Properties of temporomandibular joint nociceptors in normal and inflamed tissue

MISSING IMAGE

Material Information

Title:
Properties of temporomandibular joint nociceptors in normal and inflamed tissue
Physical Description:
xi, 130 leaves : ill. ; 29 cm.
Language:
English
Creator:
Loughner, Barry Allen, 1943-
Publication Date:

Subjects

Subjects / Keywords:
Nociceptors -- physiology   ( mesh )
Temporomandibular Joint -- physiology   ( mesh )
Department of Oral Biology thesis Ph.D   ( mesh )
Dissertations, Academic -- College of Dentistry -- Department of Oral Biology -- UF   ( mesh )
Genre:
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 1992.
Bibliography:
Includes bibliographical references (leaves 118-129).
Additional Physical Form:
Also available online.
Statement of Responsibility:
by Barry Loughner.
General Note:
Typescript.
General Note:
Vita.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 026897059
oclc - 27245086
Classification:
lcc - RK470 .L68 1992
System ID:
AA00025765:00001

Full Text










PROPERTIES OF TEMPOROMANDIBULAR JOINT NOCICEPTORS
IN NORMAL AND INFLAMED TISSUE













BY


BARRY LOUGHNER


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY


UNIVERSITY OF FLORIDA


1992




























I dedicate this work to my daughter, Carina Beth.














ACKNOWLEDGEMENTS


I thank Dr. Brian Cooper, Ph.D., for his technical guidance in the research

involved in this dissertation. I thank Ms. Martha Oberdorfer for her technical assistance. I

thank Dr. Parker Mahan, D.D.S., Ph.D., and Dr. Arnold S. Bleiweis, Ph.D. for their

strategic guidance. I thank Dr. Robert Bates, D.D.S., M.S. for his administrative

assistance.













TABLE OF CONTENTS


LIST OF TABLES ...........................................................................

LIST OF FIGURES .........................................................................

KEY TO SYMBOLS ........................................................................

ABSTRACT ...................................................................................

CHAPTERS

1 INTRODUCTION .............................................................

2 METHODS ......................................................................
Subjects ......................................................................
Exposure of the TMJ and Trigeminal Ganglion .......................
Recording Procedures .....................................................
Characterization of TMJ Afferents .......................................
Statistics ................................................................... *
Experimental Inflammation ................................................

3 RESULTS ......................................................................
Identification of TMJ Nociceptors ......................................
Characterization of TMJ Nociceptors in Normal Tissue ..............
Properties of TMJ Nociceptors in Inflamed Tissue .....................
Experiment I: Properties of units in the Previously Inflamed TMJ...
Experiment II: Properties of Units in the Acutely Inflamed TMJ .....

4 DISCUSSION ..................................................................
Sensitization of TMJ Nociceptors ........................................
Factors Contributing to Sensitization .....................................

APPENDICES

A DEVELOPMENT AND MAINTENANCE OF TMJ PAIN ..............

B PSYCHOPHYSICS OF TMJ DISORDERS ................................

C NEUROANATOMY OF THE TMJ .........................................

D CENTRAL REPRESENTATION OF TMJ AFFERENTS ................

E TRIGEMINAL GANGLION ..................................................

F GROSS ANATOMY ...........................................................








G SUMMARY TABLES OF TMJ REACTIVITY .................. 90

LIST OF REFERENCES ......................................................... 118

BIOGRAPHICAL SKETCH .................................................... 130














LIST OF TABLES


Table 3-1

Table 3-2


Table 3-3

Table 3-4



Table 3-5

Table G-l 1

Table G-2

Table G-3

Table G-4


Table G-5

Table G-6

Table G-7

Table G-8

Table G-9

Table G-10


Table G-1 I


Table G-12


Table G-13


TMJ force/movement relationships in the vertical plane......................

Properties on nociceptors that demonstrate vertical plane or horizontal
plane reactivity ..................................................................

Mean values of properties of TMJ nociceptors ...............................

Qualitative improvements in reactivity for nociceptors that were
characterized in normal tissue and then tested again subsequent to
carrageenan injection ...........................................................

Changes in reactivity for 8 units after exposure to saline ..................

Reactivity in vertical plane (normal tissue) ..................................

Reactivity in horizontal plane (normal tissue), left lateral ................

Reactivity in horizontal plane (normal tissue), right lateral ................

Reactivity in the vertical plane in previously inflamed tissue. Reactivity
in horizontal plane in previously inflamed tissue, left lateral ..............

TMJ nociceptor reactivity in normal tissue, vertical plane ................

TMJ nociceptor reactivity in normal tissue, horizontal plane ..............

TMJ nociceptor reactivity in previously inflamed tissue ...................

TMJ nociceptor reactivity in acutely inflamed tissue .......................

TMJ nociceptor reactivity in saline injected tissue .........................

Effect of canrrageenan in acutely inflamed tissue reactivity in the vertical
plane, unit #1 ....................................................................

Effect of carrageenan in acutely inflamed tissue reactivity in the vertical
plane, unit #2 ....................................................................

Effect of carrageenan in acutely inflamed tissue reactivity in the vertical
plane, unit #3 ....................................................................

Effect of carrageenan in acutely inflamed tissue reactivity in the vertical
plane, unit #4 ....................................................................


25


28

37



46

51

91

92

93


94

95

96

97

98

99


100


101


102


103








Table G-14


Table G-15


Table G-16


Table G-17


Table G-18


Table G-19


Table G-20


Table G-21

Table G-22

Table G-23

Table G-24

Table G-25

Table G-26

Table G-27


Effect of carrageenan in acutely inflamed tissue reactivity in the horizontal
plane, unit #1 .................................................................... 104

Effect of carrageenan in acutely inflamed tissue reactivity in the horizontal
plane, unit #2 ................................................................... 105

Effect of carrageenan in acutely inflamed tissue reactivity in the horizontal
plane, unit #3 ................................................................... 106

Effect of carrageenan in acutely inflamed tissue reactivity in the horizontal
plane, unit #4 .................................................................... 107

Effect of carrageenan in acutely inflamed tissue reactivity in the horizontal
plane, unit #5 .................................................................... 108

Effect of can-rrageenan in acutely inflamed tissue reactivity in the horizontal
plane, unit #6 .................................................................... 109
Effect of carrageenan in acutely inflamed tissue reactivity in the horizontal
plane, unit #7 .................................................................... 110

Effect of saline reactivity in the horizontal plane, unit #1 ................... I11

Effect of saline reactivity in the horizontal plane, unit #2 ................... 112

Effect of saline reactivity in the horizontal plane, unit #3 ................... 113

Effect of saline reactivity in the horizontal plane, unit #4 ................... 114

Effect of saline reactivity in the horizontal plane, unit #6 ................... 115

Effect of saline reactivity in the horizontal plane, unit #7 ................... 116

Effect of saline reactivity in the horizontal plane, unit #8 ................... 117








LIST OF FIGURES

Figure 2-1 Mandibular movements were produced by a hand held probe ............ 8

Figure 2-2 Dynamic and static test series for force, displacement and velocity ...... 10

Figure 2-3 Demonstration of sensitization as a shift of stimulus-response functions.. 14

Figure 3-1 Receptive field distribution in the lateral and posterior capsule of the
temporomandibular joint for normal tissue .................................. 18

Figure 3-2 Receptive field distribution in the lateral and posterior capsule of
the temporomandibular joint for previously inflamed tissue, acutely
inflamed tissue and saline control tissue ...................................... 20

Figure 3-3 Relationships between applied force and vertical plane movements of
the m andible ................................................................ 23-24

Figure 3-4 Dynamic reactivity in normal tissue .......................................... 30

Figure 3-5 Dynamic reactivity in previously inflamed tissue ......................... 33

Figure 3-6 Averaged dynamic response functions of nociceptors fitted in normal
and previously inflamed tissue ................................................ 41

Figure 3-8 Reactivity of nociceptors exposed to saline ................ ................ 43

Figure 3-9 Comparison between pre-inflamed and post-inflamed reactivity in
scatter plot forms ......................................................... 48-49

Figure F-l Schematic diagram of the human temporomandibular joint. Sagittal and
frontal views .................................................................... 87














KEY TO SYMBOLS

* Vertical plane or previously inflamed tisue.

A Horizontal plane or acutely inflamed tissue.

+ Saline control tissue.














Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

PROPERTIES OF TEMPOROMANDIBULAR JOINT NOCICEPTORS
IN NORMAL AND INFLAMED TISSUE

By

Barry Loughner

May, 1992



Chairman: Parker Mahan, D.D.S., Ph.D.
Cochairman: Brian Cooper, Ph.D.
Major Department: Oral Biology


Mechanical response properties of nociceptors of the goat temporomandibular joint

(TMJ) were characterized by unit recordings from the trigeminal ganglion. Populations of

afferents were sampled in normal tissue and in tissue previously inflamed (PI) with

carrageenan for at least 6 hours prior to testing. In other experiments, nociceptors were

characterized in normal tissue and then retested for up to 3 hrs subsequent to exposure to

carrageenan (AI). To produce mandibular movement in the vertical and horizontal planes,

dynamic and static forces were applied to the mandibular pole. Stimuli were quantified by

force and angular displacement transducers. Receptive fields were small, single spots (2-3

mm) and located on the lateral or posterior capsule of the TMJ. Conduction velocities were

shown to be in the group III and IV range. Stimulus-interval data was best described by

power functions. Assessment of force-movement relationships indicated that capsular

nociceptors responded exclusively to intense forces that produce extreme displacements of

the mandible.









In normal tissue, 24 of 36 nociceptors transduced dynamically applied stimuli

(mean activation threshold for dynamic force = 15.9 + 11.3 N and for force velocity =

17.3 + 18.9 N/s). Relatively fewer units transduced static force (8/36) or position

(1/36).

In PI conditions, tests revealed enhanced reactivity for dynamically applied

stimuli. Mean power functions for dynamic tests in PI conditions (LnISI = -1.3 LnF +

7.3) had steeper slopes compared to those in normal tissue (LnISI = -1.1 LnF + 5.4)

and were shifted graphically to regions indicating greater reactivity (below and to the

left of those in normal tissue). These shifts were suggestive of sensitization.

In AI conditions, 8 of 11 nociceptors acquired either dynamic or static coding

ability. Acquired coding ability also appeared in nociceptors (n = 2) that were without

transducing capacity in normal tissue. Acquired reactivity for vertical plane movement

was also seen for those nociceptors (n = 3) that were reactive only to movement in the

horizontal plane in normal tissue. Acquisition of coding ability following carrageenan-

induced inflammation suggested afferent sensitization.













CHAPTER 1


INTRODUCTION



The temporomandibular joint (TMJ) is the site of a number of disorders that burden

the health care system and extract a large cost in human suffering (Helkimo, 1979; Solberg,

1987). Afferent groups presumed to mediate TMJ pain are likely to play a role in a variety

of TMJ disorders. Many theories have been proposed to explain TMJ pain and dysfunction

(Deboever, 1973; Dubner et al., 1978, Yemm, 1979). These theories propose both intra-

and extra-articular origins of TMJ pain. Intra-articular theories argue that pressure or

tension on articular tissues results in TMJ pain. Extra-articular theories propose that

muscles or other contiguous structures refer pain to the TMJ (see Appendix A).

Fundamental to an evaluation of intra-articular theories is an understanding of the response

properties of TMJ nociceptors.

Our understanding of nociceptors innervating articular tissues has been advanced in

recent years (Schaible and Schmidt, 1983a, b, 1988b; Guilbaud et al., 1985; Birrell et al.,

1990). Most detailed information concerning the response properties of joint nociceptors

has come from studies of the cat knee and rat ankle joint. While there is considerable

information regarding the neuroanatomy of the TMJ (see Appendix C), little is known

about the physiology of nociceptors of the temporomandibular joint.

Early studies examining properties of afferents innervating the cat knee joint

reported activation of slowly conducting, myelinated fibers only by forceful rotational

movements beyond the physiological range of motion (Burgess and Clark, 1969; Clark and

Burgess, 1975). In contrast, Schaible and Schmidt (1983b) identified populations of









small fibers that were activated by passive movement of the knee joint, both within and

beyond the normal range of motion. These afferents were categorized into four populations

on the basis of their response range. Afferents of populations I and II responded to

innocuous joint movement in the normal range of motion, but population II responded best

to forceful rotation. Population III units were activated exclusively by forceful movement

and population IV had no response to movement. Based on movement sensitivity,

afferents that responded within the normal range of motion were not considered

nociceptors, while those that respond beyond the normal range were nociceptors.

Subsequent studies by Schaible and Schmidt (1985, 1988b) concentrated on populations

HIII and IV. However, despite the importance of range for classifying afferents, distinctions

between normal and non-normal ranges have been ancedotal. Quantification of the

relationship between range of motion and afferent activity is still pending.

There are no studies that have examined properties of TMJ nociceptors. Previous

reports have examined the discharge characteristics of afferent traffic in the

auriculotemporal nerve that may contribute to position sense and control of mandibular

movement (Kawamura et al., 1967; Klineberg et al, 1970, 1971; Kawamura and Abe,

1974; Lund and Matthews, 1981). While TMJ nociceptors may have properties similar to

those described for cat knee joint, there are considerable functional and anatomical

differences between these two joints that indicate that properties of TMJ nociceptors should

be determined (see Appendices C and F).

Changes in the properties of TMJ nociceptors (sensitization) may account for some

of the symptoms of TMJ dysfunction. These symptoms include pain during mandibular

movement, pain on local joint palpation, and tonic pain. Changes in nociceptor properties

are likely to be necessary to the development of these forms of pain. As pain on movement

is a primary symptom of TMJ dysfunction, a precise determination of the relationship

between movement and afferent discharges is valuable. Quantification of force and

movement also permits more direct specifications of movements which are outside the









normal range of motion, and identification of afferents that respond specifically in these

ranges.

Sensitization of TMJ receptors may be an important defense mechanism that

protects the joint from excessive movement which may cause injury. The TMJ is often the

site of inflammations associated with various forms of TMJ dysfunction (e.g. arthritis and

synovitis). An understanding of how inflammation may modify TMJ afferent function

would reveal the form and range of TMJ afferent sensitization and lead to a better

understanding of both etiology and treatment.

Sensitization of nociceptors has been documented in the cat knee joint (Coggeshall

et al., 1983; Schaible and Schmidt, 1985, 1988b; Grigg et al., 1986) and in the rat ankle

joint (Guilbaud et al., 1985) following chemically-induced acute or chronic arthritis. Acute

arthritis in the cat knee joint, induced by carrageenan and kaolin, produced changes in the

discharge properties in joint afferents. These changes include an increase in the proportion

of small fibers that could be activated by innocuous joint movement, recruited units, and

the appearance of spontaneous activity. Studies of sensitization of TMJ nociceptors may

provide uniquely important information regarding short and long term changes occurring in

TMJ afferents following local trauma, arthritis or chronic disc displacement.

In the following experiments, we examined properties of TMJ nociceptors in both

normal and inflamed tissue of the goat. The goat is an excellent model for TMJ physiology

because of functional and structural similarities to humans including range of motion and

innervation of the auriculotemporal nerve. (See Appendices C and F).














CHAPTER 2


METHODS



Subjects


Experiments were performed on 59 goats, ages 6- 12 months and weighing 30 60

lbs. All goats were kept in pens at the veterinary hospital. Goat health was monitored and

maintained by animal care technicians, veterinary staff and faculty. All procedures were

approved by an internal review board (protocol 7136).


Exposure of the TMJ and Trigeminal Ganglion


Rompun (3 mg/kg) and Ketamine (11 mg/kg) were given as preanesthetics.

