Abstract
To assess the diagnostic and behavioural overlap of headache patients with temporomandibular disorders (TMD), individuals recruited from the general population with self-described headaches were compared with non-headache controls. The examination and diagnostic procedures in the Research Diagnostic Criteria (RDC) for TMD were applied to both sets of subjects by a blinded examiner. Following their examination, subjects used experience sampling methods to obtain data on pain, tooth contact, masticatory muscle tension, emotional states and stress. Results showed that a significantly higher proportion of the headache patients received an RDC/TMD diagnosis of myofascial pain than non-headache controls. Headache patients also reported significantly more frequent and intense tooth contact, more masticatory muscle tension, more stress and more pain in the face/head and other parts of the body than non-headache controls. These results are similar to those reported for TMD patients and they suggest that headache patients and TMD patients overlap considerably in diagnosis and oral parafunctional behaviours.
Keywords
Introduction
Patients diagnosed with painful temporomandibular disorders (TMD) often report having headaches as well (1). The pain reported by TMD patients is typically located in the muscles of mastication, in the preauricular area, or in the temporomandibular joint (TMJ) (2) and TMD patients may also report headache and other facial pains. The headache symptoms described by TMD patients are similar to those reported by patients diagnosed as having tension-type or migraine headaches according to the nosology developed by the International Headache Society (3).
An emerging line of evidence shows that parafunctional activities (activities of the masticatory muscles and jaw that do not involve chewing, talking or swallowing) are important in the myofascial pain of TMD. Experimental studies show that deliberate clenching for up to 20 min/day for up to 8 days increases pain and can lead to a diagnosis of TMD, particularly myofascial pain and/or arthralgia (4–7). When experience sampling methods are used to repeatedly obtain information about behaviours and psychological states from TMD patients and non-TMD controls in their natural environments, TMD patients with myofascial pain (with/without arthralgia, as defined in the Research Diagnostic Criteria (RDC) for TMD (8)) show higher levels of parafunctional activity, muscle tension, pain, distress and stress (8). Treatments deliberately designed to reduce parafunctions show therapeutic effects (9–11). These findings point to parafunctional activities as a potentially important factor in the aetiology and maintenance of TMD pain.
The role that muscle activity plays in headache is uncertain. Some studies have shown a difference between resting electromyographic (EMG) levels in the frontal and/or neck regions of subjects with tension-type headache vs. subjects without such headaches, and others have not shown differences (12). Some studies have shown a difference between EMG activity between periods of headache and times when no headache is present; others have not (12). Some patients show a relationship between changes in office-recorded EMG activity across sessions and reduced headaches; other patients do not show this relationship (12). There is no clear or consistent relationship between EMG activity and pain intensity or pain frequency in patients experiencing headache.
One reason for the failure to find EMG differences in headache vs. non-headache individuals is that investigators have rarely considered the role that masticatory muscles, particularly the temporalis, could play in headache and have failed to use an appropriate technology to assess the question. When mild to moderate contact exists for a large proportion of an individual's waking hours (8), this may cause pain in the activated muscle that can be interpreted by the patient as headache pain. The use of laboratory-based measurements to measure briefly EMG states may not be an appropriate methodology for investigating activity that occurs in a patient's everyday environment.
Experience sampling methods (ESM) are an alternative to laboratory-based methods. The hallmark of ESM is the collection of repeated momentary assessments from participants in their natural environments (13). Because participants respond in their natural environments, the ability to generalize the findings to a patient's actual behaviour is enhanced. The relatively high rate of sampling during an individual's waking hours also makes it possible to estimate with increased accuracy the rate of a behaviour such as oral parafunctions.
We hypothesized that headache patients have diagnostic signs and symptoms similar to TMD patients and that oral parafunctional behaviours characteristic of TMD patients are also characteristic of headache patients. In this study, blinded examiners performed a physical examination using procedures specified in the RDC for TMD (14) on headache patients and non-headache controls. Following the examination, ESM techniques were used to assess oral parafunctional activities and other states.
