Synergist coactivation and substitution pattern of the human masseter and temporalis muscles during sustained static contractions

Clinical Neurophysiology 120 (2009) 190–197

Contents lists available at ScienceDirect

Clinical Neurophysiology journal homepage: www.elsevier.com/locate/clinph

Synergist coactivation and substitution pattern of the human masseter and temporalis muscles during sustained static contractions M. Farella a,d,*, A. Palumbo b, S. Milani c, S. Avecone a, L.M. Gallo d, A. Michelotti a a

Department of Oral, Dental and Maxillo-Facial Sciences, Section of Orthodontics and Clinical Gnathology, University of Naples ‘‘Federico II’’, Italy Department of Electronics, Computer Science and Systems, University of Calabria, Rende, Italy c Institute of Medical Statistics and Biometry ‘‘G.A. Maccacaro’’, University of Milan, Italy d Clinic for Removable Prosthodontics, Masticatory Disorders, and Special Dental Care, University of Zurich, Plattenstrasse, 11, 8032 Zurich, Switzerland b

a r t i c l e

i n f o

Article history: Accepted 1 October 2008 Available online 20 November 2008 Keywords: Clenching Masticatory muscles Bite force Electromyography Pain

a b s t r a c t Objective: Previous reports indicated that between-muscle substitution of active motor unit pools can be found in a variety of synergist muscles, including shoulder and leg muscles, but little information is available for the masticatory muscles. We hypothesized that, during a prolonged clenching effort performed at low- to moderate-bite force levels, a substitution pattern of activity can be found also in the masseter and anterior temporal muscles. Methods: Ten healthy volunteers were recruited and were asked to clench unilaterally on a force transducer for 10 min at 10%, 15%, and 20% of the maximum bite force. During each session, bite force, perceived muscle pain and electromyographic activity were continuously assessed. Data analyses were performed by means of cross-correlation and periodogram analyses. Results: During sustained static contractions, different contraction patterns of jaw elevator muscles could be identified. These included a coactivation pattern, a substitution pattern, and several intermediate situations between coactivation and substitution. Conclusions: The findings support the concept that the masticatory muscles are functionally heterogeneous and provide evidence that the neuromuscular strategies used by the masticatory system to perform sustained static contractions differ between individuals. Significance: Individual neuromuscular strategies might play a role in the development of masticatory muscle pain conditions. Ó 2008 International Federation of Clinical Neurophysiology. Published by Elsevier Ireland Ltd. All rights reserved.

1. Introduction Muscle hyperactivity has long been considered an initiating or perpetuating factor of masticatory muscle pain (Glaros et al., 1998; Laskin, 1969; Moller et al., 1984; Sheikholeslam et al., 1982). The hyperactivity hypothesis has been tested by attempting to overload the masticatory muscles under experimental controlled conditions (Bakke et al., 1996; Clark et al., 1984, 1989; Clark and Carter, 1985; Farella et al., 2001; Glaros and Burton, 2004; Jow and Clark, 1989; Svensson and Arendt-Nielsen, 1996; Svensson et al., 2001; Christensen, 1989). The findings of these studies showed that moderate to high (i.e. P25% of maximum voluntary contraction; MVC) sustained static contractions of the masticatory muscles leads to unbearable pain and fatigue that impede the sub-

* Corresponding author. Address: Clinic for Removable Prosthodontics, Masticatory Disorders, and Special Dental Care, University of Zurich, Plattenstrasse, 11, 8032 Zurich, Switzerland. Tel.: +41 446343263; fax: +41 446344302. E-mail addresses: [email protected], [email protected] (M. Farella).

jects from continuing the contraction task (Christensen, 1989; Clark and Carter, 1985; Clark et al., 1989). Both pain and fatigue, however, disappear very quickly after task cessation and the masticatory muscles do not develop post-exercise tenderness, even when the endurance tests are repeated several consecutive days (Svensson and Arendt-Nielsen, 1996). On the other hand, there is evidence that prolonged static contractions at lower contraction levels (i.e. <25% MVC) may significantly reduce bite force and jaw opening capacity (Svensson et al., 2001), and lead to pain and tenderness in the masticatory muscles (Glaros et al., 2000; Glaros and Burton, 2004), which are the most common signs and symptoms of temporomandibular disorders (TMD). Tooth clenching is frequently reported by patients experiencing persistent masticatory muscle pain (Huang et al., 2002; Velly et al., 2003). It should be noted, however, that so-called bruxers (i.e. people who clench or grind the teeth at wake and/or sleep time) do not always feel masticatory muscle pain (Dao et al., 1994), and may react to bruxism with an enlargement and/or hypertrophy of the masticatory muscles (Balatsouras et al., 2004; Da Silva and Mandel, 2006; Kato et al., 2001; Mandel and Kaynar, 1994). Thus, it is still

1388-2457/$34.00 Ó 2008 International Federation of Clinical Neurophysiology. Published by Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.clinph.2008.10.004

M. Farella et al. / Clinical Neurophysiology 120 (2009) 190–197

unclear why hyperactivity may result in persistent masticatory muscle pain only in some individuals. One possibility is that the development of muscle injury and pain might depend upon the individual neuromuscular control strategies used for maintenance of the sustained contraction. These include, for instance, a substitution of the motor unit drive either between- or within-muscles in order to optimize the workload and to prevent damage to muscle fibers (Westgaard and de Luca, 1999; Zennaro et al., 2003a,b). Indeed, a between-muscle substitution of active motor unit pools has been found in a variety of synergist shoulder and leg muscles (Holtermann et al., 2005; Kouzaki and Shinohara, 2006; McLean and Goudy, 2004). A rotation of synergistic activity between the masseter and anterior temporalis muscles during sustained static contractions has been reported only in one study (Hellsing and Lindstrom, 1983). During a clenching test performed at high force level, five subjects experienced a sudden relief of pain, which corresponded to a marked reduction of EMG amplitude in the masseter muscle and to an increase of EMG amplitude in the anterior temporal muscle. An alternate switch of activity between the two synergistic muscles was found in all the subjects after repeated tests (i.e. a facilitation effect occurred over consecutive trials). The authors concluded that a trade-off mechanism turns part of the load over to the more rested synergist, when a jaw elevator muscle is fatigued and painful. Unfortunately, the findings arising from this study were mostly based about a qualitative description of the phenomenon under study and subjective reports of the subjects investigated. The aim of the present investigation was to quantitatively study the co-contraction pattern of the masseter and anterior temporal muscles during sustained static contractions performed at lowto moderate-bite force levels. It was hypothesized that, during these tasks, the main load can alternately switch between the two most important synergistic jaw elevator muscles.

