Craniocervical and Cervical Spine Features of Patients with Temporomandibular Disorders: A Systematic Review and Meta-Analysis of Observational Studies

To assess neck disability with respect to jaw disability, craniocervical position, cervical alignment, and sensorimotor impairments in patients with temporomandibular disorders (TMD), a systematic review and meta-analysis of observational studies trials were conducted. The meta-analysis showed statistically significant differences in the association between neck disability and jaw disability (standardized mean difference (SMD), 0.72 (0.56–0.82)). However, results showed no significant differences for cervical alignment (SMD, 0.02 (−0.31–0.36)) or for the craniocervical position (SMD, −0.09 (−0.27–0.09)). There was moderate evidence for lower pressure pain thresholds (PPT) and for limited cervical range of motion (ROM). There was limited evidence for equal values for maximal strength between the patients with TMD and controls. There was also limited evidence for reduced cervical endurance and conflicting evidence for abnormal electromyographic (EMG) activity and motor control in TMD patients. Results showed a clinically relevant association between cervical and mandibular disability in patients with TMD. Regarding sensory-motor alterations, the most conclusive findings were observed in the reduction of PPT and cervical ROM, with moderate evidence of their presence in the patients with TMD. Lastly, the evidence on impaired motor control and cervical EMG activity in patients with TMD was conflicting.


Introduction
Temporomandibular disorders (TMD) include a set of musculoskeletal disorders involving the temporomandibular joint (TMJ), masticatory musculature, and associated orofacial structures [1]. TMD is the leading cause of chronic nonodontogenic orofacial pain [2].
Pain in the temporomandibular region occurs in approximately 10% of the population older than 18 years and is more prevalent in young and middle-aged adults [3]. In addition, the associated signs (mixed, myogenic, or arthrogenous); (3) reported somatosensory, motor, and disability variables of the cervical spine. Studies were excluded if they included patients with primary headaches or non-specific chronic neck pain.

Search Strategy
The search was conducted by two independent reviewers using the same methods; any differences that emerged during this phase were resolved by consensus. Reference sections from original studies were screened manually.
We conducted a search of observational and comparative studies that included cohort, case-control, and cross-sectional studies using MEDLINE (from 1950  ). This strategy was combined with the following free terms and descriptors: 'craniocervical posture', 'cervical spine alignment', 'pressure pain threshold', 'cervical strength', 'cervical motor control', 'disability'.
In addition, bibliographic references of identified publications and published bibliographic reviews were searched by hand for potentially relevant articles.

Selection Criteria and Data Extraction
Two independent reviewers (AHG and BMM) performed the first phase, assessing the relevance of the studies. This first analysis was performed based on information from each study's title, abstract, and keywords. If the abstracts did not contain sufficient information, the full text was reviewed. During the second phase, we reviewed the full text and checked whether the studies met all of the inclusion criteria. A third reviewer (ARV) acted as a mediator when there were differences between the two reviewers, with the 3 reviewers conducting a consensus [21]. The data described in the results were extracted by means of a structured protocol that ensured the most relevant information from each study was obtained.

Methodological Quality Assessment
We assessed the methodological quality of the selected cohort, cross-sectional, and case-control studies using the modified version of the Newcastle-Ottawa Quality Assessment Scale (NOS) [22]. NOS is appropriate for reviews involving a large number of studies because of its brevity, and it presents moderate inter-rater reliability [23]. The NOS scores 3 criteria with a range of 0 to 4 stars: grade selection of participants, assessment of exposures, outcomes, and comparability, and control of confounding variables, based on 9 questions. The tallied stars provide 4 categories of study quality: (1) poor, 0 to 3 stars; (2) fair, 4 to 5 stars; (3) good, 6 to 7 stars; (4) excellent, 8 to 9 stars [24]. For the analysis of the methodological quality of the cross-sectional studies, we used the NOS modifications proposed by Fingleton et al. [25] with only 3 items: (1) 3/3 was considered good quality; (2) 2/3 was fair; (3) 1/3 was poor quality.
Two independent reviewers examined the quality of the selected studies using the same methods; disagreements between the reviewers were resolved by a consensus that included mediation by a third reviewer. The inter-rater reliability was determined using the Kappa coefficient: (1) κ > 0.7 meant a high level of agreement between the assessors; (2) κ = 0.5-0.7 meant a moderate level of agreement; (3) κ < 0.5 meant a low level of agreement [26].

