Muscle Pain Modulates Mastication:An Experimental Study in Humans

Peter Svensson, DDS, PhDAssociate ProfessorCenter for Sensory-Motor InteractionOrofaciai Pain LaboratoryAalborg UniversityAalborg, DenmarkDepartment of Prostbetic Dentistry and

Stomatognathic PhysioiogyUniversity of AarbusAarhus, Denmark

LarsAretidt-Nielsen, PhD. Dr SeiProfessor

Lars Houe. MScEEResearcb AssistantCenter for Sensory-Motor Interaction

Aalborg UniversityAalborg, Denmark

Correspondence to;Dr Peter SvenssonCenter for Sensory-Mo tor InteractionOrofacial Pain LaboratoryAalborg UniversityFredrik Bajersvej 7 D-3DK-9220 Aalborg EDenmarkE-mail: psv@miba.auc.dk

¡n this study, pain was induced in the masseter muscle by tonic in-fusion of hypertonic saline (5%) for up to 800 seconds in 12healthy men. Subjects continuously scored the pain intensity on a10-cm visual analogue scale. Mastication ipsilaterai and contralat-eral to the infusion side was quantitatively assessed with the use ofjaw-tracking and electromyograph recordings of jaw-closing mus-cles before, during, and after periods of constant muscle pain in-tensity. The maximum voluntary occiusai force (MVOF) duringshort static contractions also was monitored. Jaw movements andelectromyographic data were divided into single masticatory cyclesand analyzed on a cycle-by-cycle basis to account for intercyclevariability. In al! subjects, tonic infusion caused a deep localizedpain at a clinically relevant intensity (mean VAS * SE, 4.6 * .3cm). MVOF was significantly affected by muscle pain (P < .0005),with significantly lower MVOF during pain compared to prepainand postpain (P < .05). In a significant number of masticatory cy-cles, the averaged electromyograph activity of all jaw-closing mus-cles during their agonist function was decreased for both ipsilateraiand contralateral painful mastication (P < .05). These electromyo-graphic changes are probably a reflection of the natural bilateralrecruitment pattern of jaw-closing muscles during mastication.Significant changes in jaw movements during painful masticationcould not be detected with the present jaw-tracking device, but fur-ther studies with more accurate and sensitive devices are needed.J OROFACIAL PAIN 1998;12;7-16,

key words; mastication, experimental muscle pain, eleccro-myography, jaw movemenrs, sensorimotor interaction

Studying cause-effect relationships in painful temporomandibu-lar disorders {TMD) is not a trivial task. To explain the clini-cal effect of myogetious TMD pain on jaw motor perfor-

mance, it was recently suggested that activity in nociceptiveafférents, via local circuits in the brain stem, could cause facilita-tion of inhibitory pathways to the agonist motorneuron pool andof excitatory pathways to the antagonist motorneuron,'"^ The con-sequences of such neural circuits would be reduction of the agonistelectromyographic (EMG) burst and an increase of the antagonistEMG burst, which tn turn could cause slower movements withsmaller amplitudes and, presumably, avoid further damage to thesystem- Schwartz and Lund"* presented experimental data from de-cerebrated rabbits to support thts hypothesis; they showed that fic-tive mastication during noxious pressure stimulation of the zygomawas associated with smaller EMG bursts of the jaw-closing musciesand with smaller and slower jaw movements. Svensson et al'

Journal of Orofacial Pain 7

Svensson et al

showed that an acute bolus itijection of hypertonicsaline into the masseter muscle caused similarchanges in the mean profiles of human masticationpatterns. Thus, experimental anitnal and humanpain models tnay he used to gain sotne insight intothe baste effects of muscle pain on motor functioti.However, it has heen argued that changes in motorperformance observed during acute saline-inducedpain can be difficult to interpret because the inten-sity of the pain is not constant.*" Furthermore, tonicinfusion of hypertonic saline may produce musclepain that is more comparable to clinical pain thanacute bolus injections.'' The choice of expertmentalpatn model may therefore be important.

The next probletn is to provide a quantitative de-scription of jaw motor function. Many studies havedescribed masticatory patterns in animals and in hu-matis {for a review, see Eund'). One conclusionfrom the revtewed studtes was that mastication ischaractenzed hy a considerable cycle-to-cycle vari-abiltty. Thts variability probably reflects a continu-ous adjustment of a rhythmic function by sensoryfeedback generated by the movement and thebolus.^ The jaw-closing muscles are activated bilat-erally during mastication, reflecting a substantial de-gree of functtonal symtnetry even though side-to-side differences also are apparent in the naturalmasttcatory patterns.''•^ The variability of humanmasticatioti has not yet been taken into accountwhen the effect of experimental muscle pain hasheen studied.

The atms of the present study were to determinethe effect of constant experimental muscle pain onjaw motor function, performed as masticationboth tpstlateral and contralateral to the infusionside, and to extend the analysis of the masticationpatterns to a cycle-hy-cycle level. The final aimwas to study the influence of muscle pain on iso-metric jaw motor function.

was ohtaitied from all subjects prior to study entry.The study was approved by the local EthicsCommittee. Subjects received $50 for their partici-pation.

