An evaluation of temporomandibular joints and jaw muscles after orthodontic treatment involving premolar extractions K. K. Kundinger, DDS, MS, B. P. Austin, DDS, PhD, L. V. Christensen, DDS, MS, PhD, S. J. Donegan, DDS, MS, and D. J. Ferguson, DMD Milwaukee, Wis.
Experimental subjects (n = 29) were patients who had undergone orthodontic treatment in combination with extraction of maxillary or mandibular premolar teeth, or both. Control subjects (n = 29) were healthy dental students with no orthodontic or extraction experience. Sagittal (corrected axis) tomograms of the TMJs were used to determine the narrowest linear distances between the anterior and posterior outlines of the TMJ condyle and the TMJ fossa, expressed as the joint space ratio. There were no significant (p > 0.05) differences between the control and experimental ratios. Bipolar surface electromyograms of the masseter and anterior temporalis muscles were used to determine the isometric contraction velocities of these muscles until 50% and 100% voluntary isometric contraction effort (teeth clenching) was achieved. There were no significant (p > 0.05) differences between the control and experimental subjects. Electromyogr~.ms were also used to determine the relative contribution of the masseter and anterior temporalis muscles to the bite force developed during brief maximum voluntary tooth clenching, expressed as the activity index. There were no significant (p > 0.05) differences between the control and experimental subjects. (AM J OnTHOD DENTOFACORTHOP 1991 ;100:110-5.)
T h e extraction of premolars as a practical form of orthodontic therapy has been accepted for many years. Recently the orthodontic community has been questioned about this form of therapy, and assertions have been made that the temporomandibular joints (TMJs) are being adversely affected as a result of extraction treatment. The contention has been that premolar extraction, with retraction of the premaxillas, displaces the mandible and the mandibular condyles posteriorly; "there is an extremely high incidence of temporomandibular joint problems seen in patients with Class II, division 2 deep bite malocclusions, or patients with overretracted maxillary anteriors due to premolar extractions. ''''2 Farrar and McCarthy 3 were also of the opinion that a predisposing factor in TMJ anterior disk displacement is orthodontic therapy after extraction of premolar teeth. On the other hand, in their study of condylar position and TMJ status, Gianelly et al. 4 found no difference in the positions of the condyles of untreated patients and extraction-treated patients. Relative to premolar extraction therapy, the objective of this study was to evaluate the TMJs in terms
From the Marquette University School of Dentistry. 8/1/22729
110
of joint space ratios, often referred to as condylar position/displacement. Another objective was to evaluate jaw muscle function in terms of electromyography (EMG) of voluntary isometric contractions--a method not applied previously in this context.
MATERIALS AND METHODS The subjects were volunteers who had given informed consent to participate in the study. The sample of experimental subjects consisted of orthodontic patients in retention who had been treated with extraction of premolar teeth. Experimental subjects were 10 male and 19 female patients, ranging in age from 15 years 5 months to 36 years 1 month (mean age, 22 years 7 months). Eight of the experimental subjects had only maxillary" first premolars extracted (Angle Class II occlusion), wl{ereas the remaining 21 experimental subjects had four premolars extracted (Angle Class I occlusion). The average number of permanent teeth present was 24. The control subjects were 16 male and 13 female dental students who had not undergone orthodontic/extraction treatment. Their ages ranged from 21 years 2 months to 48 years 4 months (mean age, 27 years 3 months). Twenty-one control subjects had Class I occlusion, two had Class II, subdivision left malocclusion, and five had Class II, Division 1 malocclusion; one control subject had a Class III malocclusion. In the control group, the average number of permanent teeth present was 28.
Voh~me 100 Number 2
Premolar extraction therapy
Radiographic assessment of the TMJs included a vertical (submentovertex) projection followed by a lateral projection (sagittal tomography). Submentovertex radiographs were taken to obtain mediolateral condyle angulations and selective depth-of-cut measurements, complying with the protocols of Williamson and Wilson5 and Beckwith et al. 6 This projection used the Quint x-ray Sectograph (Denar Corporation, Anaheim, Calif.), and the subject-to-film distance was constant for each subject at I I mm. TMJ tracings were made on transparent acetate paper with a 0.5 mm lead pencil, and all measurements were taken from the machine midline. Subsequently, corrected right and left tomograms (one cut per side) were taken through the anatomic center of the mandibular condyle with the subject biting in the intercuspal position (IP). Each tomogram was cut and mounted in a conventional 35 mm slide holder and enlarged by means of a Kodak slide projector with a 103 mm lens. At each viewing, the screento-projector distance was calibrated to a fixed distance ensuring a projected in'age of × 10 magnification. The tomographic joint space was determined by the consensus of two examiners. Linearmeasurements were made of the narrowest radiographic (tomographic) anterior and posterior joint space, the shortest linear distance between the tomographic outline (contour) of the mandibular condyle and the tomographic outline (contour) of the glenoid fossa. 7.s The two linear distances were measured with the aid of the Mitutoya digimatic calipers read to the nearest 0.01 mm. If the shortest distance became difficult to read, that particular slide was remeasured at another time. If the measurements were within ---5 mm, the two measurements were averaged. Those joint space measurements that were greater than ± 5 mm were again measured, and the points were agreed on by three examiners. It was not possible to determine the joint spaces in one experimental subject and in three control subjects. According to the protocol of Pullinger and Hollender,7-*the subjective narrowest joint space (JS) ratio was expressed as Posteriorjs - A n t e r i o r , s × 100. Posterior~s + Anterior~s The JS ratio ranges from ±100% to - I 0 0 % , with + 100% indicating the presence of a posterior joint space only and - 100% indicating the presence of an anterior joint space only. Because a ratio of zero indicates equal anterior and posterior narrowest joint spaces, a positive ( + ) ratio indicates that the posterior joint space is larger than the anterior space at the closest approximation of the condyle to the fossa, and a negative ( - ) ratio indicates the opposite. For EMG, subjects were seated upright, with no headrest, in a natural head posture and with the feet flat on the floor. A verbal instruction was given to perform, briefly and rapidly, vertically directed maximum voluntary tooth clenching (MVC) in the mandibular position of IP. The isometric MVC exercise, which occurred within about 1 second was repeated after an interval of about 3 minutes, complying with the protocol of Christensen.9 Bipolar surface EMGs were obtained from the right and left masseter (MA) and anterior temporalis (AT) muscles. Circtflar Ag/AgCI electrodes were
111
placed over the palpated main bulks of the muscles, and a ground electrode was attached to the neck. The myoeleetrie signals were sampled on-line, followed by analog-to-digital conversion (total A I D conversion time of 1.5 microseconds per channel, with a resolution of 8 bits). Quantification of EMG (amplification of 3125 × ; gain of 200 p.V) was done by the mean voltage (I.tV) of 2000 milliseconds data records (BioResearch Inc., Milwaukee, Wis.). The EMG method has been described in detail elsewhere) °-~2 For each EMG, the following variables were determined by hand: the time, in milliseconds (ms), to peak (100%) mean voltage (p.V) of MVC in IP; the time (ms) to one-half peak (50%) mean voltage of MVC in IP. For the right and left MA and AT muscles, the two measurements allowed calculation of the velocity of isometric contraction (p.V/ms) from 0% to 100% and from 0% to 50% MVC in IP. The onset of MVC activity was determined on concurrent raw- and mean-voltage EMGs. To determine the relative contribution of the right and left MA and AT muscles to synergistic muscle activity (IxV) during MVC, the following activity index was used. EMG.~tA - EMGAT x 100 EMG.,IA + EMGAr The index ranges from + 100% to - I00%, with activity in MA only = + 100%, equal activity in MA ahd AT = 0, and activity in AT only = - 100%. u'13 Because.the frequency distributions of the ratios pertaining to joint spaces and MA/AT synergistic activities were not distributed normally, arcsine (sin -z) transformations were used to obtain gaussian frequency distribiations of all percentage data. 9't~ For data reductions, the two-tailed variance ratio test (F test) was used, followed by the two-tailed twosample Student's t test. The F test was applied to detect either homogeneity or heterogeneity of variances, and the t test was applied to detect significant differences between arithmetic mean values. In cases of heterogenous variances, Welch's approximate t test with corrected degrees of freedom was used? 5 The level of statistical significance was set at 5% (0.05).