Qualitative Analysis
For the qualitative analysis of the selected observational studies, we employed an adaptation of the classification criteria provided by van Tulder et al. [21] for randomized controlled trials. The results were categorized into 5 levels depending on the methodological quality: (1) strong evidence, consistent among multiple high-quality case-control/cohort/cross-sectional studies (at least 3); (2) moderate evidence, consistent findings from multiple low-quality case-control/cohort/cross-sectional studies and/or one high-quality case-control/cohort study; (3) limited evidence, one low-quality case-control/cohort studies and/or at least two cross-sectional studies; (4) conflicting evidence, inconsistent findings among multiple studies (case-control/cohort/cross-sectional studies); (5) no evidence, no case-control/cohort/ cross-sectional studies reported.

Data Synthesis and Analysis
The statistical analysis was performed using meta-analyses with interactive explanations (MIX, version 1.7) with the data comparing patients with TMD to asymptomatic participants [27].
We employed the same inclusion criteria for the systematic review and the meta-analysis but added two criteria: (1) the Results section contained detailed information on the comparative statistical data (mean, standard deviation, and/or 95% confidence interval) of the main variables and (2) data for the analyzed variables were represented in at least 3 studies. We presented the summary statistics in the form of forest plots [28], which consisted of a weighted compilation of all standardized mean differences (SMDs) and corresponding 95% confidence intervals (CI) reported by each study and provided an indication of heterogeneity among the studies.
The statistical significance of the pooled SMDs was examined using Hedges' g, to account for possible overestimation of the true population effect size in small studies [29]. The magnitude of g was interpreted according to a 4-point scale: (1) <0.20, negligible effect; (2) 0.20-0.49, small effect; (3) 0.50-0.79, moderate effect; (4) ≥0.80, large effect [30]. We estimated the degree of heterogeneity among the studies by employing Cochran's Q statistic test (p < 0.1 was considered significant) and the inconsistency index (I 2 ) [31]. I 2 > 25% is considered to represent low heterogeneity, I 2 > 50% is considered medium, and I 2 > 75% is considered to represent large heterogeneity [32]. The I 2 index is complementary to the Q test, although it has a similar problem of power as the Q test with a small number of studies [32]. Therefore, a study was considered heterogeneous when it fulfilled one or both of these conditions: (1) the Q-test was significant (p < 0.1), and (2) the result of I 2 was >75%. We performed a random-effects model, as described by DerSimonian and Laird [33], in the meta-analysis of the heterogeneous studies to obtain a pooled estimate of effect. To detect publication biases and to test the influence of each study, we performed a visual evaluation of the funnel plot and exclusion sensitivity plot, searching for any asymmetry. We also employed Egger's regression test to determine the presence of bias [34,35].

Results
The study search strategy was presented in the form of a flow diagram (Figure 1). A total of 32 articles met the inclusion criteria (three case-control studies and 29 cross-sectional studies) . Seventeen articles had been included in three separate meta-analyses. The first meta-analysis included six articles and assessed the correlation between neck disability and the presence of TMD. The second meta-analysis included five articles and dealt with the craniocervical position. The third meta-analysis included six articles and evaluated the position of the head relative to the neck. Table 1 lists the epidemiological characteristics, the results, and the conclusion of each article.

Temporomandibular (TMD) Diagnosis Criteria
More than half of the selected studies used the Research Diagnostic Criteria for Temporomandibular Disorders (RDC/TMD) as the preferred diagnostic method [36][37][38][39][40]43,[46][47][48][50][51][52][53][54][55]63,67]. The RDC/TMD was first published in 1992 by Dworkin and LeResche, and it provides an assessment of the most common TMD conditions taking into consideration both the clinical condition (Axis I) and the psychosocial status and pain-related disability (Axis II) [16]. Other validated instruments were also used to diagnose TMD: the American Association of Orofacial Pain questionnaire [59,61], Helkimo's index of mandibular mobility [42], and Conti's questionnaire of TMD subjective symptoms [48]. On the other hand, four studies carried out a physical exploration in order to give a TMD diagnosis. Finally, in four studies, the samples had already been diagnosed with TMD [45,57,60,64], and in three studies, the TMD diagnosis criteria were not specified [41,58,66].

