Abstract
To compare the jaw-stretch reflex and pressure pain thresholds (PPT) in chronic tension-type headache (CTTH) patients and healthy controls, 30 patients (15 male and 15 female) and 30 age- and sex-matched healthy subjects were investigated. Stretch reflexes were recorded in the temporalis and masseter muscles and PPT was determined in the anterior temporalis, splenius capitis and masseter muscles. The results showed that the amplitude of the stretch reflex in CTTH patients was higher compared with control subjects (P < 0.045), and higher in women compared with men in the right and left anterior temporalis muscles (P < 0.009). There were no differences in the PPT value between CTTH and control subjects (P > 0.509), whereas women showed significantly lower PPT measurements (P < 0.046). The results demonstrated a facilitation of the stretch reflex pathways in CTTH patients that is unrelated to measures of pericranial sensitivity.
Introduction
Tension-type headache (TTH) is extremely prevalent and represents a major health problem (1–4). Nevertheless, its pathophysiological mechanisms are largely unknown. Sustained muscle contraction has been suggested as an important source of pain in TTH (5). In recent years, however, central mechanisms and sensitization have been favoured (6–13). Substantial evidence for any of the suggested pathophysiological mechanisms is not yet available.
Stretch reflexes in the jaw-closing muscles have been used extensively as a tool to examine the excitability of the trigeminal motor system in various craniofacial pain disorders. There is a long history of a hypothesized association between excessive levels of muscle tension in the head and neck and TTH (14–16). Increased electromyographic (EMG) activity is included as a diagnostic criterion for the association with disorders of the pericranial muscles in the International Classification of Headache Disorders (17). Our previous studies have consistently shown that the jaw-stretch reflex is facilitated by experimental pain in both the masseter muscle (18–22) and posterior temporalis muscle (23). These results may suggest that the stretch reflex is a suitable technique for probing trigeminal nociceptive pathways. Increased sensitivity of the fusimotor system at this level of static muscle excitation is suggested as a possible mechanism, which could contribute to an increased stiffness of the jaw-closing muscles during pain. So far it is not known if the jaw-stretch reflex is influenced by TTH.
Furthermore, recordings of pressure pain thresholds (PPT) are recommended as one of the diagnostic criteria for TTH associated with disorders of the pericranial muscles (17). Increased tenderness of pericranial muscles has been reported in patients with TTH (24–26). However, when more specific measures are used, conflicting results appear. For example, the least amount of force or pressure necessary to elicit a report of pain (PPT) has been reported as lower in chronic tension-type headache (CTTH) (8, 10, 27), but other reports indicate that there are no differences in TTH or migraine (26, 28–30). The studies mentioned do not discriminate if the sensitivity changes are inherent to the person, if they are secondary to the headache, or if they simply accompany the headache.
It also needs to be considered that there is a female preponderance of TTH sufferers (2, 29). The male : female ratio in a l-year period prevalence of TTH is 3:4. When the chronic subtypes are analysed separately, the male : female ratio is 2:5 (31). These figures correspond widely with the male : female ratio reported from most population studies and from clinical populations, where women also have a higher prevalence of TTH than men (32–34).
Based on the available information (7, 8, 27), we hypothesized that the jaw-stretch reflex and PPTs in masticatory muscles would differ between CTTH and healthy controls in a sex-dependent way; however, it was also anticipated that there might be a graded response related to clinical characteristics of the CTTH (intensity, duration). Thus, this study set out to investigate (i) whether jaw-stretch reflex and PPT of pericranial muscles in patients with CTTH differed from that of healthy individuals, (ii) whether differences in the above measurements could be found between male and female subjects, and (iii) whether there would be associations between clinical characteristics of CTTH and the stretch reflex and PPT values.
