- Email: [email protected]
Jaw muscle pain and its effect on gothic arch tracings
Jaw muscle pain and its effect on gothic Ales Obrez, DMD, PhD,a and Christian S. Stohler, DMD, School of Dentistry, University of Michigan, Ann Arbor, Mich.
Dr Med
arch tracings Dentb
Perceived changes in occlusion and decreased range of motion are often expressed by patients with masticatory muscle pain. The adverse loading of craniomandibular tissues that results from an inadequate maxillomandibular relationship in combination with the coexisting dysfunction is widely regarded as the cause of pain. This study was designed to test whether pain can cause significant changes in position of the mandible and therefore form the basis for any perceived changes in the maxillomandibular relationship. A second objective was to determine whether pain can cause changes in the mandibular range of motion. Five subjects who rated pain intensity on a visual analog scale were used in a single-blind, randomized, repeated-measures study design. Tonic muscle pain was induced by infusion of 5% hypertonic saline solution into the central portion of the superficial masseter muscle. Isotonic saline solution was used as a control, with subjects blinded to the type of substance given. The effect of pain on the position of the apex of the gothic arch tracing, the direction of the lateral mandibular border movements, and the mandibular range of motion was studied in a horizontal plane with minimal occlusal separation. Pain significantly affected the position of the apex of the gothic arch tracing in anterior (F = 11.46, p = 0.03) and transverse (F = 35.0, p = 0.004) directions. Similarly, pain affected the orientation of the mandibular lateral border movements (F = 12.44, p = 0.02) and their magnitude (F = 14.97, p = 0.01). All pain-induced effects proved to be reversible. The observed effect of pain can explain the perceived change of bite that is frequently noted by patients with orofacial pain. This study provided evidence of an alternative causal relationship between pain and changes in occlusal relationship and questions occlusal therapy as treatment, directed toward the elimination of the underlying cause in patients with masticatory muscle pain. (J PROSTHET DENT 1996;75:393-8.)
C ommon
complaints of patients with masticatory muscle pain include problems with their bite and decreased range of mandibular motion.l, 2 Occlusal adjustment is a widely accepted treatment modality directed toward the elimination of adverse occlusal contacts. This therapy is based on the occlusal disengagement theory, which is the avoidance of an offending tooth contact that leads to masticator-y muscle pain.3 Because orofacial patients often complain that their “teeth no longer fit together properly,” this study examined whether pain itself could be responsible for this symptom. A reliable model of experimental tonic muscle pain was used to produce pain.4 The null hypothesis that pain had no effect on the maxillomandibular relationship and range of mandibular motion was adopted. Because both the maxillomandibular relationship and the range of motion are readily observed by recording the gothic arch tracing, this widely used dental procedure was selected to test the hypothesis.
Supported by Grant R01 DE 8606-07 from the National Institute of Dental Research of the National Institutes of Health. “Assistant Professor, Department of Restorative Dentistry, College of Dentistry, University of Illinois at Chicago. bProfessor and Chairman, Department of Biologic and Materials Sciences, School of Dentistry, University of Michigan. Copyright 0 1996 by The Editorial Council of THE JOURNAL OF PROSTHETIC DENTISTRY. 0022.3913/96/$5.00 + 0.
APRIL1996
10/l/70394
MATERIAL
AND
METHODS
Five healthy adults (two men and three women) with an average age of 27 years (range 22 to 37 years) volunteered for these experiments involving experimental pain. The volunteers were reimbursed $50.00 (U.S.) for their participation. The experimental protocol was approved by the institutional review board for use of human subjects. Each subject signed a written consent form. None of the subjects exhibited symptoms or signs of temporomandibular joint disorder and did not have any clinical pain in any other part of the body at the time of the experiment. None of the subjects were receiving medications.
Apparatus Mandibular range of motion for each subject was documented by means of the intraoral gothic arch tracing, which was recorded by an appliance that consisted of a maxillary and mandibular device (Fig. 1). The maxillary part covered the palate and included an adjustable pin that was attached in the center of the palatal plate, which coincided with the line connecting both maxillary first molars. With the mouth closed, the tip of the pin was in contact with the mandibular plate of the gothic arch assembly. By use of the adjustable pin, the bite was raised to prevent any contact between the opposing dentition in all laterotrusive and protrusive excursions. The mean vertical opening was 6.0 mm (SD 0.4 mm), measured as the separation between the maxillary and mandibular inci-
THEJOIJRNALOFPROSTHETICDENTISTRY
3%
THE JOURNAL
OF PROSTHETIC
OBREZ
DENTISTRY
AND STOHLER
IL PR
CL
3
G
Fig. 1. Experimental
instrumentation. Automated stimulus (S) delivery system for producing tonic muscle pain. Arrows, Directions of mandibular movements with respect to site of stimulus application (IL, ipsilateral; PR, protrusive; CL, contralateral). Movements were registered by means of central bearing point (CBP) on gothic arch plate (G). Insert, Example of recorded data.
sors. The pin was thus used to maintain a minimal maxillomandibular vertical relationship that avoided centric and eccentric tooth contacts.