Temporomandibular (TMD) Diagnosis Criteria
More than half of the selected studies used the Research Diagnostic Criteria for Temporomandibular Disorders (RDC/TMD) as the preferred diagnostic method [36][37][38][39][40]43,[46][47][48][50][51][52][53][54][55]63,67]. The RDC/TMD was first published in 1992 by Dworkin and LeResche, and it provides an assessment of the most common TMD conditions taking into consideration both the clinical condition (Axis I) and the psychosocial status and pain-related disability (Axis II) [16]. Other validated instruments were also used to diagnose TMD: the American Association of Orofacial Pain questionnaire [59,61], Helkimo's index of mandibular mobility [42], and Conti's questionnaire of TMD subjective symptoms [48]. On the other hand, four studies carried out a physical exploration in order to give a TMD diagnosis. Finally, in four studies, the samples had already been diagnosed with TMD [45,57,60,64], and in three studies, the TMD diagnosis criteria were not specified [41,58,66].

Experimental group
18-65 years, diagnosis of chronic cervico-craniofacial pain of muscular origin, disability, and pain in these regions according to the CF-PDI, diagnosis of myofascial pain according to RDC/TMD and bilateral pain of the masticatory and cervical muscles

Disability
Self-reported, craniofacial disability using CF-PDI, and neck disability using NDI Cervical spine alignment Head posture using the CROM TM device. The sternomental distance using plastic digital caliper with a five-digit LCD display

Disability
There was no association between craniocervical posture and pain-related disability A strong correlation between the neck and craniofacial disability was found Cervical spine alignment A moderate positive correlation was observed between craniocervical posture variables for both groups

Disability
Self-reported, neck disability using NDI, and jaw disability using JDI Sensory-motor impairments PPT in masticatory structures, cervical muscles, and the extracephalic site using a manual pressure algometer

Disability
The jaw disability was significantly higher than neck disability in patients with TMD

Sensory-motor impairments
There was a significant increase in the tenderness of the masticatory and cervical muscles in the TMD patients compared to the healthy subjects

Experimental group
Diagnosis of chronic musculoskeletal disorders, such as a painful CMD or CSD

Cervical spine alignment
Head posture using lateral photographs and lateral X-ray examination of the head and cervical spine

Cervical spine alignment
No difference was found related to head posture between the CMD with and without CSD patients, the CSD patients, and the healthy subjects. For the photographs, increasing age was associated with a more anteroposition of the head

Experimental group
Pain in the masticatory muscles/TMJ of at least 3 months, moderate or severe baseline pain score of ≥30 mm using a 100 mm VAS

Sensory-motor impairments
Cervical flexion force and endurance using a cervical flexion force device and stopwatch

Results of the Methodological Quality
The agreement between the two evaluators, according to the Kappa coefficient, was high (κ = 0.756). The intervention of a third evaluator was necessary to achieve consensus on the quality of 14 studies.
One case-control study showed a fair methodological quality, with a score of 5 [47]; the other two case-control studies achieved a score of 3 or lower, which is considered poor methodological quality [41,61]. The mean total score for the methodological quality was 3, with a standard deviation of 2.0 and a range of 1-5 points. In most cases, the methodological quality score was affected by the lack of representativeness in the cases. Neither of the case-control studies presented non-response rates for the participants. Seventeen cross-sectional studies showed fair methodological quality, with a score of 2 [37][38][39][40]43,48,[50][51][52][53][54][55]57,62,63,66,67]; the other 12 studies achieved a score of 1 or lower, which is considered poor methodological quality [36,42,[44][45][46]49,56,[58][59][60]64,65]. The mean total score for methodological quality was 1.48, with a standard deviation of 0.68, and a range of 0-2. In most cases, the methodological quality score was affected by the lack of representativeness of the exposed cohort. Tables 2 and 3 show the numerical results of the NOS scale.
Given that TMD is more frequent in women than in men, the female sex was predominant in most of the studies' samples. There was only one study in which women comprised the minority [51]. In the rest of the studies, the proportion of women ranged from 55% to 100%, barring one study that did not specify the participants' sex [41]. The age range defined in the inclusion criteria of the TMD groups was between 18 and 60 years.