Materials and methods
Subjects
Thirty patients with CTTH (15 male 48.4 ± 7.9 years and 15 female 44.8 ± 3.2 years) and 30 age- and sex-matched healthy control subjects (15 male 50.8 ± 1.4 years and 15 female 44.8 ± 2.2 years) were included. All 30 patients were recruited with the use of an announcement in the local newspaper and had experienced headache > 16 days/month and fulfilled the diagnostic criteria of the International Headache Society (17) for CTTH—headaches with the following pain characteristics: pressing or tightening (non-pulsating) quality, mild to moderate intensity (non-prohibitive), no aggravation from walking stairs or similar routine activities, no nausea or vomiting, no photophobia or phonophobia. The patients were instructed to avoid taking medication before 1 day of the experiment. All control subjects were healthy and did not take any medication. Their history and clinical examinations revealed no signs or symptoms of any type of headache (17) at the time of testing. The study was conducted in accordance with the Declaration of Helsinki, approved by the local ethics committee (Project No. VN 2001/128), and written informed consent was obtained from all participants prior to inclusion.
Experimental protocol
The experiments were performed to compare the jaw-stretch reflex and PPT in CTTH and healthy controls. For the stretch reflex recordings, EMG activity in both left and right masseter (MAL, MAR) and anterior temporalis (TAL, TAR) muscles was recorded. The subjects were asked to perform three maximal clenches each lasting up to 3 s on the bar with their incisor teeth to obtain the mean EMG value of the maximal voluntary contraction (MVC) in the four muscles. The EMG activity in the TAL was used for feedback. During the experiment, subjects were guided by visual feedback to keep their EMG level at 15% MVC. Twenty trials with an interstimulus interval of approximately 10 s were recorded, and the data were stored on a computer for later analysis.
Preasure pain threshold was determined with a handheld electronic algometer (SomedicAB, Stockholm, Sweden; range 0–2000 kPa with accuracy of +3% of reading) mounted with a 1-cm diameter circular rubber probe calibrated in kPa. The PPT was defined as the pressure at which the subject's perceived sensation changed from pressure to the first sensation of pain. To assess the PPT, the probe was held perpendicularly and pressure increased at a constant rate of approximately 30 kPa/s. The PPT recordings were made in MAL, MAR, TAL, TAR and in the left and right splenius capitis (SPL, SPR). The subjects were sitting comfortably with the stop button in one hand.
The recordings were made in CTTH patients during a period with headache intensity > 2 on a 0–10 numerical rating scale (NRS). Patients had been drug free at the time of testing for ≥ 24 h.
Stretch reflex
Stretch reflexes were evoked in the jaw-closing muscles with a muscle stretcher (22). Briefly, subjects bit onto metal bars with their incisor teeth. The vertical position of the lower bar was controlled precisely with a powerful electromagnetic vibrator whose moving core was under servo control (1-mm displacement, 10-ms ramp time) (22). Stretch reflex responses were recorded with the use of bipolar disposable surface electrodes (4 × 7 mm recording area, 720-01-k; Neuroline, Medicotest, Ølstykke, Denmark) placed 10 mm apart along the central part of the masseter and the anterior temporalis muscles on both sides. The skin over the recording positions was cleaned with alcohol. A ground electrode soaked with saline was attached to the right wrist. The EMG signals were amplified 2000–5000 times (Counterpoint MK2, Dantec, Denmark), filtered with a band pass of 20 Hz to 1 kHz, sampled at 4 kHz and stored for off-line analysis. Subjects were instructed to contract their muscle at a steady EMG level corresponding to 15% MVC. To help them achieve this they were shown a screen display of the root-mean-square value in 200-ms intervals of their EMG. The screen also showed the levels of EMG corresponding to 13.5 and 16.5% of the EMG recorded during the MVC. The display of their EMG level changed from green to red when it crossed the upper and lower limits of the window (22). The programme automatically triggered the jaw-muscle stretcher when the EMG activity remained within the preset window for > 400 ms. A total of 300-ms EMG activity was recorded with 100 ms pre-stimulus and 200 ms post-stimulus.