Experimental
procedure
The experiment followed a single-blind, randomized, repeated-measures design. Pain was induced by means of an automated system infusing hypertonic saline solution (LO%), whereas the control infusion consisted of isotonic saline solution (0.9%). Subjects were blinded with respect to sequence of the substance given. The sequence and the side of delivery of the isotonic and hypertonic solutions were randomized. Hypertonic saline solution was injected on the left side and isotonic saline solution on the right side, or vice versa. The central part of the masseter muscle was chosen as the site of infusion (Fig. 1). Further details of this system can be found elsewhere.5 Throughout the l&minute duration of the experiment, subjects were asked to rate pain intensity every 15 seconds by use of a computerized visual analog scale. The scale was 100 mm long and allowed grading of pain intensity from 0 (no pain) to 100 (most pain imaginable). Each subject was seated upright against the back of the dental chair with the head rest in place. The subject’s head was positioned so that the plane of occlusion was horizontal. After placement of the gothic arch assembly, the subject was asked to close the mandible until the pin touched the lingual plate. Subjects were instructed to produce their maximal protrusive and lateral mandibular excursions without losing contact between the pin and the lingual plate (Fig. 1). Each subject was asked to produce these
394
movements at least three times for each of the experimental conditions (baseline, control infusion, algesic infusion [pain], and after pain). The mandibular excursions were recorded between the eighth and twelfth minute into each experimental stage. To establish the error of the method, five separate arch tracings that consisted of three protrusive and three lateral mandibular border movements were obtained for each subject to determine the reproducibility of measurement. This occurred in a session before the pain experiment.
Analysis
and statistical
treatment
Tracings of the mandibular movements were photographed, enlarged (magnification x lo), and digitized. Each tracing consisted of the maximum protrusive excursion and lateral mandibular border movements (Fig. 2). The most posterior position from which these movements were made was designated as the apex of the gothic arch. The latter also represented the origin of two reference lines: (1) a line parallel to the midline and (2) a line perpendicular to the former at the point of the apex of the gothic arch. These lines served as a reference for the superimposition of tracings from different experimental stages. Measurement parameters included (1) the position of the apex of the gothic arch in anteroposterior and transverse directions, (2) the angles between the left and right lateral border movements and the transverse reference line (angles (Y and p, respectively, quantifying the orientation of the lateral border movements), and (3) the length of each excursive tracing measured from the apex of the gothic arch (Fig. 2). Variability of the protrusive and lateral movements at
VOLUME
75
NUMBER
4
ORRRZ
TEE JOURNAL
AND STOKLER
OF PROS~C
Deploy
Most posterior position from which lateral movementscambe made Reference
line
R
I&f Reference mark.3 used for superimposition
Fig. 2. Measurement parameters. Reference lines (RL): (1) line parallel to midline and (2) line perpendicular to former at point of apex of gothic arch. These lines served as reference for superimposition of tracings from different experimental stages. Transverse reference line connects reference black marks ~~~ drawn at baseline recording. baseline was determined from the analysis of the reproducibility data set. The individual means of the measurement parameters (lengths, angles, except for the position of the apex of the gothic arch) were assigned as 100%. The tracings of five trials of each subject were normalized to a percent of that value, and the mean and range of variability was calculated for each of the experimental conditions. In the first part of this study the statistical comparison of the raw data pertaining to the after-pain condition and baseline was included. Two subjects made themselves available 2 hours after the pain experiment. Wilcoxon matched-pairs sign-ranks tests were therefore used for the analysis. The results of this test enabled us to reduce the number of experimental conditions used in this study of primary effects from four to three (baseline, control, and pain). Finally, the left- and right-side dependent variables (lengths and orientations ofthe mandibular border movements) were examined to verify the feasibility of reducing the number of independent variables from three (experimental condition, side of parameter measured, side of substance injection) to two (~pe~rnen~ condition, side of substance injection relative to side of parameter measured). The study of primary effects included statistical comparison of the measurements by use of a multifactor repeated-measures analysis of variance (MANOVA) (SPSS Inc., Chicago, 111.).The within-subject experimental factors in NOVA were the expe~mental condition (baseline, control, and pain), and the side of infusion of isotonic or hypertonic saline solution relative to the mandibular movement (ipsilateral, contralateral). Whenever the analysis of variance F values were statistically significant, post hoc multiple comparisons were performed with the Bonferroni test. A probability of ~0.05 was chosen as statistically significant.
APRIL
1996
RESULTS Reproducibility
of measurements
The average positional variation (mean, range) of the apex of the gothic arch was 0.09 mm (0.0 to 0.22 mm) in the anterior and 0.06 mm (0.0 to 0.36 mm) in the transverse directions. The average variations of the orientation of the mandibular border movements, expressed as angles 01and l3 (mean, range) in percentages of the mean, were 2.31% (1.39% to 3.04%) and 2.41% (1.58% to 3.98%), respectively. The average variation of the lengths of the protrusive and right and left mandibular border movements (mean, range), expressed as a percentage of the mean, were 5.78% (2.22% to 12.47%), 4.73% (2.12% to 6.83%), and 4.88% (1.44% to 7.96%), respectively.
Comparison
of baseline
and after-pain
states
Example of gothic arch tracings recorded during each of four experimental conditions is shown in Fig. 3, and the raw data are presented in Table I. By use of the Wilcoxon matched-pairs sign-ranks test to compare baseline and after pain, no statistically significant differences between the two experimental conditions for all the dependent variables were observed: (I) position of the apex (anteroposterior direction, z = 0.0, p = 1.0; transverse direction, z = 0.0, p = l.O), (2) o~entation of the man~b~ar border movements (right, z = -1.34, p = 0.18; left, z = -1.0, p = 0.52), and (3) lengths of the mandibul.ar border movements (protrusive, z = -0.447, p = 0.65; right, z = -1.0, p = 0.317; left, z = -0.447,~ = 0.65). Because no differences were observed, a decrease in number of experimental conditions from four to three was justified in the subsequent MANOVA for the main effect.