Association between Cervical and Mandibular Disability
The association between neck disability and jaw disability in patients with TMD was assessed in six studies [36,51,53,57,58,62]. All of the studies employed the neck disability index for cervical disability. Two of the studies used the jaw function scale for TMJ disability [36,62], and the rest of the studies employed the craniofacial pain and disability inventory [51,53,57,58]. All studies showed significant associations between neck disability and jaw disability in patients with TMD [36,51,53,57,58,62]. The strongest association was found by Silveira et al. [62] (r = 0.915), and the lowest association was found by Greghi et al. [53] (r = 0.40). The meta-analysis for the association between neck disability and jaw disability in patients with TMD showed statistically significant correlations with a moderate clinical effect (six studies [36,51,53,57,58,62], 548 patients; SMD, 0.72; 95% CI 0.56-0.82) and heterogeneity (Q value, 42.07; p < 0.001; I 2 , 88%). The shape of the funnel plot appeared to be symmetrical in the dominant model, as judged by visual examining the intensity of the pain (Figure 2). The influence of each individual study was assessed with a sensitivity exclusion analysis. We obtained statistically strong results because the analysis suggested that no individual study significantly affected the pooled SMD. The similarity found among the pooled estimates suggested that there was no single study influencing the results of the meta-analysis (Annexes; Figure A1).
The meta-analysis for the craniocervical position showed no statistically significant differences (five studies [48,55,59,61,65], 226 patients; SMD, −0.09; 95% CI −0.27 to 0.09) and heterogeneity (Q, 3.12; p = 0.96; I 2 , 0%), and there was evidence of publication bias for the meta-analysis (SE, 0.03; T, −7.0; p < 0.001). The shape of the funnel plot seemed to be asymmetrical in the dominant model, as judged by visually examining the craniocervical position ( Figure 3). The influence of each individual study was assessed with a sensitivity exclusion analysis. We obtained statistically strong results because the analysis suggested that no individual study significantly affected the pooled SMD. The similarity found among the pooled estimates suggested that there was no single study influencing the results of the meta-analysis. Accordingly, we applied Egger's test of asymmetry, and the results suggested significant evidence of publication bias for the analysis of the craniocervical position (intercept, 2.08; t, 6.96; p < 0.001) ( Figure A2).
The meta-analysis for the cervical spine alignment showed no statistically significant differences (six studies [44,48,56,57,65,66], 404 patients; SMD, 0.02; 95% CI-0.31-0.36) and heterogeneity (Q, 55.18; p < 0.001; I 2 , 79%), and there was no evidence of publication bias for the meta-analysis (SE, 0.03; T, 0.53; p = 0.6). The shape of the funnel plot appeared to be asymmetrical in the dominant model as judged by visually examining the position of the head relative to the neck (Figure 4). The influence of each study was assessed with a sensitivity exclusion analysis. We obtained statistically strong results because the analysis suggested that no individual study significantly affected the pooled SMD. The similarity found among the pooled estimates suggested that there was no single study influencing the results of the meta-analysis. Accordingly, we applied Egger's test of asymmetry, with the results suggesting no significant evidence of publication bias for the analysis of the head position relative to the neck (intercept, 0; t,−0.01; p = 0.99) ( Figure A3).

Pressure Pain Thresholds in the Craniocervical Region
These five studies assessed the mechanosensitivity of masticatory and cervical muscles and orofacial structures using pressure pain thresholds (PPTs) [43,47,62,63,67]. Four of the aforementioned studies employed manual pressure algometers [47,62,63,67], and one employed a digital dynamometer to measure and compare PPTs in patients with TMD and asymptomatic controls [43]. In three studies, the TMD group had concurrent neck disability or neck pain [43,62,63]. PPTs were recorded bilaterally at various anatomical points. In the craniomandibular region, five studies chose the masseter and temporalis muscles [43,47,62,63,67], and one study chose the lateral pole of the TMJ [47]. In the cervical region, four studies used the upper trapezius [47,62,63,67] and sternocleidomastoid muscles [43,47,62,63], two studies used the suboccipital muscles [43,67], and one study used the middle trapezius [43]. The hypothenar and thenar region of the hand and the Achilles tendon were selected as distal points in three studies [43,47,63].

