1.40(.17)		.88
REST_2	1.37(.18)	1.43(.18)		.39
Theta			1.221	.303
REST_1	1.20(.12)	1.23(.12)		.43
MI_1	1.20(.10)	1.22(.10)		.59
MI_2	1.21(.11)	1.25(.11)		.39
REST_2	1.19(.14)	1.27(.14)		.17
Alpha			3.893	.022♯
REST_1	1.21(.13)	1.37(.13)		.003*
MI_1	1.21(.15)	1.38(.15)		.006*
MI_2	1.24(.15)	1.36(.15)		.045*
REST_2	1.23(.15)	1.33(.15)		.115
Beta			0.218	.865
REST_1	.83(.05)	.78(.05)		.011
MI_1	.83(.05)	.78(.05)		.024
MI_2	.82(.06)	.77(.06)		.057
REST_2	.83(.06)	.78(.06)		.033
ANOVAof repeated measures. ♯P < .05 F-test statistics presented here for the interaction between conditions and groups. *P < .05 Between-group comparisons (Bonferroni correction) are in bold. TMD/CP = Temporomandibular disorder and chronic pain, HC = healthy controls, SD = standard deviation, REST_1 = initial resting state, MI_1 = non-painful motor imagery, MI_2 = painful motor imagery, REST_2 = final resting state.

3.2.2 Effects of group and conditions on brain topography of alpha power density

The second set of analyses aimed to examine the topographical distribution of the “group” and "condition x group" effects in the alpha frequency band. For this purpose, separate repeated-measures ANOVAs were performed for each brain region of interest (frontal, central, parietal, temporal, and occipital) with the between-subjects "group" factor (TMD patients with chronic pain versus healthy controls) and within-subjects. Subject factors “condition” (initial resting-state vs. non-painful imagery vs. painful imagery vs. final resting-state), and controlling for anxiety and/or depression. Table 4 summarizes the descriptive statistics (mean and standard deviation) of alpha power density separated by groups, conditions and brain regions.

Frontal. The ANOVA showed significant group differences in alpha power density [F(3,87) = 4.087, P = .019, GG epsilon = .725]. Post-hoc analysis revealed that TMD patients with chronic pain had decreased alpha power density during initial rest (mean difference = .185, P = .004), painful (mean difference = .200, P = .005), and non-painful motor imagery (mean difference = .150, P = .039) conditions in comparison to healthy controls.

Central. The ANOVA showed no differences due to “group” or “condition x group” effects in alpha power density [F(3,87) = 2.599, P = .065].

Parietal. The ANOVA showed significant differences due to “condition x group” in alpha power density [F(3,87) = 3.326, P = .032]. Post-hoc analysis indicated decreased alpha power density in TMD patients with chronic pain during the initial rest (mean difference = .173, P = .012) and the non-painful motor imagery (mean difference = .177, P = .015) conditions in comparison to healthy controls.

Temporal. The ANOVA showed significant group differences in alpha power density [F(3,87) = 3.456, P =.027]. Post-hoc analysis showed decreased alpha power density in TMD patients with chronic pain during the initial rest (mean difference=.126, P=.011) and the non-painful motor imagery (mean difference =.121, P =.022) conditions in comparison to healthy controls.

Occipital. The ANOVA showed no differences due to “group” or “condition x group” effects in alpha power density [F(3,87) = 2.948, P = .059].

Table 4. Relative alpha power density for the interaction between regions of interest and groups, controlling for symptoms of anxiety and/or depression, in women with TMD/CP and HC
Regions of Interest (study condition)	TMD/CP Participants (n=16) Mean (±SD)	HC Participants (n=16) Mean (±SD)	F (3,87)	P-value
FRONTAL			F = 4.087	.019#
REST_1	1.17(.13)	1.32(.14)		.004*
MI_1	1.19(.15)	1.34(.16)		.005*
MI_2	1.21(.15)	1.34(.16)		.039*
REST_2	1.20(.15)	1.30(.13)		.133
TEMPORAL			F = 3.456	.027#
REST_1	1.22(.12)	1.31(.09)		.011*
MI_1	1.22(.13)	1.31(.10)		.022*
MI_2	1.25(.11)	1.32(.11)		.187
REST_2	1.23(.14)	1.29(.10)		.274
CENTRAL			F = 2.599	.065
PARIETAL			F = 3.326	.032#
REST_1	1.25(.13)	1.40(.16)		.012*
MI_1	1.25(.13)	1.39(.18)		.015*
MI_2	1.26(.13)	1.38(.17)		.069
REST_2	1.26(.15)	1.36(.15)		.125
OCCIPITAL			F = 2.948	.059
ANOVAof repeated measures. #P < .05 F-test statistics presented here for the interaction “brain regions” study “conditions”. P < .05 Conditions that presented significant differences at the Post-hoc* tests (with Bonferroni correction) are in bold. TMD/CP = Temporomandibular disorder and chronic pain, HC = healthy controls, SD = standard deviation, REST_1 = initial resting state, MI_1 = non-painful motor imagery, MI_2 = painful motor imagery, REST_2 = final resting state.