2. Material and methods 2.1. Subjects Ten subjects (five males and five females) were recruited among post-graduate students and staff members attending the Dental Clinic of the University of Naples Federico II. The group mean age ± SD was 28.2 ± 7.6 years. Eligible subjects were screened for temporomandibular disorders (TMD) according to Axis I of the Research and Diagnostic Criteria (Dworkin and LeResche, 1992). To be included in the study, the subject had to fulfil following criteria: (a) adult age (>18 years); (b) pain-free active mouth opening >40 mm (including overbite), pain-free active protrusion and laterotrusion >7 mm; (c) difference between active and passive opening 62 mm; (d) positive overjet and overbite between 0 and 4 mm; (e) willingness to participate to the experiment and to sign a written informed consent. Exclusion criteria were: (a) any TMD diagnosis; (b) chronic pain conditions (>3 month duration) in other parts of the body; (c) current orofacial inflammatory conditions; periodontal diseases; (d) removable dental prostheses; (e) absence of any teeth (except third molars); (f) neurological and movement disorders; and (g) habitual intake of drugs influencing the activity of the central nervous system. The subjects were carefully informed about the experimental protocol, but were left naïve to tested hypothesis. 2.2. Bite force equipment Subjects were assisted in the clenching tests by visual feedback from bite force. This was recorded by a steel housing strain

191

gauge force transducer (15  11  4 mm, length  width  height) with a working range between 0 and 1000 N, connected to an amplifier with an analog output (Floystrand et al., 1982). The force transducer was covered with a plastic foil to prevent moisture and then fitted into a protective polymethylene bite block (Eggen, 1969). 2.3. Pain rating The pain perceived in the masseter and anterior temporal muscles during clenching efforts was assessed by means of electronic visual analog scales (VAS; Scott and Huskisson, 1976) with the anchors points ‘‘no pain’’ and ‘‘worst pain imaginable’’. The equipment consisted of 100-mm linear cursors (variable resistors). Four cursors, each corresponding to one of the four masticatory muscles, were placed in a custom-made case (Stebi srl, Napoli, Italy). The location of each pain site was indicated to the subject by means of a schematic drawing of the face and masticatory muscles attached to the front of the case. 2.4. Electromyographic equipment Muscle activity was picked up bilaterally from the masseter and anterior temporal muscles by means of self-adhesive pre-gelled disposable bipolar surface electrodes (Duotrode, Myotronics Inc. Seattle, Washington, US). Electromyographic (EMG) signal was low-pass filtered (1000 Hz) and amplified (3000) by means of BM623 (Biomeccanica Mangoni, Pisa, Italy) with a noise level of less than 1 lV and a common mode rejection ratio higher than 100 dB. 2.5. Software and data acquisition The bite force, raw EMG, and VAS signals were A/D-converted at 2 kHz by means of a multifunction data acquisition board (PCI 6024; National Instrument, Austin, Texas, USA) connected to a computer motherboard (Intel Pentium IV). Digital signals were then processed using custom-made software developed in the LabView environment (LabView 8.0; National Instrument, Austin, Texas, USA). This software provided the subjects with visual feedback for keeping constant the bite force level. This consisted of a vertical bar whose height was proportional to the force produced. A green lamp was switched on when the subjects kept the force level within ±20% of the value. Conversely, excessive or insufficient bite force level was indicated by the appearance of a red arrow pointing up or down, respectively. Raw EMG signals were digitally filtered using a narrow band-stop centred at 50 Hz and were converted to the root mean square amplitude values (EMGRMS) using a 3 s window. The output of the software consisted of nine signals: one bite force signal, four EMGRMS signals, and four VAS signals. Recorded signal were all stored as ASCII files for subsequent off-line analyses. 2.6. Procedure The study included a clinical session and three experimental sessions, separated by at least one-week. The study protocol was fully respectful of the principles of the Helsinki Declaration and received approval from the ethical committee. During the clinical session, subjects were asked to report their preferred chewing side (later this was the ipsilateral side of the experiment). Lacking a preferred side, the right side was chosen for the application of the force transducer. For all subject, dental impressions (Extrude XP and Wash, Kerr, Romulus, MI, USA) of the lower arch were taken (SA) and poured with stone (VelMix Stone type IV, Kerr, Scafati, Italy). A 1–2 mm thick layer of