Anesthesia was induced with a-chloralose (7.5 mg/ml) via a lactated ringers drip. If

necessary, pentabarbitol was periodically administered intravenously to maintain deep

anesthesia. Goats were artificially respirated at a rate of 10-20 breaths per minute. Blood

pressure, body temperature and end-tidal CO2 were continuously monitored and maintained

within physiological limits. At the conclusion of each experiment, the goats were

euthanized with saturated KCL.

To expose the right TMJ, the skin was removed from the lateral posterior aspect of

the face. The parotid gland, pinna and the zygomaticoauricularis muscles were removed to

access the posterior portion of the TMJ capsule. The styloid process and associated

muscular and ligamentous attachments were clipped off at the sphenoid bone. The origin of









the sternomastoid and caudal digastric muscles were excised from the mastoid process and

jugular process of the occiput, respectively. The sphenomandibular ligament and fatty

tissues were removed. Care was taken to preserve the microvasculature of the TMJ

capsule. The exposed capsular tissue was kept moist by repeated applications of

physiological saline.

Several nerves were cut distal to the trigeminal ganglion in order to avoid recording

activity other than TMJ afferents. The buccinator nerve was severed as it passed the

anterior border of the masseter muscle. The masseteric nerve was cut as it entered the deep

and superficial masseter muscle. The posterior trunk of the mandibular nerve was cut distal

to the branching of the auriculotemporal nerve. As a consequence, afferents of the lingual,

inferior alveolar and mylohyoid nerves were eliminated.

The trigeminal ganglion was accessed via its position caudal to the foramen ovale.

Removal of the thin, lateral plate of bone of the tympanic bulla and the tympanic plate

exposed the ossicles of the middle ear and internal surface of the tympanic membrane.

Removal of the inferior aspect of the greater wing of the sphenoid and the inferior aspect of

the petrous bone exposed the caudal surface of the trigeminal ganglion. Access to the

ganglion is improved by removing the sphenomandibular ligament and fatty tissue in the

infratemporal fossa.



Recording Procedures


Extracellular recordings were made by penetration of the trigeminal ganglion with

tungsten microelectrodes (Microprobe, Inc.) until unit potentials were evoked by passive

jaw movement. Most units were found in the posterior region of the trigeminal ganglion.

(See Appendix E). Receptive fields were then identified on the lateral or posterior capsule

of the TMJ. Amplified output was led to and monitored by a digital oscilloscope and an

audio speaker. Data was digitized and stored on video tape (Vetter Instruments, Model No.









3000A). Data were recovered from tape for analysis with an RC computerscope (RC

Electronics) or captured on a line by a thermal printer (Astromed, Dash IV).

After a unit was characterized, the conduction velocity was determined by

measuring the latency of responses to suprathreshold electrical stimulation applied to the

receptive field on the TMJ capsule. Electrical stimulation was performed with a Grass S88

stimulator and isolated, constant current unit. Square-wave pulses of 0.2 to 2 msec

duration were applied with monopolar electrodes. Conduction distance was estimated in

situ.



Characterization of TMJ Afferents



Once an afferent was isolated, passive jaw movements were used to determine

which type of movement was preferred (preferred movements were those which produced
,
the most vigorous response). Jaw movements included: 1) vertical jaw movements from

the closed to the open position. Afferent responses to vertical jaw movements were termed

"vertical plane reactivity" (VP); 2) lateral jaw movements from the closed to extreme lateral

position. Afferent responses to the lateral jaw movements were termed "horizontal plane

reactivity" (HP).

A force probe was used to produce jaw movement. The probe consisted of an axial

force transducer (Entran Instruments, Model ELF TC500-20) embedded in a plexiglass

rod. The transducer monitored forces used to move the jaw in the vertical or horizontal

plane. The probe was applied to fixed points of a bit which was mounted to the incisal

teeth of the mandible at the midline. Force and jaw movement were recorded on video tape

along with afferent activity. Jaw movement was monitored by an angular displacement

transducer (ADT, Trans-Tek Model # 604-001) mounted above the right TMJ (Figure 2-1).

Using the probe, both dynamic and static forces were applied to the mandible (Figure 2-2).

















































Figure 2-1.


Mandibular movements were produced by a hand held probe. Force
applied to the mandible was quantified by a force transducer. Vertical
plane mandibular movement was quantified by an angular displacement
transducer mounted above the right TMJ and attached to the mandibular
pole by a slide assembly. Vertical plane movement of the mandible was
produced by applying force at point A on the jaw movement apparatus.
Horizontal plane movement was produced by applying force at point B.
Afferent recordings were made following exposure and penetration of the
trigeminal ganglion proximal to the foramen ovale.




































Oscilloscope






















































Figure 2-2. Dynamic and static test series for force, displacement and velocity. A)
Dynamic test series. Sawtooth patterns of successively increasing
velocities. B) Static test series. Stepwise pattern of successively
increasing force and displacement.

















TEST FOR DYNAMIC REACTIVITY


UNITS


FORCE (N)


DISPLACEMENT (d"g)


i1 I a -


TIME


TEST FOR STATIC REACTIVITY


UNITS









FORCE (N)







DISPLACEMENT (d"g)


I II i1 H


TIME









Isolated units were characterized by two series of jaw manipulations. The first

series consisted of five tests of reactivity to dynamic force. The first, at a low velocity,

was followed by jaw movements of progressively higher velocities. The range of

velocities was approximately 0.3 to 250.0 N/sec. The second series consisted of five

stepwise, static force applications. Each step was maintained for approximately 2 seconds.

Forces up to 120 N were applied.

To confirm that jaw movement activated afferents of the TMJ capsule, receptive

fields on the capsule were identified and characterized for each case. Thresholds for

activation of mechanosensitive afferents on the capsule of the TMJ were determined by the

bending force of von Frey filaments (monofilament nylon) of different stiffnesses. The

method of limits was used to determine threshold. Nylon probes were applied to the

capsular field until the minimum force required to activate the afferent was determined. In

some cases, no receptive field could be found. This group of cases was treated separately

from afferents with identified receptive fields.



Statistics


Nociceptors were defined as those afferents that responded preferentially to

extremes of jaw movement. Relationships between force applied and jaw movement were

assessed to determine whether the afferents reacted at movement extremes. Force-

movement data was plotted for all units (n = 10) that responded preferentially to vertical

jaw movement and produced significant functions. Momentary relationships between

force, jaw movement and unit activity were determined. Force-interspike interval,

movement-interspike interval and velocity-interspike interval records were fit to linear,

logarithmic, exponential or power functions using simple linear regression (Statistical

Analysis System, SAS). The best fits were determined by comparison of coefficients of

determination.









For analysis of static reactivity, regression data were "segmented and balanced." A

segment (or step) of force was added to the regression model until the coefficient of

determination was maximal. The regression was "balanced" by using only six scores for

each stepwise force segment This procedure avoids weighting the regression differentially

at higher steps, where more scores typically bias the regression. These procedures were

not possible on tests of dynamic force reactivity. Therefore, for dynamic force or

movement tests, all spike intervals were used until a minimum interval was obtained.

Several characteristics of regression lines were subjected to analysis. These

included force-activation thresholds, movement-activation thresholds, force-frequency

asymptotes, movement-frequency asymptotes, mean response intervals, slopes and

intercepts. Force-activation thresholds (FAT) and movement-activation thresholds (MAT)

were defined as the instantaneous force or degree of movement at the first response

interval. Force-frequency asymptotes (FFA) and movement-frequency asymptotes (MFA)

were defined as the force or movement at which the frequency response was maximal (i.e.,

minimum response interval). The FAT or MAT paired with the FFA or FMA were used as

the lower and upper boundary values to plot functions derived from regressions. The mean

response interval (MRI) was calculated from the regression equation by using the mean

force or movement value calculated from the boundary values (FAT + FFA/2 or MAT +

MFA/2). Force-frequency thresholds and movement-frequency thresholds (FFTs and

MFTs) were determined for each unit that was tested for static reactivity. The FFT1' and

MFT were defined as the minimum force or movement which would maintain a sustained

discharge of the afferent. Slopes, intercepts and coefficients of determination were

generated only from significant regressions. Unit reactivity that could not be described by

significant functions were called "non-coding" or "non-transducing" afferents.

"Sensitization" refers to the enhanced excitability ofTMJ afferents that may occur

following carrageenan-induced inflammation. Enhanced excitability is reflected

quantitatively as changes in response properties of TMJ afferents in normal and in inflamed









tissue. Sensitization was assessed qualitatively by comparisons of functions fit to units in

normal and inflamed tissues.

Shifts of stimulus-response coding functions were used as an indication of

enhanced reactivity. Significant functions generated from responses of afferents in

inflamed tissue that were located graphically below and/or to the left of functions fit in

normal tissue were suggestive of sensitization (See Figure 2-3). Significant functions that

were acquired in inflamed tissue but not expressed in normal tissue also were considered to

represent sensitization. Sensitization was assessed quantitatively by statistical comparisons

of function characteristics generated from stimulus tests on units in normal and inflamed

conditions. Function characteristics (slope, intercept, thresholds, asymptotic activity,

coefficients of determination and mean interval) were compared in normal and inflamed

conditions using a one-way ANOVA (Li, 1967) for between subject and t-tests for within

subject's comparisons. Only significant functions contributed to the analysis.


Experimental Inflammation


Inflammation of the TMJ was induced by injection of 200 l of 2% carrageenan

(Sigma Chem.). In some cases, units were characterized in normal tissue and then exposed

to carrageenan. If the receptive field was located on the lateral side of the capsule,

carrageenan was dropped directly onto the receptive field as well as injected into the area of

the receptive field. If the receptive field was located on the posterior side of the capsule,

carrageenan was injected into the area of the receptive field. The remainder of the

carrageenan was injected into the superior joint space. Subsequent testing was performed

after at least 30 minutes and continued for up to 3 hours. In other cases, units characterized

in normal tissue were treated with saline. The injection of saline and subsequent retests

were identical to those used with carrageenan. In other cases, response properties of

separate populations of TMJ afferents were examined only in normal or pre-inflamed


















DEMONSTRATION OF SENSITIZATION
SHIFT OF STIMULUS-RESPONSE CODING FUNCTION
6 Ln interval (nsec)

5

4

3

2


0 . . L . i . . . . . .
0 1 2 3 45

Ln Force O-ewtonrs)
-v Fuiction A -*- Function B
















Figure 2-3. Demonstration of sensitization as a shift of stimulus-response
functions. Shifts of significant stimulus-response functions to graphic
zones of enhanced reactivity (below and to the left) suggests
sensitization. For downward shifts, each point on line B represents a
higher rate of discharge (shorter interval) at an equivalent force. For
leftward shifts, reactivity appears at forces which previously were
ineffective.





15


tissue. In this instance, inflammation was induced at least 6 hours before recording began.

Carrageenan (2%, 200 p.1l) was injected into the lateral and posterior aspects of the TMJ

capsule as well as the superior joint space.













CHAPTER 3


RESULTS


Identification of TMJ Nociceptors


Receptive fields for units with VP or HP reactivity were found on the lateral and

posterior capsule of the TMJ (Figures 3-1 and 3-2). The mean von Frey thresholds for VP

and HP units were 27 39 g and 15 + 20 g, respectively. The receptive fields of these

units were single, small spots (approximately 2-3 mm). In some cases, receptive fields

could not be identified. It is possible that the receptive fields of these units were located on

the medial capsule, posterior attachment tissues or peripheral portions of the articular disc.

Alternatively, it is possible that units without identified receptive fields were not of TMJ

origin but, instead, of muscular origin (lateral pterygoid or temporalis muscles).

Conduction velocities were determined to be in the group III and IV fiber range (0.4 to 8.5

m/s). Due to the inaccessibility of many of the receptive fields, conduction velocities could

only be obtained for 30% of the units (19 of 63 units).

The response characteristics of TMJ afferents were examined. Mechanical

reactivity to jaw movement was determined using both dynamically and statically applied

forces. In order to distinguish between joint afferents acting as proprioceptors from those

acting as nociceptors,we searched for afferents with a preference for intense stimuli. It was

predicted that TMJ nociceptors would respond in the noxious range of mandibular motion,

where the noxious range of motion was defined as that passive mandibular motion where

the mandible encounters considerable resistance imposed by constraints of soft tissue.



















































Figure 3-1.


Receptive field (RF) distribution in the lateral and posterior capsule of the
TMJ for normal tissue. A) Distribution of RFs in the lateral capsule. B)
Distribution of RFs in the posterior capsule. RFs (n=36) of afferents in
A) and B) responded to either vertical plane (.) or horizontal plane (A)
movement of the mandible. POST, posterior, ANT, Anterior, MED,
medial; LAT, lateral.







A




POST







B


MED


LAT


















































Figure 3-2. Receptive field (RF) distribution in the lateral and posterior capsule of the
TMJ for previously inflamed tissue (-), acutely inflamed tissue (A&) and
saline control tissue (+). A) Lateral capsule. B) Posterior capsule.
POST, posterior, ANT, anterior;, MED, Medial; LAT, lateral.







A





POST







B


MED


LAT









Therefore, relationships between applied force and jaw movement were examined to

determine the noxious range and compare it with afferent discharge properties.

Due to physical constraints of the preparation, force-movement relationships could

only be examined for VP units. Characterization of force-movement relationships in the

vertical plane revealed a biphasic relationship between applied force and mandibular

movement (Figure 3-3). Linear force-movement relationships gave way to exponential

force-movement relationships when the mandible reached an extreme of opening. It is

likely that this exponential phase corresponds to the noxious range. Force-movement

curves of VP units (n = 10) were best described by power functions (Table 3-1). Power

functions were the best fit in 10 of 10 cases (R2 = 0.92 + 0.06). The functional range of

VP units may be graphically assessed by observing activation thresholds plotted on force-

movement curves. As can be seen in Figure 3-3, nearly all afferents began to respond in

the transition zone between the linear and exponential phases of the curves. This can be

examined statistically by examining the functions fit below the point of activation. For the

10 cases in which power functions were fit, activation threshold occurred during portions

of the curves which were exponentially accelerating (6/10 cases, R2 = 0.86 + 0.17). In 4

cases, activation thresholds occurred in portions of the curve in which force was increasing

linearly (R2 = 0.88 0.19). The activation thesholds of these 4 units occurred at forces

approaching the power phase of the curve. The observed preference for high forces (mean

activation threshold for force = 20.6 + 11.0 N; mean frequency asymptote = 49.6 + 20.7

N) at extremes of the range of motion (mean activation threshold at 13.7 + 5.4; mean

frequency asymptote = 20.4 + 5.9) suggested that these TMJ afferents were nociceptors.


Characterization of TMJ Nociceptors in Normal Tissue


Units were classified according to their preferred movement plane. Preferred

movement was defined as that plane of motion in which mandibular excursion evokes the




















































Figure 3-3. Relationships between applied force and vertical plane movements of the
mandible. Ten biphasic curves indicated that low forces were sufficient to
open the jaw through most of its range of motion. High forces produce
minimal degrees of opening where the joint encounters considerable
resistance from soft tissue. In each case, the TMJ nociceptor begins to
respond at the point indicated by the arrow.