RESULTS In the right TMJ (Table I), the joint space ratio of the control subjects (n = 26) showed a mean value of 1.84% ( - SD 39.28, ._ SE 7.90), indicating that the mean anterior joint space was larger. In the experimental.subjects (n = 28) the joint space ratio showed a mean of + 3 . 9 6 % ( ± S D 23.07, _ S E 4.39), indicating that the mean posterior joint space was larger (Table I). However, the difference between the two mean ratios was not significant (0.80 < p < 0.90). In the left TMJ (Table I), the mean joint space ratio of the control subjects showed a value of + 2 . 8 8 % ( ± S D 32.95, + SE 6.58), indicating that the mean posterior joint space was larger. In the experimental subjects the joint space ratio showed a mean of - 1 . 3 9 % (___SD -
112
Kundinger et al.
Am. J. Orthod. Dentofac. Orthop. August 1991
Table I. Tomographic joint space ratios for fight
and left temporomandibular joints in 26 control and 28 experimental subjects Joint space ratio Subjects
Right side
Control Experimental
- 1 . 8 4 % (39.28) + 3.96% (23.07)
I
Table II. Velocities of isometric contraction until achievement of 50% and 100% voluntary tooth clenching by fight and left anterior temporalis and masseter muscles in 28 control and 29 experimental subjects
Left side +2.88% (32.95) - 1.39% (23.18)
Data are arithmetic mean values (+-SD). p > 0.05.
23.18, ± S E 4.41), indicating that the mean anterior joint space was larger (Table I). Again, the difference between the two mean ratios was not significant (0.50 < p < 0.60). Arithhaetic mean values and their standard deviations for the velocities of isometric contraction are listed in Table II. In all cases, variance ratio testing (F test) showed homogeneity (p > 0.05). In other words, all variances pertaining to the control and experimental data were homogenous, so the control and experimental groups must be assumed to have come from populations with identical variances. In the fight AT muscle of the control and experimental subjects, the contraction velocities up to 100% MVC showed mean values of 2.45 IxV/ms (=+_SD 1.62, ± S E 0.22) and 2.49 IxV/ms ( ± S D 1.93, ± S E 0.25), respectively; a difference of 0.04 IxV/ms was not significant (0.80 < p < 0.90). In the right AT muscle of the control and experimental subjects, the mean velocities of contraction up to 50% MVC were 3.43 I.zV/ms ( ± S D 2.13, ± S E 0.29) and 3.38 p.V/ms ( ± S D 1.91, ± S E 0.25), respectively; a difference of 0.05 IxV/ms was not significant (0.80 < p < 0.90). In the left AT muscle of the control and experimental subjects, the contraction velocities up to 100% MVC showed mean values of 2.59 IxV/ms ( ± SD 1.43, ± SE 0.19) and 2.70 p.V/ms ( _ SD 1.78, ± SE 0.23), respectively; a difference of 0.11 IxV/ms was not significant (0.70 < p < 0.80). In the left AT muscle of the control and experimental subjects, the mean velocities of contraction up to 50% MVC were 3.76 p.V/ms ( ± S D 1.92, ± S E 0.25) and 3.46 IxV/ms (+--SD 1.99, ± S E 0.26), respectively; a difference of 0.30 ixV/ms was not significant (0.40 < p < 0.50). In summary and in regard to isometric contraction velocities of the AT muscle, it was not possible to demonstrate a significant difference between the treated (experimental) and untreated (control) subjects. In the fight MA muscle of the control and experimental subjects, the velocities of contraction up to 100% MVC showed mean values of 2.65 p.V/ms
Contraction velocity (I.tVIms) Muscle
50% MVC
[ I
100% MVC
Control rAT IAT rMA IMA
3.43 3.76 4.05 4.07
(2.13) (i.92) (2.06) (2.00)
2.45 2.59 2.65 2.52
(1.62) (1.43) (1.71) (1.55)
3.38 3.46 3.66 3.88
(1.91) (1.99) (2.41) (2.16)
2.49 2.70 2.43 2.69
(1.93) (1.78) (I.77) (1.75)
Experimental rAT IAT rMA IMA
Contraction velocities determined through bipolar surface electromyography of maximum voluntary teeth clenching (MVC). Data are arithmetic mean values ( - S D ) . p > 0.05.