Methods
Subjects
Subjects (initial N = 44) were selected from staff and faculty of the Kansas City University of Medicine and Biosciences or recruited from the general population of the Kansas City metropolitan area. Recruitment techniques included e-mail messages and posting of flyers requesting participation in the study. Only individuals ≥18 years old participated. Individuals with a definitive diagnosis of arthritis or a history of trauma to the head, neck, shoulder or back were excluded, as were those who resided outside the coverage area provided by the paging company. Four individuals were excluded on the basis of these criteria. Those who qualified for the study were assigned to the chronic headache (HA) group (N = 23) or to the non-headache control (C) group (N = 17) based on self-report alone. Participants who reported current, daily use of any analgesic, antidepressant or muscle relaxant medication were included if they agreed not to change the use of these medications during their participation in the study, except in exigent circumstances. All medical information, including classification of headache type, was reviewed by a physician; no individual was excluded from participation based on physician review. All participants signed an informed consent form approved by the university's Institutional Review Board.
Instruments
Headache Screening Questionnaire
The Headache Screening Questionnaire (HSQ), published by the National Headache Foundation (15), contains 50 questions. Participants were instructed to check a box when an item pertained to them. The questionnaire yields four scores: Tension Type Headaches, Cluster Headaches, Migraine Headaches, and Other. The maximum value for each subscale is 10, 7, 34 and 3 for Tension Type Headaches, Cluster Headaches, Migraine Headaches, and Other, respectively.
History Questionnaire
The History Questionnaire contains 31 questions and is part of the assessment defined in the RDC for TMD (14). Each item consists of a question regarding facial pain and its relationship to the history of the patient. The History Questionnaire contains questions on general health, facial/jaw pain, injury to the jaw and face, activities limited by jaw pain, emotional distress and several demographic variables such as work status, age, income, ethnicity, gender and marital status. Individuals taking the questionnaire were instructed to answer questions by checking a box associated with each response option.
Medical and Dental History Questionnaire
The Medical and Dental History Questionnaire assessed general health information. Items were arranged into 16 broad categories and participants were instructed to check the box corresponding to the appropriate signs/symptoms and to list the duration of each sign/symptom. Participants were also asked to provide information on current medications, hospitalizations and surgeries.
Structured Interview
All individuals assigned to the HA group participated in a structured interview to obtain more information about their headaches. The areas covered in the interview included the current complaint of headaches, history of the headaches, detailed description of symptoms, relevant medical history, prior treatment and impact on daily living. The information provided by each subject was used to complete a checklist of symptoms which were used to classify the headaches into tension-type, cluster or migraine according to the nosology of the International Headache Society (IHS) (ICHD-II) (16).
Examination
All research participants were assessed using the RDC/TMD. A trained examiner, blinded to the participant's group assignment, performed the examination. The examiner palpated a series of 16 muscle sites extraorally (left and right anterior temporalis, middle temporalis, posterior temporalis, origin of masseter, body of masseter, insertion of masseter, posterior mandibular region and submandibular region) and four muscle sites accessible intraorally (left and right temporalis tendon and lateral pterygoid areas) according to the RDC/TMD protocol. During the palpations, subjects would report pressure (scored as 0) or pain, scored as either mild (1), moderate (2) or severe (3). The presence of reproducible clicking or crepitus on vertical opening, closing, lateral excursion and protrusion was determined by palpation. Pain in the TMJ was determined by palpation and rated on the same scale used to score pain to palpation in muscles. Pain-free unassisted mandibular opening, maximum unassisted opening, left and right lateral excursions, protrusion, midline deviation and incisor overlap were all measured in mm. During these measurements, the participant was asked whether they experienced pain and whether it was in the joint, muscle or both. The opening and closing pattern was also observed for midline deviations.
Experience sampling methodology
To obtain more accurate measures of pain and other states, experience sampling methodology (ESM) was used. Subjects carried pagers in this study. The pagers were one-way devices that beeped or vibrated when contacted. A custom-programmed executable (.exe) derived from the Paradox® database was used to place calls to pagers. The mean time between calls was 120 min, with a 40-min window of variability within which a specific call could be placed; a specific call to a subject could occur up to 20 min earlier or up to 20 min later than would be expected on a fixed, invariant schedule. The variability of calls was based on a random number generator that produced an equal distribution of values on either side of the expected call time. Variability in calling schedules reduced the possibility that subject behaviour would be affected by the anticipation of a call at a fixed point in time.