3.2.3. Post-effects of motor imagery (initial and final resting-state conditions compared)

The third set of analyses aimed to further examine group differences in alpha power densities during initial and final resting conditions. For this purpose, separate repeated-measures ANOVAs were performed for each brain region of interest (frontal, central, parietal, temporal, and occipital) with "group" as between-subjects factor (TMD patients with chronic pain versus healthy controls) and “condition” as within-subjects factor (initial resting vs. final resting condition), and controlling for anxiety and/or depression. A significant "condition x group" effect [F(1,116) = 5.097, P = .037] was yielded. Post-hoc analyses of this interaction effect showed that TMD patients with chronic pain had more reduced power densities than healthy controls during the initial resting condition in all regions of interest: frontal (mean difference = .185, P = .004), temporal (mean difference = .126, P = .011), central (mean difference = .220, P < .0001), parietal (mean difference = .173, P = .012) and occipital (mean difference = .127, P = .025). However, no significant group differences were observed during the final resting condition (Figure 3).

Post-hoc analyses of the interaction effect also revealed no significant differences between the initial and the final resting condition within TMD patients with chronic pain, whereas more reduced alpha power densities were observed within healthy controls at temporal (mean difference = .042, P = .037), central (mean difference = .053, P = .036) and parietal regions (mean difference = .047, P = .024) during the final than during the initial resting condition (Figure 4).

3.3 Small-world properties of brain networks

The inter-group comparisons for the distributions of the S parameter are presented in Figure 5 for each group, condition and EEG frequency band (delta, theta, alpha and beta). The values of the S parameter were significantly different between groups for the delta, theta, and beta frequency bands across conditions (except for the theta band in the final resting condition). S medians were overall higher in healthy controls than in TMD patients with chronic pain for delta and theta frequency bands, whereas they were higher in TMD patients with chronic pain than in healthy controls for beta frequency band. Furthermore, no significant group differences were observed in the S parameter for the alpha frequency band (except during the painful motor imagery condition).

Our objective was to characterize brain activity and small-world properties of brain networks at rest and during painful and non-painful imagery tasks in patients with temporomandibular disorders (TMD) and chronic pain of myofascial origin in comparison to healthy controls. Statistical analyses were also controlled for symptoms of anxiety and depression, as they could be confusing variables in the interpretation of the results. Partially in accordance with our predictions, results indicated that TMD patients with chronic pain had lower relative power density than healthy controls in the alpha frequency band at rest, specially at frontal, temporal, and parietal brain regions. Furthermore, TMD patients with chronic pain displayed greater attenuation of alpha power densities than controls during both painful and non-painful motor imagery tasks. Finally, we observed altered properties of small-world parameters of brain networks in TMD patients with chronic pain as compared to healthy controls for the delta, theta, alpha and beta frequency bands of the EEG.

4.1 Relative EEG Power Density

4.1.1 TMS patients with chronic pain displayed reduced alpha power density at rest

We found that TMD patients with chronic pain displayed decreased alpha power density in comparison to healthy controls at rest. Previous studies have found reduced alpha power densities at frontal, central, temporal and parietal brain regions in pain patients with spinal cord injury (Jensen et al. 2013), including tetraplegia (Tran et al. 2004), and mixed musculoskeletal and neuropathic pain patients (Camfferman et al. 2017). In this last study alpha power densities at the frontal and centro parietal areas of the cortex were negatively correlated with both poor sleep quality and pain intensity (Camfferman et al. 2017). However, other studies have found opposite findings in patients with central neuropathic pain associated with spinal cord injury (Sarnthein et al. 2006; Vuckovic et al. 2014), chronic pancreatitis (Drewes et al. 2008; Olesen et al. 2011), persistent pain after breast cancer (van den Broeke et al. 2013), and rheumatoid arthritis (Meneses et al. 2016). Also, decreased alpha power density is also present in other conditions such as drowsiness (Hughes and Crunelli 2005), or even by sleep deprivation (Kim et al. 2001; Strijkstra et al. 2003; Kaida et al. 2006), and depression (Hughes and Crunelli 2005), that are many times comorbidities in chronic pain patients. These discrepant results make it difficult to find a coherent behavior for alpha power density across different painful conditions. As we did not control the quality of sleep in our study, our results may be influenced by this variable, and future studies should consider assessing it.

In the present study, we specifically found group differences in alpha power density in frontal, temporal, and parietal brain regions, even when anxiety and depression (which are closely related to sleep disturbance) were controlled for, but not in other EEG frequency bands (delta, theta, beta). Furthermore, these group differences were not limited to the resting condition, but were maintained even during painful and non-painful motor imagery tasks. Thus, it appears that reduced power density of the alpha frequency band could be an intrinsic feature of brain activity in TMD patients, as it occurs in other functional pain syndromes.

Consistent with the view that reduced alpha power density might be an intrinsic feature of brain oscillations in TMD, we found no difference between initial and final resting conditions or between imagery tasks in these patients, whereas a significant reduction of alpha power density over frontal, motor, temporal and parieto-occipital cortices was found from initial to final resting condition in healthy controls. On the one hand, these findings may suggest that the motor imagery tasks had a significant physiological post-effect in healthy controls, specifically reducing power density in the alpha frequency and not in other EEG frequency bands. This is partially in accordance with previous research showing that motor imagery alters brain synchronization patterns, resulting in specific changes of power spectral density of EEG in alpha and beta frequency bands (Pfurtscheller and Neuper 1997). Accordingly, some studies have shown reduced alpha (and beta) power over the sensorimotor cortex when imaging walking movements (Kaneko et al. 2021). On the other hand, the lack of differences in power density between initial and final resting conditions for all frequency bands in TMD may further suggest that altered brain oscillations could be considered a stable characteristic of neuroplasticity associated with the presence of pain over time in these patients, as it occurs with other physiological markers of brain function (Yin et al. 2020). In this sense, previous studies have found altered functional connectivity between insula and several regions of the resting default-mode network (e.g., cingulate cortex and amygdala) which could be reflecting maladaptive neuroplasticity of pain-attention, endogenous pain inhibitory, affective and motor systems in TMD patients (Yin et al. 2020). Thus, for instance, enhanced activation of some nodes of the default-mode network (e.g medial prefrontal and posterior cingulate cortices) has been positively linked to pain rumination in TMD patients (Weissman-Fogel et al. 2011; Kucyi et al. 2014).