TMJ FORCEIMOVEMENT RELATIONSHIPS: VERTICAL PLANE

D


0 20 40 60 sO 100 120 140
FORCE (Newtons)
-,- TEST --- TEST -*- TEST -0- TEST
1 2 3 4






B


0 15 30 45 80 75 0O 105 120 135 150
FORCE (Newtmn)
=m TEST-,,- TESTm-m TESTm TEST-*" TEST
1 2 3 4 6






C


is OVEMENT (dear***)

12







N042090H


0 10 20 30 40 so50


sO


FORCE (N*wtons)
,,, TEST ,* TESTM, TnEST-- TESTmA* TEST
1 2 3 4 5






E


0 15 30 45 s0 75 90 10S 120 135 150
FORCE (Newton )
-A- TEST.- TEST-*- TEST-U- TEST-- TEST
1 2 3 4 5


0 10 20 30 40 50 0 5 10 15 20 25 30 35
FORCE (Newton.) FORCE (Newton<)
a-A TEST E-*- TEST-- TEST-@- TEST-- TEST -A- TEST TEST E
1 2 3 4 5 1 2 3


40 45


:Sq













TMJ FORCE/MOVEMENT RELATIONSHIPS: VERTICAL PLANE



I


0 10 20 30 40 so50 60 70 80

FORCE (Nowtons)
-As TEST -*- TEST -*- TEST -m- TEST
1 2 3 4


0 10 20 30 40 SO 80


FORCE (Newtons)
"A- TEST TEST
1 2


I OVEMENT fdeorees) MOVEMENT (decrees)


25

20



10

___ _St
092791K 101691F
S . . . . . . . . . . . . . . . . . . . . . . . . . .
0 10 20 30 40 50 60 70 0 10 20 30 40 so50 so

FORCE (Newtons) FORCE (Newton.)
-Am TEST -- TEST -*- TEST U-- TEST -A- TEST -0- TEST
1 2 3 4 1 2


Figure 3.3. (cont'd)


* TEST
3









Table 3-1
Force/Movement Relationships in the Vertical Plane


2
FILE FUNCTION FULL RANGE LOW RANGE R P

042090H POW LnI = 0.69 LnF + 4.8 0.93 0.0001
042090H LIN I = 0.15 F + 1.8 0.90 0.0001
042090H POW Lnl = 0.8 LnF + 0.89 0.96 0.0001
042090H LIN I = 0.2F + 0.87 0.95 0.0001
110990E POW Lnl = 0.8 LnF + 2.1 0.97 0.0001
110990E LIN I = 0.6 F + 2.3 0.96 0.0001
110990E POW LnI = 1.1 LnF + 0.15 0.92 0.0001
110990E LIN I = 1.0 + 0.19 0.97 0.0001
111490D1 POW LnI = 0.4 LnF + 1.1 0.91 0.0001
111490D1 LIN 1=0.IF+ 8.5 0.87 0.0001
111490D1 POW LnI = 1.7 LnF+ 2.2 0.92 0.0001 1
111490D1 LIN I = 1.0F+ 4.2 0.86 0.0001
111490D2 POW LnI = 1.3 LnF + 1.2 0.91 0.0001
111490D2 LIN I = 0.7F+ 0.22 0.87 0.0001
111490D2 POW LnI = 2.3 LnF + 3.5 0.94 0.0001
111490D2 LIN I= 1.4F + 7.4 0.97 0.0001
022791A POW LnI = 0.3 LnF + 2..2 0.93 0.0001
022791A LIN I = 0.77F + 11.0 0.79 0.0001
022791A POW LnI = 0.15 LnF +2.0 0.80 0.04
022791A LIN I = 1.8F + 5.9 0.97 0.0001
032891X POW LnI =0.37 LnF + 1.1 0.99 0.0001
032891X LIN I = 0.23F + 3.7 0.88 0.0001
032891X POW LnI = 0.4 LnF + 0.01 0.98 0.0001
032891X LIN I = 0.55F + 2.3 0.88 0.0001
080591 IC POW Lnl = 0.34 LnF + 1.9 0.87 0.0001
080591C LIN I = 0.28F + 12.0 0.77 0.0001
080591C POW LnI = 0.46 LnF + 1.8 0.57 0.0003
080591C ULIN I = 1.5 F + 5.3 0.61 0.0001
091691A POW LnI = 0.54 LnF + 1.3 0.87 0.0001
091691A LIN I = 0.42F+ 9.0 0.83 0.0001
091691A POW LnI = 0.64 LnF + 1.2 0.52 0.01
091691A LIN I = 2.3 F + 1.3 0.37 0.04
092791K POW LnI = 0.48 LnF + 1.4 0.96 0.0001
092791K LIN I = 0.37 F + 8.6 0.88 0.0001
092791K POW LnI = 0.59 LnF + 1.3 0.90 0.0001
092791K LIN I = .97F + 4.1 0.88 0.0001
101891F POW LnI = 0.5 LnF + 1.3 0.83 0.0001
101891F LIN I = 0.29F + 11.0 0.73 0.0001
101891F POW LnI = 0.74 LnF + 1.0 0.86 0.0001
101891F LIN I = 0.9 F + 5.6 0.64 0.0001

Note: POW, power; LIN, linear.









strongest response from an afferent. Thirty-six afferents with receptive fields in the TMJ

capsule were activated by either vertical plane (VP or opening) or horizontal plane (HP or

lateral displacement) motion. Eighteen afferents responded preferentially to movement in

the vertical plane and 18 responded best to movement in the horizontal plane. A few units

were activated by right lateral displacement (n = 5) or protrusive (n = 2) movement but

most TMJ afferents responded preferentially to either opening (VP) or left lateral

displacement (HP).

Once preferred movement was determined, tests of dynamic and static reactivity

were conducted. Relationships between force, position or movement and response interval

were quantified. The responses of TMJ nociceptors were quantified by regressions fit

between instantaneous dynamic force, force velocity, movement, static force or position

and instantaneous interspike interval.

As it was not clear which function would best describe TMJ afferents, linear,

power, logarithmic and exponential models were explored. By comparison of the relative

proportions of variance accounted for by each model, it was determined that power

functions accounted for the greatest percentage of variance (mean R2 = 0.46 + 0.26; n = 71

cases). Power functions were best fits compared to other functions for both dynamic

(26/59 cases) and static tests (10/12 cases). Power functions were second best fits in 20

out of 59 cases for dynamic tests and 1 out of 12 cases for static tests. Linear functions

were superior in 7 cases, logarithmic in 12 cases and exponential in 16 cases. Since power

functions were the best or second best fit in 46 of 71 cases (mean R2 = 0.45 .30), power

functions were used to represent the data in normal tissue. In inflamed tissue (both PI and

AI), power functions accounted for the greatest percentage of variance in 101 of 146 cases

(mean R2 = 0.50 + 0.21).

TMJ nociceptors could be found which transduced (or coded) all aspects of force,

movement or position, and many transduced more than one variable. "Code" indicates that









a unit's reactivity was proportional to force or some other stimulus variable and that the

relationship could be described by a significant power function.

The great majority of TMJ nociceptors (n = 24) coded dynamic aspects of force

applied to the mandible, and relatively few coded for static aspects of these stimuli.

Twenty-three of 36 units that responded to either VP or HP jaw movement coded for

dynamic force (mean activation threshold = 20.1 17.5 N) and 15 of 36 units coded for

force velocity (mean activation threshold = 15.5 + 20.5 N/s; Table 3-2). In contrast, only

8 of 36 units coded for static force (mean activation threshold 17.8 + 18.3 N), while 5 of

35 units coded jaw movement (mean activation thresholds of 11.0" + 4.0, and only 1 of

36 units coded for jaw position. In some cases, unit reactivity could be described by more

than one function. Twelve of 23 units that could be fit to dynamic force were also

significantly fit to force velocity. Six of these 12 units were significantly fit to force

velocity, and 6 of these 12 units were significantly fit to static force as well. Ten of 36 units

could not be fit to any function. Fourteen additional units had jaw movement reactivity but

no receptive field could be found.

Mean functions were generated from individually fitted functions by pooling

slopes, intercepts and boundary values. Functions presented in Figure 3-4 represent the

principle transducing properties of TMJ afferents: dynamic force and force velocity.



Properties of TMJ Nociceptors in Inflamed Tissue



The response properties of afferents were examined in capsular tissue that was

previously inflamed (PI) by carrageenan. In these experiments, TMJ tissues were exposed

to carrageenan for at least 6 hours prior to testing. In other experiments, afferents were

tested in normal tissue and then tested again after an acute injection of carrageenan or

saline. Afferent reactivity was examined for up to 3 hours subsequent to exposure to

carrageenan.










TABLE 3-2


TMJ NOCICEPTOR REACTIVITY IN NORMAL TISSUE
with RF (n=36) without RF (n=14)
Test Conditions
code (n=25) no code (n=10) code (n=--9) no code (n=5)


Dynamic Force 23 12 6 6



Force Velocity 15 19 4 8



Movement 5 10 2 3


Static Force 8 23 1 4



Static Position 14 0
1 14 0 4


Conduction 0.4 to 7.5 0.5 to 1.5
mrn/sec m/sec
Velocity (n =6) (n = 3)
Post-test
Spontaneous 3 0 0 0

Activity

Note: Receptive fields (RF) in the TMJ capsule were identified for 36 units. Twenty-five
units with RFs coded for one or more of the stimuli. Most units (n = 23) with an
RE on the TMJ capsule coded for dynamic force. Fourteen additional units had
either VP or HP reactivity but an RF could not be found. "Code" indicates that a
unit's reactivity could be described by a significant power function. Of the 36
afferents with RFs in the TMJ capsule, 3 units demonstrated spontaneous
activity after dynamic or static testing. Five units with RFs in the TMJ capsule
responded preferentially to right lateral displacement. Most of these units (3 of
5 cases) coded dynamic force. Appendix G contains descriptive statistics for
these units.














I-" ; '2 0

0 C. *- 0t
II >~g II C)
(u C) C





i 0 u


o1 o~ 8 B ,
.-'--,l ^



Q oI =,.6
a,0



















r,. 0 2 ,,;"*0 0 ... .., --
0-g. .1d :+1 r: ^
44 0














" -- 0 g^ 4 .
n, o







o co-p +C '* +1o
as v.

. 0 E
C6.


C)li
-o-~00













Tt ( y 'it '> 11l


,Z 52 E ,M 641
u 0 C ) C U14
M.-. C)
C) n-8~ 00 +1





*~8CU >- '


00
C ) +1 0 o1
4W .- 4
0 o E +1





30





------- C-------------------------------------------

I 4i
Z S n .
uj z U Z |


"J -7 IT Ii n /.

((, / ^|2 /,
a. I, a.
... 0
ZW I' o ,
<- I I
o4Jw 00 z
00~ ow-
FE 0 a~ l "0
0 I V W -

-4 L ti i( 8r z
>- 4 I 1
_ 5
-- I I I- 4
0 >

c ... ... ... ... .. ..* * -* o I I . . .. . ** *** *** < I I
m < 'o
w (


a: :




,7, .,,,, / //..,. .' _!".,,
/ '"> 4 f.
" I O A I-









z I I +
- >" oa -
C n CS "i i Nfy *-
CL mQ





*f/'w
Cj C)






i -- /r*' o -1 .* o I I
(P 0 a 0S

UJ 0 0
za a *0 u <
> II> I
0-J I UI


I- < 5
U0 w 1
Ow j 5I E 0



- I I I I
.. .. I .. .. .. . . I o I I ~ .. .. ... ...1. ...* I I
U1I C < 0 Ul.f C (








Experiment 1: Properties of Units in the
Previously Inflamed TMJ


In the PI condition, transduction of the dynamic aspects of stimulus force was best

described by power functions. Power functions were the best fit compared to other

functions (7 of 13 cases; R2 = 0.33 0.25) and were best or second best fit in 9 of 13

cases. Therefore, exposure to carrageenan did not modify the nature of functions which

described afferent activity. Individually fitted power functions generated during dynamic

and static response testing in inflamed tissue are illustrated in Figure 3-5.

Comparison of units in normal and inflamed tissue (see Figure 3-6) suggested

improved reactivity for units that responded preferentially to movement in the vertical

plane. Increased reactivity to dynamic stimuli was particularly prominent. In normal

tissue, 11 of 18 units with RFs on the TMJ capsule coded for dynamic force or velocity;

while proportionally more units (5/5) transduced dynamic force or velocity in the PI

condition. Mean functions for dynamic force in inflamed and noninflamed conditions were

calculated by averaging slopes and intercepts of individual functions (LnI = -1.3LnF + 7.3,

(n = 25); LnI = 1.1 LnF + 5.4, (n = 6)] for inflamed and noninflamed, respectively). The

mean functions for all VP and HP units fitted during dynamic testing in normal and

inflamed tissue suggested improvement of dynamic reactivity (Figure 3-6). The shifts of

the mean function for units in PI condition to a graphic position below the mean function

for normal tissue was suggestive of sensitization to dynamic forces.

Mean functions were also calculated for afferents transducing force velocity. A

shift of the mean function of PI units that coded for force velocity [Lnl = -0.7 LnF + 3.5,

(n = 6)] was also observed in comparison to the mean functions in normal tissue [LnI = -

0.38 LnF + 4.7, (n = 16)]. These functions are illustrated in Figure 3-6. A shift of the

mean function for afferents in inflamed tissue below and to the left of the mean function for

normal tissue also suggested sensitization to velocity of applied forces.



















































Figure 3-5.


Dynamic reactivity in previously inflamed tissue. Power functions best
described the relationship between instantaneous dynamic force and force
velocity with instantaneous response interval in the vertical or horizontal
plane. A) Six of 7 units transduced dynamic force (mean slope = -1.1 +
0.5; range = 7.6 + 10.2 to 33.3 + 32.1 N; mean R2= 0.24 + 0.16). B)
Six of 7 units transduced force velocity (mean slope = -0.7 0.8; range =
4.8 + 4.5 to 40.2 + 22.2 N/s; mean R2 = 0.33 + 0.28).











REACTIVITY IN PREVIOUSLY INFLAMED TISSUE:
DYNAMIC FORCE
Ln Interval (msec)


5 .N



3

2

1
0 U I | *
-1.5 0.0 1.5 3.0 4.5 6.0

Ln FORCE (Newtons)
-V- T7 -*- E1 -m- J1 -*- E3 -A- T1 -+- 02



REACTIVITY IN PREVIOUSLY INFLAMED TISSUE;
FORCE VELOCITY
SLn Interval (msec)
6

5
~A

4

3

2

1


-1 0 1 2 3 4 5 6

Ln Velocity (Newtons/sec)
-+- B2 -A- T7 -0- El -- J1 -A- E3 -- T1

















































Figure 3-6.


Averaged dynamic response functions of nociceptors fitted in normal and
previously inflamed tissue (PI). A) Mean function for dynamic force
fitted in PI tissue (mean interval = 29.9 + 7.9 msec) falls below the mean
dynamic force function fitted in normal tissue (mean interval = 44.7 +
16.5 msec) suggesting sensitization. B) Mean function for force velocity
in PI tissue (mean interval = 22.2 + 3.3 msec; mean activation threshold =
17.3 + 18.9 N/s) falls below and to the left of mean force velocity
function fitted in normal tissue (mean interval = 7.4 + 4.8 msec; mean
activation threshold = 4.8 + 4.5 N/s) suggesting enhanced reactivity
(sensitization)











CHANGES IN REACTIVITY TO DYNAMIC FORCE
FOLLOWING CARRAGEENAN INFLAMMATION
Ln Interval (msec)
6

5

4


3N
3------*^---

1---------

1 I ---------------------


-V-- N (n=25)


Ln Force (Newtons)
-*- INF (n=6)


CHANGES IN REACTIVITY TO FORCE VELOCITY
FOLLOWING CARRAGEENAN INFLAMMATION
Ln Interval (msec)
6

5

4
3 ,







5 -,-------------- I p -- I ----------


-Tv- N (n=16)


2 3

Ln Velocity (Newtons/sec)
9- INF (n=6)









Statistical comparisons between function characteristics in normal and inflamed

tissues confirmed that dynamic reactivity improved in inflamed tissues (See Table 3-3).