192

M. Farella et al. / Clinical Neurophysiology 120 (2009) 190–197

self-curing acrylic resin (Duralay, Reliance Dental, IL, US) was applied to the lower surface of the bite block in order to ensure reproducible and stable seating of the transducer. Thereafter, the bite block was placed upon the dental cast so that the transducer centre coincided with the mesio-buccal cusp of the lower first molar. After removal of resin excesses and undercuts, a second layer of acrylic resin was applied to the upper surface of the bite block; this was then fitted intra-orally to the lower teeth, and the subject was asked to close the mouth slowly until the upper teeth contacted the resin layer. The force transducer in place caused a jaw opening (i.e. the interincisal distance corrected for overbite) ranging from 13.4 to 20.1 mm (mean = 17.6 mm) and was used for all clenching tests. After bite block construction, the subjects got acquainted with the procedures used in the study including the assessment of maximal bite force, the use of the force visual feedback, and the use of VAS equipment. The three experimental sessions included three clenching tasks performed at 10%, 15%, and 20% of the maximum voluntary bite force (MVBF), respectively, with the order of tasks being randomized. At the beginning of each experiment, the bite block was placed into the mouth. Since a temperature-dependent offset shift of force signal had been observed in a pilot study, the subject was asked to keep the bite block-transducer in the mouth without clenching for at least 5 min. Then, the skin overlying the masseter and anterior temporal muscles was rubbed vigorously by cotton pads soaked into 90% ethanol. The surface EMG electrodes were placed bilaterally along the main direction of the muscle fibres as determined by palpation with a center to center distance of 21 mm (Farella et al., 1999). The reference electrode was attached to the ipsilateral mastoid process. In order to allow reproducible repositioning of the electrodes across each experimental session, a transparent pliable plastic template was aligned to the ear, to the labial margin and to the eye, and the location of the electrode sites was marked. The subjects were then asked three times to clench as hard as possible. The duration of each maximum clench was 1–2 s with an inter-clench rest interval of 30 s. The maximum peak value of bite force and of EMGRMS obtained during clenching efforts were stored and used as estimates of the maximum voluntary bite force level (MVBF) and of the maximum voluntary contraction activity (MVC), respectively. After maximal clenching efforts, the subjects were allowed to rest for 3 min. Thereafter, subjects were asked to perform the clenching task at the required force level (i.e. 10%, 15%, or 20% MVBF) by the aid of visual feedback. Subjects had to rate continuously the intensity of the pain perceived in each of the four muscle throughout the whole experiment. During the experiment the subjects were not allowed to look at the amount of EMG amplitude produced during contraction in each muscle. Each clenching task had to be performed for 10 min and the subjects were verbally encouraged to endure even if they experienced pain and fatigue. However, the subjects had also been allowed to stop clenching if they felt too exhausted. After the clenching task, the subject rested for 30 s and then the recording equipment was disconnected. 2.7. Data analysis and statistics Each experimental session included an initial rest period, three maximum voluntary efforts, the clenching test, and a final rest period. The bite force signal typically showed an initial jump from zero to the target force, a sustained clenching effort, and a sudden decrease back to zero at task cessation, the sustained clenching effort being the segment of interest in the present investigation. This segment was selected by using leading and lagging zero values as initial and final markers of the task. The first and the last 2 s of the clenching segment were always dis-

carded in order to avoid boxcar artefacts during subsequent harmonic analyses. EMGRMS and VAS signals were also cut along the bite force segment selected. EMGRMS amplitude was expressed as proportion of the EMGRMS peak values obtained during maximum clenching (i.e. relative EMG amplitude = EMG%MVC) within each muscle. Preliminary analyses consisted of descriptive statistics, normality tests, and tests for homogeneity of variances. Means and standard deviations were computed for each outcome variable (i.e. bite force, EMG%MVC, and VAS) across subjects and across the different experimental factors (i.e. clenching force level, time). Bite force, EMG%MVC, and VAS were modelled as a function of the factors involved in the experiment. The following experimental factors and their first-order interactions were entered into a repeated measurements model: subject (block: random factor), clenching force level (fixed factor: three levels) and time (fixed factor: 10 levels consisting of 1-min intervals). Least square means and their standard errors were used to plot the temporal profile of the response variables. Orthogonal contrasts up the fourth degree were also calculated to estimate the time-related linear and non-linear effects of each response variable. The time series of EMG%MVC signal was then detrended only with respect to the linear component, since the quadratic and cubic components might be representative of a substitution pattern. Henceforth in this report, detrended EMG relative amplitude was indicated as EMGdetr%. The occurrence of a substitution pattern between synergist muscles during the clenching test was analysed using three approaches. The first approach consisted of computation of unlagged cross-correlation coefficients between EMGdetr% of each pair of masseter and anterior temporalis muscles obtained across the entire recording. The second approach consisted of calculation of correlation coefficients by segmenting the recording in ten 1-min segments. Significant positive correlation coefficients indicated a coactivation pattern between synergist muscles occurring during the clenching test whereas significant negative correlation coefficients were indicative of a substitution pattern. The third approach consisted of a periodogram analysis of the difference of EMGdetr% between the two synergistic muscles (i.e. diffEMG(t) = EMGdetr%(t) masseter EMGdetr%(t) anterior temporalis) using a fast Fourier transform (FFT) algorithm (Monro and Branch, 1976). In the case of a substitution pattern, it can be expected that the diffEMG will fluctuate with a cyclic pattern showing significant periodogram peaks. Periodogram intensities of diffEMG were calculated for each subject, for each force level, and for each pair of synergistic muscles. Each periodogram had a temporal resolution of 6 s and was normalized by dividing each intensity estimate by the sum of all periodogram ordinates (i.e. the overall variance), thus allowing the estimation of the proportion of variance provided by each rhythmic component (or frequency band) represented in the periodogram. Periodogram intensities were plotted as a function of period length (from 0 to 10 min). The g statistic was used to test the significance of periodogram peaks (Russel, 1985). The sum of normalized periodogram intensities for the band extending from 2 to 10 min was used as rhythmicity index to test the effect of bite force level, and of the order of the clenching sessions (i.e. facilitation factor) on the periodic components of diffEMG signals. The relationship between pain and EMG activity was analyzed using repeated measurement ANOVAs, where EMG%MVC obtained from each muscle investigated represented the dependent variables and VAS scores and time were entered into the model as response variables. All data processing and statistical analyses were performed using SAS package (SAS 8.01, SAS Institute, Cary, NC USA). Type I risk error for all statistical tests was set at 0.05 with the only exception of g statistics (a = 0.001).

193

M. Farella et al. / Clinical Neurophysiology 120 (2009) 190–197

eral and contralateral facial sides. Perceived pain was significantly influenced from the clenching intensity level (F P 3.1; p 6 0.01), being higher during the 15% and 20% MVBF tasks. After adjusting for the effect of time, there was not significant relationship between VAS scores and EMG%MVC (F 6 1.1; p P 0.44).