Furthermore, a similar reduction of alpha power density (event-related desynchronization) has also been reported in healthy individuals when experimental pain was elicited or anticipated (Babiloni et al. 2008). The authors further argued that decreases in alpha power could reflect an anticipatory facilitation of nociceptive thalamo-cortical and cortico-cortical channels. Furthermore, several authors have suggested that activation of the attentional network for pain occurs when attention is directed towards pain (Pessoa et al. 2003; Percio et al. 2006; Wiech et al. 2008; May et al. 2012). Thus, focusing attention would result in increased perception and pain discomfort, leading to decrease in alpha power densities. Moreover, it has been argued that pain stimuli could be associated with increased connectivity of somatosensory cortex and thalamus leading to suppression of alpha rhythmic oscillations (Percio et al. 2006; Anderson and Ding 2011; Liu et al. 2011). Finally, it has been observed that TMD patients with chronic pain present hypervigilance in relation to malocclusion of the mandible, meaning that they may be abnormally aware of the body sensations (McDermid et al. 1996; Hollins et al. 2009). Thus, our finding could also be the expression of an abnormal level of somatic awareness in TMD patients with chronic pain, manifested by hypervigilance in relation to occlusion, which could have physiological similarities to acute pain.

Regarding theta power density, no significant differences were found between TMD patients with chronic pain and healthy controls. Our hypothesis was based on the mechanism called 'thalamo-cortical dysrhythmia' (TCD) (Llinas et al. 1999; Stern et al. 2006) which has been discussed by several authors studying EEG activity in patients with chronic pain (Vuckovic et al. 2014; Ridder et al. 2015). Nevertheless, although there is some evidence of increased theta power density in TCD patients with chronic pain (Bjørk et al. 2009), findings in chronic pain are still controversial (Sarnthein et al. 2006; Olesen et al. 2011; Meneses et al. 2016; Camfferman et al. 2017). Thus, future studies should examine whether TCD is a characteristic of only some types of chronic pain or not.

4.1.2 TMD patients with chronic pain displayed reduced alpha power density during painful and non-painful motor imagery tasks

A further aim of the present research was to investigate how the brain of TMD patients with chronic pain reacts to painful and non-painful motor imagery tasks. Our initial hypothesis was that imagining the contraction of the jaw muscles or the temporomandibular joint would be associated with pain perception and abnormal brain activity due to the recruitment of attention circuits in TMD. Physiologically, decreased alpha power density in response to motor imagery has been mainly observed in the primary motor and somatosensory cortices (Pfurtscheller and Lopes da Silva 1999; Brinkman et al. 2014; Meirovitch et al. 2015). Moreover, previous studies have also shown that this reduction in the power density of alpha band after imagery tasks seems to be greater in individuals with CP, specifically in paraplegic patients with CP (Vuckovic et al. 2014) reflecting an enhanced activation of the sensorimotor cortex area (Jensen and Mazaheri 2010; Anderson and Ding 2011). The findings of the present study are partially in agreement with previous studies, since TMD patients with chronic pain had more reduced alpha power density than healthy controls in response to both painful and non-painful motor imagery tasks. Nevertheless, we failed to observe differential effects in brain responses to painful and non–painful imagery tasks in TMD patients with chronic pain. One possible explanation for this lack of effects in TMD is consistent with decreased brain responsiveness to sensory and cognitive stimulation, such as reduced amplitudes brain responses to painful stimuli (Romaniello et al. 2003), as have been repeatedly seen in other chronic pain syndromes (Veldhuijzen et al. 2006; Montoya et al. 2006; Pinheiro et al. 2016). Further research should explore if amplitudes and oscillations of event-related brain potentials would be able to discriminate between TMD patients with and without pain.

4.2 Brain connectivity

The range of S-values in TMD patients with chronic pain and healthy controls is consistent with those obtained previously (see tables II and VII in Smit et al [2008]). We hypothesized that there would be a change of effective brain connectivity characterized by the `small world` network parameter between TMD patients with chronic pain and healthy controls. Considering S as the `small world` parameter, the hypothesis was confirmed with a significant decrease of S in TMD patients with chronic pain in the delta (all conditions), theta (baseline resting state, non-painful MI and painful MI conditions) and alpha bands (painful MI condition). In the beta band, the results showed an increase of S in all conditions in TMD patients with chronic pain compared to healthy controls.

Studies of functional connectivity have raised new concepts of brain functionality. A significant number of studies on functional brain mechanism with fMRI has grown in recent decades as in Alzheimer's disease, schizophrenia and currently in chronic pain (Farmer et al. 2012). One approach is that the brain behaves as an interconnected network respecting a standard of effectiveness and stability characterizing 'small world' networks. This network architecture may be optimal for synchronizing neural activity between brain regions (Smit et al. 2008).