( _ S D 1.71, _ S E 0 . 2 2 ) and 2.43 p.V/ms (-4-SD 1.77, _ SE 0.23), respectively; a difference of 0.22 I~V/ms was not significant (0.40 < p < 0.50). Also in the fight MA muscle, the contraction velocities up to 50% MVC showed a mean control value of 4.05 lzV/ms (-4- SD 2.06, --- SE 0.27) and a mean experimental value of 3.66 p.V/ms ( + S D 2.41, ± S E 0.31); a difference of 0.39 tzV/ms was not significant (0.20 < p < 0.30). In the left MA muscle of the control and experimental subjects, the mean velocities of contraction up to 50% MVC were 4.07 IxV/ms ( _ S D 2.00, ± S E 0.26) and 3.88 p.V/ms ( ± S D 2.16, _ S E 0.28), respectively; a difference of 0.19 tzV/ms was not significant (0.60 < p < 0.70). Also in the left MA muscle, the contraction velocities up to 100% MVC showed a mean control value of 2.52 p.V/ms ( - S D 1.55, _ S E 0.20) and a mean experimental value of 2.69 IxV/ms ( ± S D 1.75, ± S E 0.23); a difference of 0.17 IxV/ms was not significant (0.50 < p < 0.60). In summary and in regard to the isometfic contraction velocities of the MA muscle, there were no significant differences between control and experimental subjects. Special attention was given to the eight experimental subjects with bilateral extraction of the maxillary first premolars. In the right AT muscle of the 28 control subjects and the eight experimental subjects, the velocities of contraction up to 100% MVC showed mean values of 2.45 tzV/ms (Table II) and 2.93 I~V/ms (.4-SD 1.99, - S E 0.49), respectively; a difference of 0.48 p.V/ms was not significant (0.30 < p < 0.40).
VolumeI00 Number 2 Also in the right AT muscle, the contraction velocities up to 50% MVC showed a mean control value of 3.43 lzV/ms (Table II) and a mean experimental value of 3.25 p.V/ms (+_SD 1.47, ___SE0.36); a difference of 0.18 l~V/ms was not significant (0.70 < p < 0.80). In the left AT muscle, the velocities of contraction up to 100% MVC showed a mean control value of 2.59 IxV/ms (Table II) and a mean experimental value of 3.17 IxV/ms ( _ S D 2.30, _ S E 0.57); a difference of 0.58 p.V/ms was not significant (0.20 < p < 0.30). Also in the left AT muscle, the contraction velocities up to 50% MVC showed a mean control value of 3.76 p,V/ms (Table II) and a mean experimental value of 3.86 p.V/ms ( - S D 2.41, ± S E 0.60); a difference of 0.10 IxV/ms was not significant (0.80 < p < 0.90). In the right MA muscle of the 28 control subjects and the selected 8 experimental subjects, the velocities of contraction up" to 100% MVC showed mean values of 2.65 IxV/ms (Table II) and 2.45 0 N / m s (--- SD 2.45, _ SE 0.61), respectively; a difference of 0.20 pN/ms was not significant (0.60 < p < 0.70). The contraction velocities until 50% MVC of the right MA muscle showed a mean control value of 4.05 ixV/ms (Table II) and a mean experimental value of 3.46 I-tV/ms ( _ S D 2.95, - S E 0.73); a difference of 0.59 tzV/ms was not significant (0.20 < p < 0.30). In the left MA muscle, the contraction velocities up to 100% MVC showed a mean control value of 2.52 IxV/ms (Table II) and a mean experimental value of 2.10 p.V/ms ( - S D 1.87, "4-SE 0.46); a difference of 0.42 p.V/ms was not significant (0.30 < p < 0.40). The contraction velocities up to 50% MVC of the left MA muscle had a mean control value of 4.07 p.V/ms (Table II) and a mean experimental value of 3.65 IxV/ms ( ± S D 2.08, _ S E 0.52); a difference of 0.42 p.V/ms was not significant (0.40 < p < 0.50). In summary, extraction of the maxillary first premolars did not significantly affect the isometric contraction velocities of the AT and MA muscles. In regard to the relative activity of the right MA and AT muscles (Table III), the control subjects showed a mean activity index of - 1 . 2 % ( ± S D 24.7, ± S E 4.7), and the experimental subjects showed a mean activity index of + 3 . 1 % ( ± S D 15.1, ± S E 2.8). The difference was not significant (0.80 < p < 0.90). In other words, during brief MVC the experimental subjects showed an insignificant higher contractile activity in the MA muscle, and the control subjects showed the insignificantly higher contractile activity in the AT muscle (Table III). In regard to the left MA and AT muscles, the control subjects showed a mean activity index of - 2 . 8 % (___SD 22.8, ± S E 4.4) and the experimental subjects show a mean activity index of - 3.0%
Premolar extraction therapy 113 Table III. Electromyographic activity indices for
right and left masseter and anterior temporalis muscles in 28 control and 29 experimental subjects exercising maximum voluntary tooth clenching
I S,tbjects Control Experimental
Activii index Right side Left side - 1.2% (24.7) +3.1% (15.1)
- 2 . 8 % (22.8) - 3 . 0 % (21.2)
Data are arithmetic mean values (_+SD). p > 0.05.
( _ S D 21.2, ± S E 4.0); the difference was not significant (0.80 < p < 0.90). In other words, both the control and experimental subjects showed a slightly higher contractile activity in the AT muscle during MVC (Table III). DISCUSSION
This study evaluated aspects of the "static form" and the "dynamic function" of the mandibular locomotor system in orthodontically treated (experimental) subjects and untreated (control) subjects. The TMJs were evalu~ited through corrected axis tomography, and the mandibular elevator muscles were evaluated through EMG. Tomography was chosen for examination of the anterior and posterior joint spaces of the TMJs in the static position of maximum intercuspation (IP) of the teeth. Each joint space was measured as the subjective shortest linear distance between the mandibular condyle and the TMJ fossa as suggested by Pullinger and Hollender. 7 This method yields a fairly reliable joint space ratio (expressed as a percentage), the frequency distribution of which can be normally distributed through arcsine transformations. 9"~ Note also that the joint space ratio in the majority of, if not all, cases is an expression of indiyidual condylar morphology when the teeth are in IP and not an expression of condylar displacement as often stated in the litera'tureY 6"" The reason for this is that in the horizontal plane the individual condyle shows great variation in its contour, which, when imaged sagittally on a tomogram, will express itself as variable anterior as well as posterior linear distances between condyle and fossa, t8 In the right TMJ, the joint space ratio of the control subjects showed a mean value of - 1 . 8 4 % (Table I), suggesting that, on the average, the anterior joint space was the larger; and in the experimental subjects the joint space ratio showed a mean of + 3.96% (Table I),
114
Kundinger et al.