Jaw Movement Recordings

A Sirognathograph (Sictnens, Munich, Germany)was used to track jaw movements in a vertical axis{z), a lateral axis (y), and an anteroposterior axis(x).^' Since the output of the Sirognathograph hasheen shown to be distorted,'^-'^ the performance ofthe present system was tested in three dtmensions. Alightweight magnet (I g) was centered (0, 0, 0) in athree-dimensional stereotaxic bench that allowedplacement in 7 (x axis) X 7 ¡y axts) X 17 (z axis)standardized positions. The x, y, and z axes spanned48 mm, 40 mm, and 52 mm, respectively. The out-put signal of the Sirognathograph was recorded foreach of the corresponding 833 postttons. The actualposition of the magnet tn three-dtmenstotial spacewas then related to electrognathographic (EGG) sig-nals, and the distortion could be compensated hy acomputer algorithm. Within a radius of 15 mm fromthe center of the aerial, the uncompensated distor-tion oí the Sirognathograph is on average 5.9%.̂ ^Within the same range of movements and after com-pensation, the distortion irt the present system wasless than 3.5%. The EGG stgnais were recorded withthe subject sitting upright in a chair of nonferromag-netic material with the Frankfort horizontal parallelto the floor.^- With a conventional neck rest sup-porttng the head, subjects were asked to fix theirglance at a point on the wall and to avoid move-ments of the head during mastication. Before eachrecording, the output was zero-adjusted (0, 0, 0) inthe maximum intercuspal position. The F.GG signalswere sampled at 1024 Hz (DT2801-A, DataTranslation, Marlboro, MA).

Materials and Methods

Subjects

Twelve healthy, unmedicated men (mean age 24.4 ±1.0 years [SE]; range 18 to 31 years) were recruitedfor the study from among the students at AalborgUniversity. (Only men responded to the advertise-ment in the university paper.) The absence of TMDin these subjects was verified by screening proce-dures described by the American Academy of Oro-factal Pain.!*̂ All subjects were fully dentate, withthe possible exception of third molars. Informedconsent, according to the Helsinki-II Declaration,

Electromyographic Recordings

Self-fabricated bipolar surface silver electrodes, withan active area of 12 X 6 mm and with thetr longaxes in parallel, were arranged 10 mm apart on apiece of adhesive tape and coated with conductivepaste (DanTENS, Copenhagen, Denmark). The skinwas cleansed with ethanol and the electrodes wereplaced with their long axes rransverse to the main di-rection of the muscle fibers in the central part of theright and left masseter and anrerior temporalis mus-cles. Electrode placemenr was based on palpation ofthe muscles during full effort, as previously describedby MßUer.̂ A saline-soaked ground electrode waswrapped around the neck. The EMG signals were

a Volume 12, Number 1. 1998

Svensson et al

amplified differentially (5,000 to 20,000 times) (Disa15C01, Copenhagen, Denmark), filtered (20 to 500Hz), sampled at 1024 Hz, digitized, and stored.Subjects were asked to make a forced choice of theirpreferred chewing side. The test food consisted ofone piece of chewing gum (1-5 g, Sorbits), which wassoftened for 3 to 4 minutes. Three orthodontic elas-tics {Energy Pak Elastics, 280) were then introducedinto the bolus to give it a constant consistency.

Recordings of Maximum Voluntary Occlusal Force

A U-shaped occlusal force meter (Aalborg University,Aalborg, Denmark), based on a double-strain gaugeprinciple, was used to measure the maximum volun-tary occlusal force (MVOE). To protect the teeth, theocclusal area (1,1 X 1.1 cm) was covered with plas-tic rubes. The total height of the occlusal force meterwith plastic tubes was 12 mm. The particular fea-tures of this apparatus were insensitivity to varia-tions in temperature and placement of loading on theocclusal area. Calibration curves showed a stronglinear relation between standardized loading andreadings up to 1000 N (coefficient of determination,R- = 0.9999), Subjects were instructed to clench theirteeth as hard as they could for 3 to 4 seconds, whileMVOF was measured between the second premolarson both sides. Verbal encouragement was given toobtain the maximum effort, but no visual feedbackwas provided- The MVOF was determined as thepeak value and stored on a display- This was re-peated at least three rimes. About 15 to 30 secondselapsed between repeated measurements- The aver-age of the three peak MVOE measurements for eachside was used for further calculations.

Induction of Jaw Muscle Pain

A plastic catherer inside a hypodermic needle (Ven-flon, 22 g/25 mm, BOC Ohmeda AB, Helsingborg,Sweden) was inserted through the anesthetized skin(0.1 mL lidocaine) into the masseter musde on thepreferred chewing side (denoted ipsilateral), 1 cm an-teriorly and I cm caudally to the EMG electrode.Placement deep in the muscle was secured by bonecontact before the needle was fully retracted and theplastic catheter left in the muscle with the positionsecured by tape on the skin. The catheter was con-nected via a tube (extension set with polyethyleneinner line, Ivac G30303O, San Diego, CA) to a com-puter-controlled syringe pump (Ivac, model 770)with a 10 mL syringe,''' A bolus injection of 0-2 mL5% salme was given over 30 seconds and followedby a steady inhision at a rate of 66 pL/minute, Thisinfusion rate was semiautomatically increased or de-