Networks that maintain normal brain function can be disrupted by the presence of diseases, including chronic pain. The Default Mode Network (DMN) is one of the most studied networks in the resting state and reflects a coherence set of brain regions that are preferentially affected by the presence of chronic pain such as seen in fibromyalgia, chronic back pain, and diabetic neuropathy (Baliki et al. 2008; Cauda et al. 2009; Napadow et al. 2010; Cifre et al. 2012). Studies of functional connectivity in TMD patients also showed alterations of the DMN in the fronto medial and posterior cingulate cortex regions (Weissman-Fogel et al. 2011; Kucyi et al. 2014), as well as reduced functional connectivity of corticostriatal networks (He et al. 2018), suggesting relevant deficits in motor control, pain processing and cognition in these patients. Studies of pain-evoked responses using EEG found decreased functional connectivity in theta band, suggesting top-down modulation in the pattern observed by frontocentral node network (Taesler and Rose 2016). In the present study, we found that TMD patients with chronic pain had reduced values of the `small world` parameters in three EEG frequencies (delta, theta and alpha), indicating significant reductions of local nodes together with enhancements of long-distance node interactions, and suggesting a low-effective connectivity network. By contrast, ‘small world’ parameters in beta frequency during all conditions in TMD were characteristics of an effective network. Future investigations should elucidate if these “small-world” parameters extracted from EEG and functional connectivity obtained from fMRI are complementary measures of the local and global connectivity networks which are disrupted in TMD patients with chronic pain.

4.3 Limitations

Our study has limitations that must be considered when interpreting the presented results. We evaluated only participants with myofascial TMD, which limits the external validity of our findings. We chose myofascial because it is the most severe type of TMD in terms of pain intensity and duration (Auerbach et al. 2001). Thus, the results cannot be extrapolated to all types of TMD, and future research should include other TMD classifications.

Our sample size did not allow studying subgroups of participants with or without different natures of CP (e.g. neuropathic and non-neuropathic) or other factors that might have influenced the results, such as quality of sleep, and degree of pain hypervigilance. Future studies could try to subgroup these conditions in order to explore the possible existence of a common thread within these classifications.

TMD patients with chronic pain of myofascial origin show decreased alpha power densities during baseline at rest, but no changes during painful and non-painful motor imagery tasks, which may correspond to a difficulty to change brain states according to different tasks. Confirming these findings, TMD patients with chronic pain displayed a less effective `small world` connectivity as extracted from EEG. These results reinforce the idea that chronic pain in TMD patients could be associated with maladaptive plasticity in the brain, and that these changes can be detected by qEEG, a method which may be very important because of its characteristics of good temporal resolution and possibility to be performed in naturalistic setups.

Acknowledgements
We are grateful for the funding from FAPESB, CAPES and CNPq; in which a master's scholarship was made possible and the contribution of a visiting professor from the University of Balearic Islands, Palma de Mallorca, Spain was provided. We thank Lucas G. França, Raphael Rosário, and José Garcia V. Miranda for their support in data analysis.

Data availability

The data that support the findings of this study are available on request from the corresponding author, AFB. The data are not publicly available due to ethical restrictions, as data sharing was not requested when the initial project was submitted to ethical evaluation.

Funding: This study was funded by FAPESB, also financed by CAPES (PVE A002_2013), and CNPq (Edital Universal 2014).

Conflicts of interest/Competing interests: The authors have no conflicts of interest to declare that are relevant to the content of this article.

Ethics approval: The study was approved by the Research Ethics Committee of Maternity Climério de Oliveira, Salvador-Bahia, Brazil (# 1234340).

Consent to participate: Research participants signed a free and informed consent form for the study.

Consent for publication: Research participants declare to be aware of the publication of research results.

Availability of data and material: Data and study materials are available if requested for analysis.