3. Results 3.1. Bite force, endurance and pain Mean ± SD maximal unilateral bite force was 420.7 ± 55.7 N. No polynomial component of the time-related profile of bite force level was significant (F 6 3.6; p P 0.06; Fig. 1A), thus indicating that the mean bite force was not affected by time. All subjects could sustain the 10% MVBF trial for 10 min, but endurance time became shorter at higher clenching level being 9.9 ± 0.15 min at 15% MVBF, and 7.5 ± 2.1 min at 20% MVBF. One out of the 10 subjects investigated (i.e. subject 8) was not able to sustain the 15% MVBC for 10 min, whereas only three subjects (i.e. subjects 2, 4, and 5) could sustain the 20% MVBC trial for 10 min. With the only exception of subject 4, all the other subjects experienced moderate to severe pain 4–5 min after starting of the clenching tests; when endurance time was shorter than 10 min, the subjects reported that unbearable pain was the reason for stopping the task. Endurance time and peak pain values experienced during the 20% MVBC trial were negatively correlated (R2 = 0.68; p = 0.003). Mean values of VAS pain scores for pain gradually increased throughout the clenching task for all muscle sites. Orthogonal polynomial decomposition of time-related profiles of VAS scores indicated a significant linear positive trend in the amount of perceived pain (4.15 6 F 6 137.0; 0.0001 < p 6 0.045). Higher order components, however, were not statistically significant (F 6 1.4; p P 0.23; Fig. 1B). Perceived pain during the clenching tasks was more intense at the masseter than at the anterior temporal muscles sites (p < 0.001), but it did not differ between ipsilat-

Force (N)

80

BITE FORCE

A

80

BITE FORCE

B

80

60

60

40

40

40

20

20

20

40

IpsMass IpsTemp CntMass CntTemp

40

50 IpsMass IpsTemp CntMass CntTemp

40 30

30

30

20

20

20

10

10

10

60 50 40

60 IpsMass IpsTemp CntMass CntTemp

30

0 1 2 3 4 5 6 7 8 9 10

0 1 2 3 4 5 6 7 8 9 10

0 1 2 3 4 5 6 7 8 9 10

50 40

60 IpsMass IpsTemp CntMass CntTemp

50 40

30

30

20

20

20

10

10

10

0

0

0 1 2 3 4 5 6 7 8 9 10

Time (min)

IpsMass IpsTemp CntMass CntTemp

0

0

0

C

0 1 2 3 4 5 6 7 8 9 10

0 1 2 3 4 5 6 7 8 9 10 50

50

BITE FORCE

0

0 0 1 2 3 4 5 6 7 8 9 10

EMG (%MVC)

Mean EMG relative amplitude (i.e. EMG%MVC) gradually increased throughout the clenching tasks for all muscles investigated (Fig. 1C), with the anterior temporalis ipsilateral to the bite force transducer showing the highest EMG%MVC values. EMG%MVC of both masseter and anterior temporalis signifi-cantly increased with time (51.8 6 F 6 137.0; p 6 0.0001), the mean square (MS) due to the linear component being by far higher than that due to the quadratic and cubic components (1478.8 6 MS 6 3755.2, 28.8 6 MS 6 88.9, and 37.0 6 MS 6 323.1, respectively). Visual examination of all individual recordings showed that the EMGdetr% amplitude fluctuations of synergist muscles exhibited a variable degree of synchronization among different subjects and among different trials (Fig. 2). The magnitude and significance level of unlagged cross-correlation coefficients between the detrended EMG relative amplitudes of synergist muscles during the whole clenching tasks for each subject, force level, and facial side are summarized in Table 1. In most recordings of synergist muscle pair (48 out of 60) the EMGdetr% values were positively interrelated (Table 2). Only in 4

60

0

VAS (mm)

3.2. Coactivation and substitution pattern

0 1 2 3 4 5 6 7 8 9 10

Time (min)

IpsMass IpsTemp CntMass CntTemp

0 0 1 2 3 4 5 6 7 8 9 10

Time (min)

Fig. 1. Mean values (±standard error of the means) of bite forces, VAS scores and relative EMG amplitude (EMG%MVC) averaged across subjects, and time interval (each time interval = 1 min) obtained during the clenching tasks at 10% (A), 15% (B), and 20% (C) of maximum voluntary bite force. Abbreviations: EMG, electromyographic; VAS, visual analog scales; IpsMass, ipsilateral masseter; IpsTemp, ipsilateral anterior temporalis; CntMass, contralateral masseter; CntTemp, contralateral anterior temporalis.

194

M. Farella et al. / Clinical Neurophysiology 120 (2009) 190–197

A

Time (min)

Detrended EMG (%)

0

1

Masseter 2

3

4

the 1-min segments were positive, while the remaining coefficients were not significant (Table 2). No negative correlation coefficient was ever found across all 1-min segments (Table 2). Negative cross-correlation coefficients across the entire recording were found only in two subjects (i.e. subjects 2 and 5), and more frequently in subject 5 (Table 1). In these subjects, significant negative correlation coefficients across the 1-min segments occurred much more frequently than in all the other subjects (Table 2). The remaining five subjects (i.e. subjects 3, 6, 8, 9, and 10) showed less straightforward results, as a number of positive as well as of non-significant correlation coefficients across the entire recording were found in each subject (Table 1). The average strength of positive correlation coefficients was low (<0.5), and in four (e.g. subjects 3, 6, 8, and 0) out of these five subjects, at least one negative correlation coefficient was found across the 1-min segments (Table 2). Periodogram analysis of diffEMGs showed the occurrence of significant intensity peaks in 4 out of 10 subjects (i.e. subjects 2, 5, 8, and 10) investigated and most frequently in subject 5 (Table 3). The mean period (±SD) of these rhythmic components was 5.7 min (±2.4 min) and ranged from 4.1 to 10 min. The rhythmicity index of diffEMGs was not affected by the bite force level (F = 0.69; p = 0.42) and by the order of the clenching sessions (F = 0.22; p = 0.64).