suggesting that, on the average, the posterior joint space was the larger. When one considers that the coefficients of variation (CV = SD/~) were extremely large (CVs of 2134% and 582%, respectively), it makes sense that there was no significant difference between the two ratios. This, in turn, can be attributed to highly variable individual condylar morphology) 8"19 In the left TMJ, the joint space ratio of the control subjects showed a mean value of + 2.88% (Table I), suggesting that, on the average, the posterior joint space was the larger; in the experimental subjects the ratio showed a mean of - 1.39% (Table I), suggesting that, on the average, the anterior joint space was the larger. Again, the difference between the two ratios was not significant, and the reversal of positive and negative values suggests highly variable individual condylar morphology. It is, of course, impossible to state whether the individual condylai" morphology of the experimental and control subjects was due to normal variability or individual remodeling, but it is Well known that condylar morphology shows large individual variations. ~8"-° On the method of TMJ tomography, the present study disclosed nosignificant differences between control and experimental subjects. This finding, made in the present cross-sectional study, is in agreement with the longitudinal study by Gianelly et al. 4 They found no difference in the so-called condylar positions of untreated and extraction-treated subjects. Electromyography was chosen for examination of the isometric contraction velocities and isometric motor innervation patterns (activity index) of the masseter and anterior temporalis muscles in the control and experimental subjects. The isometric contraction velocity, until 50% and 100% voluntary teeth clenching effort (MVC) is achieved, is an expression of the rapidity of recruitment of motor units. That is to say, as MVC rapidly increases so does the number of active motor units and their firing rates. 9~2 Table II shows that during the first half (50%) of the MVC contraction, the velocity of contraction was about 1.5 times faster than that of the entire (0% to 100%) contraction. This is consistent with other observations) °,~2 On the objective of this study, it should be noted that there were no significant differences between the control and experimental subjects. In other words, this aspect of EMG provided no evidence of adverse effects of orthodontic treatment combined with extraction of premolar teeth. The activity index of MVC is an expression of the relative contribution of the masseter and anterior temporalis muscles to maximum voluntary bite force. At high levels of bite force, the index either shows positive values or fluctuates around zero, implying that, in com-
Am. J. Orthod. Dentofac. Orthop. August 1991
parison with low levels of force, masseter.muscle activity predominates over that of the anterior temporalis muscle. 1°-~3In this study, both the control and experimental subjects showed mean index values that fluctuated around zero (Table III), and there were no significant differences between the two groups. In other words, this aspect of EMG also provided no evidence of adverse effects of orthodontic treatment combined with extraction of premolar teeth. At the time of this article, there have been no published reports other than that of Gianelly et al., 4 specifically evaluating the effects of extraction treatment, but their results and those of the present study, applying TMJ radiography and jaw muscle EMG, complement each other. That is to say, the combined results do not support the contentions t3 that extraction of premolar teeth and subsequent orthodontic therapy lead to irreparable damage to the temporomandibular joints and the jaw muscles. REFERENCES i. Witzig JW, Spahl TJ. The clinical management of basic maxillofacial orthopedic appliances. Vol. 1. Mechanics. Boston, Mass.: PSG Publishing, 1987:156-7, 183. 2. Witzig JW, Spahl TJ. The clinical management of basle maxillofacial orthopedic appliances. Vol. 2. Diagnosis. Boston, Mass.: PSG Publishing, 1987:221-4. 3. Farrar WB, McCarthy WL. A clinical outline of temporomandibular joint diagnosis and treatment. Montgomery, Ala.: Walker Publishing, 1983:84-8. 4. Gianelly AA, Hughes HM, Wohlgemuth P, Gildea G. Condylar position and extraction treatment. AM J OR~IOD DENTOFACOR'rHoP 1988;93:201-5. 5. Williamson EH, Wilson CW. Use of a submental-vertex analysis for producing quality temporomandibularjoint laminagraphs. AM J OR'I'ItOD 1976;70:200-7. 6. Beckwith PJ, Monfort DR, Williams BH. Accurate depth of cut in temporomandibular joint laminagraphs. Angle Orthod 1'980; 50:16-22. 7. Pullinger AG, Hollender L. Assessment of mandibular condyle position: a comparison of transcranial radi~raphs and linear tom , r a m s . Oral Surg Oral Med Oral Pathol 1985;60:329-34. 8. Pullinger AG, Solberg WK, Hollender L, Petersson A. Relationships of mandibular condylar position to dental occlusion factors in an asymptomatic population. AM J OR'mOP DENTOFACORTHOP 1987;91:200-6. 9. Christensen LV. Experimental teeth clenching in man. Swed Dent J 1989, Suppl 60, 10-54. 10. Christensen EV, Donegan SJ. Observations in the time and frequency domains of surface electromyograms of experimental brief teeth clenching in man. J Oral Rehabil 1990;17:473-86. I 1. Christensen LV, Carr AB, Donegan SJ, Ziebert GJ. Observations on the motor control of brief teeth clenching in man. J Oral Rehabil 1991;18:15-29. 12. Donegan SJ, Carr AB, Christensen LV, Ziebert GJ. An electromyographic study of aspects of 'deprogramming' of human jaw muscles. J Oral Rehabil 1990;17:509-18. 13. Naeije M, McCarroll RS, Weijs WA. Electromyographic activity of the human masticatory muscles during submaximal clenching in the intercuspal position. J Oral Rehabil 1989;16:63-70.
Volume I00 Number 2
Premolar extraction therapy
14. Bartlett MS. the use of transformations. Biometrics 1947;3:3952. 15. Zar JH. Biostatistical analysis. Englewood Cliffs, N.J.: PrenticeHall, 1984:130-1. 16. WeinbergLA. What we really see in a TMJ radiograph. J Prosthet Dent 1973;30:898-913. 17. Kundert M. Limits of perceptibility of condyle displacements on temporomandibularjoint radi~raphs. J Oral Rehabi11979;6: 375-83. 18. Raustia AM, Pyhtinen J. Morphology of the condyles and mandibular fossa as seen by computed tomography. J Prosthet Dent 1990;63:77-82.
115
19. Oberg T, Carlsson GE. Macroscopic and microscopic anatomy of the temporomandibularjoint. In: Zarb GA, Carlsson GE, eds. Temporomandibular joint function and dysfunction. St Louis: CV Mosby, 1979:101-18. 20. Carlsson GE, Oberg T. Remodelling of the temporomandibular joint. In: Zarb GA, Carlsson GE, eds. Temporomandibularjoint function and dysfunction. St Louis: CV Mosby, 1979:155-74. Reprint requests to:
Dr. L.V. Christensen School of Dentistry Marquette University Milwaukee, WI 53233
BOUND VOLUMES AVAILABLE TO SUBSCRIBERS Bound volumes of the AMERICAN JOURNAL OF ORTHODONTICSAND DENTOFACIAL ORTHOPEDICS are available to subscribers (only) for the 1991 issues from the Publisher, at a cost of $48.00 ($62.36 Canada and $59.00 international) for Vol. 99 (January-June) and Vol. 100 (July-December). Shipping chaiges are included. Each bound volume contains a subject and author index and all advertising is removed. Copies are shipped within 60 days after publication o f the last issue in the volume. The binding is durable buckram with the journal name, volume number, and year stamped in gold on the spine. P a y n z e n t m u s t a c c o m p a n y a l l orders. Contact M o s b y - Y e a r Book, Inc., Circulation Department, 11830 Westline Industrial Drive, St. Louis, M O 63146-3318, USA; telephone (800)325-4177, ext. 4351.