This was manifested as changes in activation threshold, frequency asymptotes and function

intercepts. The mean activation threshold (8.3 13.2 N) for units in inflamed tissue (n =

4) that responded best to jaw opening and transduced dynamic force was significantly

lowered relative to the mean activation threshold (24.1 + 11.2 N) for VP units (n = 10) in

normal tissue (F = 5.36, df = 13, p < 0.04). For those same units, the mean intercept in

inflamed tissue (5.0 + 3.0) was significantly decreased relative to the mean intercept in

normal tissue (9.0 + 3.1). Similar changes were observed for afferents that transduced

force velocity. The mean frequency asymptote (37.5 23.6 N/s) for these units was

significantly less than the mean frequency asymptote (148.0 + 79.6 N/s) for units (n = 8),

in normal tissue (F = 8.89, df = 12, p < 0.01). The mean force-frequency asymptote (40.2

+ 22.2 N/s) for PI units (n = 6) that coded for force velocity in both VP and HP jaw

movements were significantly less than the mean force-frequency asymptote (156.5 + 83.1

N/s) for VP and HP units (n = 15) in normal tissue (F = 9.57, df = 21, p < 0.01).

The changes we have observed in dynamic reactivity of TMJ nociceptors are

suggestive of sensitization. However, experiments of this type (PI tissue) are subject to

sampling errors. The best form of evidence for nociceptor sensitization comes from

observations of single affferents making transitions from the normal to the sensitized state,

following acute injections of pro-inflammatory substances.


Experiment II: Promperties of Units in the
Acutely Inflamed TMJ


The findings of enhanced dynamic reactivity observed in populations of afferents

sampled from normal and subacutely inflamed TMJ tissues were complemented by

observations made on units in which testing was performed before and after carrageenan-



















+1 +1 +1 +1
N- -40\ o'0"




0000 .
+I +I +I +I I
W 00 o 0

A




odo

o\
m;b




o n r-^ AD
on 00 (Sl
+I +I +I +C


a

* 06 &n 06

+I +I +I +I b
ON> on oo r-*-






+1 +1 +1 +1

Vl At lt Un


0q v! at un
n 0 ki- 00
6666 1 lo

(S lO(S\0 w
6666


c > N o -

N^___


+1+1 +1 -1




cqcn ON CD0
\0 00 *-4


+1+1+1
C^le CO e 0







1-4
dod
CO~C




+1 +1
. .+11






ONCIt Cl


(r! "0q


-400 ul-
+1+1 k+1t







cD0 019







+1'+l q +1
r-^ A? ON

N^ooc
-O d ('
+| +| I-+


C A0 W


cl
of


U
U
I-
U





U
01
Cd,


oo on o i
000 0

+ +1+1+i+1+i+!+1+A1m
In ci q qOi vUi i ul of)


c o r- CS 00000000
l +1 +1 +1 +1 +1 +1 +1 +1 6 6 ON

o00O0000 t00



bOO n o
S^ . . + -H o r
.14
,0irV



.-< cl *- '--


ON ot00 Cl tON -
L r +I+I +I+I +I+I +I +I











'U +i+i+i+i+i+i+1+j1Cf
4- +1+1 +1 +1 +1 +1 +1 +10 IS
0 '
-0-4 ICn
t 0' N4 DIton V






-44
v, Tr- 41-oomsoI
O~NONoduo V W & W^








ONC5 ON*
S+1 +1 +1 +1 +1 +1 +1 +l2
U ON a;*t CD 0 c; r7 Cq olv

-l Ce 0\Clr-C4O -0 o




0 +1+1+1+1+1+1 +1+l '100
It ul At A; oo op at- at r- s


"acyC t- l on on CDo cc

C5 5 D A-i '-i
^00




0oo st~c- o H0 oo 0-
I I I01'0- Q


(-) ( 1

OUc cc ro c 'i-
H


. -1
60 U
LLS


-U
-~ I~
U -~


c


U3
H
















0



.4)
*>;







4.o
or











.-0
S










40.

u uN


















-4




0
cz
>* 0


*4)
z-









induced inflammation (acutely inflamed or AI). Outcomes of experiments using PI

methods suggested that individually characterized afferents would manifiest improved

dynamic reactivity.

Observations of quantitative shifts in force interval functions were confirmed, and

additional qualitative changes were observed. Eleven units were characterized in normal

tissue and then retested 1-3 h after exposure to carrageenan. Four units responded

preferentially to VP jaw movement and 7 responded best to HP movement For 9 units that

transduced stimulus intensity in both normal and inflamed tissue, 6 demonstrated enhanced

reactivity for dynamic force (Figure 3-7). Four of these 6 units had post-inflammatory

functions that shifted below the pre-inflammatory function. One of these also showed

leftward shift. Three of 6 units that demonstrated enhanced dynamic force reactivity

showed a leftward function shift relative to pre-inflammatory functions but failed to shift

below the control case. Mean function characteristics for these nociceptors are presented in

Table 3-3.

Experiments were performed to determine whether changes in AI reactivity

occurred as a result of carrageenan or other causes, such as repeated stimulation. In these

control experiments, saline (100 ill) was injected into the area of the RF and into the TMJ

capsule. Reactivity was examined in 8 units after exposure to saline. Five of 8 units

transduced dynamic force in both the pre- and post-saline conditions. While quantitative

changes in reactivity were small following saline injections, there was some indication that

repeated testing could produce qualitative changes in reactivity similar to those observed

following carrageenan injection (See Figure 3-8).

Following saline injection, mean activation threshold was reduced from 14.1 +

15.0 N in pre-saline condition to 11.5 + 7.5 N. These changes were relatively small

compared to changes observed following carrageenan (11.2 + 9.9 to 5.1 + 3.7 N).

Statistical comparisons indicated that the mean changes in activation threshold was

significantly less in saline than carrageenan treated cases (T = -2.2; DF = 15; p = 0.04)).




















































Figure 3-7.


Demonstration of sensitization (enhanced reactivity). Six of 11 units
were characterized in normal tissue and then tested again after acutely
inflamed (AI) with carrageenan. Power functions generated from
retesting in acutely inflamed tissue were positioned below and/or to the
left of power functions fitted on normal tissue. N, normal, Min,
inflamed; HR, inflamed.





















REACTIVITY IN VERTICAL PLANE: NORMAL TO INFLAMED
DYNAMIC FORCE: UNIT #1

6 b wr' 1-- Iiu---

5

4


3
2

1


0 1 2 3 4 a 6


-- N


Ln Force #mwtcnx
-*- I -6- 3 W


REACTIVITY IN VERTICAL PLANE: NORMAL TO INFLAMED
DYNAMIC FORCE: UNIT #3






4

3

2



0
0 1 2 3 4 6


REACTIVITY IN HORIZONTAL PLANE: NORMAL TO INFLAMED
DYNAMIC FORCE: UNIT #2

6r t-


0 1 2 3 4

Ln Frc* t>osu
S-- N --1K


6 6


a- 2


REACTIVITY IN HORIZONTAL PLANE: NORMAL TO INFLAMED
DYNAMIC FORCE: UNIT #5

:n htw-... hoo



4

3

2
:^




0
0 1 2 3 4 6 6


V- N


Ln Fore* fJw.rd)
-- 1 W


REACTIVITY IN VERTICAL PLANE: NORMAL TO INFLAMED
DYNAMIC FORCE: UNIT #4
n ftw~l vr________________
6 ,r I ,,.I e

s
4


3

2




01
0 .. . . . . i . . . . " "
0 1 2 3 4 S 6

Ln Foce t.il6I
-w- N *- 90 MB -U- 2 W


V N


Ln F ce. f-tom
-- 2. R


REACTIVITY IN HORIZONTAL PLANE: NORMAL TO INFLAMED
DYNAMIC FORCE: UNIT #6

6 "r-,* (nI '-




4
"-----:--- ---


2


0
0 . . . . I . . . . 1 . . . .
0 1 2 3 4 S

Ln Forc tlatonm
-v- N --- I t






















































Figure 3-8.


Reactivity of nociceptors exposed to saline. Five units transduced
dynamic force in both the pre- and post-saline conditions. A) and B)
demonstrate quantative changes in dynamic force reactivity. C), D) and
E) show no enhanced reactivity.

















A
REACTIVITY IN HORIZONTAL PLANE: SALINE CONTROL
DYNAMIC FORCE: LEFT LATERAL UNIT #3

: n Wa *lr4W-----------






2
S -----------------


-w- N


1 2 3
Ln Force* Nwle)
-- S1a


4 5


B
REACTIVITY IN HORIZONTAL PLANE: SALINE CONTROL
DYNAMIC FORCE: LEFT LATERAL UNIT #6





4
3

2



0
0 1 2 3 4


-- N


C
REACTIVITY IN HORIZONTAL PLANE: SALINE CONTROL
DYNAMIC FORCE: LEFT LATERAL UNIT #2
L. IV&NtW 61-d


0 . . . . 6 "
0 1 2 3 4


-- N


Ln For** It*.ol
-0- Sil -a- S


Ln F*c. 04Wewd
-0- SIaR


D
REACTIVITY IN HORIZONTAL PLANE: SALINE CONTROL
DYNAMIC FORCE: LEFT LATERAL UNIT #7





4
3

2

1

0 1 2 3 4 5


Ln Frve Ntowla
--- SIm -6- S2mF


-Y- N


E
REACTIVITY IN HORIZONTAL PLANE: SALINE CONTROL
DYNAMIC FORCE: RIGHT LATERAL UNIT #1
a, ,In b-vo-. w m



4
:=S
3

2


.
0 . .-----.---.----- ..----


- N


2 3 4 5
Ln Faooe twtcmn
-- si. -- sa6









Therefore while shifts in activation threshold could occur following repeated testing, the

changes that followed carrageenan injection were significantly greater.

In experiment I, it was determined that most TMJ nociceptors transduced dynamic

force and velocity, but few transduced either static force, movement or position.

Following acute exposure to carrageenan, qualitative changes in coding capacities were

noted; that is, afferents coded for stimulus features not previously transduced.

Qualitative improvements were observed for most units for either dynamically or

statically applied stimuli. Eight of 11 nociceptors acquired either dynamic or static coding

ability (Table 3-4). Four of 4 units that preferred VP jaw movement and transduced

dynamic force in normal tissue acquired the ability to transduce force velocity (n = 2), static

force (n = 1) or jaw position (n = 1) in inflamed tissue. Two of 5 AI units that preferred

HP movement and transduced dynamic force in normal tissue also acquired the ability to

transduce static force. Acquired coding also appeared in nociceptors that were without

transducing capacity in normal tissue. Two of 7 AI units that preferred HP movement

acquired the capacity to transduce both dynamic and static force. Other changes appeared

as de novo reactivity. Three of 7 units that had only HP jaw movement initially acquired

similar reactivity for VP movement (dynamic force). Six units demonstrated spontaneous

activity after stimulus testing.

The finding that a large proportion (8/11) of nociceptors acquired coding ability

after exposure of carrageenan suggested a qualitative form of sensitization. The nature of

these qualitative changes can be best appreciated by comparing scatter plots representing

comparisons between pre-inflamed and post-inflamed reactivity. Two types of trends (shift

and rotation) were observed. Shift indicates a movement of a scatter of points into graphic

zones of greater sensitivity. Rotation indicates the development of a significant slope

associated with the scatter field of points. Of the 9 of 11 nociceptors that acquired coding

ability, all 9 showed shifts and/or rotation. Figure 3-9 illustrates significant functions of

units that demonstrated shifts and units that demonstrated rotation of scatter point fields.





45




Table 3-4.

EFFECT OF CARRAGEENAN ON TRANSDUCING CAPACITY

Qualitative improvements in reactivity for nociceptors that were
characterized in normal tissue and then tested again subsequent to
carrageenan injection. Eight of 11 nociceptors acquired either dynamic or
static coding ability. Coding ability was also acquired for nociceptors (n =
2) that had no transducing capacity in normal tissue. Three units that
responded to horizontal plane movement acquired vertical plane reactivity in
acutely inflamed tissue. Six units demonstrated spontaneous activity after
stimulus testing in acutely inflamed tissue.



UNIT PLANE PRE-INFLAMED POST-INFLAMED CODING OTHER POST-TEST
SHIFT MOVEMENT SPONTAN.
____________________________ACTIVITY

1 VP DYNAMIC FORCE DYNAMIC FORCE YES YES
FORCE VELOCITY
MOVEMENT MOVEMENT
STATIC FORCE STATIC FORCE
POSITION POSITION

2 VP DYNAMIC FORCE DYNAMIC FORCE YES
STATIC FORCE YES


3 VP DYNAMIC FORCE DYNAMIC FORCE
FORCE VELOCITY YES NOW
STATIC FORCE STATIC FORCE

4 VP DYNAMIC FORCE DYNAMIC FORCE
FORCE VELOCITY
MOVEMENT MOVEMENT WE
POSITION YES


Note: VP, vertical plane











Table 3-4 (cont'd).


UNIT PLANE PRE-INFLAMED POST-INFLAMED CODING OTHER POST-TEST
SHIFT MOVEMENT SPONTAN.
ACTIVITY

5 HP DYNAMIC FORCE YES
FORCE VELOCITY YES YES
STATIC FORCE YES ____________

6 HP DYNAMIC FORCE DYNAMIC FORCE
FORCE VELOCITY FORCE VELOCITY
STATIC FORCE YES YES


7 HP DYNAMIC FORCE DYNAMIC FORCE OPEN: CODE N3NE
FORCE VELOCITY FORCE VELOCITY OPEN: CODE
STATIC FORCE YES OPEN: NON-CODE


8 HP DYNAMIC FORCE DYNAMIC FORCE YES
FORCECVELOCITY FORCE VELOCITY
STATIC FORCE STATIC FORCE

9 HP DYNAMIC FORCE DYNAMIC FORCE OPEN: CODE YES


10 HP DYNAMIC FORCE DYNAMIC FORCE
STATIC FORCE STATIC FORCE
NOWE

11 HP DYNAMIC FORCE YES OPEN: NON-CODE
FORCE VELOCITY YES OPEN: NON-CODE
STATIC FORCE YES OPEN:NON-CODE N3NE


Note: HP, horizontal plane














IL) w) C:


10 jg


4-o Q,





0 o








I-.0 CU
CU 0


oo
L*sU











.0 C




-Cd 41.4










0 C



Cd Cd



Qd Cd





0















I-
|PP






48






o 0
w w
uJ-

z z S
11 '* '


0 0


0,- < U
0z 0 5
,',:3 ,- -, |; ,,"3 ::I




Oil O .
S .o Ic* *Cg'



0" < <0
N j. N 0-0 0
a: 4'* E h



Z Z Z z
>- > -
51 . 0 5- .0 0

0 0
F- 5 .... ... .. .. ^. ^... . . o *i....I....I....I....I....I....I..... o
> Is W l OI O h lO OT I t q' N O







w w
z o
a *
LU UJ


< <.
0i .0 z '




< <
'C 'Co: o







00

9-
-- -N wlt -< ?
0 0 +, 1 0 --





S0 0
z z








++ + z is z
>" ,, >- I
-t ....... o 0 .... ... o 0

0 0 0e
< <
< wJ caf5^ o ujw


c:c ccb e
>,, 31....I....I.. n ...I 1 I....I....I.. .. .
rr o o ^ n c r o o ^ n N














a
w

U-
u...
z
0
t-



z_

CUJ
< L

5,o



10
01-









N1-
z
0 <











0
!--

o
a:




0





a:
X






0-
I-







z
0
<-


2


Er
w



IL
40
1-4


I-
5
0*
41

Ww


a: I


8


tL
5


I
3



.. .




*0 4a





















cc
o 4*






L









0 *
0





,

I



c 0 ~ ( d *


a
w
UL

U.
z
0
--



zi_
-o







-r
0
z
z





i.-
i-

0





a:





-J
0








I-"
ZC~L

04^
m
(E -
Z



























i>-
P













I-



0














>0
CLC

0 :












w -



CLa
_J UJ


0-1
cc 0

z >-
0 >
a:
0


;-U.