Anterior Temporalis 5

6

7

8

9

10

30 20 10 0 -10 -20 -30

Detrended EMG (%)

B

Detrended EMG (%)

C

30 20 10 0 -10 -20 -30 30 20 10 0

4. Discussion

-10 -20 -30

Fig. 2. Examples of different individual patterns of detrended EMG relative amplitude (EMGdetr%) obtained from the masseter and anterior temporal muscles ipsilaterally to the bite force transducer in three subjects (subjects 4, 5, and 10, respectively) during a clenching test performed at 10% of maximal voluntary bite force: (A) coactivation pattern, (B) substitution pattern, and (C) intermediate pattern in between coactivation and substitution. Note that negative values of EMG values are the consequence of linear trend removal.

out of 60 recordings of synergist muscle pair significant negative correlation coefficients were obtained (Table 2). The examination of the magnitude and of the signs (+, , and =) of correlations coefficients indicated that in three subjects (i.e. subjects 1, 4, and 7), all the unlagged cross-correlation coefficients obtained across the entire recording were positive, significant, and greater than 0.5 (Table 1). In these subjects, more than 70% of coefficients obtained across

Maximum unilateral bite forces obtained from the sample investigated were in agreement with previous data obtained in healthy subjects (Bakke et al., 1989, 1990). The observation that bite force levels did not vary significantly throughout the clenching tasks indicated that the visual feedback helped the subjects in maintaining a steady level of biting force throughout the whole experimental session. Endurance time decreased with increasing clenching force levels and pain was the reason for stopping the higher level clenching tasks. This pain can be ascribed to the impairment of blood flow due to an increased intramuscular pressure and subsequent ischemia (Clark et al., 1989; Rasmussen et al., 1977), and was more pronounced in the masseter muscles than in the temporalis muscles. Mean values of the EMG relative amplitude continuously increased throughout the submaximal clenching tasks. This agrees with previous findings obtained from the same muscles (Svensson et al., 2001) and has been explained through facilitated motor unit recruitment (Moritani et al., 1986), increase in the degree of synchronization between motor units (Krogh-Lund and Jorgensen,

Table 1 Unlagged cross-correlation coefficients between detrended EMG relative amplitude of the masseter and of the anterior temporalis muscle ipsilaterally and contralaterally to the bite force transducer, obtained over the whole recording Bite force level

10% MVBF

Facial side

Ipsilateral

Subject Subject Subject Subject Subject Subject Subject Subject Subject Subject

1 2 3 4 5 6 7 8 9 10

0.82** 0.45** 0.58** 0.88** 0.36** 0.70** 0.66** 0.31** 0.37** 0.48**

MVBF, maximum voluntary bite force. * p < 0.05. ** p < 0.01.

15% MVBF Contralateral 0.59** 0.19* 0.16 0.62** 0.88** 0.07 0.68** 0.29** 0.07 0.37**

Ipsilateral 0.89** 0.43** 0.47** 0.88** 0.19 0.80** 0.84** 0.46** 0.68** 0.67**

20% MVBF Contralateral 0.78** 0.51** 0.51** 0.81** 0.14 0.21* 0.85** 0.34** 0.52** 0.12

Ipsilateral 0.76** 0.08 0.20 0.88** 0.02 0.85** 0.88** 0.75** 0.33** 0.50**

Contralateral 0.68** 0.04 0.56** 0.74** 0.56** 0.06 0.72** 0.61** 0.19 0.46**

195

M. Farella et al. / Clinical Neurophysiology 120 (2009) 190–197

Table 2 Signs counts of correlation coefficients between detrended EMG relative amplitude of the masseter and of the anterior temporalis muscle ipsilaterally and contralaterally to the bite force transducer obtained sequentially over 1-min intervals Bite force level

10% MVBF

Facial side

Ipsilateral

Sign of coefficients

+

Subject Subject Subject Subject Subject Subject Subject Subject Subject Subject

10 5 1 10 3 8 5 4 6 7

1 2 3 4 5 6 7 8 9 10

0 0 0 0 3 0 0 0 0 0

15% MVBF Contralateral =

+

0 5 9 0 4 2 5 6 4 3

9 3 10 8 6 3 7 3 4 8

0 4 0 0 2 0 0 0 0 1

20% MVBF

Ipsilateral =

+

1 3 0 2 2 7 3 7 6 1

10 5 4 10 1 9 9 6 8 5

Contralateral

0 0 0 0 2 0 0 1 0 0

=

+

0 5 6 0 7 1 1 3 2 5

10 0 3 8 7 6 7 4 4 2

0 3 1 0 0 0 0 0 0 0

Ipsilateral =

+

0 7 6 2 3 4 3 6 6 8

6 7 2 10 2 5 7 5 3 7

0 0 0 0 1 0 0 0 0 0

Contralateral =

+

4 3 4 0 5 0 1 1 4 3

8 1 3 10 3 0 7 3 2 4

= 0 0 1 0 2 0 0 0 0 1

2 9 2 0 3 5 1 3 5 5

MVBF, maximum voluntary bite force; ‘+’, number of positive correlation coefficients; ‘ ’, number of negative correlation coefficients; ‘=’, number of non-significant correlation coefficients.