Subjects were instructed to fill out a preprinted 7.5 × 12.5 cm card each time they were paged, unless doing so would jeopardize their safety. Subjects were asked to report on pain in the jaw, face or head; pain elsewhere in the body; the presence and intensity of tooth contact; tension in the jaw, face or head; mood; and stress. Except for tooth contact, all measures were recorded on an 11-point (0–10) numerical rating scale. The anchors for the pain measures were No pain and Severe pain, the anchors for the tension measure were Completely relaxed and Extremely tense, and the anchors for the two mood measures were Irritable and Cheerful for one scale (Mood 1) and Happy and Sad for the other (Mood 2). The two mood scales were reversed-scored as a measure of response validity. The anchors for stress were ‘No stress’ and ‘Extremely high stress’. Tooth contact was scored on a four-point scale, ranging from ‘no contact’ to ‘strong clenching’. Subjects were instructed that they could turn off the pager if they went to bed early or stayed in bed late or were in an environment in which paging would be dangerous or disruptive.
Procedure
Subjects completed the questionnaires and, for the headache subjects, the structured interview. A second assistant, blinded to the diagnosis of the subject, was introduced to the subject and performed the examination following the protocol for the RDC/TMD.
After the physical examination was completed, an assistant instructed the subject on the use of the pager following a written script. Each subject was given multiple opportunities to interact with the device and to demonstrate competence in turning on the device, responding to a page, changing the battery and filling out the questionnaire form. Subjects were asked when they become fully alert and capable of responding to a page after awakening, and they were also asked when they typically retired for the evening. This allowed for individual flexibility in generating call schedules appropriate for each subject.
The first day of paging for each subject varied randomly from Monday through Sunday and continued for 1 week. The times when subjects were willing to receive a page in the morning and were no longer willing to receive a page in the evening were biased toward values on the hour (e.g. 07.00 h) or on the half-hour (e.g. 22.30 h). Values requested by subjects were therefore randomized by up to 15 min on either side of the requested time prior to being entered into the dialer program. Subjects were not contacted during their normal sleep hours.
Results
Demographic information on the subjects is provided in Table 1. Groups did not differ in age, education, gender composition or race.
Demographic characteristics of subjects
Standard deviations in parentheses.
Individuals in the headache group had significantly higher scores on the migraine, tension headache and cluster headache scales of the HSQ (Table 2) than the non-headache controls, F(1,38) = 142.25, 93.17 and 12.53, P ≤ 0.001, adjusted R 2 = 0.78, 0.70 and 0.23, for migraine, tension headache and cluster headache, respectively. Of the 23 individuals assigned to the headache group, all met the criteria for at least one headache type as defined in the ICHD-II; 56.5%, 52.2% and 34.8% met the criteria for tension-type headache, migraine and cluster headache, respectively.
Headache Screening Questionnaire scores
Numbers in parentheses are maximum possible values for subscales.
HSQ, Headache Screening Questionnaire.
Group differences in the number of individuals who received an RDC/TMD diagnosis were examined for each diagnostic subtype. Individuals assigned to the headache group were significantly more likely to be diagnosed as having myofascial pain (without limited opening) than individuals assigned to the non-headache control group, Fisher's exact test = 0.005. Three additional individuals in the headache group received a diagnosis of myofascial pain with limited opening, and they are included under the myofascial pain diagnosis in Table 3. Groups did not differ significantly for all other RDC diagnoses.
RDC/TMD diagnoses by group
The myofascial pain group combines Research Diagnostic Criteria (RDC)/temporomandibular disorder (TMD) diagnoses for both myofascial pain and myofascial pain with limited opening.
The diagnosis of myofascial pain requires that a patient report a current pain problem and report pain to palpation in three or more muscles. Individuals assigned to the headache group reported pain in a larger number of sites palpated than the non-headache controls (Table 4), F(1,38) = 8.07, P < 0.01, adjusted R 2 = 0.15. Group differences were present for muscle sites palpated extraorally, F(1,38) = 8.47, P < 0.01, adjusted R 2 = 0.16, but not intraorally.
Number of muscle sites with pain to palpation
Maximum values in parentheses.
Data from the questionnaire cards completed by subjects during the EMS phase of the study were summarized as means, shown in Table 5. Three of the individuals assigned to the headache group did not complete this phase of the study. The means derived from the questionnaire cards were analysed by one-way analysis of variance. These analyses showed that groups differed in the degree of facial pain, pain elsewhere in the body, tooth contact, muscle tension in the face and head, and stress.