creased, depending on the subject's scores on a 10-cm electronic visual analogue scale (VAS), The lowerend of the scale was marked "no pam" and theupper end was marked "the worst imaginable pain,"The analogue signal from the VAS box was A/D con-verted and fed to the computer at a sampling fre-quency of 0,2 Hz, One investigator followed theVAS scores on a monitor and responded to changesby either increasing or decreasing the infusion rate ina stepwise procedure, A VAS between 3 and 5 wasdesignated as the target level because it is compara-ble to the pain levels reported by a majority of TMDpatients,'^ Based on pilot experiments in the authors'laboratory, a step was determined to be an increaseor decrease of the infusion rate by a factor of 2, Thechange in infusion rate was initiated if the VASscores in two consecutive samples (10 seconds) indi-cated increased or decreased pain intensity outsidethe target level. The correction of the infusion ratewas repeated 30 seconds later if the VAS scores didnot change. With use of this standard protocol, itwas possible to keep a relatively constant pain inten-sity during recordings of mastication and MVOE.After completion of saline infusion, subjects wereasked to fill out a Danish version of the McGill PainQuestiormaire'̂ and to draw the distribution of thepain on anatomic maps.

Experimental Procedure

After insertion of the catheter, EGG and EMG sig-nals during gum chewing ipsilateral and contra lat-eral to the preferred side were recorded for 60 sec-onds (prepain series). The infusion of 5% hypertonicsaline was then started. When the pain intensity wasat a constant level (VAS from 3 to 5), two series of60-second ipsilateral and contralateral masticationwere recorded in random order followed by mea-surements of MVOE (pain series)- The infusion wasstopped, and sufficient time was allowed for thepain to disappear. When subjects had been pain-freefor at least 10 minutes, recordings of ipsilateral andcontralateral mastication and MVOE were repeated(postpain series). The session lasted about 2 to3 hours.

Data Analysis

In the present study, the analysis of the EGG sig-nals was focused on the maximum displacementsin three dimensions during mastication. These jawmovements normally have a frequency of about 1to 2 Hz, Accordingly, the EGG signals weresmoothed with a low-pass finite impulse responsefilter (filter order = 20; cut-off frequency = 10 Hz)

Journal of Orofacial Pain 9

Svensson et al

Fig 1 Recordings of jaw movements in vertical (z), anreroposterior (x), and lateral (y) di-rections, and EMG activity of ipsilateral and contralateral masseter (ll-mas, cl-mas) and an-terior temporalis ¡il-tem, cl-iem) muscles. Masticatory cycles were divided into fast-opening¡Fo), fast-closing (Fc), and slow-closing (Sc) occlusal phases.

because no attempt was made to study physiologictremor, which has a peak frequency of about 10Hz.'^ The vertical position was differentiated withrespect to time and provided an estimate of thevertical velocity. This was used to determine theonset of the ]aw-opcning phase, defined as thezero-crossing velocity where the velocity changedfrom positive (|aw-closing) to negative (jaw-open-ing). When the mean velocity in a moving 20-sam-pie window was above a defined threshold of 10mm/second, an onset was marked.

Attempts were made to subdivide the closingphase into fast-closing and slow-closing (occlusal)phases similar to the detailed descriptions of ani-mal masticatory patterns provided by Schwartz etal."^ The fast-closing phase in rabbit masticationcorresponds to a swift upward movement, whereasduring the slow-closing phase the mandible eitherstops moving vertically or is closed slowly as thefood is crushed.'^ In the present authors' experi-ence, the distinction between a fast- and a slow-closing phase was less clear in human than in ani-mal mastication, and modified criteria thereforehad to be applied. In the present study, the onsetof the slow-closing phase was defined as the timepoint at which the mean vertical velocity in the

moving 20-sample window was below 10 mm/sec-ond. Thus, the slow-closing phase was essentiallythe onset of an occlusal phase even though themandible may not have entered the maximum in-tercuspal position as a result of the unhteakahleelastics in the chewing gum. The onset of the fast-closing phase was determined as the maximumvertical value (negative) during jaw-opening. Theperiod from one jaw-opening to the next jaw-opening was termed a masticatory cycle (Fig 1).'^The jaw-opening phase corresponds to the antago-nist phase and the jaw-closing phase to the agonistphase of the jaw-closer muscles.

The EMG and EGG signals were divided intosingle masticatory cycles, which were analyzed ona cycle-hy-cycle basis. Since subjects were in-structed to avoid swallowing, ali masticatory cy-cles could be detected and included in the analysis.A total of 49 consecutive and unselected mastica-tory cycles were analyzed for all subjects. Becauseof constraints in the statistical software, no morethan 49 cycles could be included in the repeatedmeasurement analysis. For each masticatory cycle,the following parameters were calculated: maxi-mum displacement in three dimensions (mm)'maximum velocities during jaw-opening and fast-

10 Volume 12. Number 1, 1998

Svensson et al

Pain Intensity

400 600

Seconds

Fig 2a Pain intensity induced by tonic infusion of hy-pertonic saline in 12 subjects (mean ± SE), An electronic0 to 10 cm visual analogue scale was used. The horizon-tal stippled line indicates the mean period with record-ings of mastication.

~- 500 1

~ 300UJ" 200 •ra 100

á 0

y-J"^^L*j:~í^U,Rj^J'Hj;T

0 400 eoo

Seconds

Fig 2b Infusion rate of hypertonic saline measured bythe micrnin fusion pump during the e.xperiment (mean iSE). The horizontal stippled line indicates the mean in-fusiun period.

clositig phases (mm/second); duration of the open-ing, fast-closing, and slow-closing phases (milli-seconds); and the root-mean-square (RMS) ampli-tude (pV) of the rectified EMG activity in theopening and closing (fast + slow) phase (Fig 1).