Subscriptions must be in force to qualify. Bound volumes are not available in place of a regular Journal subscription.
Experimental subjects (n = 29) were patients who had undergone orthodontic treatment in combination with extraction of maxillary or mandibular premolar teeth, or both. Control subjects (n = 29) were healthy dental students with no orthodontic or extraction experience. Sagittal (corrected axis) tomograms of the TMJs were used to determine the narrowest linear distances between the anterior and posterior outlines of the TMJ condyle and the TMJ fossa, expressed as the joint space ratio. There were no significant (p > 0.05) differences between the control and experimental ratios. Bipolar surface electromyograms of the masseter and anterior temporalis muscles were used to determine the isometric contraction velocities of these muscles until 50% and 100% voluntary isometric contraction effort (teeth clenching) was achieved. There were no significant (p > 0.05) differences between the control and experimental subjects. Electromyogr~.ms were also used to determine the relative contribution of the masseter and anterior temporalis muscles to the bite force developed during brief maximum voluntary tooth clenching, expressed as the activity index. There were no significant (p > 0.05) differences between the control and experimental subjects. (AM J OnTHOD DENTOFACORTHOP 1991 ;100:110-5.)
T h e extraction of premolars as a practical form of orthodontic therapy has been accepted for many years. Recently the orthodontic community has been questioned about this form of therapy, and assertions have been made that the temporomandibular joints (TMJs) are being adversely affected as a result of extraction treatment. The contention has been that premolar extraction, with retraction of the premaxillas, displaces the mandible and the mandibular condyles posteriorly; "there is an extremely high incidence of temporomandibular joint problems seen in patients with Class II, division 2 deep bite malocclusions, or patients with overretracted maxillary anteriors due to premolar extractions. ''''2 Farrar and McCarthy 3 were also of the opinion that a predisposing factor in TMJ anterior disk displacement is orthodontic therapy after extraction of premolar teeth. On the other hand, in their study of condylar position and TMJ status, Gianelly et al. 4 found no difference in the positions of the condyles of untreated patients and extraction-treated patients. Relative to premolar extraction therapy, the objective of this study was to evaluate the TMJs in terms
From the Marquette University School of Dentistry. 8/1/22729
110
of joint space ratios, often referred to as condylar position/displacement. Another objective was to evaluate jaw muscle function in terms of electromyography (EMG) of voluntary isometric contractions--a method not applied previously in this context.
MATERIALS AND METHODS The subjects were volunteers who had given informed consent to participate in the study. The sample of experimental subjects consisted of orthodontic patients in retention who had been treated with extraction of premolar teeth. Experimental subjects were 10 male and 19 female patients, ranging in age from 15 years 5 months to 36 years 1 month (mean age, 22 years 7 months). Eight of the experimental subjects had only maxillary" first premolars extracted (Angle Class II occlusion), wl{ereas the remaining 21 experimental subjects had four premolars extracted (Angle Class I occlusion). The average number of permanent teeth present was 24. The control subjects were 16 male and 13 female dental students who had not undergone orthodontic/extraction treatment. Their ages ranged from 21 years 2 months to 48 years 4 months (mean age, 27 years 3 months). Twenty-one control subjects had Class I occlusion, two had Class II, subdivision left malocclusion, and five had Class II, Division 1 malocclusion; one control subject had a Class III malocclusion. In the control group, the average number of permanent teeth present was 28.
Voh~me 100 Number 2
Premolar extraction therapy
Radiographic assessment of the TMJs included a vertical (submentovertex) projection followed by a lateral projection (sagittal tomography). Submentovertex radiographs were taken to obtain mediolateral condyle angulations and selective depth-of-cut measurements, complying with the protocols of Williamson and Wilson5 and Beckwith et al. 6 This projection used the Quint x-ray Sectograph (Denar Corporation, Anaheim, Calif.), and the subject-to-film distance was constant for each subject at I I mm. TMJ tracings were made on transparent acetate paper with a 0.5 mm lead pencil, and all measurements were taken from the machine midline. Subsequently, corrected right and left tomograms (one cut per side) were taken through the anatomic center of the mandibular condyle with the subject biting in the intercuspal position (IP). Each tomogram was cut and mounted in a conventional 35 mm slide holder and enlarged by means of a Kodak slide projector with a 103 mm lens. At each viewing, the screento-projector distance was calibrated to a fixed distance ensuring a projected in'age of × 10 magnification. The tomographic joint space was determined by the consensus of two examiners. Linearmeasurements were made of the narrowest radiographic (tomographic) anterior and posterior joint space, the shortest linear distance between the tomographic outline (contour) of the mandibular condyle and the tomographic outline (contour) of the glenoid fossa. 7.s The two linear distances were measured with the aid of the Mitutoya digimatic calipers read to the nearest 0.01 mm. If the shortest distance became difficult to read, that particular slide was remeasured at another time. If the measurements were within ---5 mm, the two measurements were averaged. Those joint space measurements that were greater than ± 5 mm were again measured, and the points were agreed on by three examiners. It was not possible to determine the joint spaces in one experimental subject and in three control subjects. According to the protocol of Pullinger and Hollender,7-*the subjective narrowest joint space (JS) ratio was expressed as Posteriorjs - A n t e r i o r , s × 100. Posterior~s + Anterior~s The JS ratio ranges from ±100% to - I 0 0 % , with + 100% indicating the presence of a posterior joint space only and - 100% indicating the presence of an anterior joint space only. Because a ratio of zero indicates equal anterior and posterior narrowest joint spaces, a positive ( + ) ratio indicates that the posterior joint space is larger than the anterior space at the closest approximation of the condyle to the fossa, and a negative ( - ) ratio indicates the opposite. For EMG, subjects were seated upright, with no headrest, in a natural head posture and with the feet flat on the floor. A verbal instruction was given to perform, briefly and rapidly, vertically directed maximum voluntary tooth clenching (MVC) in the mandibular position of IP. The isometric MVC exercise, which occurred within about 1 second was repeated after an interval of about 3 minutes, complying with the protocol of Christensen.9 Bipolar surface EMGs were obtained from the right and left masseter (MA) and anterior temporalis (AT) muscles. Circtflar Ag/AgCI electrodes were
111
placed over the palpated main bulks of the muscles, and a ground electrode was attached to the neck. The myoeleetrie signals were sampled on-line, followed by analog-to-digital conversion (total A I D conversion time of 1.5 microseconds per channel, with a resolution of 8 bits). Quantification of EMG (amplification of 3125 × ; gain of 200 p.V) was done by the mean voltage (I.tV) of 2000 milliseconds data records (BioResearch Inc., Milwaukee, Wis.). The EMG method has been described in detail elsewhere) °-~2 For each EMG, the following variables were determined by hand: the time, in milliseconds (ms), to peak (100%) mean voltage (p.V) of MVC in IP; the time (ms) to one-half peak (50%) mean voltage of MVC in IP. For the right and left MA and AT muscles, the two measurements allowed calculation of the velocity of isometric contraction (p.V/ms) from 0% to 100% and from 0% to 50% MVC in IP. The onset of MVC activity was determined on concurrent raw- and mean-voltage EMGs. To determine the relative contribution of the right and left MA and AT muscles to synergistic muscle activity (IxV) during MVC, the following activity index was used. EMG.~tA - EMGAT x 100 EMG.,IA + EMGAr The index ranges from + 100% to - I00%, with activity in MA only = + 100%, equal activity in MA ahd AT = 0, and activity in AT only = - 100%. u'13 Because.the frequency distributions of the ratios pertaining to joint spaces and MA/AT synergistic activities were not distributed normally, arcsine (sin -z) transformations were used to obtain gaussian frequency distribiations of all percentage data. 9't~ For data reductions, the two-tailed variance ratio test (F test) was used, followed by the two-tailed twosample Student's t test. The F test was applied to detect either homogeneity or heterogeneity of variances, and the t test was applied to detect significant differences between arithmetic mean values. In cases of heterogenous variances, Welch's approximate t test with corrected degrees of freedom was used? 5 The level of statistical significance was set at 5% (0.05).