395
THE JOURh’AL
OF PROSTHETIC
DENTISTRY
ORREZ
AND STOHLER
/ BASELINE
ISGIONIC SALINE PPI= OS
PH= 00
HYPERTONIC SALINE
AFTER PAIN
PPI=35
PPI=W
Fig. 3. Example of gothic arch tracings recorded during each of four experimental conditions. PPI, Present pain intensity, rated on scale from 00 (no pain) to 100 (most pain imaginable). Length of calibration bar represents 8.5 mm.
Subject 4 (No F%in/Placebo
Subject
Subject 5
identical)
1 (No Pain/Placebo
identical)
:____--- .-_-,- ^.I "-.ll^--.-"-l
-0.4-
1 Subject 3 (No ~~~Placebo
0
I
I
I
I
.I
.2
.3
.4
i 0 Isotonic saline infusion I 1 x Hypertonic saline i&&n /. .” . . ._ . .
identic~) I
.6 mm
Fig. 4. Graphic representation of change in position of apex of gothic arch tracings as consequence of placebo infusion (isotonic saline solution, o) and pain (h~ertonic saline solution, x). Origin of coordinate system (0,O) is position of apex of gothic arch recorded at baseline and used as origin to measure experimental effects. X nxis, distance (in millimeters) from origin in anterior direction; y axis, distance (in millimeters) from origin in transverse direction. Example of superimposed raw data is given in insert (anterior is to right). Length of calibration bar represents 9.25 mm. Border right)
movements
and effects
of sides (left,
Length and orientation of the right and left excursive border movements and their orientations were analyzed with repeated~meas~es NOVA to determine whether the experimental condition had a significant effect. There was no significant difference between the left and right excursive border movement for both dependent variables (length,F = 0.04,~ = 0.856; orientation, F = 2.3,~ = 0.204). The results therefore enabled us to pool both sides and to consider each dependent variable only with respect to the side where the substance was injected (ipsilateral, contralateral). The number of independent variables was thus reduced from three to two (Experimental condition
396
[3] x Mandibular substance 121). Primary
side relative to the side of the injected
effects
This portion of the analysis is presented in four parts. The first part details the effect of the experimental condition on all dependent variables, whereas the other parts specifically address the effect of pain on individual variables. Raw data are presented in Table I. Efiecct enta~*~n
of experimental and &m&z
condition on apex and protrusive
of border
and orimove-
men&. A significant primary effect of the experimental condition (baseline, control, pain) on position of the apex of the gothic arch, length, and orientation of the mandibular
VOLT
75
BIER
4
OBBEZ
AND STOHLER
Table
I. Summary
THE JOBS
Experimental condition
Baseline Control Ipsilateral Contralateral Pain Ipsilateral Contralateral Baseline Control Ipsilateral Contralateral Pain Ipsilateral Contralateral After pain Baseline Control Ipsilateral Contralateral Pain Ipsilateral Contralateral BItseline Control Ipsilateral Contralateral Pain Ipsilateral Contrala~ral Baseline Control Ipsilateral Contralateral Pain Tpsilateral Contralateral After pain
1
2
3
4
5
of movement (mm)
Protrusive 10.2
1996
Position
Lateral
Orientation (degrees)
8.8
35.0
7.8
33.0
7.2
37.0
6.2
13.2 11.9
6.6 4.8 10.5
34.0 33.0 40.0
10.3 10.3
43.0 36.7
7.9 9.8 12.4
34.0 39.5 40.0 42.0
12.3 11.9
45.0 42.0
10.2 11.9 14.1
38.2 38.2 41.0
13.1 13.1
40.0 38.7
12.2 13.3 11.4
37.7 41.5 47.7
11.4 7.6
40.5 45.0
10.8 10.9 10.9
47.5 46.0 47.5
8.6
13.7 13.5 12.3
11.1
12.9
17.2 14.1
14.9
10.9 10.0
11.8
10.8
of apex bd
8.4
border movements was found by use of repeated-measures NOVA (F = 21.73, p = 0.016). There was also a significant interaction effect between the experimental condition and the dependent variables (F = 2.53,~ = 0.03). Pain had the most significant effect on all the dependent variables (F = 11.095, p = 0.02). Effect of pain on the apex of the gothic arch trueing. The results of the multiva~ia~ tests of the repeatedmeasures MANOVA (Experimental condition 131 x Mandibular side relative to the side of injected substance [2]) are summarized as follows. The position of the apex of the gothic arch differed significantly as a consequence of the experimental condition (8’ = 33.371,~ = 0.009). The significant average effect of the experimental condition was observed in both the anterior (F = 7.71, p = 0.014) and
APRIL
DENTISTRY
of raw data Length
Subject
OF POETIC
A-P
Transverse
0.0 0.0
0.0 0.0
0.6
0.2
0.0
0.2
0.0 0.0
0.4
0.2
0.0 0.0 0.2
0.0 0.0 0.0
0.3
0.4
0.0 0.0
0.0 0.0
0.0
0.4
0.0 0.1
0.0
0.3
0.2
0.0
0.0
0.3
transverse (F = 9.06, p = 0.009) directions. Pain was the expe~montal condition that &ected the changes most (F = 25.45, p = 0.00). More specifically, the observed effect was expressed in the presence of pain in the anterior (F = 11.46, p = 0.03) and transverse (F = 35.0, p = 0.004) directions, whereas the control infusion did not affect the position of the apex to a statistically si~~cant degree (anterior, F = 2.48, p = 0.19; transverse, F = 1.0, p = 0.374) (Fig. 4). Whether the substance was injected ipsilaterally or contralaterally had no statistically significant effect on the position of the apex (F = 0.4, p = 0.564). This result is further supported by lack of an interaction effect CF = 0.09, p = 0.915). Eflect of pain on the o~entation of ~ndibular border movements. The results of the multivariate tests
397
THE JOURNAL
OF PROSTHETIC
of the repeated-measures MANOVA (Experimental condition [3] x Mandibular side relative to the side of injected substance [2]) demonstrated a significant effect of the experimental condition on the orientation of the mandibular border movements (F = 10.493,p = 0.04). Pain was the experimental condition that changed the orientation most significantly (8’ = 12.44, p = 0.02). The side of stimulation proved to be statistically insignificant for all experimental conditions (F = 0.786, p = 0.53). Effect of pain on the length of border and protrusive movements. The results of the repeated-measures MANOVA (Experimental condition [3] x Mandibular side relative to the side of the injected substance [2]) demonstrated a significant effect of the experimental condition on range of mandibular motion (F = 5.2, p = 0.036). Further post hoc comparison tests of the individual experimental conditions revealed that both control (8’ = 47.15,~ = 0.002) and pain (F = 11.32, p = 0.03) contributed significantly to the change of the length of the lateral border and the protrusive movements. When only lateral border movements were considered, the experimental condition as the primary effect remained a signifkant factor (F = 6.75, p = 0.02), with pain as the most effective (F = 14.971, p = 0.01). Protrusive movements were not significantly affected by the experimental condition (F = 3.53, p = 0.07). It is noted that the relative side of the stimulus application did not affect the protrusive (F = 0.34, p = 0.2) or the ipsilateral or contralateral border movements to a significant degree (F = 0.10, p = 0.767).