0
LL c


40 4 t t') Cd .- 0



















0 0


0

0
0




I
0


to in


cm -


>


5*
C.

5


in
40 0

*' *


8

I L

I






z
0


0


'0



40


' I


PI'




5









Of the 9 units exhibiting rotation, 4 also shifted. It was possible that shift was a more

mature form of sensitization that followed rotation. In units which were examined over

several hours, 2 of 2 units that exhibited rotations subsequently manifested shifts of their

scatter plot field (See Figure 3-9).

Control tests indicted that qualitative changes could also be observed after saline

injection. In 1 of 8 cases, a qualitative improvement was observed in dynamic coding

capacity (Table 3-5). Thus, improved dynamic reactivity is possible as the result of saline

injection. However, the tendency towards acquired coding appeared to be greater for

carrageenan-induced inflammation (8 of 11 cases).

In summary, experiments were performed to characterize the response properties of

presumed nociceptors in normal and inflamed capsular tissue of the TMJ. Most TMJ

afferents preferred either movement in the horizontal or vertical plane. Their reactivity

functions were best described by power functions fit between dynamic force, force velocity

and response interval. In normal tissue, TMJ afferents had properties of nociceptors, in

that the activation threshold was at the extreme range of jaw movement, either in the

exponential portion or in the transition zone between the linear and exponential force-

movement curves that described jaw movement. It was predicted that TMJ nociceptors

would sensitize following exposure to carrageenan. To test this, populations of afferents

were sampled in previouslyly inflamed tissue and their responses compared to those of

afferent populations in normal tissue. Improved reactivity was observed, especially

reactivity to dynamic stimuli (dynamic force and force velocity). To avoid sampling errors

that could have biased these observations, units were characterized in acutely inflamed

tissue after being characterized in normal tissue. These afferents manifested similar

quantitative improvements in dynamic reactivity. Additionally, qualitative improvements of

coding capacity were observed. Acquisition of coding capacity following acute

inflammation suggested afferent sensitization.











TABLE 3-5
EFFECT OF SANE ON TRANSDUCING CAPACITY
UNIT PLANE PRE-CODING POST-CODING CODIN OTHER POST-TEST
SHIFT MOVEMENT SPONTAN.
ACTIVITY

I I-lP DYNAMIC FORCE DYNAMIC FORCE
FORCE VELOCITY FORCE VELOCITY NONE
STATIC FORCE STATIC FORCE


2 HP DYNAMIC FORCE DYNAMIC FORCE
FORCE VELOCITY YES NONE
STATIC FORCE STATIC FORCE


3 HP DYNAMIC FORCE DYNAMIC FORCE
STATIC FORCE STATIC FORCE NONE


4 HP DYNAMIC FORCE
FORCE VELOCITY
MOVEMENT NONE
STATIC FORCE
POSITION


5 HP NON-CODING NON-CODING NOE




6 HP DYNAMIC FORCE DYNAMIC FORCE
NONE



7 HP DYNAMIC FORCE DYNAMIC FORCE
STATIC FORCE STATIC FORCE NONE




8 liHP STATIC FORCE STATIC FORCE NONE


Note: Changes in reactivity for 8 units after exposure to saline. One of 8 units acquired coding ability
for force velocity. HP, horizontal plane.













CHAPTER 4


DISCUSSION



Afferents were isolated from the trigeminal ganglion and efforts were made to

determine whether they had nociceptive properties. Tests included quantification of

response range, identification of relevant stimulus variables and the tendency to sensitize.

Afferents that are nociceptors should transduce noxious stimuli (Sherrington, 1947), while

other joint afferents (proprioceptors) would be expected to be most reactive in the normal

range of motion, be bidirectionallly sensitive, and exhibit spontaneous activity (Burke et

aL, 1988; Macefield et al., 1990, Dorm et al., 1991).

The relationship between applied force and mandibular movement was examined to

determine the "noxious range" and compare it with the threshold of afferents. The noxious

range is likely to correspond to that portion of the force/jaw movement relationship where

forces increased exponentially with jaw movement. TMJ afferents were activated at or near

the beginning of the exponential phase and their firing became asymptotic in the exponential

phase. Thus, a preference for high forces (mean activation threshold = 20.6 + 1 1.ON) at

extremes of the mandibular range of motion (mean activation threshold = 13.7 5.4; mean

frequency asymptote = 49.6 + 20.7 N) suggested that these TMJ afferents were

nociceptors.

In contrast, a small group of units (n = 3) had responses to both jaw opening and

closing. Their properties are distinct from populations of afferents which may be important

in signalling tissue damage. Their activation thresholds (1.1 + 0.9 N) and range of

reactivity (1.1 0.9 to 10.4 + 4.1 N) are considerably different from that of nociceptors

described above, and respond chiefly to innocuous joint movement within the physiological









range of motion. Afferents with similar properties have been observed in the cat knee joint

(Dom, et al., 1991). In this study, large diameter afferents (Group II) responded

exclusively in the normal range of motion, were directionally sensitive, and demonstrated

no resting activity. Whereas activity in nociceptors is likely to contribute to joint pain,

these low threshold afferents may contribute to proprioception or sensations of pressure

(Burke et al., 1988; Macefield et al., 1990; see Appendix B).

In other laboratories, recordings have been made from group III and IV joint

afferents innervating the cat knee (Burgess and Clark, 1969; Clark, 1975; Schaible and

Schmidt, 1983b). Schaible and Schmidt (1983b) divided their joint afferents into 4

subgroups which were differentiated, in part, by their qualitative response to movement.

They defined nociceptors as those afferents that responded through, beyond or exclusively

beyond the normal range of joint motion. These afferents had a steep rise in discharge

frequency during forceful rotational movements, similar to those described previously

(Burgess and Clark, .1969; Coggeshall et al., 1983). These observations compare

favorably with ours, in that intense mechanical stimulation is needed to achieve a maximum

response. It is difficult to make detailed comparisons between the response properties of

afferents described by Schaible and Schmidt and others with those in our experiments, as

no attempt was made by them to quantify the properties of afferents other than local

pressure thresholds. Our experiments represent the first quantification of nociceptor

reactivity across a range of suprathreshold stimuli with verification of the relationship

between movement, response minima and maxima.

Early experiments created some confusion as to the basic properties of joint

afferents. Studies that have examined the activity of mechanoreceptors supplying cat joint

tissues (Burgess and Clark, 1969; Clark and Burgess, 1975; Rossi and Grigg, 1982)

primarily emphasized proprioception, and addressed the questions as to whether position

sense is coded by joint or muscle afferent fibers. They determined that joint afferents are

only responsive at the extremes of the movement range, seemingly ruling out any useful









role in proprioception and clouding the functional distinction between free nerve endings

(nociceptors) and well recognized encapsulated Golgi and Ruffini type end organs which

are found throughout the capsule (Thilander, 1961; Ishibashi, 1966). Yet, others have

reported that many afferents respond within the normal range movement (Klineberg et al.,

1971; Lund and Matthews, 1981; Schaible and Schmidt, 1983b; Dom et al., 1991).

Microneurographic studies appear to have resolved this confusion in favor of the findings

of Schmidt and company. These studies have examined the behavior of neuronal

populations that signal joint position and mechanosensibility in humans. Burke et al.,

(1988) recorded responses of joint, muscle and cutaneous mechanoreceptors associated

with finger joints to position and movement. Whereas most responded at the limits of joint

rotation, a few afferents responded within the normal range. It is unclear whether the

former group could code for pain, as pain was never induced by the movements, but the

latter group appears to be appropriate for proprioceptive function. The relatively small

population (3 of 16 cases) that provided a proprioceptive code was not dissimilar to our

own (3 of 36 cases). Microstimulation of similar groups of joint afferents innervating

finger joints indicated that innocuous sensations of movement or deep pressure were

evoked (Macefield, et al., 1990). Thus, a proprioceptive role for joint afferents seems to

be confirmed, although they may represent a small portion of the afferent population.


Sensitization of TMJ Nociceptors


Nociceptors in all preparations will sensitize following exposure to extreme heat

(Burgess and Perl, 1967; Beck et al., 1974; Fitzgerald and Lynn, 1977; Campbell et al.,

1979), mechanical forces (Reeh et al., 1987) or proinflammatory substances (Schaible and

Schmidt, 1988a; Berberich et al., 1988; Cooper et al., 1991), and these are reflected as

changes in central reactivity (Neugebauer and Schaible, 1990; see Appendix D). Sensitized

nociceptors in skin, muscles and joints have enhanced reactivity to thermal, mechanical









and/or chemical stimuli. In our experiments, where afferents were sampled from

previously inflamed capsular tissue or where characterized afferents were retested in acutely

inflamed tissue, a variety of forms of sensitization was found. These changes included a

significant decrease in mean activation threshold, shifts in stimulus response functions and

qualitative forms of sensitization. Additional changes were observed for afferents in the PI

condition (frequency asymptote and intercept) that were not replicated in the AI condition.

Recent experiments on cat knee joint nociceptors (Schaible and Schmidt, 1988b)

have also produced evidence of changes in reactivity following carrageenan injection. Over

half of the units that were classified as nociceptors in normal tissue responded to innocuous

joint movements in the normal range of motion after inflammation. In effect, such

enhanced reactivity, from extremes of movement to the normal range of motion indicates a

decreased activation threshold. In our experiments, decreases in activation threshold were

observed following carrageenean-induced inflammation. These decreased activation

thresholds fell within the linear phase of force-movement relationship.

In our study, qualitative forms of sensitization were also observed. These appeared

both in carrageenan treated nociceptors (8 of 11 cases), and infrequently in saline treated

controls (1/8). The nature of these qualitative changes can be best appreciated by

comparing scatter plots representing comparisons between preinflamed and postinflamed

reactivity. It is important to examine scatter plots because the acquisition of transducing

ability could have represented a distorted outcome. For example, outliers could have

rendered functions insignificant where they were otherwise significant. However,

examination of plots following carrageenan showed new relationships between force and

response interval and shifts to greater reactivity.

Two types of trends were observed: rotation and shift. Rotational trends were

indicated by a change in slopes which suggested increased proportional reactivity. A shift

to the left represented a decrease in activation threshold that was seen consistently in AI and

PI conditions for units with extant coding capacity. The shifts that were observed, were









less likely to be related to the development of qualitative changes than rotation. Rotation

represents a new functional reactivity in the form of a proportional response to imposed

forces. The basis of the qualitative changes could be either receptor based metabolic

alteration secondary to the generation of proinflammatory substances, or could result from

improved mechanical coupling between the receptor and the applied force.

Edema is a major component of inflammation that may impact on mechanical

coupling. Afferent terminals of small fibers that are located in the area of inflammation are

subject to an altered physical environment as a consequence of fluid extravasation. What

effect local changes from edema may have in the responsiveness of joint receptors is

unclear. In other preparations, edema may act to enhance or deter afferent responsiveness,

the direction of the effect being dependent on tissue properties that are peculiar to each site

(Cooper et al, 1990).

Acquisition of transducing capacity could be similar to observations of recruited

activity described by other laboratories. This "recruitment" of afferents was observed in

the cat knee joint preparation (Schaible and Schmidt, 1985, 1988b) and also in cutaneous

preparations (Handwerker et al., 1991). A distinction between our observations and those

of Schaible and Schmidt is that our sensitized afferents all have some pre-existing

reactivity. We did not observe recruitment as silent units which "wake up." Given this,

recruited activity and acquired coding may be distinct forms of sensitization. Our failure to

observe recruitment may be due to methodological constraints of our preparation. In our

experiments, extracellular recordings of neural activity in the trigeminal ganglion limits the

number of neurons that could be recorded at any site. Very few cell bodies are in the

functional range of the microelectrode. On the other hand, in teased fiber preparations such

as those used by Schaible, Schmidt and Handwerker, electrodes have essentially zero

impedance, therefore many more afferents can be recorded from simultaneously. This

makes it more likely for recruitment to be observed.









Factors Contributing to Sensitization



Mechanical sensitization of joint afferents, following inflammation, is likely to be

due to the production and release of endogenous substances. Injection of carrageenan

produces a rapid inflammation that has been characterized in a number of behavioral and

physiological preparations (Vinegar et al., 1987). Various factors may contribute to

sensitization of joint afferents during carrageenan-induced inflammation, including both

edema and the formation and release of algesics.

Changes in the response properties of joint afferents during inflammation have been

attributed to the actions of bradykinin, prostaglandins, and serotonin. These mediators

may alter mechanical responsivity of afferents, modify chemical responsivity, activate

afferents or a combination of the three. Changes in reactivity of small fibers in inflamed

tissue to bradykinin have been observed (Kanaka et al., 1985; He et al., 1990; Grubb et

al., 1991). Bradykinin activates most group III and IV units that responded to passive joint

movement. Bradykinin also activates afferents otherwise classified as "silent" (Heppleman

et al., 1987).

Joint afferents can also be activated by prostaglandin (E or I). Prostaglandins E1 or

E2 activated small afferents which responded to passive joint movement (Heppleman et al.,

1985; Schaible and Schmidt, 1988a), and modified mechanical responses of these

afferents. Following exposure to PGE2, most group III and some group IV units showed

enhanced joint movement reactivity and responded within the normal range of motion.

These PGE2-evoked increases in unit responsiveness paralleled those observed during

acute arthritis (Schaible and Schmidt, 1985, 1988a). Similar properties have been

prescribed for PGI2, and it has been suggested that PGI2 may be the dominant PG in joint

afferent sensitization (Birrell et al., 1991; McQueen et al., 1991).

Interactions between pro-inflammatory mediators are also likely to be important, but

they have not been determined in much detail. Prostaglandin E1 (Schaible, 1983) and









prostaglandin E2 (Grubb et al., 1991) have been shown to sensitize joint nociceptors to

bradykinin. This may impact both on movement sensitive and tonic forms of joint pain.

Serotonin is also important in nociceptor sensitization. The effect of 5-HT on

response characteristics of small afferent fibers have been studied in normal and inflamed

cat knee joints (Heppelman et al., 1987) and rat ankle joints (Birrell et al., 1990; Grubb et

al., 1988). Most group III and IV fibers that showed reactivity to cat knee movement were

also activated by 5-HT. Units recorded from normal rat ankle joints were excited by 5-HT

and showed enhanced activity in arthritic joints.

The consequences of carrageenan injections are very complex, and have not been

determined for joints specifically (Vinegar et al., 1987). One sequence of events that

occurs in skin, during the initial stage of inflammation (0-60 min) suggests a divergence

from mechanisms proposed for sensitization of joint afferents. Immediately after injection

of carrageenan, small amounts are absorbed by mast cells. The cytotoxic action of

carrageenan initiates the arachidonate metabolic cascade and mast cell degranulation.

Endoperoxides and serotonin are released. The endoperoxides are not believed to be

prostaglandins but are believed to be reactive intermediates, and bradykinin is not believed

to play a role. This stands in contrast to studies directly implicating bradykinin, PGE's and

PGI2 in afferent sensitization and at the very least implicates that additional factors may

play a role in carrageenan inflammation. The role of 5-HT in carrageenan induced arthritis

is also unclear. While 5-HT is released by carrageenan in rats, it is unlikely to be directly

released in other species due to the absence of 5-HT in mast cells. Serotonin may be more

important when vascular damage occurs (Zeller et al, 1983; Garcia-Leme, 1989). This

may, in fact, be more relevant to TMJ dysfunction.