RESULTS In the right TMJ (Table I), the joint space ratio of the control subjects (n = 26) showed a mean value of 1.84% ( - SD 39.28, ._ SE 7.90), indicating that the mean anterior joint space was larger. In the experimental.subjects (n = 28) the joint space ratio showed a mean of + 3 . 9 6 % ( ± S D 23.07, _ S E 4.39), indicating that the mean posterior joint space was larger (Table I). However, the difference between the two mean ratios was not significant (0.80 < p < 0.90). In the left TMJ (Table I), the mean joint space ratio of the control subjects showed a value of + 2 . 8 8 % ( ± S D 32.95, + SE 6.58), indicating that the mean posterior joint space was larger. In the experimental subjects the joint space ratio showed a mean of - 1 . 3 9 % (___SD -
112
Kundinger et al.
Am. J. Orthod. Dentofac. Orthop. August 1991
Table I. Tomographic joint space ratios for fight
and left temporomandibular joints in 26 control and 28 experimental subjects Joint space ratio Subjects
Right side
Control Experimental
- 1 . 8 4 % (39.28) + 3.96% (23.07)
I
Table II. Velocities of isometric contraction until achievement of 50% and 100% voluntary tooth clenching by fight and left anterior temporalis and masseter muscles in 28 control and 29 experimental subjects
Left side +2.88% (32.95) - 1.39% (23.18)
Data are arithmetic mean values (+-SD). p > 0.05.
23.18, ± S E 4.41), indicating that the mean anterior joint space was larger (Table I). Again, the difference between the two mean ratios was not significant (0.50 < p < 0.60). Arithhaetic mean values and their standard deviations for the velocities of isometric contraction are listed in Table II. In all cases, variance ratio testing (F test) showed homogeneity (p > 0.05). In other words, all variances pertaining to the control and experimental data were homogenous, so the control and experimental groups must be assumed to have come from populations with identical variances. In the fight AT muscle of the control and experimental subjects, the contraction velocities up to 100% MVC showed mean values of 2.45 IxV/ms (=+_SD 1.62, ± S E 0.22) and 2.49 IxV/ms ( ± S D 1.93, ± S E 0.25), respectively; a difference of 0.04 IxV/ms was not significant (0.80 < p < 0.90). In the right AT muscle of the control and experimental subjects, the mean velocities of contraction up to 50% MVC were 3.43 I.zV/ms ( ± S D 2.13, ± S E 0.29) and 3.38 p.V/ms ( ± S D 1.91, ± S E 0.25), respectively; a difference of 0.05 IxV/ms was not significant (0.80 < p < 0.90). In the left AT muscle of the control and experimental subjects, the contraction velocities up to 100% MVC showed mean values of 2.59 IxV/ms ( ± SD 1.43, ± SE 0.19) and 2.70 p.V/ms ( _ SD 1.78, ± SE 0.23), respectively; a difference of 0.11 IxV/ms was not significant (0.70 < p < 0.80). In the left AT muscle of the control and experimental subjects, the mean velocities of contraction up to 50% MVC were 3.76 p.V/ms ( ± S D 1.92, ± S E 0.25) and 3.46 IxV/ms (+--SD 1.99, ± S E 0.26), respectively; a difference of 0.30 ixV/ms was not significant (0.40 < p < 0.50). In summary and in regard to isometric contraction velocities of the AT muscle, it was not possible to demonstrate a significant difference between the treated (experimental) and untreated (control) subjects. In the fight MA muscle of the control and experimental subjects, the velocities of contraction up to 100% MVC showed mean values of 2.65 p.V/ms
Contraction velocity (I.tVIms) Muscle
50% MVC
[ I
100% MVC
Control rAT IAT rMA IMA
3.43 3.76 4.05 4.07
(2.13) (i.92) (2.06) (2.00)
2.45 2.59 2.65 2.52
(1.62) (1.43) (1.71) (1.55)
3.38 3.46 3.66 3.88
(1.91) (1.99) (2.41) (2.16)
2.49 2.70 2.43 2.69
(1.93) (1.78) (I.77) (1.75)
Experimental rAT IAT rMA IMA
Contraction velocities determined through bipolar surface electromyography of maximum voluntary teeth clenching (MVC). Data are arithmetic mean values ( - S D ) . p > 0.05.
( _ S D 1.71, _ S E 0 . 2 2 ) and 2.43 p.V/ms (-4-SD 1.77, _ SE 0.23), respectively; a difference of 0.22 I~V/ms was not significant (0.40 < p < 0.50). Also in the fight MA muscle, the contraction velocities up to 50% MVC showed a mean control value of 4.05 lzV/ms (-4- SD 2.06, --- SE 0.27) and a mean experimental value of 3.66 p.V/ms ( + S D 2.41, ± S E 0.31); a difference of 0.39 tzV/ms was not significant (0.20 < p < 0.30). In the left MA muscle of the control and experimental subjects, the mean velocities of contraction up to 50% MVC were 4.07 IxV/ms ( _ S D 2.00, ± S E 0.26) and 3.88 p.V/ms ( ± S D 2.16, _ S E 0.28), respectively; a difference of 0.19 tzV/ms was not significant (0.60 < p < 0.70). Also in the left MA muscle, the contraction velocities up to 100% MVC showed a mean control value of 2.52 p.V/ms ( - S D 1.55, _ S E 0.20) and a mean experimental value of 2.69 IxV/ms ( ± S D 1.75, ± S E 0.23); a difference of 0.17 IxV/ms was not significant (0.50 < p < 0.60). In summary and in regard to the isometfic contraction velocities of the MA muscle, there were no significant differences between control and experimental subjects. Special attention was given to the eight experimental subjects with bilateral extraction of the maxillary first premolars. In the right AT muscle of the 28 control subjects and the eight experimental subjects, the velocities of contraction up to 100% MVC showed mean values of 2.45 tzV/ms (Table II) and 2.93 I~V/ms (.4-SD 1.99, - S E 0.49), respectively; a difference of 0.48 p.V/ms was not significant (0.30 < p < 0.40).