Alonso AA, Koutlas IG, Leuthold AC, et al (2010) Cortical processing of facial tactile stimuli in temporomandibular disorder as revealed by magnetoencephalography. Experimental Brain Research 204:33–45. https://doi.org/10.1007/s00221-010-2291-6
Anderson KL, Ding M (2011) Attentional modulation of the somatosensory mu rhythm. Neuroscience 180:165–180. https://doi.org/10.1016/j.neuroscience.2011.02.004
Auerbach SM, Laskin DM, Frantsve LM, Orr T (2001) Depression, pain, exposure to stressful life events, and long-term outcomes in temporomandibular disorder patients. J Oral Maxillofac Surg 59:628–33; discussion 634. https://doi.org/10.1053/joms.2001.23371
Babiloni C, Capotosto P, Brancucci A, et al (2008) Cortical Alpha Rhythms Are Related to the Anticipation of Sensorimotor Interaction Between Painful Stimuli and Movements: A High-Resolution EEG Study. The Journal of Pain 9:902–911. https://org/10.1016/j.jpain.2008.05.007
Baliki MN, Geha PY, Apkarian AV, Chialvo DR (2008) Beyond Feeling: Chronic Pain Hurts the Brain, Disrupting the Default-Mode Network Dynamics. Journal of Neuroscience 28:1398–1403. https://doi.org/10.1523/JNEUROSCI.4123-07.2008
Bassett DS, Bullmore E (2006) Small-world brain networks. Neuroscientist 12:512–523. https://org/10.1177//1073858406293182
Berni KCDS, Dibai-Filho AV, Rodrigues-Bigaton D (2015) Accuracy of the Fonseca anamnestic index in the identification of myogenous temporomandibular disorder in female community cases. J Bodyw Mov Ther 19:404–409. https://org/10.1016/j.jbmt.2014.08.001
Bjelland I, Dahl AA, Haug TT, Neckelmann D (2002) The validity of the Hospital Anxiety and Depression Scale. Journal of Psychosomatic Research 52:69–77. https://org/10.1016/s002-3999(01)00296-3
Bjørk MH, Stovner LJ, Engstrøm M, et al (2009) Interictal quantitative EEG in migraine: a blinded controlled study. The Journal of Headache and Pain 10:331–339. https://doi.org/10.1007/s10194-009-0140-4
Boord P, Siddall PJ, Tran Y, et al (2008) Electroencephalographic slowing and reduced reactivity in neuropathic pain following spinal cord injury. Spinal Cord 46:118–123. https://doi.org/10.1038/sj.sc.3102077
Brinkman L, Stolk A, Dijkerman HC, et al (2014) Distinct roles for alpha- and beta-band oscillations during mental simulation of goal-directed actions. J Neurosci 34:14783–14792. https://doi.org/10.1523/JNEUROSCI.2039-14.2014
Bullmore E, Sporns O (2009) Complex brain networks: graph theoretical analysis of structural and functional systems. Nat Rev Neurosci 10:186–198. https://doi.org/10.1038/nrn2575
Camfferman D, Lorimer Moseley G, Gertz K, et al (2017) Waking EEG Cortical Markers of Chronic Pain and Sleepiness. Pain Medicine 18:1921–1931. https://doi.org/10.1093/pm/pnw294
Carrara SV, Conti PCR, Barbosa JS (2010) Termo do 1^o Consenso em Disfunção Temporomandibular e Dor Orofacial. Dental Press J Orthod 15:114–120
Case LK, Pineda J, Ramachandran VS (2015) Common coding and dynamic interactions between observed, imagined, and experienced motor and somatosensory activity. Neuropsychologia 79:233–245. https://doi.org/10.1016/j.neuropsychologia.2015.04.005
Castro MMC, Quarantini L, Batista-Neves S, et al (2006) Validade da escala hospitalar de ansiedade e depressão em pacientes com dor crônica. Revista Brasileira de Anestesiologia 56. https://doi.org/10.1590/s0034-70942006000500005
Cauda F, Sacco K, Duca S, et al (2009) Altered Resting State in Diabetic Neuropathic Pain. PLoS ONE 4:e4542. https://doi.org/10.1371/journal.pone.0004542
Cifre I, Sitges C, Fraiman D, et al (2012) Disrupted functional connectivity of the pain network in fibromyalgia. Psychosom Med 74:55–62. https://doi.org/10.1097/PSY.0b013e3182408f04
Corbett DB, Simon CB, Manini TM, et al (2018) Movement-evoked pain. Pain 1. https://doi.org/10.1097/j.pain.0000000000001431
Cruccu G, Frisardi G, Pauletti G, et al (1997) Excitability of the central masticatory pathways in patients with painful temporomandibular disorders. Pain 73:447–454. https://doi.org/10.1016/S0304-3959(97)00139-5
Dahan H, Shir Y, Velly A, Allison P (2015) Specific and number of comorbidities are associated with increased levels of temporomandibular pain intensity and duration. J Headache Pain 16:528. https://doi.org/10.1186/s10194-015-0528-2
de Godoi Gonçalves DA, Dal Fabbro AL, Campos JADB, et al (2010) Symptoms of temporomandibular disorders in the population: an epidemiological study. J Orofac Pain 24:270-8.
de Vries M, Wilder-Smith OH, Jongsma MLA, et al (2013) Altered resting state EEG in chronic pancreatitis patients: toward a marker for chronic pain. J Pain Res 6:815–824. https://doi.org/10.2147/JPR.S50919
Dinh ST, Nickel MM, Tiemann L, et al (2019) Brain dysfunction in chronic pain patients assessed by resting-state electroencephalography. Pain 160:2751–2765. https://doi.org/10.1097/j.pain.0000000000001666
Drewes AM, Krarup AL, Detlefsen S, et al (2008) Pain in chronic pancreatitis: the role of neuropathic pain mechanisms. Gut 57:1616–1627. https://doi.org/10.1136/gut.2007.146621