Table 3 Number of significant peaks for each periodogram obtained from the difference of detrended EMG relative amplitude of the masseter and of the anterior temporalis muscle ipsilaterally and contralaterally to the bite force transducer Bite force level

10% MVBF

Facial side

Ipsilateral

Contralateral

Ipsilateral

Contralateral

Ipsilateral

Contralateral

Subject Subject Subject Subject Subject Subject Subject Subject Subject Subject

0 1 0 0 1 0 0 0 0 0

0 1 0 0 2 0 0 0 0 0

0 0 0 0 1 0 0 1 0 1

0 0 0 0 1 0 0 0 0 0

0 1 0 0 0 0 0 0 0 1

0 0 0 0 1 0 0 0 0 0

1 2 3 4 5 6 7 8 9 10

15% MVBF

1991), and decrease of action potential conduction velocity along the muscle fibers (Lindstrom et al., 1977). In order to assess the synchronicity between masseter and temporalis in each subject during the clenching tests we used analytic approaches of progressive refinement. The first approach consisted in determining the unlagged cross-correlation coefficients between the detrended activities of each pair of jaw elevator muscles (i.e. ipsilateral and contralateral masseter and temporalis muscles) obtained during the whole clenching task. The second approach consisted in calculating the cross-correlation coefficients by segmenting the entire EMG recording in 1-min intervals. As a last approach, we used the periodogram analysis, looking for the occurrence of cyclic components in the difference of detrended EMG amplitude between the two synergist muscles investigated. Even though the findings resulting from these different analytic approaches were not totally in agreement, they all indicated that individuals differ in the co-contraction patterns of the synergist masseter and anterior temporalis during sustained clenching efforts. Based upon critical evaluation of the findings, we could describe several patterns of activity. Indeed, 3 out of 10 subjects (subjects 1, 4, and 7) demonstrated a marked synchronicity with continuous coactivation of the two elevator muscles, which was reflected in positive sign, magnitude, and significance of the cross-correlation coefficients, and in the absence of any significant peak emerging from the periodogram analysis. Conversely, significant negative cross-correlation coefficients and significant periodogram peaks were found more frequently in 2 out of the 10 subjects investigated (subjects 2 and 5). These findings indicated that an activity substitution during the clenching test occurred in these subjects with a cycle period of about

20% MVBF

5–6 min. This phenomenon was similar to the so-called ‘‘rotation’’ of synergist activity (Hellsing and Lindstrom, 1983), and could be interpreted as a protective mechanism, as a trade-off of activity between synergist muscles during prolonged fatiguing contractions would allow time for the contractile elements to recover, while the overall force being produced by the masticatory system remains constant. The observation that the subjects showing this phenomenon could hold the whole 20%MVBF clenching task may further corroborate this hypothesis. Unfortunately, in the present study, duration of clenching sessions was limited to 10 min at maximum. It would be interesting in future studies, to investigate substitution activity patterns in relation to endurance time resulting from different clenching efforts performed up to exhaustion. Contrary to previous findings (Hellsing and Lindstrom, 1983), which reported a rotation of activity in the whole sample investigated, the substitution phenomenon occurred in only 2 out of the 10 subjects participating to the present study. Furthermore, we did not find a facilitation effect over consecutive trials, as the factor ‘‘order of clenching sessions’’ did not affect significantly the periodogram peaks, and the substitution phenomenon occurred since the first clenching session. Also, we could not demonstrate any significant relationship between EMG relative amplitude values and the amount of current pain experienced at each muscle investigated. Due to a high content of type I muscle fibers, the jaw muscles are less susceptible to pain and fatigue than limb muscles (Eriksson and Thornell, 1983). Indeed, the pain perceived during the clenching tests, became moderate or severe only 4–5 min after starting the 15% and 20% MVFB tasks. We cannot exclude that the substitution phenomenon might occur more frequently with stronger pain. This hypothesis may be tested in future studies using higher

196

M. Farella et al. / Clinical Neurophysiology 120 (2009) 190–197

clenching forces and/or injecting algesic substances in the masticatory muscles during the clenching tasks. It needs to be emphasized, however, that we did not find any relationship between the substitution pattern and the clenching intensity, at least in the range of bite force investigated in the present study (i.e. 10–20% of MVBF), which could be considered representative of low- to moderateclenching efforts. The EMG co-contraction pattern of the masseter and anterior temporalis in 5 out of 10 subjects was difficult to interpret, as neither a marked synchronous coactivation nor an evident substitution could be identified. These patterns were considered as ‘‘intermediate’’ situations between coactivation and substitution, and may reflect spatial inhomogeneous activation of jaw elevator muscles during sustained static contractions of the masticatory muscles. In other words, it is possible that different part of the masseter and anterior temporalis muscles are predominantly active during the clenching tests and that only the most active parts are picked up by the bipolar surface electrodes as they were placed in the present study. Since the masticatory system is mechanically redundant (van Eijden et al., 1990) an infinite number of muscle activity patterns is theoretically capable of producing a specific task. The observation that the motor strategy employed for the performance of the clenching tasks varied among different individuals is therefore not surprising and is consistent with previous EMG studies showing heterogeneous activation in masticatory muscles during other tasks (Blanksma and van Eijden, 1995; Farella et al., 2002; Turp et al., 2002), and with histochemical and immunohistochemical studies showing that these muscles exhibit an uneven fiber type distribution and an architectural and nervous compartmentalization (Eriksson and Thornell, 1983; Korfage and van Eijden, 2003). It is also possible that substitution or rotation phenomena occurred with other synergist not investigated in the present study (e.g. medial pterygoid muscle) or within the motor unit pool of each masticatory muscle. This last hypothesis would be consistent with previous findings obtained in the trapezius (Westgaard and de Luca, 1999), in the limb muscles (Bawa et al., 2006) and in the masseter muscle (Nordstrom and Miles, 1991). It is also possible that during prolonged clenching, a number of agonist and antagonistic jaw muscles undertook a progressive cocontraction. This type of co-contraction, for instance, has been previously shown for several limb muscles during fatiguing contractions (Levenez et al., 2005; Psek and Cafarelli, 1993). It needs to be emphasized, however, that our analytical approach aimed to investigate the synchronicity between synergist muscles after linear detrending of all EMG signals. Therefore, a possible co-contraction between antagonistic jaw muscles should not affect our results. A limitation of the present study is the potential confounding factor of cross-talk between the jaw elevators muscles investigated. This cross-talk might have some impact on the magnitude of unlagged cross-correlation coefficients, particularly for the subjects exhibiting a marked synchronicity of EMG signals. To the best of our knowledge, no investigation has ever tried to estimate the cross-talk between the masseter and anterior temporalis muscles during sustained contractions, and therefore we cannot infer from the previous literature on this argument. It may be worth mentioning, however, that a recent study (Jaberzadeh et al., 2007) has investigated the EMG cross-talk between right and left digastric muscles, using focal transcranial magnetic stimulation. The crosstalk between surface EMG recordings obtained from these two muscles was rather low (i.e. estimated at approximately 10%). Since masseter and anterior temporalis are anatomically well distinct muscles, and since unilaterally they are at greater distance than left and right digastric muscles, we consider that their