Statistics

The data were analyzed using tbe Statistica soft-ware (StatSoft 5.1, Tulsa, OK). Mean values + SEare presented in the text and figures. Multivariateanalysis of variance (MANOVA) with repeatedtneasures was used to analyze the effect of the fac-tors: experimental condition (prepain, pain, post-pain), masticatory cycle (49 levels), and muscles(4). The kinetnatic paratneters were analyzed withrespect to experimental condition (3) and mastica-tory cycle (49 levels). The factors in the analysis ofMVOF were experimental condition (3] and side ofbite (2). Student-Neuman-Keuls (SNK) post-hoctests were used to compensate for multiple compar-isons. Significance was accepted at P < .05.

Results

Subjective Description of ExperimentalMasseter Pain

All subjects experienced deep, local pain originatingfrom the site of infusioti in the masseter muscle. Themean VAS-time curve is shown in Fig 2a. The meanpain intensity across subjects was 4.6 + .3 cm duringrecording of ipsilateral mastication, and 3.9 ± ,4 cmduring recording of eontralateral mastication. TheseVAS scores were not significantly different (paired ítest: P > .05). The changes in mean infusion rate dur-ing the course of the experiment are shown in Pig

Fig 2c Distribution of self-perceived pain during tonicinfusion of hypertoiiic saline into the masseter muscie.Tracings of original pain drawings from each subject.

2h. The infusion rate increased from 66 pL/minuteup to about 200 pUminute, with a total infusionvolume of 2.3 i .2 mL saline (Fig 2b). The local painwas associated with a spread of pain to adjacentareas: the temple, the TMJ area, the maxillary molarteeth, and the hasis of the mandible (Eig 2c).

Tlie McGill Pain Questionnaire (MPQ) meanpain rating indices of sensory, affective, evaluative,and miscellaneous dimensions of pain were 10.3 ±1.3, 2.1 ± .7, 2,5 ± .3, and 5.3 ± .9, respectively.Among the words chosen from the MPQ by at least40% of the suhjccts were "taut" (67%), "drilling"(42%), "intense" (42%), and "penetrating" (42%).

Motor Consequences of ExperimentalMasseter Pain

According to the EMG data, there were no signifi-cant tnain effects of experitnental condition, tnusclerecording site, or masticatory cycles. However, the

Journal of Orofaciai Pain 11

Suensson et ai

Table 1 Kinematic Parameters Before, During, and After Induction of Pain in the Masseter Muscle"

Displacement (mm)Verticai (z)La te rial fy)Anterior-posterior (x)

Maximal velocity (mm/s)OpeningFast-ciosing

Du ration (ms)OpeningFast-ciosingSiow-closing

Before

16.4 ± 1.14.6 ± 0.52.1 ±0.2

93 ± 1299± 10

361 ± 24349 ± 17124 ± 12

Ipsiliiteriil rtiasticarion

During

16,4 ± 0.94,3 ± 0,42,4 ± 0,3

92± 1296 ± g

369 ± 22357 ± 2197 ± 10

After

17.4 ±0.94.5 ± 0.42 .2±02

106 I 13107 ± 11

335 ± 21340 ± 19106 ± 10

P

.076410195

.065

.159

.175

.470,053

Before

16.5 t 1.14,8 ±0.52,3 ±0.2

91 ± 1098 ± 10

361 ±24340 ± 19145 ± 19

Contralateral

During

16.0 ± 0 94.5 t 0,43,5 ± 0.2

93 ± 1196 t 9

342 ± 19334 ± 131 20 ± 1 2

masriMtion

After

1 7 34.92 3

112107

322332129

± 0 7±0,5±0.2

± 13± 10

± 2 3± 24± 17

P

.166,440,431

.039

.099

.055

.825

.364

values ± SE oí 49 masticatory cycies from 12 subjects are presented Pi, i from thie Ar»JOVA indicate the effect of eiperimentai condition

MANOVA analysis of the EMG activity of the jaw-closing muscles during their agonist phase showedhighly significant interactions between experimen-tal condition and masticatory cycles for both ipsi-laterai and contralateral masttcatton (F[96,6336] =2,12, P < ,0001; F[96,é336] = 3.08, P < ,0001). Foripsilaterai painful mastication, a significant reduc-tion of jaw-closing EMG activity in the agonistphase was detected by post-hoc tests in most of themasticatory cycles when compared to prepain andpostpain masticatory cycles (SNK: P < .05) (Eig3a), A stTialler number of masticatory cycles wtthsignificantly reduced EMG activtty was found forcontralateral painíul mastication (Eig 3b|,

There were no significant main effects of experi-mental condition on the various kinematic parame-ters except for the maximum opening velocitj' durttigcontralateral mastication (Table 1|. Eor this parame-ter, there was a significant interaction berween exper-imental condition and masticatory cycles (E[96,3072J= 1.71, P < .0001), but post-hoc tests could not iden-tify the masticatory cycles with signtficant changes.