Dworkin SF, Huggins KH, LeResche L, et al (1990) Epidemiology of signs and symptoms in temporomandibular disorders: clinical signs in cases and controls. J Am Dent Assoc 120:273–281. https://doi.org/10.14219/jada.archive.1990.0043
Fardo F, Allen M, Jegindø E-ME, et al (2015) Neurocognitive evidence for mental imagery-driven hypoalgesic and hyperalgesic pain regulation. NeuroImage 120:350–361. https://doi.org/10.1016/j.neuroimage.2015.07.008
Farmer MA, Baliki MN, Vania Apkarian A (2012) A dynamic network perspective of chronic pain. Neuroscience Letters 520:197–203. https://doi.org/10.1016/j.neulet.2012.05.001
Ferreira CLP, da Silva MAMR, de Felício CM (2016) Sinais e sintomas de desordem temporomandibular em mulheres e homens. CoDAS 28:17–21. https://doi.org/10.1590/2317-1782/20162014218
Festa F, Rotelli C, Scarano A, et al (2021) Functional Magnetic Resonance Connectivity in Patients With Temporomadibular Joint Disorders. Front Neurol 12:629211. https://doi.org/10.3389/fneur.2021.629211
Gesch D, Bernhardt O, Alte D, et al (2004) Prevalence of signs and symptoms of temporomandibular disorders in an urban and rural German population: results of a population-based Study of Health in Pomerania. Quintessence Int 35:143-50
Greene CS, Klasser GD, Epstein JB (2010) Revision of the American Association of Dental Research’s Science Information Statement about Temporomandibular Disorders. J Can Dent Assoc 76:a115.
Harper DE, Schrepf A, Clauw DJ (2016) Pain Mechanisms and Centralized Pain in Temporomandibular Disorders. J Dent Res 95:1102–1108. https://doi.org/10.1177/0022034516657070
He S, Li F, Gu T, et al (2018) Reduced corticostriatal functional connectivity in temporomandibular disorders. Hum Brain Mapp 39:2563–2572. https://doi.org/10.1002/hbm.24023
Hollins M, Harper D, Gallagher S, et al (2009) Perceived intensity and unpleasantness of cutaneous and auditory stimuli: An evaluation of the generalized hypervigilance hypothesis. Pain 141:215–221. https://doi.org/10.1016/j.pain.2008.10.003
Hughes SW, Crunelli V (2005) Thalamic Mechanisms of EEG Alpha Rhythms and Their Pathological Implications. The Neuroscientist 11:357–372. https://doi.org10.1177/1073858405277450/
Ioannides AA (2007) Dynamic functional connectivity. Current Opinion in Neurobiology 17:161–170. https://doi.org/10.1016/j.conb.2007.03.008
Jensen MP, Sherlin LH, Gertz KJ, et al (2013) Brain EEG activity correlates of chronic pain in persons with spinal cord injury: clinical implications. Spinal Cord 51:55–58. https://doi.org/10.1038/sc.2012.84
Jensen O, Mazaheri A (2010) Shaping functional architecture by oscillatory alpha activity: gating by inhibition. Front Hum Neurosci 4:186. https://doi.org/10.3389/fnhum.2010.00186
Kaida K, Takahashi M, Åkerstedt T, et al (2006) Validation of the Karolinska sleepiness scale against performance and EEG variables. Clinical Neurophysiology 117:1574–1581. https://doi.org/10.1016/j.clinph.2006.03.011
Kaneko N, Yokoyama H, Masugi Y, et al (2021) Phase dependent modulation of cortical activity during action observation and motor imagery of walking: An EEG study. Neuroimage 225:117486. https://doi.org/10.1016/j.neuroimage.2020.117486
Kim H, Guilleminault C, Hong S, et al (2001) Pattern analysis of sleep-deprived human EEG. Journal of Sleep Research 10:193–201. https://doi.org/10.1046/j.1365-2869.2001.00258.x
Kucyi A, Moayedi M, Weissman-Fogel I, et al (2014) Enhanced medial prefrontal-default mode network functional connectivity in chronic pain and its association with pain rumination. J Neurosci 34:3969–3975. https://doi.org/10.1523/JNEUROSCI.5055-13.2014
Kuner R, Flor H (2016) Structural plasticity and reorganisation in chronic pain. Nat Rev Neurosci 18:20–30. https://doi.org10.1038/nrn.2016.162/
Lin C-S (2014) Brain signature of chronic orofacial pain: a systematic review and meta-analysis on neuroimaging research of trigeminal neuropathic pain and temporomandibular joint disorders. PLoS One 9:e94300. https://doi.org/10.1371/journal.pone.0094300
Liu C-C, Ohara S, Franaszczuk PJ, et al (2011) Attention to painful cutaneous laser stimuli evokes directed functional interactions between human sensory and modulatory pain-related cortical areas. Pain 152:2781–2791. https://doi.org/10.1016/j.pain.2011.09.002
Liu J, Zhang F, Liu X, et al (2018) Altered small-world, functional brain networks in patients with lower back pain. Sci China Life Sci 61:1420–1424. https://doi.org/10.1007/s11427-017-9108-6
Llinas RR, Ribary U, Jeanmonod D, et al (1999) Thalamocortical dysrhythmia: A neurological and neuropsychiatric syndrome characterized by magnetoencephalography. Proceedings of the National Academy of Sciences 96:15222–15227. https://doi.org/10.1073/pnas.96.26.15222
Lorduy KM, Liegey-Dougall A, Haggard R, et al (2013) The prevalence of comorbid symptoms of central sensitization syndrome among three different groups of temporomandibular disorder patients. Pain Pract 13:604–613. https://doi.org/10.1111/papr.12029
Lucena LBS de, de Lucena LBS, Kosminsky M, et al (2006) Validation of the Portuguese version of the RDC/TMD Axis II questionnaire. Brazilian Oral Research 20:312–317. https://doi.org/10.1590/s1806-83242006000400006