DISCUSSION On the basis of the results of this study, it was concluded that experimentally induced tonic masticatory muscle pain affected the mandibular border movements as observed in the horizontal plane. Experimental muscle pain also significantly affected the location of the most posterior mandibular position from which the lateral border movements could be made. The mean and the range (mean 0.22 mm, range 0.2 to 0.6 mm) within which the positional change occurred in the anterior and transverse directions were significantly greater than those observed during the repeated trials in a pain-free state in this (mean 0.07 mm, range 0.0 to 0.3 mm) and other studies6, 7 Such changes could explain the complaint of the reported “inability to find a comfortable intercuspal position” or “teeth no longer fit together right,” frequently expressed concerns of paof tients with facial pain. l, 8 Ten minutes after resolution pain, the position of the apex of the gothic arch coincided with its location at the baseline. This confirmed that the change was reversible and was indeed the consequence of the painful stimulus. A change was also observed in the orientation of the border movements relative to the transverse reference line while subjects were in pain. With the mandible slightly protruded in pain, the angle within which lateral excursive movements could be made became significantly wider, suggesting an alteration of the occlusal guidance during lateral excursive movements.
398
OBREZ
DENTISTRY
AND STOHLER
Finally, pain was shown to significantly decrease the maximum range of motion. This finding is consistent with the subjective reports and objective findings in patients with myalgia who commonly complain of a limited range of mandibular motion.2
CLIN-ICAL
SIGNIFICANCE
This study demonstrated that pain in the muscles of mastication caused a change in the maxillomandibular relationship. This change could account for the complaint of patients with pain that the bite no longer feels right. If this pain causes a change in the bite, the question arises as to what biologic advantage is to be gained from adjusting the occlusion. Instead, treatment should be directed toward the control of pain.
SUMMARY This study showed that experimentally induced tonic masticatory muscle pain could reversibly alter the gothic arch tracings in healthy volunteers. It is suggested that pain has a bearing on static and dynamic occlusal contact relationships. This challenges the occlusal disengagement theory, which assumes the reverse, namely, a causal relationship between an occlusal interference and muscle pain. On the basis of these findings, most notably the reversal of pain-induced changes with resolution of pain, this study challenges the necessity of occlusal therapy as the treatment of choice in cases of masticatory muscle pain. Instead, the results suggest that treatment emphasis be directed toward the control of pain. REFERENCES 1. Bell WE. Temporomandibular disorders: classification, diagnosis, management. 2nd ed. Chicago: Year Book Medical, 1986. 2. Dworkin SF, Huggins KH, Le Resche L, et al. Epidemiology of signs and symptoms in temporomandibular disorders: clinical signs in cases and controls. J Am Dent Assoc 1990;120:273-81. 3. Ramfjord SP, Ash MM. Occlusion. 3rd ed. Philadelphia: WB Saunders, 1983. 4. Zhang X, Ashton-Miller J, Stohler CS. A closed-loop system for maintaining constant experimental muscle pain in man. IEEE Tram Biomed Eng 1993;40:344-52. 5. Stohler CS, Zhang X, Ashton-Miller JA. An experimental model ofjaw muscle pain in man. In: Davidovitch Z, ed. The biological mechanisms of tooth movement and craniofacial adaptation. Birmingham: EBSCO Media, 1991:261-‘7. 6. Smith HF Jr. A comparison of empirical centric relation records with location of terminal hinge axis and apex of the gothic arch tracing. J PROSTHET DEB 1975;33:511-20. 7. Simon RL, Nicholls JI. Variability of passively recorded centric relation. J PROSTHET DENT 1980;44:21-6. 8. Tryde G, Frydenberg 0, Brill N. An assessment of the tactile sensibility in human teeth: an evaluation of a quantitative method. Acta Odont Stand 1962;20:233-56.