The maximum occiusai force was significantly af-fected by e.xperimental condition (E[2,22] = 11,10,P<.0O0,S) butnot by side of clench (E[ 1,111 = 1.51,P = .24), The MVOE was stgnificatitly lower duringpainful clenches (597 ± 19 N) as compared to pre-patn (618 ± 19 N; SNK: P < .05) and postpainclenches (650 + 21 N; SNK: P < .05),

Discussion

Subjective Description of ExperimentalMasseter Pain

The subjective description of jaw muscle pain in-dticed by tonic infusion of hypertonic saline in

healthy volunteers was quite similar to the descrip-tion obtaitied from chrontc myogenous TMD pa-tients.^ Erequently chosen words from the MPQwere the evaluative adjective "intense" and the sen-sory-miscellaneous adjectives "taut/tender." TheDanish version of the MPQ has been validated,"•and it supported the coticept of expertmental mus-cle pain as a multidimensional experience with sen-sory-discriminative, cognitive-evalttative, and affec-tive components.'^ It may be possible to separateout an esttmate of patn tntenstty and an esttmate ofunpleasantness-''; however, rattngs of these two di-mensions of pain are clearly interrelated becatiseboth ratings increase with increasing stimulus in-tensity.'°--- In the present study, no attempts weremade to dtsttnguish between pain intensity and un-pleasantness; subjects simply used a VAS labeledwith "worst pain imaginable." It has also beenshown that the word descrtptor "worst pain imag-inable" on a VAS is the most suitable choice formeasuring dental pain.'^

The present infusion paradigm allowed periodswith relatively stable pain ratings (Fig 2a), althoughmore sophtsticated tnfusioti paradtgms wtth feed-back-control have heen developed and should bepreferred when possible.-'̂ A major advantage ofthe tontc tnfusioti of hypertonic saline is that sub-jects avoid a very rapid change in pain intensity,and the induced pain mimics clinical pain morethan after acute bolus tn)ecttotis of hypertonicsaline.-^ Another similarity between experimentalmuscle pain atid chronic TMD is the spread of painto adjacent areas. Clinical and experimental pain inlimb muscles is usually referred to areas distal tothe region wirh local pain.•'''•̂ -̂-̂ Pain in the mas-seter muscle is usually spread to posterior and cra-nial regions but rarely in an anterior direc-tion.^•^'^^•'' The stereotyped patterns of referred

12 Voiume 12, Number 1, 1998

Svensson et al

m

Œc

Me.

120 I

110 .

100

90

80

70

ni/ ^

0

A.

10

l/v++

Ipsiiateial masticatia

H- ++

20 30

Masticalcry cycles

R

•vA+

40

—1

—1

LJ

3 —

» —

D—

kD

Beiore

During

Atler

150

Fig 3a Effect of tonic infusion of hypertonic saline on the mean EMG activity from themasseter and anterior temporalis muscles in the ¡avv-closing phase dnrjng ipsiiateral masti-cation. Recordings of 49 consecutive masticatory cycles before, during, and after infusion in12 subjects. + indicares significant difference betviieen the painful condition and the beforeand after condition (SNK: F < .05).

Fig 3b Effect of tonic infusion nf hypertonic saline on the mean EMG activity from themasseter and anterior temporalis muscles ¡n the jaw-closing phase during contralateral mas-tication. Recordings of 49 consecutive masticatory cycles before, during, and after infusionin 12 subjects. + indicates significant difference between the painful condition and the be-fore and after condition (SNK: ? < .05).

Journal of Orofacial Pain 13

Svensson et al

pain or spreading of pain from muscles may reflecta distinct topographic organization of centra! neu-rons and may he used diagnostically to locate thesource of pain. The spread or referral of trigeminalpain is most likely explained by central conver-gence of nociceptive afférents onto common wide-dynamic-range (WDR) neurons in rhe subnucleuscauda I is.-"^^^ Extensive convergence has also beenobserved onto spinal cord WDR neurons" and hasbeen suggested to cause a mislocalization of periph-eral noxious srimuli by higher central centers.^''Neuroplastic changes in the central nervous systemwith formation of new connections between adja-cent neurons may also play a critical role for the re-ferral of pain,̂ "*

Motor Consequences of ExperimentalMasseter Pain

The overall effect of experimental jaw muscle paininduced by acute bolus infusions is a reduction ofagonist EMG activity.^ In the present study, therewas no statistical main effect of the experimentalcondition, but there were significant interactionsbetween experimental condition and masticatorycycle, and hence a significant reduction of agonistEMG activity was seen in a majority of masticatorycycles during tonic infusion of hypertonic saline.Stohier et ai'*" described masticatory cycles in TMDpatients and found only painful cycles to be associ-ated with higher EMG activity in the antagonistphase. This shows that human mastication is linkedto a considerable cycle-to-cycle variability and tbatnot all masticatory cycles may be changed duringpainful mastication. Thus, analysis of multiple mas-ticatory cycles seems to be necessar}' to describe theeffect of muscle pain on mastication and otherrhythmic functions.

Symmetric and empty open-close jaw movementsm humans have been reported to be slower andwith smaller amplitude during experimental pain inthe masseter muscle,' Our cycle-by-cycle analysisfailed to reveal a significant effect of jaw musclepain on the maximum amplitudes of movementsduring a more natural motor task, ie, mastication(Table 1). The magnitude of reduction in decere-brated rabbits is in the range of 3 to 7%''; takingthe variability of human masticatory cycles and thelimitations of the present recording system into ac-count, this may partly explain the lack of statisticalsignificance for kinematic parameters in the presentstudy. It is likely that chewing on a larger boluswith larger movement amplitudes (> 20 mm) and amore intense pain would have caused significant ef-fects- We observed a trend of faster and larger jaw

movements in the postpain conditions (Table l)iwhich could be related to adaptive changes in mas-tication. In future studies, jaw-tracking devicesmore accurate and sensitive'̂ '̂ *" than the binignatho-graph should be used to detect kinematic changesduring painful mastication.