Pressure pain thresholds
The PPT was assessed at six sites by a blinded examiner. PPT assessment was block randomized to start on the left or right side, but always followed the same sequence with masseter, temporalis and splenius capitis muscles. About 30 s elapsed between consecutive measures and allowed sufficient time to avoid any sensitization of the test sites. Subjects were instructed to press the handheld button when the sensation of pressure first changed to pain. The force was released immediately following the tone produced by pressing the handheld button. The single-point measurements of the muscles were done twice on both sides of the head for each individual. The average of the two different measurements was calculated and used in subsequent statistical calculations.
Analysis
For the stretch reflex, a special-purpose computer programme processed the reflex responses evoked in the EMG. First, the mean EMG in the pre-stimulus interval (−100 to 0 ms) of the averaged and rectified signal was calculated. The onset and peak-to-peak amplitude of the early reflex component, which appeared as a biphasic potential in the average of the non-rectified recordings, was measured (22). The normalized peak-to-peak amplitudes were expressed as a percentage of pre-stimulus EMG activity (20–22) (Fig. 1). For PPT, the average of the two measurements on each point with 2-min interval was calculated and used in subsequent statistical calculations
(A) Averaged reflex responses (20 sweeps) evoked by fast jaw-stretches in a single subject. Arrows show the onset and offset of reflex. The amplitude was measured as the peak-to-peak value of the positive and negative potential. (B) Displacement signal of the stretch with 1-mm displacement and 10-ms ramp.
Statistics
Two-way analyses of variance (
Results
Clinical characteristics
The headache pain intensity was from 2 to 8 with an average of 5.7 ± 1.8 on a 0–10 NRS on the day when the experiment was performed. The headache pain duration was from 3 to 30 years with an average of 15.6 ± 12.1 years. The headache pain was described as ‘pressing’ (20/30), ‘taut’ (21/30), ‘hurting’ (15/30) and ‘exhausting’ (17/30). The pain was located in the forehead, occipital region and temple, often with a spread or referral towards the neck (19/30), and in some patients towards the masseter muscles (6/30).
Stretch reflex
Two-way
(A–F) Pre-stimulus electromyographic activity, peak-to-peak amplitude and normalized peak-to-peak amplitude of the short-latency reflex response from left and right masseter (MAL, MAR) and anterior temporalis muscles (TAL, TAR) in chronic tension-type headache (CTTH) vs. healthy controls and female vs. male subjects. Mean values ±
The mean onset latency of the reflex response evoked by the fast stretches in the EMG was 8.3 ± 0.7 ms in CTTH and 8.4 ± 0.8 ms in control subjects; 8.1 ± 0.7 ms in women and 8.5 ± 0.7 ms in men. There were no differences in onset, offset latencies and duration between CTTH and control subjects or between men and women (P > 0.454).
The mean amplitude of the stretch reflex in the CTTH and control subjects is shown in Fig. 2C. Two-way
The mean amplitude of the jaw-stretch reflex in female and male subjects is shown in Fig. 2D. Post hoc analysis showed significantly higher peak-to-peak amplitudes for women compared with men in the TAL and TAR (Tukey: P < 0.013). The normalized peak-to-peak amplitude of female was higher compared with male subjects in MAR, TAL and TAR (Tukey: P < 0.009) (Fig. 2F).
Pressure pain thresholds
The mean values of the PPT in the CTTH and control subjects are shown in Fig. 3A. The PPTs did not differ between CTTH and control subjects at any of the test sites (F < 0.448; P > 0.509).
(A–B) Pressure pain threshold (PPT) values from left and right masseter (MAL, MAR), anterior temporalis (TAL, TAR) and splenius capitis (SPL, SPR) muscles in chronic tension-type headache (CTTH) vs. healthy controls and female vs. male subjects. Mean values ±
The mean values of the PPT in women and men are shown in Fig. 3B. Women showed significantly lower PPT values in MAL, MAR, TAL and SPL compared with men (Tukey: P < 0.046).