Outcome measures obtained from experience sampling
Values in parentheses are numerical scale ratings available to participants completing the data cards. Except for the Mood 2 scale, larger values indicate more pain, contact, tension, emotional distress, stress, and effort. Smaller values in the Mood 2 scale indicate more emotional distress. For all F-ratios, d.f. = 1,35.
P < 0.05;
P < 0.01;
P < 0.00.
The proportion of time that participants reported that their teeth were in contact is also reported in Table 5. The headache group reported significantly more contact than non-headache controls. A composite variable expressing the amount of ‘effort’ exerted by subjects was computed by multiplying the mean tooth contact intensity by the percentage of time the subject reported tooth contact. Groups differed significantly on this measure.
To determine the degree to which headache type may have influenced the results, the analyses for the ESM data were re-run using HSQ scores (for migraine, tension and cluster) as covariates. All formerly significant differences between the headache and control groups were eliminated, and a significant group effect for Mood 1 emerged, F(1,29) = 5.57, P < 0.05, partial η2 = 0.161. The HSQ migraine score accounted for a significant proportion of the variance in measures of tooth contact, F(1,29) = 4.29, P < 0.05, partial η2 = 0.129, and muscle tension, F(1,29) = 5.94, P < 0.05, partial η2 = 0.170. The HSQ cluster score accounted for a significant proportion of the variance in tension, F(1,29) = 5.69, P < 0.05, partial η2 = 0.164, and Mood 2, F(1,29) = 4.70, P < 0.05, partial η2 = 0.140.
Discussion
Individuals reporting chronic headaches reported significantly more sites painful to palpation than non-headache controls, and they were significantly more likely to meet the criteria for the RDC/TMD diagnosis of myofascial pain (with or without limited opening). Palpation sites accessible extraorally were significantly more frequently reported as painful by individuals in the headache group than by those in the non-headache control group. Individuals in the headache group were somewhat more likely to receive a diagnosis of arthralgia based on reports of pain to palpation in the temporomandibular joint.
Participants in the headache group also reported significantly more frequent and more intense tooth contact than those in the control group. Subjects in the headache group reported more pain elsewhere in the body, more tension in the jaw, face and head, and stress. Addition of covariates, scores from the HSQ, accounted for significant proportions of the variance in some outcome measures. In particular, the migraine score from the HSQ accounted for variance in tooth contact and muscle tension scores, and the cluster score accounted for variance in muscle tension and Mood 2. In contrast, the Tension score from the HSQ did not account for variability in the ESM outcome measures. Introduction of covariates into the analysis eliminated significant group effects. These findings suggest that different types of headache may affect outcomes obtained via ESM. Given the strong relationship between scores on the HSQ and group membership, the elimination of group effects may also be a statistical artefact caused by the use of covariates that are highly correlated with group assignment. The use of more homogeneous subtypes in the sample of headache patients may provide greater clarity on the relationship between headache, oral parafunctional activity and TMD. A larger sample size would also clarify the relationship between headache and TMD.
These findings clearly replicate those reported for patients diagnosed with TMD. Glaros and colleagues have shown that individuals diagnosed with myofascial pain and myofascial pain plus arthralgia report more frequent and intense tooth contact, more tension, greater irritability, and greater stress (8). The proportion of time in tooth contact for subjects in the headache group was similar to that reported for TMD patients in an initial study (17) and lower than reported in the most recent report (8).
Other studies have reported at least some diagnostic overlap between headache and TMD patients and the findings reported here are not unique in that regard. For example, approximately one-half of individuals with recurrent headache reported pain when they were asked to open their mouths as wide as possible, and a slightly larger proportion also reported temporalis and masseter muscle tenderness (18). Both migraine and tension headache patients had similar amounts of pericranial muscle tenderness characteristic of many TMD patients with myofascial pain of the masticatory muscles (19). A screening questionnaire for use in a general medical practice had difficulty distinguishing between TMD subjects and patients reporting tension-type headache in the temporal area (20).
This study is unique in that it also investigated oral behaviours linked to TMD. Headache patients reported their teeth were touching nearly 50% of the time. This value is very much larger than the 17-min estimate provided in dental textbooks for the time that teeth are in contact during a typical day (21). Slight contact between the teeth is sufficient to raise the activity of the temporalis 3.0–3.5 times above the level recorded during a resting baseline (15, 22) and increased force between the teeth is associated with even more elevation in EMG activity.