Figure 2.
Synthesis forest plot for the association between cervical and mandibular disability. SMD, standardized mean difference. This forest plot summarizes the results of six included studies (sample size, standardized mean differences (SMDs), and weight). The small boxes with the squares represent the point estimate of the effect size and sample size. The lines on either side of the box represent a 95% confidence interval (CI). The horizontal axis represents whether the quantitative analysis is for or against the association.     All five studies reached the same conclusion: patients with TMD, regardless of the presence of neck disability or neck pain, showed significantly lower PPTs at almost all craniocervical structures when compared with the control group [43,47,62,63,67]. Altogether, there was moderate evidence of lower PPTs in patients with TMD.

Cervical Spine Range of Motion
Seven studies analyzed the cervical range of movement (ROM) in patients with TMD [42,45,46,50,54,64,67]. Three of these studies compared ROM between patients with TMD and asymptomatic controls using a cervical ROM (CROM) instrument [50], the Keno ® -cervical measurement instrument [67], and through visual evaluation [45]. Three other studies evaluated active ROM using a goniometer [42,46,67] and an inclinometer [54] on patients with TMD and those with TMD and concurrent disorders. The remaining study used CROM to measure the ROM of patients with TMD and concurrent neck or headache disorders and patients with headache or neck pain [64]. All of the aforementioned studies assessed flexion, extension, and both lateral flexions and rotations in the cervical spine.
In four studies, the patients with TMD presented significant limitations in flexion, extension [50,54,67], and both lateral flexion movements [46] compared with the asymptomatic participants.
In the remaining three studies, concise conclusions could not be drawn due to various reasons: the results could only be expressed as correlations [64] were nonexistent [45] or were expressed in percentages and related to TMD severity [42].
Furthermore, three studies included the flexion-rotation test as a measurement of ROM [50,54,67]. Two of the studies showed that both rotation movements were significantly lower in patients with TMD compared with the control group [50,54], and one study found no relevant differences between the groups [67]. In summary, there was moderate evidence of limited cervical ROM in patients with TMD.
Two studies analyzed the maximal cervical flexor strength using a cervical flexor strength device that monitored the force generated by the participants with a load cell [37,41]. In both studies, there was no significant difference in maximal cervical flexor strength between the patients with TMD and the asymptomatic participants [37,41].
Four studies evaluated the cervical endurance of patients with TMD compared with the control group [38,39,41,67]. Three studies measured the cervical flexor endurance [38,41,67], and two studies measured the cervical extensor endurance [39,41]. One study found no significant differences in cervical flexor endurance between the patients with TMD and the control group [67]. However, two studies found significant differences in holding time when the cervical flexor endurance test was performed at 25% of maximal voluntary contraction, with less holding time in the TMD group than in the control group [38,41]. Cervical extensor endurance was significantly lower in the TMD group than in the control group [39,41].
Consequently, there was limited evidence for equal values in maximal cervical flexor strength between the patients with TMD and the control group. There was also limited evidence of reduced cervical endurance in patients with TMD.

Electromyographic Activity in Cervical Muscles
Four studies compared the electromyographic (EMG) activity in the neck muscles of patients with TMD and control participants [39][40][41]60]. Three studies compared EMG activity in the superficial neck muscles [40,41,60], and one study compared the EMG activity in the neck extensor muscles [39]. Specifically, one study measured EMG activity in the sternocleidomastoid and trapezius muscles at rest [60], two studies measured EMG activity in the sternocleidomastoid and anterior scalene muscles while performing the craneo-cervical flexor test (CCFT) [40,41], and the remaining study measured EMG activity in the extensor muscles while performing the neck extensor muscle endurance test (NEMET) [39].
There were no significant differences in EMG activity in the sternocleidomastoid and anterior scalene muscles in the patients with TMD when compared with the asymptomatic participants [40,41]. However, the patients with TMD had a significantly higher resting EMG activity in the sternocleidomastoid and trapezius muscles when compared with the asymptomatic participants [60]. There were significant differences in EMG activity during the NEMET, which showed higher fatigability of the cervical extensor muscles in the patients with TMD [39]. As a result, there was conflicting evidence regarding abnormal EMG activity in patients with TMD.