Unbreakable elastics were incorporated into thechewing gum to ensure a constant consistency ofthe test bolus. As a consequence, the mandiblemay not have entered the maximum intercuspalposition, but since this is true for all the condi-tions, it is unlikely that it would have affected theoutcome of the present study,

Methodologie Considerations

The present study employed some methods thatdeserve specific comments, Eirst, only men volun-teered for the study. While this enabled a homoge-nous group, it may be controversial in light of re-ported gender differences in masticatorypatterns'^•^''•'"' and the female dominance amongTMD patients,'" However, women have lowerpain thresholds to various stimuli,*^ which couldsuggest that tonic infusion of hypertonic salinewould have been perceived as more painful inwomen than in men. Therefore, changes in jawmotor function during painful mastication couldhave been larger in a female popularion, and thepresent findings in men may at worst represent anunderestimation of the changes. Moreover, the de-sign of the study allowed a paired comparison,which minimizes interindividual and gender differ-ences. Second, a control condition was not estab-lished in terms of infusion with isotonic saline, buta recent study did not show any significant effectsof isotonic saline injections on masticatory pat-terns, suggesting little or no influence of infusedvolume.''^ Eurthermore, rhe observed changes inthe masticatory pattern during experimental paincan be interpreted as in accordance with findingsin TMD patients because they demonstrate loweragonist EMG activity and higher antagonist EMGactivity during mastication,'^'^^''*'' In the presentstudy, the mastication and MVOF recordings werefirst started when the pain had reached a constantlevel. Thus, acute psychologic effects like anxietyand nervousness are likely to have been minimizedat this time- However, all types of pain, includingexperimental pain, encompass psychologic as wellas sensory-discriminative aspects, and it is there-fore difficult in conscious humans to disregard theinfluence of higher brain centers on jaw motorfunction during pain. Nevertheless, the observedpain modulation of human mastication is in many

14 Volume 12, Number 1, 1998

SvensËon et al

ways comparable to the modulation of rhythmicmovements in decerebtated animals,^ indicatingthe involvement of brain-stem circuits in the regu-lation of motor function during pain. '

Within the constraints of the present study, exper-imental jaw-muscle pain induced by tonic infiisionof Kypertotiic saline caused a diminished capacity ofthe jaw-closing muscles to work against a load,which is in accordance with a functional adaptationto muscle pam. The biologic purpose of such adap-tation may be to allow healing of an injured area.'

Acknowledgments

The study was supported by the Danish Dental Associarion'sFUT and Caldii Foundation, Colgiite-Palmolive, Foght's Fond,and the Danish National Research Foundation.

References

1. Lund IP, Donga R, Widmer CG, Stohler CS. The pain-adaptarion model; A discussion of the relationship be-tween chronic musculoskeletal pain and tnotor activity.Can J Physiol Phartnacol 1991;69:683-694.

2. Lund JP, Stohler CS, Widmer CG. The relationship be-tiveen pain and muscle activity in fibromyalgia and simi-lar conditions, in: Vieray H, Merskey H (eds). Progress inFihromyalgia and Myofascial Pain. Amsteidami Elsevier.1993í311-327.

3. Lund JP, Sessle BJ. Neurophysiological mechanisms. In:Zarb GA, Carlsson GE, Sessle BJ, Mohl ND (eds). Tem-poromandibular Joint and Masticatory Muscle Disorders.Copenhagen: Munksgaard, 1994:188-207.

4. Schwartz G, Lund JP. Modification uf rhythmical jawmovements by noxious pressure applied to the periosteumof the zygoma in decerebrate rabhits. Pain 1995;63t153-161.

5. Svensson P, Arendr-Nielsen L, Houe L. Sensory-motor in-teractions of human experimental unilateral jaw musclepain: A quantitative analysis. Pain 19%;64:241-249.

6. Stohler CS, Lund JP. Psycho physical and orofacial motorresponses to muscle pain validation and utility of an ex-perimental model. In: Morimoto T, Matsuya T, TakadaK leds). Brain and Oral Function. Amsterdam: Elsevier,1995:227-237.

7. Lund JP. Mastication and its conrroi hy the brain stem.Crit Rev Oral Biol Med 1 9 9 1 Í 2 : 3 3 - 6 4 .

S. Mailer E. The chewing apparatus. An electro myographicstudy of the action of the muscles of mastication and itscorrelation to facial morphology. Acta Physiol Scand1966;69|suppl2S0¡.

9. Kumai T. Differences in chewing patterns between in-voived and opposite sides in patients with unilateral tem-poromandibular joint and myofascial pain dysfunction.Arch Oral Biol 1993,38:467-478.

10. Okeson JP ¡ed). Orofaciai Pain: Guidelines for Assess-ment , Diagnosis , and Management . Chicago:Quintessence, 1996.