May ES, Butz M, Kahlbrock N, et al (2012) Pre- and post-stimulus alpha activity shows differential modulation with spatial attention during the processing of pain. NeuroImage 62:1965–1974. https://doi.org10.1016/j.neuroimage.2012.05.071/
McDermid AJ, Rollman GB, McCain GA (1996) Generalized hypervigilance in fibromyalgia: evidence of perceptual amplification. Pain 66:133–144. https://doi.org/10.1016/0304-3959(96)03059-x
Meirovitch Y, Harris H, Dayan E, et al (2015) Alpha and beta band event-related desynchronization reflects kinematic regularities. J Neurosci 35:1627–1637. https://doi.org/10.1523/JNEUROSCI.5371-13.2015
Melzack R (1975) The McGill Pain Questionnaire: Major properties and scoring methods. Pain 1:277–299. https://doi.org/10.1016/0304-3959(75)90044-5
Meneses FM, Queirós FC, Montoya P, et al (2016) Patients with Rheumatoid Arthritis and Chronic Pain Display Enhanced Alpha Power Density at Rest. Frontiers in Human Neuroscience 10. https://doi.org/10.3389/fnhum.2016.003
Mills EP, Akhter R, Di Pietro F, et al (2021) Altered Brainstem Pain Modulating Circuitry Functional Connectivity in Chronic Painful Temporomandibular Disorder. J Pain 22:219–232. https://doi.org/10.1016/j.jpain.2020.08.002
Mønster D, Fusaroli R, Tylén K, et al (2016) Inferring Causality from Noisy Time Series Data - A Test of Convergent Cross-Mapping. Proceedings of the 1st International Conference on Complex Information Systems. https://doi.org/10.5220/0005932600480056
Montoya P, Sitges C, García-Herrera M, et al (2006) Reduced brain habituation to somatosensory stimulation in patients with fibromyalgia. Arthritis Rheum 54:1995–2003. https://doi.org/10.1002/art.21910
Moseley GL, Zalucki N, Birklein F, et al (2008) Thinking about movement hurts: the effect of motor imagery on pain and swelling in people with chronic arm pain. Arthritis Rheum 59:623–631. https://doi.org/10.1002/art.23580
Napadow V, LaCount L, Park K, et al (2010) Intrinsic brain connectivity in fibromyalgia is associated with chronic pain intensity. Arthritis Rheum 62:2545–2555. https://doi.org/10.1002/art.27497
Nebel MB, Folger S, Tommerdahl M, et al (2010) Temporomandibular disorder modifies cortical response to tactile stimulation. J Pain 11:1083–1094. https://doi.org/10.1016/j.jpain.2010.02.021
Nickel MM, May ES, Tiemann L, et al (2017) Brain oscillations differentially encode noxious stimulus intensity and pain intensity. Neuroimage 148:141–147. https://doi.org/10.1016/j.neuroimage.2017.01.011
Olesen SS, Hansen TM, Graversen C, et al (2011) Slowed EEG rhythmicity in patients with chronic pancreatitis. European Journal of Gastroenterology & Hepatology 23:418–424. https://doi.org/10.1097/MEG.0b013e3283457b09
Park H-J, Friston K (2013) Structural and functional brain networks: from connections to cognition. Science 342:1238411. https://doi.org/10.1126/science.1238411
Percio CD, Del Percio C, Le Pera D, et al (2006) Distraction affects frontal alpha rhythms related to expectancy of pain: An EEG study. NeuroImage 31:1268–1277. https://doi.org/10.1016/j.neuroimage.2006.01.013
Pessoa L, Kastner S, Ungerleider LG (2003) Neuroimaging Studies of Attention: From Modulation of Sensory Processing to Top-Down Control. The Journal of Neuroscience 23:3990–3998. https://doi.org/10.1523/JNEUROSCI.23-10-03990.2003
Pfurtscheller G, Lopes da Silva FH (1999) Event-related EEG/MEG synchronization and desynchronization: basic principles. Clin Neurophysiol 110:1842–1857. https://doi.org/10.1016/s1388-2457(99)00141-8
Pfurtscheller G, Neuper C (1997) Motor imagery activates primary sensorimotor area in humans. Neuroscience Letters 239:65–68. https://doi.org/10.1016/s0304-3940(97)00889-6
Pietro FD, Di Pietro F, Macey PM, et al (2018) The relationship between thalamic GABA content and resting cortical rhythm in neuropathic pain. Human Brain Mapping 39:1945–1956. https://doi.org/10.1002/hbm.23973
Pimenta CA de M, de Mattos Pimenta CA, Teixeira MJ (1996) Questionário de dor McGill: proposta de adaptação para a língua portuguesa. Revista da Associação Médica Brasileira 30
Pinheiro ES dos S, dos Santos Pinheiro ES, de Queirós FC, et al (2016) Electroencephalographic Patterns in Chronic Pain: A Systematic Review of the Literature. PLOS ONE 11:e0149085. https://doi.org/:10.1371/journal.pone.0149085
Ridder DD, De Ridder D, Vanneste S, et al (2015) Thalamocortical Dysrhythmia: A Theoretical Update in Tinnitus. Frontiers in Neurology 6. https://doi.org/10.3389/fneur.2015.00124
Romaniello A, Cruccu G, Frisardi G, et al (2003) Assessment of nociceptive trigeminal pathways by laser-evoked potentials and laser silent periods in patients with painful temporomandibular disorders. Pain 103:31–39. https://doi.org/10.1016/s0304-3959(02)00347-0
Rossini PM, Di Iorio R, Bentivoglio M, et al (2019) Methods for analysis of brain connectivity: An IFCN-sponsored review. Clin Neurophysiol 130:1833–1858. https://doi.org/10.1016/j.clinph.2019.06.006