VolumeI00 Number 2 Also in the right AT muscle, the contraction velocities up to 50% MVC showed a mean control value of 3.43 lzV/ms (Table II) and a mean experimental value of 3.25 p.V/ms (+_SD 1.47, ___SE0.36); a difference of 0.18 l~V/ms was not significant (0.70 < p < 0.80). In the left AT muscle, the velocities of contraction up to 100% MVC showed a mean control value of 2.59 IxV/ms (Table II) and a mean experimental value of 3.17 IxV/ms ( _ S D 2.30, _ S E 0.57); a difference of 0.58 p.V/ms was not significant (0.20 < p < 0.30). Also in the left AT muscle, the contraction velocities up to 50% MVC showed a mean control value of 3.76 p,V/ms (Table II) and a mean experimental value of 3.86 p.V/ms ( - S D 2.41, ± S E 0.60); a difference of 0.10 IxV/ms was not significant (0.80 < p < 0.90). In the right MA muscle of the 28 control subjects and the selected 8 experimental subjects, the velocities of contraction up" to 100% MVC showed mean values of 2.65 IxV/ms (Table II) and 2.45 0 N / m s (--- SD 2.45, _ SE 0.61), respectively; a difference of 0.20 pN/ms was not significant (0.60 < p < 0.70). The contraction velocities until 50% MVC of the right MA muscle showed a mean control value of 4.05 ixV/ms (Table II) and a mean experimental value of 3.46 I-tV/ms ( _ S D 2.95, - S E 0.73); a difference of 0.59 tzV/ms was not significant (0.20 < p < 0.30). In the left MA muscle, the contraction velocities up to 100% MVC showed a mean control value of 2.52 IxV/ms (Table II) and a mean experimental value of 2.10 p.V/ms ( - S D 1.87, "4-SE 0.46); a difference of 0.42 p.V/ms was not significant (0.30 < p < 0.40). The contraction velocities up to 50% MVC of the left MA muscle had a mean control value of 4.07 p.V/ms (Table II) and a mean experimental value of 3.65 IxV/ms ( ± S D 2.08, _ S E 0.52); a difference of 0.42 p.V/ms was not significant (0.40 < p < 0.50). In summary, extraction of the maxillary first premolars did not significantly affect the isometric contraction velocities of the AT and MA muscles. In regard to the relative activity of the right MA and AT muscles (Table III), the control subjects showed a mean activity index of - 1 . 2 % ( ± S D 24.7, ± S E 4.7), and the experimental subjects showed a mean activity index of + 3 . 1 % ( ± S D 15.1, ± S E 2.8). The difference was not significant (0.80 < p < 0.90). In other words, during brief MVC the experimental subjects showed an insignificant higher contractile activity in the MA muscle, and the control subjects showed the insignificantly higher contractile activity in the AT muscle (Table III). In regard to the left MA and AT muscles, the control subjects showed a mean activity index of - 2 . 8 % (___SD 22.8, ± S E 4.4) and the experimental subjects show a mean activity index of - 3.0%
Premolar extraction therapy 113 Table III. Electromyographic activity indices for
right and left masseter and anterior temporalis muscles in 28 control and 29 experimental subjects exercising maximum voluntary tooth clenching
I S,tbjects Control Experimental
Activii index Right side Left side - 1.2% (24.7) +3.1% (15.1)
- 2 . 8 % (22.8) - 3 . 0 % (21.2)
Data are arithmetic mean values (_+SD). p > 0.05.
( _ S D 21.2, ± S E 4.0); the difference was not significant (0.80 < p < 0.90). In other words, both the control and experimental subjects showed a slightly higher contractile activity in the AT muscle during MVC (Table III). DISCUSSION
This study evaluated aspects of the "static form" and the "dynamic function" of the mandibular locomotor system in orthodontically treated (experimental) subjects and untreated (control) subjects. The TMJs were evalu~ited through corrected axis tomography, and the mandibular elevator muscles were evaluated through EMG. Tomography was chosen for examination of the anterior and posterior joint spaces of the TMJs in the static position of maximum intercuspation (IP) of the teeth. Each joint space was measured as the subjective shortest linear distance between the mandibular condyle and the TMJ fossa as suggested by Pullinger and Hollender. 7 This method yields a fairly reliable joint space ratio (expressed as a percentage), the frequency distribution of which can be normally distributed through arcsine transformations. 9"~ Note also that the joint space ratio in the majority of, if not all, cases is an expression of indiyidual condylar morphology when the teeth are in IP and not an expression of condylar displacement as often stated in the litera'tureY 6"" The reason for this is that in the horizontal plane the individual condyle shows great variation in its contour, which, when imaged sagittally on a tomogram, will express itself as variable anterior as well as posterior linear distances between condyle and fossa, t8 In the right TMJ, the joint space ratio of the control subjects showed a mean value of - 1 . 8 4 % (Table I), suggesting that, on the average, the anterior joint space was the larger; and in the experimental subjects the joint space ratio showed a mean of + 3.96% (Table I),
114
Kundinger et al.