11. Lewin A. Electrognathographics: Atlas of Diagnostic Pro-cedures and Interpretation. Chicago: Quintessence, 1985.

Michler L, Bakke M, Mailer E. Graphic assessment ofnatural mandibdar movements. J Craniomandih DisordFacial Oral Pain 1987;1:97-] 14.Kazazoglu E, Heath MR, Ferman AM, Davis GR. Re-cording mandibular movements: Technical and clinicallimitations ofthe Sirognatograph. J Orotacial Pain1994;8:165-177.Graven-Ni el sen T, Arendr-Nielsen L, Svensson P, JensenTS. Experimental muscle pain: A quantitative study oflocal and referred pain in humans following injection ofhyperronic saline. J Musculoskel Pain 1997:5;49-69.Dao I I I , Lund JP, Lavigne GJ. Comparison of pain andquality of life in bruxers and patients with myofascialpain of masticatory muscles. J Orofacial Pain1994;8í350-3Sé.Drewes AM, Helweg-Larsen S, Petersen P, Brennum J,Andreassen A, Poulsen LH, et al. McGill Pain Question-naire translated into Danish: Experimental and clinicalfindings. Clin J Pain ]993;9:80-87.Wilding RJC, Shaikh M. Jaw movement tremor as a pre-dictor of chewing performance. J Orofacial Pain 1997;11:101-U4.Schwartï G, Enomoto S, Valiquette G, Lund JP. Masti-cation in the rabbit: A description of movement and mus-cle activity. J Neurophysiol 1989;62:273-287.Sessle BJ. The neurobiology of facial and dental pain;Present knowledge, future directions, J Dent Res 1987;66;962-981.Gracely RH, McGrath P, Dubner R. Ratio scales of sen-sory and affective verbal pain descriptors. Pain1978;5:5-18.Rainville P, Feine JS, Bushneil MC, Duncan GH. A psy-chophysical comparison of sensory and affective re-sponses to four modalities of experimental pain.Somatosens Mot Res 1992;9:2é5-277,Svensson P, Beydoun A, Morrow TJ, Casey KL. Humanintramuscular and cutaneous pain: Psychophysical com-parisons. Exp Brain Res 1997^114:390-392.Seymour RA, Simpson JM, Chariton JE, Phillips ME. Anevaluation of length and end-phrase of visual analoguescales in dental pain. Pain 1985;21:177-185.Zhang X, Ashton-Miller JA, Stohler CS. A closed-loopsysrem for mainraining constant experimental musclepam m man. IEEE Trans Biomed Eng 1993;40:344-352.Stohler CS, Lund JP. Effects of noxious stimulation of the¡aw muscles on the sensory experience of volunteer humansubjects. In: Stohler CS, Carlson DS (eds). Biological andPsychological Aspecrs of Orofaciai Pain. CraniofacialGrowrh Series 29. Ann Arbor: Center for Human Growthand Development, University of Michigan, 1994: 55-73.Travell JG, Simons DC. Myofascial Pain andDysfunction. The Trigger Point Manual. Baltimore:"Williams S: Wilkins, 19S3.Hockaday JM, Whitty CWM. Patterns of referred pain inthe normal subject. Brain 1967;90:48X-196.Kellgren JH. Observations on referred pain arising frommuscle. Clin Sei 193S;3:17S-190.Svensson P, Arendt-Nielsen L, Nielsen H, Larsen JK.Effect of chronic and experimental jaw muscle pain onpain-pressure thresholds and stimulus-response curves. JOrofacial Pain X995;9:347-356.Amano N, Hu JW, Sessle BJ. Responses of neurons in fe-line trigeminal subnucleus caudalis (medullary dorsalhorn) ro cutaneous, mtraoral, and muscle afferent stimuli.J Neurophysiol 198É;55:227-243.

Journal of Orofaciai Pain 15

Svensson et al

31. Sessle BJ, Hu JW, Amaino N, Zliong G, Convergence ofcutaneous, tooth pulp, visceral, neck and muscle afferenrsonto nociceptive and non-nociceptive neurones in trigemi-nal subnucleu.i caudalis (medullary dorsal horn) and itsimplications for referred pain. Pain 1986;27;219-235.

32. Sessle BJ, Hu JW, Yu XM. Brainstem mcch;inisnis uf re-ferred pain and liyperalgesia in the orofacial and remporo-mandibular region. In: Vecchict L, Alhe-Fcssard D,Lindblom U (eds). New Trends in Referred Pain andHyperalgesia. Am.stcrdam; Elsevier, 1993:59-71.

33. Foreman RD, Sclimidi RP, Willis WD. Convergence ofmuscle and cutaneous input onto primare spinochalaniictract neurons. Brain Res I977:I24;555-56O.

34. Mense S. Nociception from skeletal muscle in relation toclinical muscle pain. Pain 1993;54:24]-289.

35. Ramfiord SP, Ash MM, Occlusion. Philadelphia: WBSaunders, 1983,

36. Stohler CS, Ashton-Miller JA, Carlson DS, The effects ofpain from the mandibular joint and muscles on mastica-tory motor behavior in man, Archs Oral Biol1988;33:175-182.

37. Zhang X, Ashton-Miller JA, Stohler CS, Three-dimen-sional unilateral method for the bilateral measurement ofcondyiar movements. ] Biomech I995;28:1007-I011.

38. Galb LM, Airoldi GB, Airoldi RL, Palla S, Description ofmandibular finite helical axis pathways in asymtomaticsubiects, J Dent Res 1997;76:704-713

39. Wilding RJC, Lewin A, The determination oí optimalhuman iaw movements based on their association withchewing performance. Arch Oral Biol 1994;38:5!i9-59é.