Reprint
requests
to:
DR. ALES OBREZ DEPARTMEW OF RESTORATIVE DENTISTRY COLLEGE OF DENTISTRY THE UNIVERSITY OF ILLINOIS AT CHICAGO 801 S. PAULINA ST. CHICAGO, IL 60612-7212
VOLUME
75
NUMBER
4
Dr Med
arch tracings Dentb
Perceived changes in occlusion and decreased range of motion are often expressed by patients with masticatory muscle pain. The adverse loading of craniomandibular tissues that results from an inadequate maxillomandibular relationship in combination with the coexisting dysfunction is widely regarded as the cause of pain. This study was designed to test whether pain can cause significant changes in position of the mandible and therefore form the basis for any perceived changes in the maxillomandibular relationship. A second objective was to determine whether pain can cause changes in the mandibular range of motion. Five subjects who rated pain intensity on a visual analog scale were used in a single-blind, randomized, repeated-measures study design. Tonic muscle pain was induced by infusion of 5% hypertonic saline solution into the central portion of the superficial masseter muscle. Isotonic saline solution was used as a control, with subjects blinded to the type of substance given. The effect of pain on the position of the apex of the gothic arch tracing, the direction of the lateral mandibular border movements, and the mandibular range of motion was studied in a horizontal plane with minimal occlusal separation. Pain significantly affected the position of the apex of the gothic arch tracing in anterior (F = 11.46, p = 0.03) and transverse (F = 35.0, p = 0.004) directions. Similarly, pain affected the orientation of the mandibular lateral border movements (F = 12.44, p = 0.02) and their magnitude (F = 14.97, p = 0.01). All pain-induced effects proved to be reversible. The observed effect of pain can explain the perceived change of bite that is frequently noted by patients with orofacial pain. This study provided evidence of an alternative causal relationship between pain and changes in occlusal relationship and questions occlusal therapy as treatment, directed toward the elimination of the underlying cause in patients with masticatory muscle pain. (J PROSTHET DENT 1996;75:393-8.)
C ommon
complaints of patients with masticatory muscle pain include problems with their bite and decreased range of mandibular motion.l, 2 Occlusal adjustment is a widely accepted treatment modality directed toward the elimination of adverse occlusal contacts. This therapy is based on the occlusal disengagement theory, which is the avoidance of an offending tooth contact that leads to masticator-y muscle pain.3 Because orofacial patients often complain that their “teeth no longer fit together properly,” this study examined whether pain itself could be responsible for this symptom. A reliable model of experimental tonic muscle pain was used to produce pain.4 The null hypothesis that pain had no effect on the maxillomandibular relationship and range of mandibular motion was adopted. Because both the maxillomandibular relationship and the range of motion are readily observed by recording the gothic arch tracing, this widely used dental procedure was selected to test the hypothesis.
Supported by Grant R01 DE 8606-07 from the National Institute of Dental Research of the National Institutes of Health. “Assistant Professor, Department of Restorative Dentistry, College of Dentistry, University of Illinois at Chicago. bProfessor and Chairman, Department of Biologic and Materials Sciences, School of Dentistry, University of Michigan. Copyright 0 1996 by The Editorial Council of THE JOURNAL OF PROSTHETIC DENTISTRY. 0022.3913/96/$5.00 + 0.
APRIL1996
10/l/70394
MATERIAL
AND
METHODS
Five healthy adults (two men and three women) with an average age of 27 years (range 22 to 37 years) volunteered for these experiments involving experimental pain. The volunteers were reimbursed $50.00 (U.S.) for their participation. The experimental protocol was approved by the institutional review board for use of human subjects. Each subject signed a written consent form. None of the subjects exhibited symptoms or signs of temporomandibular joint disorder and did not have any clinical pain in any other part of the body at the time of the experiment. None of the subjects were receiving medications.
Apparatus Mandibular range of motion for each subject was documented by means of the intraoral gothic arch tracing, which was recorded by an appliance that consisted of a maxillary and mandibular device (Fig. 1). The maxillary part covered the palate and included an adjustable pin that was attached in the center of the palatal plate, which coincided with the line connecting both maxillary first molars. With the mouth closed, the tip of the pin was in contact with the mandibular plate of the gothic arch assembly. By use of the adjustable pin, the bite was raised to prevent any contact between the opposing dentition in all laterotrusive and protrusive excursions. The mean vertical opening was 6.0 mm (SD 0.4 mm), measured as the separation between the maxillary and mandibular inci-
THEJOIJRNALOFPROSTHETICDENTISTRY
3%
THE JOURNAL
OF PROSTHETIC
OBREZ
DENTISTRY
AND STOHLER
IL PR
CL
3
G
Fig. 1. Experimental
instrumentation. Automated stimulus (S) delivery system for producing tonic muscle pain. Arrows, Directions of mandibular movements with respect to site of stimulus application (IL, ipsilateral; PR, protrusive; CL, contralateral). Movements were registered by means of central bearing point (CBP) on gothic arch plate (G). Insert, Example of recorded data.
sors. The pin was thus used to maintain a minimal maxillomandibular vertical relationship that avoided centric and eccentric tooth contacts.