Clearly, reactivity to a host of algesics is shared by most small diameter afferents

innervating joint tissue. Bradykinin, prostaglandins and perhaps 5-HT may play a role in

afferent sensitization in inflamed joints. However, in each instance, it is not clear what is

meant by normal range of motion and it is not clear to what extent shifts produced by





59



bradykinin, PGE2 or serotonin are the basis for changes produced with carrageenan or

other proinflammatory substances. The lack of detail in this respect is partly due to the

absence of quantification of stimulus and response variables. Quantification of neural

activity in respect to threshold, range and salient stimulus features (static or dynamic

reactivity) should permit a fuller evaluation of role played by these substances and others in

modification of joint nociceptor reactivity.














APPENDIX A


DEVELOPMENT AND MAINTENANCE OF TMJ PAIN


Many theories have been advanced to account for the production of TMJ pains.

(For reviews, see Deboever, 1973; Dubner et al., 1978, Yemm, 1979). TMJ dysfunction

may account not only for pain in the joint itself, but also for pain arising from associated

facial structures.


Intra-articular Origins of TMJ Pain


1) Pressure Theory

Pressure theories have been the most prevalent explanations for the production of

TMJ pain. These theories suggest that disruption of the normal articular mechanics may

impose abnormal pressures on structures of the TMJ. Costen (1934) postulated that facial

pain associated with the TMJ was a consequence of pressure on the auriculotemporal nerve

brought about by impaction of the condyle against the tympanic plate. Anatomically, this

theory is untenable since the auriculotemporal nerve ramifies before reaching the TMJ. The

articular branches of the auriculotemporal nerve arise at the medialposterior aspect of the

neck of the condyle and ascend into the soft tissue of the joint inferior to the tympanic plate.

The auriculotemporal nerve is vulnerable to pressure at its proximal course. The

auriculotemporal nerve arises from the posterior trunk of the mandibular nerve along the

medial surface of the lower belly of the lateral pterygoid muscle. Entrapment of the

auriculotemporal nerve in the lateral pterygoid muscle has been demonstrated in 5% of









cadaver specimens (Loughner et al., 1990). Pressure in the form of compression by a

hyperactive lateral pterygoid muscle may produce pain and/or paresthesia. Furthermore,

patients with an internal derangement of the TMJ often have accompanying spasticity of the

lateral pterygoid muscle.

Pressure on the posterior attachment tissues by the condyle could also cause TMJ

pain (Sicher, 1955). Posterior displacement of the condyle or spastic contracture of the

superior belly of the lateral pterygoid muscle may result in anterior displacement of the

disc. As a consequence, the posterior attachment tissues are positioned where forces

generated during mastication may produce local pressure on these highly vascularized and

innervated tissues.

Local joint pressure may activate mechanoreceptive nociceptors in connective tissue

such as the posterior attachment tissues. Inflammation could result from trauma due to a

sudden, intense impact or from repeated pressure. Endogenous algesics released in the

zone of inflammation may sensitize joint nociceptors by decreasing threshold and

increasing response to mechanical stimuli (Heppelman et al., 1985).

A common procedure thought to discern anterior displacement of the disc is the

impact-loading test. Manual loading applied in an anterior-superior direction at the

posterior portion of the body of the mandible impacts the condyle into the articular fossa.

If pain is evoked, it is thought that the disc is displaced and the condyle is impacting onto

the posterior attachment tissues. However, disc displacement may exist without impact

loading pain. This is assessed by imaging techniques, such as arthrography or magnetic

resonsance imaging. Therefore, direct pressure on the disc or posterior attachment tissues

is not sufficient to explain TMJ pain in all cases.

A corollary notion related to Sicher's theory was advanced by Frost (1968) to

explain some patterns of joint pain. Frost hypothesized that bone pain may be evoked by

the pressure of two articulating, irregular, joint surfaces in the absence of an intact

interposing disc. Presumably, the irregular bony surfaces represent osteoarthritic changes.









Other investigators (Kreutziger and Mahan, 1975) have included arthroses of bony

structures as causative factors in TMJ pain. Besides the degenerative processes that may

occur in the bony anatomy of the TMJ, degenerative changes of the villi of the synovium

have been claimed to be causally related to TMJ pain (Olson, 1969). Supposedly,

disruption of the secretary function of the villi can disturb proper nutrition and lubrication

of the TMJ resulting in TMJ dysfunction.

2) Tension Theory
Tension theories of TMJ pain can be divided into static and dynamic. Static state

involves a resting position of the mandible. Dynamic state involves active or passive jaw

movement Anteriormedial displacement of the disc may generate static tension on the

posterior and lateral aspects of the capsule (Dubner et al., 1978) where the greatest member

of mechanoreceptors are located (Zimny, 1988). It is possible that resting tension will

produce prolonged discharge in TMJ nociceptors. If the anteriormedial displacement of the

"disc is advanced, there is a release of the disc attachment at the lateral pole of the condyle

which untethers the disc from the lateral aspect of the TMJ. In this case, the posterior

attachment tissues are vulnerable to tension from a disc displacement during static

conditions.

Patients suffering from pain in the TMJ often demonstrate symptoms of capsulitis;

this included pain on local palpation or pain during small movements within the normal

range of motion. Inflammatory processes in the TMJ may initiate or potentiate TMJ pain

(Dubner et al., 1978). Dynamic theories propose that pain is produced during jaw

movement by excitation of functionally distinct types of mechanoreceptors in the capsule

and posterior attachment tissues. Activation may be enhanced by capsulitis. Zimmy

(1988) advanced a conceptual model of TMJ pain during jaw movement. As the condyle

rotates and translates freely through intermediate degrees of angle, the capsule and posterior

attachment tissues are deformed. Such deformation is normally insufficient to excite

mechanoreceptors. As the condyle approaches its border movement, the capsule and









posterior attachment tissues become taut. Mechanoreceptors in the capsule and posterior

attachment tissues are activated by the resultant tension. Thus, these mechanoreceptors

appear to act as detectors that sense the safe limits of joint movement. Supposedly, jaw

movement beyond safe limits can represent real or potential tissue damaging stimuli. If

these mechanoreceptors are nociceptors, they may be sensitized following capsulitis so that

small movements in the functional range of motion produce pain. Clark and Burgess

(1975) suggested that large, myelinated afferents signal the pressure felt as the extremes of

joint movement are reached. These investigators demonstrated that slowly adapting

receptors in the capsule were excited only at extremes of flexion and extension of the knee

joint of the cat (Burgess and Clark, 1969). On the other hand, Schaible and Schmidt

(1983b) attribute a similar detection role for small fibers which are active only at the

borders of the joint movement.

Following capsulitis, the tension required to produce activation of nociceptors

might be greatly reduced. Animals studies in the cat knee joint (Schaible and Schmidt,

1988b) and rat ankle joint (Guilbaud, 1988) have demonstrated increased responsiveness

of group III and IV afferents to innocuous mechanical stimulation after induction of acute

inflammation. The relationship between nociceptor discharge and capsular tension is yet to

be demonstrated.

The reactivity of mechanoreceptors in the capsule of the cat knee have been used to

assess tension produced by joint movement (Grigg and Hoffman, 1989; Hoffman and

Grigg, 1989). The frequency of neurons with receptive fields in the posterior capsule

showed a linear relationship with tension applied (knee extension) at the area of the

receptive field. The posterior joint capsule was minimally loaded even during

hyperextension movements.








Extra-articular Theories of TMJ pain



Distinct theories of the origin of extra-articular pain have arisen, and include the

myofascial pain theory and referred pain theory.

1) Myofascial Pain Theory

The myofascial pain theory postulates chronic hyperactivity in head and neck

musculature as a causative agent in extra-articular pain. The myofascial pain theory

suggests that the chronicity of muscle pain is due to the "trigger point" (Travell, 1976).

The trigger point is defined as a localized area of chronic inflammation of fibrous

connective tissue in muscle. Histological examination of biopsy material from trigger

points in patients suffering from interstitial myofibrositis reveal degranulated mast cells and

platelet clots (Awad, 1973). The author proposes a hypothesis suggesting that trauma-

induced extravasation of platelets leads to release of 5-HT which subsequently results in

edema and vasoconstriction. The release of such endogenous algesics in muscle may play

a role in the pain and tenderness associated with trigger points. For example, activation of

group IV afferent neurons in cat muscle have been demonstrated by close arterial injection

of bradykinin, 5-HT or histamine (Mense and Schmidt, 1974). Serotonin may also

sensitize mucosal nociceptors to mechanical stimuli (Friedman et al., 1988; Cooper et aL,

1991).

Muscle pain and the reduction of pain free jaw movement are frequent complaints of

patients with facial pain. The limitation of jaw movement may involve the modulation of

the reflex control of masticatory muscles (Klineberg, 1971; Dubner et al., 1978).

Greenfield and Wyke (1966) studied the role of TMJ afferents in reflexes that modify the

activity of muscles of mastication. The experimental procedure involved detaching all the

supra- and infra-mandibular muscles at their insertions from one half of the cat mandible.

Movement of the ipsilateral half of the mandible resulted in activation or inhibition of









dissected muscles. Similar responses were produced in the detatched muscles by direct

tension on the capsule after condylectomy.

Jaw movement has been shown to modulate the firing pattern of alpha motoneurons

in the trigeminal motor nucleus in cats (Kawamura et al., 1967). These investigators

demonstrated changes in the discharge rate of motor neurons of the masseter muscle in

response to passive rotation of the ipsilateral condyle. Rotating the condyle to an open

position evoked an increase in firing pattern of masseter motoneurons. A closed position

evoked a decrease in firing pattern.

Abe and Kawamura (1973) recorded efferent discharges from branches of the

mandibular nerve which innervate the masseter and digastric muscles. Ipsilateral condylar

rotation to an open position inhibited alpha-motor units in the masseteric nerve. Local

mechanical stimulation of the posterior part of the TMJ capsule inhibited the discharge

pattern of masseteric neurons and facilitated motor fibers innervating the digastric muscle.

It is not surprising that afferent activity from mechanoreceptors in the TMJ modulate

efferent activity in masticatory muscles in order to coordinate jaw movement. The presence

of chronic activation of TMJ afferents may induce chronic reflex activation of facial

musculature. Chronic activation may produce muscle pain that limits movement.

Clinically, bruxism has been associated with muscle hyperactivity and TMJ

dysfunction. Bruxism is defined as a nonfunctional voluntary or involuntary mandibular

activity featuring habitual grinding or clenching of the teeth (Kawamura, 1980). Chronic,

repetitive mandibular movement seen in bruxism is thought to deliver excessive reaction

forces to the TMJ which could produce local inflammatory changes. Putative causes of

bruxism include occlusal stress, CNS and hereditary factors.

2) Referred pain theory

Pain may be referred to the TMJ from other head and neck regions (Bell, 1969).

Travell (1976) suggested that continuous hyperactivity of muscles from the head and neck

may refer pain to the TMJ. The explanations for the origin of referred pain have largely









ruled out nerve branching in the periphery as a source of referred pain. The convergence

theory of referred pain invokes a central locus of action in the superficial layers of the

medullary dorsal horn. Noxious input from different sources converge on transmission

cells responding to noxious input (for review, see Fields, 1987). Pain may be mistakenly

mislocated to an area distant from the site of origin due to expectation or previous

experience with pain in the referred site.

In most cases, TMJ pain is not restricted to the joint but is also associated with

extra-articular regions. The most common areas are ipsilateral eyeball, superciliary ridge,

deep masseter muscle, and angle of mandible (Mahan and Ailing, 1991). The

craniovertebral junction and upper neck are other referral sites of TMJ pain (Schellhas et

al., 1989).













APPENDIX B


PSYCHOPHYSICS OF TMJ DISORDERS



Pain and mandibular dysfunction are two.common manifestations of TMJ

disorders. Pain can occur when the mandible is at rest or during function. Disordered

movement of the mandible is a sign of mandibular dysfunction (i.e. the inability to

masticate properly). Patients suffering from TMJ disorders often complain of pain while

chewing. Whether mandibular dysfuntion is directly or indirectly due to TMJ disorder or

due uniquely to pain that results from jaw movement is unclear.
Chewing efficiency, jaw movement, biting force and perception of jaw position are

quantifiable features of mastication. A review of the literature reveals studies which have

been directed at investigating the relationship between TMJ disorders and mastication.

There has been little psychophysical experimentation involved in quantifying pain

associated with mandibular dysfunction. Even though most of these studies pertain to

treatment efficacy, understanding the relationship between TMJ disorders and mastication

may be valuable in understanding the relationship between pain and TMJ disorders.


Chewing Efficiency


Typically, the clinical appraisal of individual's chewing efficiency is limited to a

subjective report provided by the patient. Early investigators (Christiansen, 1922;

Dahlberg, 1942) developed the technique of using a series of sieves with different mesh

sizes and test materials for chewing in order to estimate chewing efficiency. A simple









method that can be used clinically to assess chewing efficiency was developed by Loos

(1963) and modified by Carlsson and Helkimo (1972). Test material is chewed for 10, 20

or 40 seconds.

Masticated test materials are then measured in millimeters. Chewing efficiency is rated

according to the size of the material and number of seconds of chewing.

A more elaborate, standardized measure of chewing efficiency, called the

Coefficient of Masticatory Performance, was developed by Manly and Braley (1950). The

Coefficient of Masticatory Performance was defined as the percentage of chewed peanuts

that pass through a 10-mesh screen sieve after being subjected to 20 masticatory strokes.

Rominger and Rugh (1986), utilizing this test, compared the chewing performance of a

group of patients with TMJ dysfunction to a group of normal controls. The patient group

had significantly lower coefficient scores than the normal control group. Lemke et al.

(1987) investigated the level of masticatory efficiency of post-TMJ surgical patients several

years after treatment. These investigators found that the post-surgery group had

significantly lower chewing efficiency scores than a group of non-treated non-

dysfunctional patients.

Taken together, these studies indicate a negative relationship between chewing

efficiency and TMJ disorders. Whether the inability to chew efficiently is due to the TMJ

pain or associated muscular pain is uncertain.


Jaw Movement



The sensory apparatus of the TMJ is thought to play a role in the synchronizing

mechanisms involved with the control of jaw movement (Sicher, 1955). The influences of

TMJ afferents on functional jaw movements has been studied in both normal subjects and

patients suffering from TMJ disorders. In normal subjects, Posselt and Thilander (1965)

anesthetized the lateral part of the TMJ capsule. They observed an increase in lateral









movement and maximum opening. The authors suggested that the lateral ligament may be

associated with a protective mechanism during extreme opening. Klineberg (1980) also

anesthetized the TMJs of normal subjects and observed an increase in size of the envelope

of function (range of motion), and suggested that mechanoreceptors of the TMJ play a

modulatory role in motor control of the mandible.

One measure of jaw movement is maximum range of motion (ROM). The

measurement is made for opening, lateral and protrusive excursions. An assessment of

ROM for normal subjects has been reported by Posselt (1968). For normal subjects (adult

males) maximum opening averages 50-60 mm; hinge opening, 20-25 mm; lateral excursion

from midline, 10 mm; and protrusive excursion, 10 mm. Anaesthesization of the lateral

capusle of the TMJ increased hinge opening and protrusion by 10-15% (Posselt and

Thilander, 1965). For patients suffering from TMJ disorders, ROM is generally limited

compared to normal subjects. Helkimo (1974) devised a pain-dysfunction index which

rates severity of mandibular function in regard to pain report for jaw movement and TMJ

palpation. This index was modified by Zarb and Carlsson (1988). Normal ROM was >39

mm for opening and >6 mm for lateral excursions. For patients, ROM was 30-39 mm for

opening and 4-6 mm for lateral excursions. Patients with this amount of impairment

reported pain during at least one jaw movement and pain on palpation of the TMJ's. For

other patients, ROM was <30 mm for opening and <4 mm for lateral excursions. Patients

with this amount of impairment reported during 2 or more jaw movements and pain on

palpation of the TMJ's.