cross-talk should be negligible. On the other hand, the alternative hypothesis that the significant cross-correlation between the two EMG signals would reflect a synergist coactivation pattern due to common drive is consistent with the ‘‘sharing principle’’ proposed by Stephens et al. (1999). According to this model, muscles acting around a common joint, such as the temporomandibular joint, receive common synaptic inputs to their motoneuron pools. The hypothesis of a central common drive to several masticatory muscles is also supported by other findings (Carr et al., 1994; Jaberzadeh et al., 2006). Another limitation of the present study is the small number of subjects investigated, which, for instance, did not allow us to investigate for potential gender differences in the contraction pattern of the masticatory muscle. It is interesting to note, however, that out of the two subjects showing a substitution activity pattern, one was male and the other was female. Despite the small sample size, our findings support the idea that neuromuscular strategies used by the masticatory system to perform sustained static contractions differ between subjects and that several distinct contraction patterns can be identified. This finding may account to explain why prolonged masticatory load of long duration such as that of bruxers may be at risk of developing muscle pain only in some individuals. Differences in the individual motor control strategies occurring during prolonged clenching contractions can be implicated in protective mechanisms aiming to reduce the potential detrimental effects of prolonged fatigue and overload of the jaw muscles. Acknowledgments We gratefully acknowledge the volunteers participating in the experiments. The study was supported by the standard financial plan of the Department of Oral, Dental and Maxillo-Facial Sciences, University of Naples ‘‘Federico II’’. References Bakke M, Michler L, Han K, Moller E. Clinical significance of isometric bite force versus electrical activity in temporal and masseter muscles. Scand J Dent Res 1989;97:539–51. Bakke M, Holm B, Jensen BL, Michler L, Moller E. Unilateral, isometric bite force in 8to 68-year-old women and men related to occlusal factors. Scand J Dent Res 1990;98:149–58. Bakke M, Thomsen CE, Vilmann A, Soneda K, Farella M, Moller E. Ultrasonographic assessment of the swelling of the human masseter muscle after static and dynamic activity. Arch Oral Biol 1996;41:133–40. Balatsouras D, Kaberos A, Psaltakos V, Papaliakos E, Economou N. Bruxism: two case reports. Acta Otorhinolaryngol Ital 2004;24:165–70. Bawa P, Pang MY, Olesen KA, Calancie B. Rotation of motoneurons during prolonged isometric contractions in humans. J Neurophysiol 2006;96:1135–40. Blanksma NG, van Eijden TM. Electromyographic heterogeneity in the human temporalis and masseter muscles during static biting, open/close excursions, and chewing. J Dent Res 1995;74:1318–27. Carr LJ, Harrison LM, Stephens JA. Evidence for bilateral innervation of certain homologous motoneurone pools in man. J Physiol 1994;475:217–27. Christensen LV. Experimental teeth clenching in man. Swed Dent J Suppl 1989;60:1–66. Clark GT, Carter MC. Electromyographic study of human jaw-closing muscle endurance, fatigue and recovery at various isometric force levels. Arch Oral Biol 1985;30:563–9. Clark GT, Beemsterboer PL, Jacobson R. The effect of sustained submaximal clenching on maximum bite force in myofascial pain dysfunction patients. J Oral Rehabil 1984;11:387–91. Clark GT, Jow RW, Lee JJ. Jaw pain and stiffness levels after repeated maximum voluntary clenching. J Dent Res 1989;68:69–71. Dao TT, Lund JP, Lavigne GJ. Comparison of pain and quality of life in bruxers and patients with myofascial pain of the masticatory muscles. J Orofac Pain 1994;8:350–6. Da Silva K, Mandel L. Bilateral temporalis muscle hypertrophy: a case report. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2006;102:e1–3. Dworkin SF, LeResche L. Research diagnostic criteria for temporomandibular disorders: review, criteria, examinations and specifications, critique. J Craniomandib Disord 1992;6:301–55. Eggen S. Simplification and standardization of intraoral radiography technics. Quintessenz 1969;20:109–12.