suggesting that, on the average, the posterior joint space was the larger. When one considers that the coefficients of variation (CV = SD/~) were extremely large (CVs of 2134% and 582%, respectively), it makes sense that there was no significant difference between the two ratios. This, in turn, can be attributed to highly variable individual condylar morphology) 8"19 In the left TMJ, the joint space ratio of the control subjects showed a mean value of + 2.88% (Table I), suggesting that, on the average, the posterior joint space was the larger; in the experimental subjects the ratio showed a mean of - 1.39% (Table I), suggesting that, on the average, the anterior joint space was the larger. Again, the difference between the two ratios was not significant, and the reversal of positive and negative values suggests highly variable individual condylar morphology. It is, of course, impossible to state whether the individual condylai" morphology of the experimental and control subjects was due to normal variability or individual remodeling, but it is Well known that condylar morphology shows large individual variations. ~8"-° On the method of TMJ tomography, the present study disclosed nosignificant differences between control and experimental subjects. This finding, made in the present cross-sectional study, is in agreement with the longitudinal study by Gianelly et al. 4 They found no difference in the so-called condylar positions of untreated and extraction-treated subjects. Electromyography was chosen for examination of the isometric contraction velocities and isometric motor innervation patterns (activity index) of the masseter and anterior temporalis muscles in the control and experimental subjects. The isometric contraction velocity, until 50% and 100% voluntary teeth clenching effort (MVC) is achieved, is an expression of the rapidity of recruitment of motor units. That is to say, as MVC rapidly increases so does the number of active motor units and their firing rates. 9~2 Table II shows that during the first half (50%) of the MVC contraction, the velocity of contraction was about 1.5 times faster than that of the entire (0% to 100%) contraction. This is consistent with other observations) °,~2 On the objective of this study, it should be noted that there were no significant differences between the control and experimental subjects. In other words, this aspect of EMG provided no evidence of adverse effects of orthodontic treatment combined with extraction of premolar teeth. The activity index of MVC is an expression of the relative contribution of the masseter and anterior temporalis muscles to maximum voluntary bite force. At high levels of bite force, the index either shows positive values or fluctuates around zero, implying that, in com-
Am. J. Orthod. Dentofac. Orthop. August 1991
parison with low levels of force, masseter.muscle activity predominates over that of the anterior temporalis muscle. 1°-~3In this study, both the control and experimental subjects showed mean index values that fluctuated around zero (Table III), and there were no significant differences between the two groups. In other words, this aspect of EMG also provided no evidence of adverse effects of orthodontic treatment combined with extraction of premolar teeth. At the time of this article, there have been no published reports other than that of Gianelly et al., 4 specifically evaluating the effects of extraction treatment, but their results and those of the present study, applying TMJ radiography and jaw muscle EMG, complement each other. That is to say, the combined results do not support the contentions t3 that extraction of premolar teeth and subsequent orthodontic therapy lead to irreparable damage to the temporomandibular joints and the jaw muscles. REFERENCES i. Witzig JW, Spahl TJ. The clinical management of basic maxillofacial orthopedic appliances. Vol. 1. Mechanics. Boston, Mass.: PSG Publishing, 1987:156-7, 183. 2. Witzig JW, Spahl TJ. The clinical management of basle maxillofacial orthopedic appliances. Vol. 2. Diagnosis. Boston, Mass.: PSG Publishing, 1987:221-4. 3. Farrar WB, McCarthy WL. A clinical outline of temporomandibular joint diagnosis and treatment. Montgomery, Ala.: Walker Publishing, 1983:84-8. 4. Gianelly AA, Hughes HM, Wohlgemuth P, Gildea G. Condylar position and extraction treatment. AM J OR~IOD DENTOFACOR'rHoP 1988;93:201-5. 5. Williamson EH, Wilson CW. Use of a submental-vertex analysis for producing quality temporomandibularjoint laminagraphs. AM J OR'I'ItOD 1976;70:200-7. 6. Beckwith PJ, Monfort DR, Williams BH. Accurate depth of cut in temporomandibular joint laminagraphs. Angle Orthod 1'980; 50:16-22. 7. Pullinger AG, Hollender L. Assessment of mandibular condyle position: a comparison of transcranial radi~raphs and linear tom , r a m s . Oral Surg Oral Med Oral Pathol 1985;60:329-34. 8. Pullinger AG, Solberg WK, Hollender L, Petersson A. Relationships of mandibular condylar position to dental occlusion factors in an asymptomatic population. AM J OR'mOP DENTOFACORTHOP 1987;91:200-6. 9. Christensen LV. Experimental teeth clenching in man. Swed Dent J 1989, Suppl 60, 10-54. 10. Christensen EV, Donegan SJ. Observations in the time and frequency domains of surface electromyograms of experimental brief teeth clenching in man. J Oral Rehabil 1990;17:473-86. I 1. Christensen LV, Carr AB, Donegan SJ, Ziebert GJ. Observations on the motor control of brief teeth clenching in man. J Oral Rehabil 1991;18:15-29. 12. Donegan SJ, Carr AB, Christensen LV, Ziebert GJ. An electromyographic study of aspects of 'deprogramming' of human jaw muscles. J Oral Rehabil 1990;17:509-18. 13. Naeije M, McCarroll RS, Weijs WA. Electromyographic activity of the human masticatory muscles during submaximal clenching in the intercuspal position. J Oral Rehabil 1989;16:63-70.
Volume I00 Number 2
Premolar extraction therapy
14. Bartlett MS. the use of transformations. Biometrics 1947;3:3952. 15. Zar JH. Biostatistical analysis. Englewood Cliffs, N.J.: PrenticeHall, 1984:130-1. 16. WeinbergLA. What we really see in a TMJ radiograph. J Prosthet Dent 1973;30:898-913. 17. Kundert M. Limits of perceptibility of condyle displacements on temporomandibularjoint radi~raphs. J Oral Rehabi11979;6: 375-83. 18. Raustia AM, Pyhtinen J. Morphology of the condyles and mandibular fossa as seen by computed tomography. J Prosthet Dent 1990;63:77-82.
115
19. Oberg T, Carlsson GE. Macroscopic and microscopic anatomy of the temporomandibularjoint. In: Zarb GA, Carlsson GE, eds. Temporomandibular joint function and dysfunction. St Louis: CV Mosby, 1979:101-18. 20. Carlsson GE, Oberg T. Remodelling of the temporomandibular joint. In: Zarb GA, Carlsson GE, eds. Temporomandibularjoint function and dysfunction. St Louis: CV Mosby, 1979:155-74. Reprint requests to:
Dr. L.V. Christensen School of Dentistry Marquette University Milwaukee, WI 53233
BOUND VOLUMES AVAILABLE TO SUBSCRIBERS Bound volumes of the AMERICAN JOURNAL OF ORTHODONTICSAND DENTOFACIAL ORTHOPEDICS are available to subscribers (only) for the 1991 issues from the Publisher, at a cost of $48.00 ($62.36 Canada and $59.00 international) for Vol. 99 (January-June) and Vol. 100 (July-December). Shipping chaiges are included. Each bound volume contains a subject and author index and all advertising is removed. Copies are shipped within 60 days after publication o f the last issue in the volume. The binding is durable buckram with the journal name, volume number, and year stamped in gold on the spine. P a y n z e n t m u s t a c c o m p a n y a l l orders. Contact M o s b y - Y e a r Book, Inc., Circulation Department, 11830 Westline Industrial Drive, St. Louis, M O 63146-3318, USA; telephone (800)325-4177, ext. 4351.
Subscriptions must be in force to qualify. Bound volumes are not available in place of a regular Journal subscription.