Many clinicians observe a decrease in ROM for patients suffering from TMJ

disorders. These patients are often hesitant to open their mouth maximally because of real

or expected pain. As a result, a measure of ROM may result in a value less than the patient

is actually capable of producing. Therefore, other components of mastication may be more

likely to be reliable measures of jaw function. Moreover, the movements of the mandible









that are made during mastication are well within the border movements of the mandible as

defined by the ROM (Okeson, 1989).


Jaw Position



One feature of mandibular function is the ability to control the position of the

mandible within the limits of one's maximum ROM. An important question in the literature

dealing with proprioception is whether position sense is signalled by joint or muscle

afferents. Part of the literature implicates joint afferents (For reviews see Rose and

Mountcastle, 1959; Skoglund, 1973; Dubner et al., 1978); another part of the literature

implicates muscle afferents (for review, see Burgess et al., 1982). It is reasonable to

explain joint position sense in terms of both joint and muscle afferent input.

Jaw position sense has been studied in both normal subjects and patients suffering

from TMJ disorders. Thilander (1961) assessed perception of mandibular position before

and after unilateral or bilateral anesthetization of the TMJ in normal subjects. The results

showed that subjects had more difficulty returning to an initial jaw position, following

anesthetization. Larsson and Thilander (1964) explored the relationship between position

sense and local mechanical pressure applied to the TMJ in normal subjects. The authors

found that position sense was not influenced by local pressure. When the local pressure

was reported to be painful, again position sense was not affected. However, when the

joint was locally anesthetized, jaw position sense deteriorated. If then, local pressure was

applied to the joint during the course of anesthesia, position sense returned to normal

levels. Presumably, local pressure recruited muscle afferents. Input from muscle

mechanoreceptors may have provided the sensory feedback necessary to restore position

sense. The authors reported that ansethetization of the TMJ was performed by infiltration

into the skin and tissues lateral to the capsule. This technique may have not anesthetized

the TMJ completely.








Interdental thickness discrimination is one method that has been used to analyze the

position sense of the mandible. Interdental thickness discrimination is the ability to detect

differences in the thickness of test materials placed between opposing teeth. Caffese et al.,

(1973) studied the possible influence of TMJ receptors in tactile occlusal perception by

assessing minimal thickness detection between opposing teeth before and after

anesthetization of the TMJ. These authors found significant differences in threshold

detection in the anesthetized joint compared to the normal joint. On the other hand, Siirila

and Laine (1972) obtained different results from their study of interdental thickness

discrimination. Anesthetization of TMJ did not significantly diminish subjects' ability to

detect differences in thickness between teeth at various degrees of mouth opening.

Morimoto and Kawamura (1978) obtained similar results and suggested that the muscle

spindles in the muscles of mastication are mainly responsible for size discrimination.

If mechanoreceptors of the TMJ play a role in jaw position sense, their influence

maybe modulated by TMJ disorders. Ransjo and Thilander (1963) studied position sense

of the mandible in patients with TMJ disorders which were thought to be of myogenic

origin or due to functional malocclusion. The authors assumed that functional

malocclusion caused impairment of TMJ receptors. Position sense was found to be poorer

for the malocclusion group compared to the normal group. Position sense of the myogenic

group was normal. Amelioration of TMJ symptoms after treatment was accompanied by

improvement in the perception of jaw position. This study is in accord with the previous

work of Thilander (1961) suggesting that TMJ mechanoreceptors play a major role in the

ability to perceive position sense of the mandible.

A more recent study by van Willigen et al., (1986) investigated perception of jaw

position in subjects with symptoms of TMJ disorders. They found that all subjects with

craniomandibular dysfunction demonstrated greater mismatches of jaw position compared

with normal subjects. On the other hand, other studies present contrasting evidence

(Christensen and Troest, 1975; Morimoto and Kawamura, 1978; Broekhuijsen and van









Willigen, 1983; van der Berghe et al., 1987). Morimoto and Kawamura (1978) tested

interdental thickness discrimination in 8 patients with unilateral or bilateral mandibular

condyles that were either surgically mobilized or removed. The authors failed to observe

any lessened capacity to discriminate thickness. These authors concluded that TMJ

receptors were not essential to size discrimination ability.

Broekhuijsen and van Willigen (1983) studied position sense of the mandible in

normal subjects before and after injection of the TMJ capsule bilaterally with local

anesthetic. Perception of mandibular position was unaffected by anesthetization of the

TMJs. Van den Berghe et al., (1987) assessed position sense of the mandible in patients

suffering from TMJ disorders before and after treatment. Patients' abilities to determine

jaw position were unchanged after successful treatment.

The relative unimportance of joint afferents in assessing position and movement is

noted for a variety of joints. In the human knee joint, injection of local anesthetic into the

joint does not deteriorate the subjects' ability to detect passive angular displacements of five

degrees (Clark et al, 1979). Moreover, total joint replacement of the hip produces little

impairment of sensations associated with joint movement (Grigg et al, 1973).

In summary, the ability to control the position of the jaw is an important aspect of

mandibular function. Position sense has been studied in both normal subjects and patients

suffering from TMJ disorders. The preponderance of the evidence suggests that sensory

input from the TMJ contributes little information in controlling the position of the mandible.


Bite Force


Different parameters of bite force, such as maximum bite force, submaximal bite

force and bite force discrimination have been investigated in an effort to assess mandibular

function.








Maximum Bite Force

The maximum bite force is the measure of the greatest force which an individual can

produce by biting on an instrument containing a force transducer. The maximal level of

bite force that an individual is capable of producing is rarely required or utilized during

chewing. Therefore, the relevance of maximum bite force as a measure of mandibular

function is uncertain. Also, a measure of maximum bite force may result in a value less

than an individual is capable or willing to produce. Clark et al., (1984) reported that the

maximum bite force of normal subjects is limited by pain tolerance. Patients suffering from

TMJ disorder may be incapable of producing maximum bite force due to the direct effect of

mandibular dysfunction or unwillingness to bite normally due to the indirect effect of pain.

Such may be the case in several studies that assessed maximum bite force in patients

suffering from TMJ disorders (Helkimo et al., 1975; Sheikolelam et al., 1982). Helkimo

et al. (1975) reported that maximum bite force was lower in a patient group with TMJ

disorders than for a control group. Patients' level of maximum bite force increased

following treatment. Sheikolelam et al., (1982) observed that maximum bite force was

significantly less for their patients with TMJ disorders than for normal controls.


Submaximal Bite Force

One important component of the normal chewing cycle is the individual's ability to

produce and control relatively low levels of biting force. Submaximal bite force is the

measure of different levels of biting force below maximum bite force. It seems reasonable

to suggest that impaired mandibular function may result in impairment of the ability to

control submaximal bite force. Several studies have investigated submaximal bite force in

patients with mandibular dysfunction (Helkimo et al., 1975; Agerberg, 1988). Bite force

was registered with different types of bite beams incorporating a force transducer. A

common feature of patient populations was pain in the TMJ. These studies reported that









patients had lower values for submaximal bite force (3-25 kg) compared to control subjects

(5-50 kg).

Bite Force Discrimination


In order for an individual to control the different levels of bite force, he/she must be

able to discriminate differences in the intensity of their own biting force. The role of TMJ

mechanoreceptors in bite force discrimination has been investigated in normal subjects

(Williams et al., 1984; Williams et al., 1989). Bite force was assessed before, during and

after anesthetization of the TMJ by auriculotemporal nerve block (Williams et al., 1984) or

infiltration into the superior joint cavity (Williams et al., 1989). These studies suggest that

the TMJ mechanoreceptors do not appear to be responsible for discrimination of bite force.

In summary, maximum bite force, submaximal bite force and bite force

discrimination are three commonly used measures of mandibular function. Studies

assessing maximum and submaximal bite force indicate that patients suffering from TMJ

disorders have decreased bite force strength. The ability to discriminate bite force levels

has not been studied in patients having TMJ disorders. Neither maximum bite force,

submaximal bite force nor bite force discrimination measures have provided clear

information on the relationship between TMJ sensory input and mandibular function.













APPENDIX C


NEUROANATOMY OF THE TMJ


Until Thilander's (1961) comprehensive, anatomical and histological study of the

innervation of the human TMJ, research on innervation of the TMJ involved macroscopic

studies. Riidinger (1857) observed that the auriculotemporal nerve coursed to the posterior

aspect of the joint where three or four articular branches entered the capsule. The anterior

aspect of the capsule received one or two twigs derived from the masseteric and deep

temporal nerves. Textbooks describing the TMJ generally accepted Riidinger's findings.

Moreover, other anatomists (Guerrier and Bolonyi, 1948; Baumann, 1951; Hromada,

1960) are in agreement with the neural topography described by Riidinger but for the

possible exception of the contribution of the facial nerve.

Some sensory fibers travel in the facial nerve, which is otherwise devoted to motor

function. These afferent fibers constitute about 15% of the total nerve fiber population

(Foley, 1960). Guerrier and Bolonyi (1948) reported that branches of the facial nerve

always terminated in the lateral surface of the TMJ. Hromada (1960) stated that a twig

from the facial nerve passed into the lateral aspect of the capsule in about half of his

dissected specimens. On the other hand, Thilander (1961) reported that, immediately

lateral to the capsule, a twig from the facial nerves courses in an anterior direction but does

not ramify into the joint. Kerr (1962) also reported that afferents of the facial nerve

terminate only in the dermatomes of Cl and C2 in the cat. Our observations (unpublished)

suggest that the facial nerve does not innervate the TMJ of the human. However, the facial









nerve was seen to communicate with the auriculotemporal nerve distal to the TMJ.

In contrast to the innervation of the human TMJ, the facial nerve of the goat

communicates with the auriculotemporal nerve proximal to the articular branches. The

functional implications of this topography are unclear. Some afferents from the TMJ may

travel centrally in the facial nerve or afferents in the facial nerve travel in the

auriculotemporal nerve.

Thilander (1961) reported that the auriculotemporal nerve is the major sensory

nerve of the TMJ. The branches from the auriculotemporal nerve innervate the posterior,

medial, and lateral aspect of the anterior capsule. These branches enter the inferior aspect

of the capsule along the neck of the condyle. This observation is not in accord with

Rudinger (1857) who reported that the articular branches of the auriculotemporal nerve

passed between the condyle and the pars tympanica. However, a recent study (Loughner et

al., 1990) confirms Thilander's report. In this laboratory, macroscopic dissections of the

TMJ of the goat (unpublished) show a similar course of the auriculotemporal nerve and

distribution of articular branches.

According to Hilton's Law (1879), nerves that innervate the muscles that move a

particular joint also contribute to innervation of the joint. Surprisingly, that contribution is

only about 30% of the total TMJ innervation (Thilander, 1961). Twigs from the masseteric

nerve innervate parts of the anterior capsule and anterior aspect of the medial capsule. The

posterior deep temporal nerve innervate the lateral anterior aspect of the capsule. The

auriculotemporal nerve supplies the remaining 70% of the TMJ.

Thilander (1961) estimated that the human TMJ is innervated by an average of 1500

peripheral afferent fibers. The auriculotemporal nerve contributes two thirds of the total

amount The greatest number of the articular fibers have a diameter between 1 and 2 gm

and probably terminate as free endings in the capsule, posterior attachment tissues and

adventitia of blood vessels associated with the joint. Other afferent fibers, though fewer in

number, and ranging from 6 to 11 m, terminate in complex neural endings.









The posterior attachment tissues contain a rich plexus of nerve fibers from the

auriculotemporal nerve. Hall et al., (1985) identified and quantified the types of nerve

fibers in the posterior attachment tissues. The authors found that the majority of axons

were unmyelinated. The lateral portion of the posterior attachment tissues contain

significantly higher percentage of unmyelinated fibers. Other investigators (Weinmann and

Sicher, 1951; Sicher, 1955) have reported abundant innervation in association with blood

vessels in the loose connective tissue between the posteror border of the articular disc and

the capsule.

A few nerve fibers have been observed in the synovial membrane of the human

TMJ (Thilander, 1961; Ishibashi, 1966). Ishibashi, (1974) found free nerve endings and

glomerular endings beneath the superficial cell layer of the synovial membrane located on

the posterior aspect of the TMJ. Bemrnick (1962) described encapsulated end bulbs in the

synovial folds in the TMJ of the rat. These observations appear to contradict those who

believe that the synovial tissue serves no sensory function (Olson, 1969). Although the

nerve endings in the synovial membrane have been reported to be closely associated with

blood vessels (Hagen-Torn, 1882; Davis, 1945), unmyelinated nerve fibers have been

described in the synovium separated from blood vessels (Kellgren and Samuel, 1950;

Rossie, 1950). Kawamura et al. (1967) found Golgi-type endings in the synovial tissue

layer of the TMJ of the cat but not in the synovial membrane itself. Whether the synovial

membrane is innervated or not, the justapositon of afferent fibers allows sensibility in the

territory of the synovium.

As reported by Thilander (1961), the articular disc contained no nerve endings.

However, other investigators (Ishibashi, 1966; Schmid, 1969; Zimny and St. Onge, 1987)

described articular nerves innervating the periphery of the disc in the human TMJ. Similar

results were reported in the monkey (Keller and Moffet, 1968), in the rat (Bemick, 1962)

and in the mouse (Frommer and Monroe, 1966). These studies are consistent with

histological studies of fetal material. Kitamura (1974) reported that, in the 5th month in









utero, nerve fibers were observed in the disc; further development was accompanied by a

reduction in disc innervation. Corroboration of the influence of growth on disc innervation

came from Hromada (1960), who observed free nerve endings in the disc of various

animals; at older ages such endings were limited to the periphery of the disc. By

comparison, the nerve supply to the meniscus of the human knee remains unclear (For

review see Zimny, 1988). The most recent study (Albright and Zimny, 1987) reports that

free and complex nerve endings were observed in outer and middle one-third of the

meniscus; whereas, no innervation was found in the inner one-third.

Innervation of the TMJ of the monkey and cat resemble that of the human joint

(Franks, 1965; Keller and Moffett, 1968). Branches from the deep temporal and

masseteric nerves supply the anterior capsule, and the articular branches of the

auriculotemporal nerve supply the TMJ posteriorly. Preliminary macrodissection of the

nerves innervating the TMJ of one goat was performed in this laboratory. The

auriculotemporal nerve gave off three articular branches along the posterior aspect of the

neck of the condyle. The articular branches ascended into the fatty and fibrous articular

tissue. The termination of the temporal branch of the auriculotemporal nerve appeared in

the facial skin inferior to the zygomatic arch. The masseteric nerve was observed to arise

from the anterior trunk of the mandibular nerve, course lateral along the anterior border of

the capsule of the TMJ and terminate in the superficial and deep masseter muscle. Twigs

from the masseteric nerve appeared to enter the anteriorlateral aspect of the capsule.

Tissues surrounding articulations possess fewer types of receptors than muscle or

cutaneous tissue. Thilander (1961) described four types of neural endings in the human

TMJ: Ruffini-like receptors, modified Pacinian endings, Golgi tendon organs and free

nerve endings. Ishibashi (1966) also reported the presence of free nerve endings and

complex terminal expansions in the human TMJ. Keller and Moffett (1968) reported free

and complex nerve endings in the TMJ of monkeys. Many investigators have identified








free and complex nerve endings in the TMJ of the cat (Kawamura and Majima, 1964;

Greenfield and Wyke, 1966; Kawamura et al., 1967; Wyke, 1967; Klineberg, 1971).