M. Farella et al. / Clinical Neurophysiology 120 (2009) 190–197 Eriksson PO, Thornell LE. Histochemical and morphological muscle-fibre characteristics of the human masseter, the medial pterygoid and the temporal muscles. Arch Oral Biol 1983;28:781–95. Farella M, Bakke M, Michelotti A, Marotta G, Martina R. Cardiovascular responses in humans to experimental chewing of gums of different consistencies. Arch Oral Biol 1999;44:835–42. Farella M, Bakke M, Michelotti A, Martina R. Effects of prolonged gum chewing on pain and fatigue in human jaw muscles. Eur J Oral Sci 2001;109:81–5. Farella M, Van ET, Baccini M, Michelotti A. Task-related electromyographic spectral changes in the human masseter and temporalis muscles. Eur J Oral Sci 2002;110:8–12. Floystrand F, Kleven E, Oilo G. A novel miniature bite force recorder and its clinical application. Acta Odontol Scand 1982;40:209–14. Glaros AG, Burton E. Parafunctional clenching, pain, and effort in temporomandibular disorders. J Behav Med 2004;27:91–100. Glaros AG, Tabacchi KN, Glass EG. Effect of parafunctional clenching on TMD pain. J Orofac Pain 1998;12:145–52. Glaros AG, Forbes M, Shanker J, Glass EG. Effect of parafunctional clenching on temporomandibular disorder pain and proprioceptive awareness. Cranio 2000;18:198–204. Hellsing G, Lindstrom L. Rotation of synergistic activity during isometric jaw closing muscle contraction in man. Acta Physiol Scand 1983;118:203–7. Holtermann A, Roeleveld K, Karlsson JS. Inhomogeneities in muscle activation reveal motor unit recruitment. J Electromyogr Kinesiol 2005;15:131–7. Huang GJ, LeResche L, Critchlow CW, Martin MD, Drangsholt MT. Risk factors for diagnostic subgroups of painful temporomandibular disorders (TMD). J Dent Res 2002;81:284–8. Jaberzadeh S, Miles TS, Nordstrom MA. Organisation of common inputs to motoneuron pools of human masticatory muscles. Clin Neurophysiol 2006;117:1931–40. Jaberzadeh S, Pearce SL, Miles TS, Turker KS, Nordstrom MA. Intracortical inhibition in the human trigeminal motor system. Clin Neurophysiol 2007;118:1785–93. Jow RW, Clark GT. Endurance and recovery from a sustained isometric contraction in human jaw-elevating muscles. Arch Oral Biol 1989;34:857–62. Kato T, Thie NM, Montplaisir JY, Lavigne GJ. Bruxism and orofacial movements during sleep. Dent Clin North Am 2001;45:657–84. Korfage JA, van Eijden TM. Myosin heavy chain composition in human masticatory muscles by immunohistochemistry and gel electrophoresis. J Histochem Cytochem 2003;51:113–9. Kouzaki M, Shinohara M. The frequency of alternate muscle activity is associated with the attenuation in muscle fatigue. J Appl Physiol 2006;101:715–20. Krogh-Lund C, Jorgensen K. Changes in conduction velocity, median frequency, and root mean square-amplitude of the electromyogram during 25% maximal voluntary contraction of the triceps brachii muscle, to limit of endurance. Eur J Appl Physiol Occup Physiol 1991;63:60–9. Laskin DM. Etiology of the pain-dysfunction syndrome. J Am Dent Assoc 1969;79:147–53. Levenez M, Kotzamanidis C, Carpentier A, Duchateau J. Spinal reflexes and coactivation of ankle muscles during a submaximal fatiguing contraction. J Appl Physiol 2005;99:1182–8.

197

Lindstrom L, Kadefors R, Petersen I. An electromyographic index for localized muscle fatigue. J Appl Physiol 1977;43:750–4. Mandel L, Kaynar A. Masseteric hypertrophy. N Y State Dent J 1994;60:44–7. McLean L, Goudy N. Neuromuscular response to sustained low-level muscle activation: within- and between-synergist substitution in the triceps surae muscles. Eur J Appl Physiol 2004;91:204–16. Moller E, Sheikholeslam A, Lous I. Response of elevator activity during mastication to treatment of functional disorders. Scand J Dent Res 1984;92:64–83. Monro DM, Branch JL. Algorithm AS 117. The chirp discrete Fourier transform of general length. Appl Stat 1976;26:351–61. Moritani T, Muro M, Nagata A. Intramuscular and surface electromyogram changes during muscle fatigue. J Appl Physiol 1986;60:1179–85. Nordstrom MA, Miles TS. Instability of motor unit firing rates during prolonged isometric contractions in human masseter. Brain Res 1991;549:268–74. Psek JA, Cafarelli E. Behavior of coactive muscles during fatigue. J Appl Physiol 1993;74:170–5. Rasmussen OC, Bonde-Petersen F, Christensen LV, Moller E. Blood flow in human mandibular elevators at rest and during controlled biting. Arch Oral Biol 1977;22:539–43. Russel RJ. Significance tables for the results of fast Fourier transforms. Br J Math Stat Psychol 1985;38:116–9. Scott J, Huskisson EC. Graphic representation of pain. Pain 1976;2:175–84. Sheikholeslam A, Moller E, Lous I. Postural and maximal activity in elevators of mandible before and after treatment of functional disorders. Scand J Dent Res 1982;90:37–46. Stephens JA, Harrison LM, Mayston MJ, Carr LJ, Gibbs J. The sharing principle. Prog Brain Res 1999;123:419–26. Svensson P, Arendt-Nielsen L. Effects of 5 days of repeated submaximal clenching on masticatory muscle pain and tenderness: an experimental study. J Orofac Pain 1996;10:330–8. Svensson P, Burgaard A, Schlosser S. Fatigue and pain in human jaw muscles during a sustained, low-intensity clenching task. Arch Oral Biol 2001;46:773–7. Turp JC, Schindler HJ, Pritsch M, Rong Q. Antero-posterior activity changes in the superficial masseter muscle after exposure to experimental pain. Eur J Oral Sci 2002;110:83–91. van Eijden TM, Brugman P, Weijs WA, Oosting J. Coactivation of jaw muscles: recruitment order and level as a function of bite force direction and magnitude. J Biomech 1990;23:475–85. Velly AM, Gornitsky M, Philippe P. Contributing factors to chronic myofascial pain: a case-control study. Pain 2003;104:491–9. Westgaard RH, de Luca CJ. Motor unit substitution in long-duration contractions of the human trapezius muscle. J Neurophysiol 1999;82:501–4. Zennaro D, Laubli T, Krueger H. Motor unit identification in two neighboring recording positions of the human trapezius muscle during prolonged computer work. Eur J Appl Physiol 2003b;89:526–35. Zennaro D, Laubli T, Krebs D, Klipstein A, Krueger H. Continuous, intermitted and sporadic motor unit activity in the trapezius muscle during prolonged computer work. J Electromyogr Kinesiol 2003a;13:113–24.