Experimental
procedure
The experiment followed a single-blind, randomized, repeated-measures design. Pain was induced by means of an automated system infusing hypertonic saline solution (LO%), whereas the control infusion consisted of isotonic saline solution (0.9%). Subjects were blinded with respect to sequence of the substance given. The sequence and the side of delivery of the isotonic and hypertonic solutions were randomized. Hypertonic saline solution was injected on the left side and isotonic saline solution on the right side, or vice versa. The central part of the masseter muscle was chosen as the site of infusion (Fig. 1). Further details of this system can be found elsewhere.5 Throughout the l&minute duration of the experiment, subjects were asked to rate pain intensity every 15 seconds by use of a computerized visual analog scale. The scale was 100 mm long and allowed grading of pain intensity from 0 (no pain) to 100 (most pain imaginable). Each subject was seated upright against the back of the dental chair with the head rest in place. The subject’s head was positioned so that the plane of occlusion was horizontal. After placement of the gothic arch assembly, the subject was asked to close the mandible until the pin touched the lingual plate. Subjects were instructed to produce their maximal protrusive and lateral mandibular excursions without losing contact between the pin and the lingual plate (Fig. 1). Each subject was asked to produce these
394
movements at least three times for each of the experimental conditions (baseline, control infusion, algesic infusion [pain], and after pain). The mandibular excursions were recorded between the eighth and twelfth minute into each experimental stage. To establish the error of the method, five separate arch tracings that consisted of three protrusive and three lateral mandibular border movements were obtained for each subject to determine the reproducibility of measurement. This occurred in a session before the pain experiment.
Analysis
and statistical
treatment
Tracings of the mandibular movements were photographed, enlarged (magnification x lo), and digitized. Each tracing consisted of the maximum protrusive excursion and lateral mandibular border movements (Fig. 2). The most posterior position from which these movements were made was designated as the apex of the gothic arch. The latter also represented the origin of two reference lines: (1) a line parallel to the midline and (2) a line perpendicular to the former at the point of the apex of the gothic arch. These lines served as a reference for the superimposition of tracings from different experimental stages. Measurement parameters included (1) the position of the apex of the gothic arch in anteroposterior and transverse directions, (2) the angles between the left and right lateral border movements and the transverse reference line (angles (Y and p, respectively, quantifying the orientation of the lateral border movements), and (3) the length of each excursive tracing measured from the apex of the gothic arch (Fig. 2). Variability of the protrusive and lateral movements at
VOLUME
75
NUMBER
4
ORRRZ
TEE JOURNAL
AND STOKLER
OF PROS~C
Deploy
Most posterior position from which lateral movementscambe made Reference
line
R
I&f Reference mark.3 used for superimposition
Fig. 2. Measurement parameters. Reference lines (RL): (1) line parallel to midline and (2) line perpendicular to former at point of apex of gothic arch. These lines served as reference for superimposition of tracings from different experimental stages. Transverse reference line connects reference black marks ~~~ drawn at baseline recording. baseline was determined from the analysis of the reproducibility data set. The individual means of the measurement parameters (lengths, angles, except for the position of the apex of the gothic arch) were assigned as 100%. The tracings of five trials of each subject were normalized to a percent of that value, and the mean and range of variability was calculated for each of the experimental conditions. In the first part of this study the statistical comparison of the raw data pertaining to the after-pain condition and baseline was included. Two subjects made themselves available 2 hours after the pain experiment. Wilcoxon matched-pairs sign-ranks tests were therefore used for the analysis. The results of this test enabled us to reduce the number of experimental conditions used in this study of primary effects from four to three (baseline, control, and pain). Finally, the left- and right-side dependent variables (lengths and orientations ofthe mandibular border movements) were examined to verify the feasibility of reducing the number of independent variables from three (experimental condition, side of parameter measured, side of substance injection) to two (~pe~rnen~ condition, side of substance injection relative to side of parameter measured). The study of primary effects included statistical comparison of the measurements by use of a multifactor repeated-measures analysis of variance (MANOVA) (SPSS Inc., Chicago, 111.).The within-subject experimental factors in NOVA were the expe~mental condition (baseline, control, and pain), and the side of infusion of isotonic or hypertonic saline solution relative to the mandibular movement (ipsilateral, contralateral). Whenever the analysis of variance F values were statistically significant, post hoc multiple comparisons were performed with the Bonferroni test. A probability of ~0.05 was chosen as statistically significant.
APRIL
1996
RESULTS Reproducibility
of measurements
The average positional variation (mean, range) of the apex of the gothic arch was 0.09 mm (0.0 to 0.22 mm) in the anterior and 0.06 mm (0.0 to 0.36 mm) in the transverse directions. The average variations of the orientation of the mandibular border movements, expressed as angles 01and l3 (mean, range) in percentages of the mean, were 2.31% (1.39% to 3.04%) and 2.41% (1.58% to 3.98%), respectively. The average variation of the lengths of the protrusive and right and left mandibular border movements (mean, range), expressed as a percentage of the mean, were 5.78% (2.22% to 12.47%), 4.73% (2.12% to 6.83%), and 4.88% (1.44% to 7.96%), respectively.
Comparison
of baseline
and after-pain
states
Example of gothic arch tracings recorded during each of four experimental conditions is shown in Fig. 3, and the raw data are presented in Table I. By use of the Wilcoxon matched-pairs sign-ranks test to compare baseline and after pain, no statistically significant differences between the two experimental conditions for all the dependent variables were observed: (I) position of the apex (anteroposterior direction, z = 0.0, p = 1.0; transverse direction, z = 0.0, p = l.O), (2) o~entation of the man~b~ar border movements (right, z = -1.34, p = 0.18; left, z = -1.0, p = 0.52), and (3) lengths of the mandibul.ar border movements (protrusive, z = -0.447, p = 0.65; right, z = -1.0, p = 0.317; left, z = -0.447,~ = 0.65). Because no differences were observed, a decrease in number of experimental conditions from four to three was justified in the subsequent MANOVA for the main effect.