Additional
Correlation analyses
There was no significant correlation between the normalized peak-to-peak and clinical characteristics of CTTH (intensity and duration) (P > 0.229). There was no significant correlation between the PPT and intensity and duration of CTTH (P > 0.159).
Discussion
Stretch reflex
Background EMG activity
The literature concerning electrical activity of pericranial muscles in primary headache disorders is inconclusive. No significant differences in EMG activity at rest were found between patients with TTH on the one hand and healthy individuals on the other (24, 35). Clark et al. (36) found no obvious relationship between change in temporalis muscle EMG level and headache pain level and concluded that, in the natural environment, self-perceived stress is not linked in any substantial way to a major elevation in temporalis muscle activity. In the present study, pre-stimulus EMG activity during biting on the stretch device in the CTTH patients was significantly lower than in the control subjects (Fig. 2A). This is consistent with findings of lower MVC values in various musculoskeletal pain conditions (for a review see (37)).
However, increased EMG values in the frontal, temporalis or neck musculature with the jaw and head in a relaxed (postural) position have been reported in patients with TTH (38, 39). Schoenen et al. (6) found that the EMG levels of pericranial muscles (temporalis, frontalis and trapezius muscles) were increased in CTTH patients compared with healthy volunteers. The finding that EMG level and pain sensitivity may vary independently at the same site also suggests that pain is not directly linked to muscle contraction, and suggests that EMG levels and pain thresholds would not necessarily vary simultaneously, but rather be peripheral ‘markers’, which, in spite of being produced by a common central dysfunction, are expressed to varying degrees at the different pericranial sites. Bodere et al. (40) have recorded temporalis and masseter EMG activity at rest and the masseteric chin-tap reflex in temporomandibular and orofacial neuropathic pain patients. They found that the EMG activities of both muscles at rest were significantly higher in the pain patient groups compared with the asymptomatic control group and increased EMG levels in masseter and temporalis on both sides, but were not related to the side of pain or the intensity of pain. These results suggested that the modulation of muscle activity was not the direct consequence of a peripheral nociceptive mechanism and seemed to indicate that a central mechanism was at work. The contrast between the increased EMG activity at rest and the reduction of the masseteric reflex amplitude may reflect modulations of motorneurons that differed in tonic vs. phasic conditions in chronic pain patients.
Electromyographic levels from the frontal and temporalis muscles have been recorded during standardized resting conditions and during MVC (28). These recordings indicated that the facial muscles were not completely relaxed and that the minimal electrical activity recorded by surface electrodes actually represented activity in motor units. EMG amplitudes during rest were increased in the temporalis and frontal muscles of subjects with CTTH, but not in those with the episodic form from the general population (41). Similarly, the EMG amplitudes were increased in the trapezius, and partly also in the temporalis of frequent headache sufferers from the clinical studies (5, 26), indicating insufficient relaxation. Decreased mean and median frequency levels during MVC were found in subjects with CTTH (41). These results may therefore support the involvement of the pericranial muscles in CTTH. However, the decreased frequency levels during MVC could not be confirmed in another clinical study, where 30 chronic and 28 episodic patients were examined in a similar way (5). Therefore, the relationship between EMG activity and pain are not simple, and the subject needs further investigation before final conclusions about cause/effect relations and pathophysiological relevance can be drawn.
Previous studies on EMG activity during resting conditions have indicated that the right frontal muscle is less active in women than in men, whereas no other significant sex variation could be detected (42). Men revealed higher amplitudes in the frontal, but not in the temporalis muscles during MVC compared with women (42). However, these results partly contrast with those reported by Visser et al. (43), where significantly higher masticatory EMG values were found in men when the temporalis and masseter muscles were studied, emphasizing the need for extensive EMG analysis during the entire force spectrum, and not just during rest and at MVC.