The three types of complicated nerve endings were sparsely distributed compared to

the free nerve endings. The most common complex ending is the Ruffini ending. Ruffini-

like endings were encountered in the lateral and posteriorlateral aspect of the capsule in

humans (Thilander, 1961). Griffin and Harris (1975) described a thinly unencapsulated

corpuscle located close to the periosteum of the neck of the condyle and in the lateral and

medial aspects of the capsule fat pads. These receptors may have been similar to the

Ruffini endings described by Klineberg (1971) in the TMJ of the cat. The Ruffini-type,

low theshold mechanoreceptors were once thought to signal joint position (Boyd and

Roberts, 1953; Boyd, 1954; Cohen, 1955; Skoglund, 1956, 1973). However, other

investigators (Burgess and Clark, 1969; McCall et al., 1974; Clark and Burgess, 1975;

Clark, 1975; Grigg, 1975; Grigg and Greenspan, 1977) have shown that Ruffini endings

are seldom active at intermediate positions of the cat knee joint. They are responsive to

extremes of joint movement. Consequently, these receptors are unlikely the peripheral

neural substrate that code for position sense. Instead, Ruffini endings appear to signal the

torque produced when the joint is extended and/or rotated at the limit of the range of

motion. (Grigg et al., 1982a).

Pacinian endings were found in the lateral aspect of the capsule (Thilander, 1961).

Paciniform-type end organs in cats have a similar distribution as Ruffini endings but less

dense (Kawamura et al., 1967; Klineberg, 1971). Golgi tendon organs were located in the

temporomandibular ligament and anterior capsule (Thilander, 1961; Griffin et al., 1965).

In the anterior capsule, the Golgi end organ was in series with the extrafusal fibers of the

lateral pterygoid muscle and tendon fasciculi of the capsular tissue or the fibrous tissue of

the pes menisci. In the temporomandibular ligament the Golgi end organ was thought to be

in series with fibers of the deep masseter muscle. Both Paciniform endings and Golgi

tendon organs are thought to respond like comparable endings found in cutaneous and









muscle tissue (Willis and Coggeshall, 1978). However, Grigg et al., (1982b) showed that

in the cat knee, Golgi tendon organs responded to local pressure applied perpendicular to

the capsule but not to tension applied parallel to the capsule.

Free nerve endings are the most abundant receptor in the capsule of the human

TMJ. Presumably, free nerve endings derived from myelinated afferents subserve fast pain

sensibility and nociceptive somatic reflexes; free nerve endings derived from unmyelinated

nerve fibers may subserve slow pain sensibility or innervate vascular elements. Free

ending receptor specializations have yet to be defined. In sympathectomized cats, sensory

nonmedullated plexuses have been found in the inner layers of the fibrous capsule, adjacent

synovial tissues and adventitia of blood vessels in the knee joint of cat (Samuel, 1952).

Free endings have been described in the joint capsule, joint ligaments and periarticular fat

pads (Greenfield and Wyke, 1966). Similar results have been reported by Freeman and

Wyke (1967). Free endings characteristic of group III and IV afferents have been

described in the Archilles tendon of the cat as well (Andres et al., 1980). Some of these

free endings may be nociceptors.

Although there are great variations in the morphology of the TMJ between species,

the innervation of the TMJ is quite similar (Kawamura, 1980). The distribution and

appearance of receptors may vary with age and species (Polacek, 1966), but the types of

receptors are consonant across species including human.

In summary, the mammalian TMJ contains four types of receptors (for review see

Skoglund, 1973; Zimny, 1988). The Ruffini and Paciniform mechanoreceptors are

innervated by medium-sized (AB) myelinated fibers, the Golgi tendon organ by large (Aa)

myelinated fibers and the free endings by fine myelinated and unmyelinated afferents.













APPENDIX D


CENTRAL REPRESENTATION OF TMJ AFFERENTS


The central distribution of TMJ afferents has been studied in the trigeminal nuclear

complex. The main sensory nuclei and the dorsal part of the rostral spinal nucleus were

found to contain neurons responsive to isolated condylar movement or pressure applied to

the joint capsule of the cat (Kawamura aand Majima, 1964; Kawamura et al., 1967). The

authors reported simple, qualitative findings describing three types of responses to

ipsilateral condylar rotation (rapidly adapting, slowly adapting and on-off types) and two

types of responses to mechanical stimulation (rapidly adapting and slowly adapting types).

In addition, response patterns of neurons in the trigeminal motor nucleus were recorded

during condylar movements or mechanical stimulation of the joint capsule. As with

responses from the sensory nuclei, no quantification of activity was attempted. Procaine

infiltration into the joint capsule abolished neural responses. Another qualitative study

reported unit activity found in the trigeminal main sensory nucleus and nucleus oralis of the

cat that responded to jaw rotation or pressure applied to the ipsilateral TMJ (Sessle and

Greenwood, 1976). One of these units was antidromically activated from the thalamus.

Other direct projections from the superficial layers of the medullary dorsal horn to the

thalamus have been demonstrated anatomically (Craig and Burton, 1981) and

electrophysiologically (Dostrovsky and Broton, 1985).

The caudal aspects of the trigeminal nucleus complex were found to contain

neurons that are driven by electrical, mechanical, and algesic chemical stimuli applied to the

TMJ (Broton and Sessle, 1988). Neurons in the subnucleus caudalis were first classified









on the basis of their responsiveness to mechanical stimulation to the skin: low-threshold

mechanoreceptors (LTM), wide dynamic range (WDR) or nociceptive specified (NS).

WDR and NS neurons responded maximally to algesic chemical and intense mechanical

stimulation applied to the joint. TMJ stimulation consisted of local probing or extreme jaw

opening. The majority of single units with TMJ input received convergent input from

cutaneous afferents. Kojima (1990) also found convergence patterns of afferent input from

the TMJ and muscle in the subnucleus caudalis. Most of the units tested were responsive

to mechanical and thermal stimulation of both the TMJ and the masseter muscle.

Extensive convergence of cutaneous and muscle inputs on second-order neurons

that respond to articular movement have been described in the cat knee (Schaible et al.,

1986). In some neurons, maximal responses to joint movement occurred with forced

extension, inward rotation or outward rotation. Other neurons were excited only by

noxious movement. Joint movement was considered noxious if the movement was made

forceably. No quantification of forces was attempted. This report compares favorably

with the effects of TMJ stimulation in neurons in the medullary dorsal horn (Broton and

Sessle, 1988), in that most second order neurons that respond to noxious joint simulation

also receive convergent input from skin and/or muscle. Such central convergence may be

important in explaining the tenderness of skin and/or muscles near the joint that are often

observed clinically when the joint is painful.

Changes in responsiveness in spinal neurons have been investigated during the

development of acute arthritis in the cat knee jont (Neugebauer and Schaible, 1990). All

neurons tested with joint input showed enhanced responsiveness to joint flexion after

induction of inflammation.

Other electrophysiological experiments performed in polyarthritic rats demonstrated

nociceptive inputs from chronically inflamed joints to the ventrobasal thalmus (Guilbaud et

al., 1980; Gautron and Guilbaud, 1982), intralaminar and medial thalamus (Kayser and

Guilbaud, 1984; Dostrovsky and Guilbaud, 1990), and somatosensory cortex (Lamour et





83



al., 1983). One consistent feature of the neuronal reactivity seen in all of these studies is

the increase responsiveness of central neurons to non-noxious stimulation of the inflamed

joints. The central mechanisms which might contribute to changes in responsiveness of

central neurons in arthritic rats is still in question. The alterations in discharge behavior

seen in CNS neurons is likely due, in part, to the changes in responsiveness of joint

receptors due to inflammatory processes.













APPENDIX E


TRIGEMINAL GANGLION


The trigeminal ganglion is situated near the apex of the petrous bone in the middle

cranial fossa. It lies in Meckel's cave near the cavernous sinus and internal carotid artery.

The sensory root portiono major) enters the pons in association with the motor root portiono

minor), which courses dorsomedially, and terminates in the trigeminal nuclear complex.

Sensory input from the face enters the trigeminal ganglion via the opthalamic,

maxillary and mandibular division of the trigeminal nerve. The cell bodies of these afferent

fibers are segregated into discrete clusters (Jerge, 1964; Kerr, 1962).

Retrograph transport of HRP injected into the cat TMJ showed that sensory

neurons originate in the trigeminal ganglion (Romfh et al., 1979; Capra, 1987). Whether

all of the cell bodies of afferent neurons innervating the TMJ are located in the trigeminal

ganglion remains unanswered. There is evidence suggesting that the peripheral spinal

neurons innervate the TMJ. Widenfalk and Wiberg (1990) injected HRP into the TMJ of

rats. Labeled cells were observed ipsilaterally in the second to fifth dorsal root ganglion.

In addition, there is evidence suggesting that the mesencephalic nucleus contains cell bodies

of neurons that terminate in the TMJ. In a preliminary communication, Limwongee (1986)

reported injection of HRP into the TMJ of rat, cat and monkey. He found labeled neurons

in the mesencephalic nucleus at the level of the caudal pons. On the other hand, Corbin

(1940) ablated the mesencephalic nucleus and examined histologically the auriculotemporal

nerve. He found no degenerated fibers. In addition, Chen and Turner (1987) injected









HRP into the TMJ and insertion of the lateral pterygoid muscle of rats. They found no

central projection to the mesencephalic nucleus.


Somatotopy


There is general agreement that the trigeminal ganglion in cats is somatopically

represented from a medial to lateral direction (Beaudreau and Jerge, 1968; Zucker and

Welker, 1969; Kerr and Lysak, 1964; Darian-Smith et al., 1965; Marfurt, 1981). The

ophthalmic division lies anteriomedially, the mandibular division lies posteriolaterally and

the maxillary division is situated in an intermediate position. Overlap between divisions

has been reported (Marfurt, 1981; Henry et al., 1986). Romfh et al., (1979) and Capra

(1987) reported that injection of HRP into the capsular tissues of the TMJ in the cat resulted

in a restricted distribution of label in the posteriolateral position of the trigeminal ganglion.

Very few labeled cells were found in the most dorsal or in the most ventral areas of the

mandibular division of the trigeminal ganglion. The majority of labeled cells is found

throughout the intermediate zone in the posteriolateral part of the ganglion.

Electrophysiological studies in cats (Capra and Gaitpon, 1981) and in rabbits (Lund and

Matthews, 1981; Appenteng et al., 1982) confirmed that TMJ afferents were located in the

posteriolateral region of the trigeminal ganglion.

In addition to the mediolateral somatotopy, studies suggest a dorsoventral

somatotopic organization (Kerr and Lysak, 1964; Marfurt, 1981; Capra, 1987).

Representation of oral and perioral structures appear ventral in the ganglion. Areas remote

from the mouth lie in the dorsum of the ganglion.













APPENDIX F


GROSS ANATOMY


The TMJ in mammals is a ginglymus-arthroidal joint, i.e., a hinge joint capable of

gliding movement (Figure F- 1). The craniomandibular articulation involves the condyle of

the mandible juxtaposed with the articular surface of the squamous portion of the temporal

bone. Interposed between the condyle and the articular fossa is an articular disk dividing

the articular space into upper and lower compartments.

The periarticular soft tissue consists of fibrous capsule, posterior attachment

tissues, synovial membrane and accessory ligaments. Except for the synovial membrane,

these tissues serve to limit the range of motion of the joint and to maintain a packed position

(Sicher, 1960; Rocabado, 1983). A packed position is the condyle-disc-fossa relationship

which juxtaposes these TMJ elements at rest and during mandibular movements. The joint

capsule is loose and composed of dense connective tissue with a collagenous matrix richly

supplied with blood vessels and nerves. The dense connective tissue of the capsule is

located primarily on the lateral and medial aspect of the joint and intimately associated with

the temporomandibular and sphenomandibular ligaments, respectively. The anterior

capsule arises from the suture line between the greater wing of the sphenoid bone and the

temporal bone. It inserts on the anteriorlateral aspect of the condylar neck. The loose

connective tissue of the capsule encompasses the joint posteriorly. It arises from the tragal

cartilage laterally and tympanic plate medially. The posterior capsule, as a meshwork of

unorganized collagen fiber, inserts and blends in with the posterior periosteum of the

condylar neck and upper capsule of the parotid gland.









SAGITTAL VIEW


FRONTAL VIEW


1. Articular Surface of Glenoid Fossa 15. Auriculotemporal Nerve
2. Superior Cavitt 16. Blood Vessels
3. Disc (Stippled Area) 17. Posterior Deep Temporal Nerve
4. Capsule 18. Squamo-Sphenoidal Suture
5. Articular Surface of Condyle 19. Parotid Gland
6. Synovial Membrane 20. Sphenomandibular Ligament
7. Squamo-Tympanic Suture
8. Spine of the Sphenoid
9. Vascular Knee of Meniscus
10. Pes Meniscus
11. Superior Belly ofLat. Pterygoid
12. Inferior Belly ofLat. Pterygoid
13. Superior Stratum of Bilaminar Zone of Meniscus
14. Inferior Stratum of Bilaminar Zone of Meniscus

Figure F-I
Schematic diagram of the human temporomandibular joint. Sagittal and
frontal views.









The dense connective tissue of the capsule reinforces the lateral position of the TMJ

and is incorporated into the temporomandibular ligament. Horizontal fibers of the

temporomandibular ligament restrict posterior displacement of the condyle and oblique

fibers tend to limit lateral and inferior displacement (Sicher, 1960; Griffin and Malor,

1974). The posterior attachment tissues consists of collagenous and elastic connective

tissues which extend from the posterior border of the disc to the tympanic plate.

Macroscopic dissections of the goat TMJ were performed in this laboratory. The

articular surface of the condyle is oval in shape, its longer dimension is concave

mediolateral and the shorter dimension is convex anterioposterior. The articular disc is

ovoid, biconcave, and divides the joint into two synovial cavities. The fibrous capsule is

moderately loose which allows for extensive lateral translation of the condyle. The capsule

surrounds the joint, with the exception of the anteriomedial portions, where the lateral

pterygoid muscle attaches to the neck of the condyle. With four exceptions the gross

anatomy of the goat TMJ is similar to the human. First, the lateral aspect of the posterior

attachment tissue extends posteriorly beyond the postglenoid process and attaches to the

inferior surface of the zygomatic process of the temporal bone. This anatomical variation is

associated with the lateralization of the bony articulation of the TMJ beyond the cranial

base. Second, the articular surface of the glenoid fossa is slightly convex in all directions,

rather than concave. To accommodate this convexity the condyle is appropriately concave

along the mediallateral axis. Third, there is no articular eminence. Finally, the lateral

pterygoid muscle attaches to the anteriormedial aspect of the neck of the condyle in a

manner similar to the human, but it does not appear to attach to the disc.


Comparative Anatomy


According to the classification of Turnbull (1970), goats are section I herbivors

while humans are section III omnivores. The TMJ of the goat has been used as an animal









model for oral surgery procedures because the anatomy and biomechanics are similar to the

human TMJ (Bifano et al., 1990).

First, for the vertebrate chewing apparatus, the range of motion is generally

confined to those movements necessary for dealing with a particular diet. Due to a coarse

diet of vegetation, ruminants possess a greater range of mandibular movement than do

carnivores. The excursive movements include hinging, lateral translation, slight protrusion

and mediolateral shift. Hinge opening and lateral translational movements are required for

forceful grinding. The lateral pterygoid muscle is the prime mover of the mandible during

lateral translational movements. The lateral excursion, initiated by contraction of either one

of the lateral pterygoid muscles, is a prerequisite to an effective grinding stroke. Both the

goat and human possess a well-developed, lateral pterygoid muscle.

Second, the elevated position of the condyle of the mandible relative to the occlusal

plane benefits the operation of the masseter and medial pterygoid muscles. They function

synergistically to provide powerful bite force needed for grinding activity.

Third, the condyle is convex anteriorposteriorly, which allows some translation

along a convex articular surface of the fossa. Such anteriorposterior translation represents

a common denominator between the TMJ of non-camivorous mammals and the human

joints. In contrast, carnivores have a deep concave articular fossa that allows only hinge

movement and prevents dislocation during seizure of prey.