395
THE JOURh’AL
OF PROSTHETIC
DENTISTRY
ORREZ
AND STOHLER
/ BASELINE
ISGIONIC SALINE PPI= OS
PH= 00
HYPERTONIC SALINE
AFTER PAIN
PPI=35
PPI=W
Fig. 3. Example of gothic arch tracings recorded during each of four experimental conditions. PPI, Present pain intensity, rated on scale from 00 (no pain) to 100 (most pain imaginable). Length of calibration bar represents 8.5 mm.
Subject 4 (No F%in/Placebo
Subject
Subject 5
identical)
1 (No Pain/Placebo
identical)
:____--- .-_-,- ^.I "-.ll^--.-"-l
-0.4-
1 Subject 3 (No ~~~Placebo
0
I
I
I
I
.I
.2
.3
.4
i 0 Isotonic saline infusion I 1 x Hypertonic saline i&&n /. .” . . ._ . .
identic~) I
.6 mm
Fig. 4. Graphic representation of change in position of apex of gothic arch tracings as consequence of placebo infusion (isotonic saline solution, o) and pain (h~ertonic saline solution, x). Origin of coordinate system (0,O) is position of apex of gothic arch recorded at baseline and used as origin to measure experimental effects. X nxis, distance (in millimeters) from origin in anterior direction; y axis, distance (in millimeters) from origin in transverse direction. Example of superimposed raw data is given in insert (anterior is to right). Length of calibration bar represents 9.25 mm. Border right)
movements
and effects
of sides (left,
Length and orientation of the right and left excursive border movements and their orientations were analyzed with repeated~meas~es NOVA to determine whether the experimental condition had a significant effect. There was no significant difference between the left and right excursive border movement for both dependent variables (length,F = 0.04,~ = 0.856; orientation, F = 2.3,~ = 0.204). The results therefore enabled us to pool both sides and to consider each dependent variable only with respect to the side where the substance was injected (ipsilateral, contralateral). The number of independent variables was thus reduced from three to two (Experimental condition
396
[3] x Mandibular substance 121). Primary
side relative to the side of the injected
effects
This portion of the analysis is presented in four parts. The first part details the effect of the experimental condition on all dependent variables, whereas the other parts specifically address the effect of pain on individual variables. Raw data are presented in Table I. Efiecct enta~*~n
of experimental and &m&z
condition on apex and protrusive
of border
and orimove-
men&. A significant primary effect of the experimental condition (baseline, control, pain) on position of the apex of the gothic arch, length, and orientation of the mandibular
VOLT
75
BIER
4
OBBEZ
AND STOHLER
Table
I. Summary
THE JOBS
Experimental condition
Baseline Control Ipsilateral Contralateral Pain Ipsilateral Contralateral Baseline Control Ipsilateral Contralateral Pain Ipsilateral Contralateral After pain Baseline Control Ipsilateral Contralateral Pain Ipsilateral Contralateral BItseline Control Ipsilateral Contralateral Pain Ipsilateral Contrala~ral Baseline Control Ipsilateral Contralateral Pain Tpsilateral Contralateral After pain
1
2
3
4
5
of movement (mm)
Protrusive 10.2
1996
Position
Lateral
Orientation (degrees)
8.8
35.0
7.8
33.0
7.2
37.0
6.2
13.2 11.9
6.6 4.8 10.5
34.0 33.0 40.0
10.3 10.3
43.0 36.7
7.9 9.8 12.4
34.0 39.5 40.0 42.0
12.3 11.9
45.0 42.0
10.2 11.9 14.1
38.2 38.2 41.0
13.1 13.1
40.0 38.7
12.2 13.3 11.4
37.7 41.5 47.7
11.4 7.6
40.5 45.0
10.8 10.9 10.9
47.5 46.0 47.5
8.6
13.7 13.5 12.3
11.1
12.9
17.2 14.1
14.9
10.9 10.0
11.8
10.8
of apex bd
8.4
border movements was found by use of repeated-measures NOVA (F = 21.73, p = 0.016). There was also a significant interaction effect between the experimental condition and the dependent variables (F = 2.53,~ = 0.03). Pain had the most significant effect on all the dependent variables (F = 11.095, p = 0.02). Effect of pain on the apex of the gothic arch trueing. The results of the multiva~ia~ tests of the repeatedmeasures MANOVA (Experimental condition 131 x Mandibular side relative to the side of injected substance [2]) are summarized as follows. The position of the apex of the gothic arch differed significantly as a consequence of the experimental condition (8’ = 33.371,~ = 0.009). The significant average effect of the experimental condition was observed in both the anterior (F = 7.71, p = 0.014) and
APRIL
DENTISTRY
of raw data Length
Subject
OF POETIC
A-P
Transverse
0.0 0.0
0.0 0.0
0.6
0.2
0.0
0.2
0.0 0.0
0.4
0.2
0.0 0.0 0.2
0.0 0.0 0.0
0.3
0.4
0.0 0.0
0.0 0.0
0.0
0.4
0.0 0.1
0.0
0.3
0.2
0.0
0.0
0.3
transverse (F = 9.06, p = 0.009) directions. Pain was the expe~montal condition that &ected the changes most (F = 25.45, p = 0.00). More specifically, the observed effect was expressed in the presence of pain in the anterior (F = 11.46, p = 0.03) and transverse (F = 35.0, p = 0.004) directions, whereas the control infusion did not affect the position of the apex to a statistically si~~cant degree (anterior, F = 2.48, p = 0.19; transverse, F = 1.0, p = 0.374) (Fig. 4). Whether the substance was injected ipsilaterally or contralaterally had no statistically significant effect on the position of the apex (F = 0.4, p = 0.564). This result is further supported by lack of an interaction effect CF = 0.09, p = 0.915). Eflect of pain on the o~entation of ~ndibular border movements. The results of the multivariate tests
397
THE JOURNAL
OF PROSTHETIC
of the repeated-measures MANOVA (Experimental condition [3] x Mandibular side relative to the side of injected substance [2]) demonstrated a significant effect of the experimental condition on the orientation of the mandibular border movements (F = 10.493,p = 0.04). Pain was the experimental condition that changed the orientation most significantly (8’ = 12.44, p = 0.02). The side of stimulation proved to be statistically insignificant for all experimental conditions (F = 0.786, p = 0.53). Effect of pain on the length of border and protrusive movements. The results of the repeated-measures MANOVA (Experimental condition [3] x Mandibular side relative to the side of the injected substance [2]) demonstrated a significant effect of the experimental condition on range of mandibular motion (F = 5.2, p = 0.036). Further post hoc comparison tests of the individual experimental conditions revealed that both control (8’ = 47.15,~ = 0.002) and pain (F = 11.32, p = 0.03) contributed significantly to the change of the length of the lateral border and the protrusive movements. When only lateral border movements were considered, the experimental condition as the primary effect remained a signifkant factor (F = 6.75, p = 0.02), with pain as the most effective (F = 14.971, p = 0.01). Protrusive movements were not significantly affected by the experimental condition (F = 3.53, p = 0.07). It is noted that the relative side of the stimulus application did not affect the protrusive (F = 0.34, p = 0.2) or the ipsilateral or contralateral border movements to a significant degree (F = 0.10, p = 0.767).