40. Visser SL, De Rukc W. Influence of sex and age on EMGcontraction patterns. Eur Neurol iy74;l 2:229-235.

4L Carlsson GE, LeResche L, Epidemiology of temporo-mandibular disorders. In: Sessle BJ, Bryant PS, Dionne RA(eds), Seattle: IASP, 1995: 211-226.

42. Lauterbacher S, Rollman GB, Sex differences in respon-siveness to painful and non-painful stimuli are dependentupon the stimulation method. Pain I993;53:255-264.

43. Svensson P, Houe L, Arendt-Nieken. Bilateral experimen-tal muscle pain changes electromyographic activity ofhuman jaw-closing muscles during mastication, Exp BrainRes 1997;116d 82-185.

44. Nielsen IL, McNeil! C, Danzig W, Goldman S, Levy J,Miller AJ. Adaptation of craniofacial muscles m suhjectswith craniomandibuiar disorders. Am J Orthod DentofaeOrthop 1990;97;2U-34.

Resumen

El dolor muscular modula la masticación: Estudio experi-mental en humanos.

Zusammenfassung

Muskelsohmer? steuert das Kauen: eine experimentelleStudie am Menschen

En este estudio, el doior fue inducido en ei músouio maseteropor medio de ia infusión tónica de una solución saiina hipertónica(5%) hasta de 800 en 12 hombres sanos Los participantes reg-istraron puntuaciones de ia intensidad del doior continuamentesobre una escala análoga visual (EAV) de 10 cm. La masticaciónípsoiateral y contralateral en el iado de ia infusión fue evaiuadacuantitatibiamente por medio dei uso de registros eiectromiográ-fieos y de rastreo mandibular de los múscuios de cierre mandibu-lar antes, durante, y después de ios periodos, con una intensi-dad de doior muscular constante. La fuerza ociusal voluntariamáxima ÍFOVM) durante las contracciones estáticas cortas tam-bién fue monitoreada. La información de ios movimientosmandibuiares y ios registnDS eiectromiográficos fue dividida enciclos masticatorios senciiios y analizada de ciclo a cicio paratener en cuenta ia vanabilidad entre los ciólos. En todos los par-ticipantes, la infusión tónica causó dolor iocali?ado profundo, conuna intensidad reievante ciinicamente (EAV media, ± desviaciónestándar, 4,6 ± 0,3 cm). La FOVM fue afectada significativa-mente por ei doior muscular tP •; 0,0005) siendo la FOVM signi-ficativamente menor durante ei dolor en comparación con antesy después dei dolor (P < 0,05), En un número significativo de cic-los masticatorios, la actividad electromiográfica promedio detodos los múscuios de cierre mandibuiar durante su Funcióncomo agonistas disminuyó durante ia masticación dotorosa,tanto ipsoiatersi como oontraiaterai (P < 0,05). Estos cambioseleotnjmiográflcos son probablemente una refiexion del patrónde reclutamiento bilateral natural de ios músculos de cierremandibular durante ia masticación. No se pudieron detectar cam-bios significativos en cuanto a bs movimientos maridibuiares du-rante la masticación dolorosa, con el dispositivo de rastreomandibular actual, pero son necesarios más estudios con dis-positivos más precisos y sensitivos.

In dieser Studie wurde ein Scbmerz im Musculus Masseter durcheine tonische Infusion von hypertonischer Sabiösung (5%) für800ms bei 12 gesunden Männern ausgelost Die Personen bevj-erteten iaufend die Schmerzintensitât auf einer 10cm visuellenAnalogskala. Das Kauen ipsilateral und kontralateral zurInfusion s Seite wurde quantitativ beurteiit mittels Aufzeichnen derUnterkieferbewegungen und der Eiektromyographie dersohiiessenden Muskeln vor, wahrend und nach Zeitabschnittenmit konstanter Wuskeischmerzintensität. Die maximale wiiikür-liche okklusale Kraft IMVOF) während kurzen statischenKontraktionen wurde ebenfalls überwacht. Kieferbewegungen undeiektromyographische Daten wurden in einzelne Kauzyklanaufgeteilt und aufgrund einer Zyklus-auf-Zyklus-Basis analysiert,um eine Veränderiicbkeit zwischen den Zyklen zu erklären. Beialien Personen verursachte die tonische infusion einen tiefen,iokaiisierten Schmerz in einer kiinisch relevanten Starke (durch-schnittliche VAS ± SE, 4.6 ± .3 cm). Die MVOF war signifikantbeeinfiusst durch den Muskelschmerz iP< ,0005), mit signifikantniedrigerer MVOF wahrend dem Schmerz verglichen mit denAbschnitten vor und nach dem Schmerz (P < .05). in einer sig-nifikanten Anzahi von Kauzyklen war die d u roh schnitt i iche eiek-tromyographische Aktivität alier sohiiessenden Muskeln wahrendderen agonisüschen Funktion erniedrigt, sowohi für das ipsiiat-erale als auch das ko n traíate raie schmerzhafte Kauen (P < .05).Diese eiektromyograophischen Veränderungen sind wahrschein-lich ein Spiegelbiid der natürlichen bilateralen Erholungsmusterder kieferschiiessenden Mtiskein wahrend des KauensSignifikante Veränderungen in den Kieferbewegungen währenddes schmerzhaften Kauens konnten mit der aktuellenKieferbewegungsvorrichtung nicht entdeckt werden, aber es sindweitere Studien mit genaueren und empfmdiicherenVorrichtungen notwendig.

16 Voiume 12, Number 1, 1998