Cervical Motor Control
Three studies evaluated the motor control of cervical flexors in patients with TMD using the CCFT [40,50,67]. Two studies found no significant differences in CCFT performance between the patients with TMD and the control group [40,67]. Meanwhile, one study showed significantly lower pressures during the CCFT performance (a finding related to poorer motor control) in the TMD group than in the control group [50]. As a consequence, conflicting evidence regarding abnormal cervical motor control was shown in patients with TMD.

Discussion
The aim of this systematic review and meta-analysis was to assess whether cervical and mandibular disabilities were related in patients with TMD and to determine the possible differences in craniocervical posture, cervical spine alignment, and cervical sensory-motor function in these patients compared with asymptomatic participants. Several studies have reported an association between cervical pain and TMD [36,39,43,47,68,69], which might be explained by the neuroanatomical link between the orofacial and cervical regions [15,69,70]. However, disability is a complex concept influenced by the patient's perception of their condition [36,62]; some patients with severe TMD have low levels of disability and low impact on their quality of life [36,62]. Therefore, the degree of disability depends only partly on the patient's signs and symptoms [38,40,50,62,64]. We, therefore, considered this systematic review relevant because it was the first to analyze the relationship between the two regions in terms of disability. The results revealed that patients with TMD presented jaw disability moderately related to their degree of cervical disability. These patients also presented sensory-motor impairments (but not postural) in the cervical region compared with the asymptomatic participants.
The association between TMD and cervical disorders has been an area of interest for many years, a relationship attributed to the neurophysiological, biomechanical, and functional link between the two regions [15,68,[70][71][72]. Our results suggested that the neurophysiological component might be more important than the biomechanical in explaining the observed disorders. For example, the study conducted by Favia et al. [73] showed the role of neuroreceptors in TMD. This hypothesis is reinforced by the lack of differences in craniocervical posture and cervical alignment between patients with TMD and asymptomatic participants. Furthermore, the reported quantitative analysis provides more evidence than that of two previous systematic reviews on the subject, which showed inconclusive results [15,74]. However, we considered that the relevance of posture in these patients should not be completely ruled out because their assessment could be influenced by the Hawthorne effect [75]. Patients might, therefore, not adopt their actual posture when asked to position themselves in a specific manner in preparation for radiography [48,55,59,61,66]. In fact, patient monitoring over a temporary period, and not just momentarily, seems to be a determining factor in identifying postural alterations [76]. These factors should be considered in future studies to establish more conclusive results.
The impact of mandibular disability on perceived neck disability is evident, with the results reflecting a relationship in the clinical impact due to the resulting size of the effect (moderate/large with a g of 0.72). However, the mechanisms underlying this relationship are currently unknown [62]. The best explanation might be the neurophysiological connection between the two regions of the trigeminocervical nucleus [70,72,77,78]. Painful afferences from the temporomandibular region would, therefore, sensitize the cervical region [79,80]. A number of articles in the literature support this hypothesis, showing an association between pain intensity and perceived disability [18,50,[81][82][83]. In contrast, however, it could be argued that disability is a phenomenon not entirely explained by pain intensity, with numerous other relevant aspects, such as psychosocial factors [18,84,85]. However, most of the analyzed studies included patients with chronic TMD, thereby showing a certain predisposition to central sensitization [86,87], as well as to cognitive/emotional maladaptive factors [88][89][90]. It is, therefore, possible that the relevance of nociceptive information gains greater prominence in explaining the relationship in disability between the orofacial and cervical regions.
Regarding sensory-motor disorders, the most conclusive findings were observed in the reduction of PPT and cervical ROM, with moderate evidence of this reduction in patients with TMD. These disorders could be due to an increase in cervical muscle activity, which a number of authors have attributed to changes in head and neck position [91,92]. However, the lack of differences at the postural level gives greater plausibility to the neurophysiological hypothesis than to the biomechanical one. Thus, the reduction in PPT could be attributed to ischemia caused by sustained contraction [93,94], which could also explain the reduction in ROM by changes in cervical neuromuscular control (e.g., co-contraction of antagonistic agonists, increased co-activation of synergistic muscles, and/or increased activity of superficial muscles at rest) for protective purposes [95][96][97][98]. Along the same lines, a number of authors have reported the so-called trigeminocervical reflex as a possible physiopathological mechanism [99], a phenomenon that demonstrates the effect of mechanoreceptors and nociceptors of TMJ on the fusimotor-muscular spindle system of the cervical muscles [100][101][102]. Abnormalities in cervical neuromuscular control could, therefore, be the result of an overload of the cervical structures due to increased muscle activity. According to the Cinderella hypothesis [103], long-lasting muscular activity and low-intensity loading can activate small type-I motor units in a selective and continuous manner [104][105][106]. The metabolic disorders produced by this event would, therefore, result in tissue damage and, most likely, pain [107,108].
However, the evidence regarding the impairment of motor control and cervical EMG activity in patients with TMD is conflicting. The number of studies was limited, as were some of the aspects from studies that established the lack of differences between these patients and asymptomatic participants. Specifically, the results of the studies conducted by Armijo-Olivo et al. [40,41] showed that the magnitude of the difference was clinically relevant, despite showing no statistically significant differences. Therefore, the lack of differences could be due to a type II error as a consequence of EMG measurements having high variability [40]. In contrast to most of the studies included in this review that studied patients with chronic TMD, the study by von Piekartz et al. [67] was conducted with acute/subacute patients with an average pain intensity of fewer than three points on the VAS. Therefore, the low intensity and duration of their symptoms might be insufficient to sensitize the trigeminocervical nucleus and cause disorders in the cervical region. Future studies that consider these aspects should, therefore, provide definite conclusions on the presence of disorders in motor control and EMG activity in patients with TMD.