Amplitude of stretch reflex
The normalized peak-to-peak amplitude of the CTTH patients was higher compared with control subjects in the right and left temporalis muscles (Fig. 2E). A clinical consequence of the increased jaw-stretch reflex in the presence of muscle pain may be protection of the painful (damaged) tissues owing to reflex-mediated muscle stiffness. This results in reduced mobility, thereby preventing further tissue damage (22).
Schoenen et al. (7) found no relation between EMG activity and headache severity, anxiety, or response to biofeedback treatment in their study of CTTH patients, and suggested that the increased pericranial EMG activity was produced by central dysfunction (7). Similarly, increases in EMG levels from jaw and neck muscles, although short-lasting, have been reported during experimental muscle pain (18–23, 40).
A peripheral mechanism of TTH is most likely in the episodic subtype, whereas secondary, segmental central sensitization and/or impaired supraspinal modulation of incoming stimuli seem to be involved in subjects with CTTH (6–13, 31). Prolonged nociceptive stimuli from myofascial tissue may be of importance for the conversion of episodic TTH into CTTH.
We believe that the present study has provided new evidence that the jaw-stretch reflex responses, in accordance with our previous studies (18–23), are facilitated during TTH. The results could also indicate that the facilitation of the stretch reflex is a consequence of a chronic painful condition rather than a pathophysiological cause of the pain; however, prospective studies would be needed to address the cause–effect relationship, and, indeed, the relationship between nociceptive activity and stretch reflex responses is still controversial. A series of animal studies has demonstrated that various algogenic substances including hypertonic saline can induce significant changes in muscle spindle afferent activity (44, 45). Thus, the changes in muscle spindle afferent activity could be mediated via fusimotor reflexes. It has also been suggested that hypertonic saline-induced changes in the proprioceptive properties of brainstem neurons are in accordance with the notion that muscle nociceptors acting through interneurons alter fusimotor drive, which in turn alters muscle spindle primary and secondary endings (46, 47). The facilitatory effects of nociceptive activity from the pericranial region may, at least partly, contribute to the increased reflex responses.
Gender differences
Significantly higher normalized peak-to-peak amplitudes for women compared with men in the right and left anterior temporalis and right masseter muscles were found in the present study (Fig. 2F) This finding is consistent with previous studies, which have reported that the peak-to-peak amplitude of the masseter and temporalis stretch reflexes is significantly greater in women than in men (48–50). Several methodological factors such as sex-related differences in skin thickness, pre-stretch EMG activity, jaw displacement distance and biting position might have influenced the magnitude of the peak-to-peak amplitude of the jaw-stretch reflex response (51). It has been suggested that decreased resistance between the skin and the EMG electrodes in women owing to their thinner skin might explain the larger peak-to-peak amplitudes recorded in women than in men (48–50). However, in the present study the peak-to-peak amplitude of the jaw-stretch reflex response was normalized to the pre-stretch EMG activity to correct for influence of pre-stretch EMG activity (51). This manipulation should also have eliminated any potential sex-related difference owing to skin thickness. Thus, it is possible that there are sex-related differences in the sensitivity of spindle afferent fibres, the excitability of pre-motor interneurons, of masseter and temporalis motorneurons and/or the descending influences that can modulate the excitability of pre-motor interneurons and motorneurons.