The similarity in results between patients who describe themselves as having TMD and those who describe themselves as suffering from chronic headaches raises some interesting questions. These data indicate that the pain reported by TMD patients and that reported by headache patients share a similar aetiological mechanism. Previous laboratory studies have shown that otherwise pain-free individuals who engage in deliberate, low-level tooth contact report greater pain than those instructed to relax their masticatory muscles (4–7). Examinations performed by blinded examiners have shown that participation in the clenching phase of these studies results in a diagnosis of arthralia, myofascial pain, or both following clenching in about 25% of the sample, while such diagnoses are never applied to those engaged in biofeedback-assisted relaxation. Individuals diagnosed with tension-type headache who engage in 30 min of sustained low-level tooth clenching are considerably more likely to develop headache in the 24-h period following the experimental task, compared with an age- and sex-matched group of controls (23). These studies suggest that oral parafunctions may be an aetiological mechanism in some headache patients, and further research is needed to identify the role that oral parafunctions play in specific types of headache. The potential role of activity in masticatory muscles that are accessible extraorally and intraorally can also be examined.
The results may also point to a common reaction to pain. Corticotropin-releasing factor (CRF) expressed in the hypothalamus plays an important role in mediating behavioural responses to stressors. Restraining the body of an animal can activate and induce enhanced expression of CRF in paraventricular neurons of the rat hypothalamus, and aggressive biting behaviour can suppress stress-induced noradrenaline secretion in the central nervous system. In one study examining the effect of biting on restraint-induced CRF expression in the rat hypothalamus, the number of CRF-expressing neurons in the paraventricular nucleus increased significantly after short-time restraint (30 or 60 min) followed by a 180-min post-restraint period. Biting of a wooden stick during the restraint stress significantly suppressed the restraint-induced enhancement of CRF expression in the paraventricular nucleus (24). At the same time, the results may provide indirect information about patterns of care-seeking in individuals who experience facial and head pain. In their study of patients who presented at a tertiary care centre for facial pain, Glaros et al. (25) reported that approximately 40% of the patients who came to a dental school where a tertiary treatment centre was located had also been seen by a physician prior to their arrival in the clinic. In addition, referral patterns across visits showed that patients who consulted physicians were more likely to be referred to physicians and those who consulted dentists were more likely to be referred to dentists. Thus, the distinction between facial pain/temporomandibular disorder and common forms of headache may in part reflect patients' conceptualization of the condition, the financial incentives provided by medical and dental insurance and patients' subsequent choice of initial provider.
The use of ESM can reduce response bias and simultaneously introduce other uncertainties. ESM, for example, can reduce retrospective bias that occurs when a research subject selects a particularly memorable event to report as evidence of a link between a behaviour (or event) and pain. This type of selective attention necessarily ignores other events that would show a weak, non-existent or negative correlation between the behaviour and pain. A form of selection bias occurs when subjects report what each has been told by their provider. If a provider tells a patient that headache is related to the ingestion of particular foods, the patient may come to believe that the relationship is true for them, despite evidence to the contrary. At the same time, subjects in this study could be certain of at least 100 min of ‘undisturbed life’ after being paged, and we could not monitor the on/off state of the pagers. Whether either of these factors could have influenced the outcome is probably unanswerable with current technology.
Studies show that treatments emphasizing reduction of oral parafunctions are effective in reducing TMD pain (9–11). Additionally, patients who decrease oral parafunctional activity the most are also those most likely to report decreased pain (26). These findings suggest that headache patients may benefit from a reduction in oral parafunctions also. The experience sampling methodology employed in this study could be modified to teach headache patients to reduce parafunctional tooth contact and reduce muscle tension in the face, jaw and head. Patients would receive instructions to check tooth position (a proxy for masticatory muscle activity) each time they were paged and to reduce these activities if they were present. Compared with standard treatments for headaches, the risks and costs to patients would be very low and patients could develop long-lasting skills that could be applied to non-treatment environments.
Acknowledgments
We thank the staff of the Clinical Research Center and Dr Paul Dew for their assistance. D.U. and J.L. received fellowship support provided by the Kansas City University of Medicine and Biosciences for this project.