Clinical Implications
Clinically, these results suggest that patients with TMD show sensorimotor but not postural impairments in the cervical region compared with asymptomatic participants. Although these results should be interpreted with caution due to the methodological quality of the included studies, the results could help increase clinicians' understanding of the effect of TMD in these patients and thereby help apply the optimal treatment.
A recent review by Gil-Martínez et al. [51] reported that neck disability was a strong predictor of craniofacial pain and disability in a subgroup of patients with TMD due to muscle pain and that neck disability had a positive correlation with orofacial pain and disability, kinesiophobia, and pain catastrophizing. These findings suggest the possibility of including a new therapeutic approach for patients with TMD. Based on our results, future interventions applied to patients with TMD should address their psychosocial behavior to improve the cervical and mandibular disability observed in these patients. However, there is limited evidence on the efficacy of an approach based on psychosocial factors in improving disability in patients with TMD, and future clinical trials addressing this issue are needed.

Limitations
This review presents a number of limitations. First, the design of the studies prevented a cause-effect relationship from being established. Future studies using cohort design and especially experimental studies are needed to better understand how TMD influence neck disorders. Secondly, the methodological quality of the studies was fair/poor, and, therefore, the results should be interpreted with caution. We could not perform quantitative analysis for the neck sensorimotor variables or a comparison between the various TMD diagnoses (mixed, myogenic, and arthrogenous) due to the scarcity of studies. Based on their possible influence on the results, future studies need to consider these aspects to establish more conclusive results. Finally, the meta-analysis for the craniocervical position showed significant evidence of publication bias, which should also be taken into account.

Conclusions
The results of this study showed a clinically relevant association between cervical and mandibular disability in patients with TMD. These patients also showed sensorimotor but not postural impairments in the cervical region compared with the asymptomatic participants. Specifically, patients with TMD experienced reduced PPT and cervical ROM (moderate evidence) and loss of cervical muscle endurance (limited evidence). However, maximal cervical musculature strength was not changed (limited evidence). Finally, there was conflicting evidence regarding the impairment of EMG activity and cervical motor control in patients with TMD. Figure A2. Publication bias heterogeneity funnel plot for the craniocervical position. A funnel plot was used to assess the risk of publication bias. The diagonal lines represent 95% confidence limits. Figure A2. Publication bias heterogeneity funnel plot for the craniocervical position. A funnel plot was used to assess the risk of publication bias. The diagonal lines represent 95% confidence limits. Figure A3. Publication bias heterogeneity funnel plot for the cervical spine alignment. A funnel plot was used to assess the risk of publication bias. The diagonal lines represent 95% confidence limits. Figure A3. Publication bias heterogeneity funnel plot for the cervical spine alignment. A funnel plot was used to assess the risk of publication bias. The diagonal lines represent 95% confidence limits.