Pressure pain thresholds
Our results indicate that on average there is no significant difference in pericranial muscle PPT between the present group of CTTH patients and healthy controls. This is consistent with other studies, in which no significant differences in PPT between the time of headache attack and the pain-free interval in patients with migraine (35) or between TTH and that of healthy subjects (29, 30, 38) can be found. The PPTs in the anterior temporalis region of 22 subjects with CTTH from the general population did not differ from the rest of the population or from the other headache groups (28). Similarly, PPTs and pain tolerances in the temporalis region of patients with CTTH were not different from those patients with episodic TTH (52) or in 30 age- and sex-matched healthy controls in the clinical study (5, 28). A decreased PPT indicates a state of allodynia, that is, pain elicited by stimuli which normally are non-painful. However, the findings of normal PPTs in CTTH patients indicate that general pain sensitivity is not permanently disturbed, as previously suggested (5). In patients with episodic TTH a likely pathophysiological mechanism is a slightly increased input from myofascial nociceptors projecting to a normal central nervous system. As CTTH usually evolves from the episodic subtype of TTH (53), it has been suggested that prolonged painful input from the periphery may sensitize the central nervous system and that the pain in CTTH associated with musculoskeletal disorders, thus, may be due to central misinterpretation of the incoming signals at the dorsal horn or trigeminal level. Muscular disorders may, therefore, be of major importance for the conversion of episodic TTH into CTTH.
In contrast to the present results, Langemark et al. (25) found significantly lower values for the PPT as well as for the heat pain threshold in patients with primary headache disorders, and higher tenderness scores in the pericranial muscles than controls (54), and concluded that the pathological tenderness in patients with TTH could be the source of nociception. However, the pain mechanisms are more complex, as evidenced by discrepancy between tenderness and pain in some patients. Because over one-third of their patients with TTH also suffered from migraine and a statistical age comparison was not done, these results need to be interpreted with caution. Fernández-de-Las-Peñas et al. (55)found that CTTH patients showed increased tenderness and decreased PPTs compared with healthy controls. Increased tenderness was negatively correlated with decreased PPTs in both the cephalic or neck regions. Finally, neither increased tenderness nor decreased PPTs were related to headache intensity, frequency or duration, or vice versa. Similarly, lowered pressure pain thresholds were found in patients with CTTH (8, 10, 24). Nevertheless, the decrease in PPTs was relatively small in large group comparisons. Thus, Bendtsen et al. (10) were unable to demonstrate any significant difference in PPTs from the temporalis regions of 40 CTTH patients compared with healthy controls, whereas significantly lower values were demonstrated on the fingers of these patients. Jensen et al. (56) have pointed out that 45 subjects are needed in order to detect a 30% difference in PPT in group comparisons, and the negative finding in the present study may possibly be due to type II statistical errors. This is owing to the fact that only 30 patients with CTTH were examined. The variations in previous studies may, however, also reflect the graded phenomenon of chronicity, as subjects from the general population (28) and patients with frequent but not daily headache (5) may be less severely affected than patients with daily or almost daily occurrence of headache (8, 10, 25). However, in the present study we did not find any significant correlation between the PPT and intensity and duration of CTTH. The duration of craniofacial pain and fluctuations in daily levels of pain may influence the outcome of PPT examinations within the painful region (31).
In the present study, PPTs from the left and right masseter, left anterior temporalis and splenius capitis muscles were lower in women than in men. This is consistent with the previous results (27, 31, 55), indicating that women in general are more sensitive to mechanical stimulation of muscle tissues.
Female sex has been identified as an important risk factor for both migraine and TTH in the Danish epidemiological follow-up study (57). The question is, what implications the sex differences might have on headache research. First, it underlines the importance of controlling for age and sex, or using age- and sex-matched control groups as we did in the present study. Second, we must continue research into sex differences in order to improve our understanding of the pathophysiology and treatment of pain disorders.
In conclusion, our investigations have indicated that normalized peak-to-peak amplitudes for CTTH patients were higher compared with healthy controls in the right and left anterior temporalis muscles, suggesting facilitation of the stretch reflex pathways in CTTH patients that is unrelated to measures of pericranial sensitivity. Women showed higher normalized peak-to-peak amplitudes and lower PPT measurements compared with men, indicating that myofascial pain sensitivity is higher in women compared with men. This sex-related difference in the jaw-stretch reflex response and the PPT may be an important clue in understanding why the prevalence of CTTH is greater in women than in men.