DISCUSSION On the basis of the results of this study, it was concluded that experimentally induced tonic masticatory muscle pain affected the mandibular border movements as observed in the horizontal plane. Experimental muscle pain also significantly affected the location of the most posterior mandibular position from which the lateral border movements could be made. The mean and the range (mean 0.22 mm, range 0.2 to 0.6 mm) within which the positional change occurred in the anterior and transverse directions were significantly greater than those observed during the repeated trials in a pain-free state in this (mean 0.07 mm, range 0.0 to 0.3 mm) and other studies6, 7 Such changes could explain the complaint of the reported “inability to find a comfortable intercuspal position” or “teeth no longer fit together right,” frequently expressed concerns of paof tients with facial pain. l, 8 Ten minutes after resolution pain, the position of the apex of the gothic arch coincided with its location at the baseline. This confirmed that the change was reversible and was indeed the consequence of the painful stimulus. A change was also observed in the orientation of the border movements relative to the transverse reference line while subjects were in pain. With the mandible slightly protruded in pain, the angle within which lateral excursive movements could be made became significantly wider, suggesting an alteration of the occlusal guidance during lateral excursive movements.
398
OBREZ
DENTISTRY
AND STOHLER
Finally, pain was shown to significantly decrease the maximum range of motion. This finding is consistent with the subjective reports and objective findings in patients with myalgia who commonly complain of a limited range of mandibular motion.2
CLIN-ICAL
SIGNIFICANCE
This study demonstrated that pain in the muscles of mastication caused a change in the maxillomandibular relationship. This change could account for the complaint of patients with pain that the bite no longer feels right. If this pain causes a change in the bite, the question arises as to what biologic advantage is to be gained from adjusting the occlusion. Instead, treatment should be directed toward the control of pain.
SUMMARY This study showed that experimentally induced tonic masticatory muscle pain could reversibly alter the gothic arch tracings in healthy volunteers. It is suggested that pain has a bearing on static and dynamic occlusal contact relationships. This challenges the occlusal disengagement theory, which assumes the reverse, namely, a causal relationship between an occlusal interference and muscle pain. On the basis of these findings, most notably the reversal of pain-induced changes with resolution of pain, this study challenges the necessity of occlusal therapy as the treatment of choice in cases of masticatory muscle pain. Instead, the results suggest that treatment emphasis be directed toward the control of pain. REFERENCES 1. Bell WE. Temporomandibular disorders: classification, diagnosis, management. 2nd ed. Chicago: Year Book Medical, 1986. 2. Dworkin SF, Huggins KH, Le Resche L, et al. Epidemiology of signs and symptoms in temporomandibular disorders: clinical signs in cases and controls. J Am Dent Assoc 1990;120:273-81. 3. Ramfjord SP, Ash MM. Occlusion. 3rd ed. Philadelphia: WB Saunders, 1983. 4. Zhang X, Ashton-Miller J, Stohler CS. A closed-loop system for maintaining constant experimental muscle pain in man. IEEE Tram Biomed Eng 1993;40:344-52. 5. Stohler CS, Zhang X, Ashton-Miller JA. An experimental model ofjaw muscle pain in man. In: Davidovitch Z, ed. The biological mechanisms of tooth movement and craniofacial adaptation. Birmingham: EBSCO Media, 1991:261-‘7. 6. Smith HF Jr. A comparison of empirical centric relation records with location of terminal hinge axis and apex of the gothic arch tracing. J PROSTHET DEB 1975;33:511-20. 7. Simon RL, Nicholls JI. Variability of passively recorded centric relation. J PROSTHET DENT 1980;44:21-6. 8. Tryde G, Frydenberg 0, Brill N. An assessment of the tactile sensibility in human teeth: an evaluation of a quantitative method. Acta Odont Stand 1962;20:233-56.
Reprint
requests
to:
DR. ALES OBREZ DEPARTMEW OF RESTORATIVE DENTISTRY COLLEGE OF DENTISTRY THE UNIVERSITY OF ILLINOIS AT CHICAGO 801 S. PAULINA ST. CHICAGO, IL 60612-7212
VOLUME
75
NUMBER
4