Santana JERS, Baptista AF, Lucena R, et al (2021) Altered Dynamic Brain Connectivity in Individuals With Sickle Cell Disease and Chronic Pain Secondary to Hip Osteonecrosis. Clin EEG Neurosci 15500594211054297. https://doi.org/10.1177/15500594211054297
Santos CC dos, dos Santos CC, Pereira LSM, et al (2006) Applicability of the Brazilian version of the McGill pain questionnaire in elderly patients with chronic pain. Acta Fisiátrica 13
Sarnthein J, Stern J, Aufenberg C, et al (2006) Increased EEG power and slowed dominant frequency in patients with neurogenic pain. Brain 129:55–64. https://doi.org/10.1093/brain/awh631
Shamsi F, Haddad A, Zadeh LN (2020) Recognizing Pain in Motor Imagery EEG Recordings Using Dynamic Functional Connectivity Graphs. Conf Proc IEEE Eng Med Biol Soc 2020:2869–2872. https://doi.org/10.1109/EMBC44109.2020.9175627
Shibukawa Y, Ishikawa T, Kato Y, et al (2007) Cerebral cortical dysfunction in patients with temporomandibular disorders in association with jaw movement observation. Pain 128:180–188. https://doi.org/10.1016/j.pain.2006.10.006
Smit DJA, Stam CJ, Posthuma D, et al (2008) Heritability of “small-world” networks in the brain: A graph theoretical analysis of resting-state EEG functional connectivity. Human Brain Mapping 29:1368–1378. https://doi.org/10.1002/hbm.20468
Stern J, Jeanmonod D, Sarnthein J (2006) Persistent EEG overactivation in the cortical pain matrix of neurogenic pain patients. NeuroImage 31:721–731. https://doi.org/10.1016/j.neuroimage.2005.12.042
Strijkstra AM, Beersma DGM, Drayer B, et al (2003) Subjective sleepiness correlates negatively with global alpha (8–12 Hz) and positively with central frontal theta (4–8 Hz) frequencies in the human resting awake electroencephalogram. Neuroscience Letters 340:17–20. https://doi.org/10.1016/s0304-3940(03)00033-8
Sugihara G, May R, Ye H, et al (2012) Detecting Causality in Complex Ecosystems. Science 338:496–500. https://doi.org/10.1126/science.1227079
Taesler P, Rose M (2016) Prestimulus Theta Oscillations and Connectivity Modulate Pain Perception. J Neurosci 36:5026–5033. https://doi.org/10.1523/JNEUROSCI.3325-15.2016
Tchivileva IE, Ohrbach R, Fillingim RB, et al (2017) Temporal change in headache and its contribution to the risk of developing first-onset temporomandibular disorder in the Orofacial Pain. PAIN 158:120–129. https://doi.org/10.1097/j.pain.0000000000000737
Tran Y, Boord P, Middleton J, Craig A (2004) Levels of brain wave activity (8-13 Hz) in persons with spinal cord injury. Spinal Cord 42:73–79. https://doi.org/10.1038/sj.sc.3101543
van den Broeke EN, Wilder-Smith OHG, van Goor H, et al (2013) Patients with Persistent Pain after Breast Cancer Treatment Show Enhanced Alpha Activity in Spontaneous EEG. Pain Medicine 14:1893–1899. https://doi.org/10.1111/pme.12216
Veldhuijzen DS, Kenemans JL, van A J, et al (2006) Processing capacity in chronic pain patients: A visual event-related potentials study. Pain 121:60–68. https://doi.org/10.1016/j.pain.2005.12.004
Visscher CM, van Wesemael-Suijkerbuijk EA, Lobbezoo F (2016) Is the experience of pain in patients with temporomandibular disorder associated with the presence of comorbidity? Eur J Oral Sci 124:459–464. https://doi.org/10.1111/eos.12295
Vuckovic A, Hasan MA, Fraser M, et al (2014) Dynamic Oscillatory Signatures of Central Neuropathic Pain in Spinal Cord Injury. The Journal of Pain 15:645–655. https://doi.org/10.1016/j.jpain.2014.02.005
Wang W-E, Ho RLM, Ribeiro-Dasilva MC, et al (2020) Chronic jaw pain attenuates neural oscillations during motor-evoked pain. Brain Res 1748:147085. https://doi.org/10.1016/j.brainres.2020.147085
Watts DJ, Strogatz SH (1998) Collective dynamics of “small-world” networks. Nature 393:440–442. https://doi.org/10.1038/30918
Weissman-Fogel I, Moayedi M, Tenenbaum HC, et al (2011) Abnormal cortical activity in patients with temporomandibular disorder evoked by cognitive and emotional tasks. Pain 152:384–396. https://doi.org/10.1016/j.pain.2010.10.046
Wiech K, Ploner M, Tracey I (2008) Neurocognitive aspects of pain perception. Trends in Cognitive Sciences 12:306–313. https://doi.org/:10.1016/j.tics.2008.05.005
Yin Y, He S, Xu J, et al (2020) The neuro-pathophysiology of temporomandibular disorders-related pain: a systematic review of structural and functional MRI studies. J Headache Pain 21:78. https://doi.org/10.1186/s10194-020-01131-4
Zhang Y, Liu J, Li L, et al (2014) A study on small-world brain functional networks altered by postherpetic neuralgia. Magn Reson Imaging 32:359–365. https://doi.org/10.1016/j.mri.2013.12.016

No competing interests reported.

Download PDF

Version 1

posted

You are reading this latest preprint version

Patients with temporomandibular disorders and chronic pain of myofascial origin display reduced alpha power density and altered Small-world properties of brain networks

Status:

Version 1

Abstract

Figures

1. Introduction

2. Materials And Methods

3. Results

4. Discussion

5. Conclusion

Declarations

References

Additional Declarations

Status:

Version 1