You are on page 1of 6

Journal of Dental Research http://jdr.sagepub.


The Three-dimensional Active Envelope of Jaw Border Movement and its Determinants
J.H. Koolstra, M. Naeije and T.M.G.J. Van Eijden J DENT RES 2001 80: 1908 DOI: 10.1177/00220345010800100901 The online version of this article can be found at:

Published by:

On behalf of:
International and American Associations for Dental Research

Additional services and information for Journal of Dental Research can be found at: Email Alerts: Subscriptions: Reprints: Permissions: Citations:

Downloaded from by guest on July 10, 2011 For personal use only. No other uses without permission.

J.H. Koolstral*, M. Naeije2, and T.M.G.J. van Eijdenl

'Department of Functional Anatomy and 2Department of Oral Function, Section on Craniomandibular Disorders, Academic Centre for Dentistry Amsterdam (ACTA), Meibergdreef 15, 1105 AZ Amsterdam, the Netherlands; *corresponding author,

The Three-dimensional Active Envelope of Jaw Border Movement and its Determinants
The excursions of the human lower jaw with respect to the skull are limited by constraints within the masticatory system. These limitations are considered to be a fundamental portion of the functional performance of the masticatory system, while all its separate functional contributions are related to these limitations (Heners, 1977). In the sagittal plane, these limits have a banana-like shape (Posselt, 1952) which is known as the socalled "Posselt figure". Such a figure can be constructed by tracing the path of the lower central incisors during a maximum open-close movement while keeping the jaw as far as possible in a forward or backward direction. Such so-called "border movements" can also be made while keeping the lower jaw to one side as far as possible. In this way, the envelope of border movement in a more or less frontal plane can be constructed (Nishigawa et al., 1991). Jaw movements are not restricted to the frontal and sagittal plane only. Border movements, therefore, can also be performed in directions oblique to these planes. The area that can be reached by the lower central incisors will have a three-dimensional shape. Presently, three-dimensional jaw movement analysis systems (e.g., Naeije et al., 1995) can be applied to trace the path of the lower incisor point during three-dimensional border movements (Fig. 1). The constraints that determine the envelope of border movement may be passive (articular surfaces, temporomandibular ligaments, passive forces of the muscles) and/or active (reflexes of muscles to protect the articular capsule) (Posselt, 1962; Lauer, 1985). At present, it is not known in what way the shape of the envelope of border movement is influenced by these factors. Therefore, although shape characteristics of the envelope may be correlated to masticatory function characteristics, no unambiguous causal relationships can be established. Unfortunately, it is not possible to study the determinants for the active envelope of border movement experimentally. Therefore, a biomechanical modeling approach has to be applied. This approach allows one to study the influence of one of the structures on the behavior of the complete system by adapting its mechanical or morphological characteristics. If model predictions can be validated experimentally, this has proved to be a powerful tool in masticatory function analysis (Koolstra and van Eijden, 1992). The purpose of the present study was to determine the influence of passive structures on the active three-dimensional envelope of border movement. This was accomplished by simulation of border movements with a three-dimensional dynamic model of the human masticatory system (Koolstra and van Eijden, 1995, 2001). Similar simulations were performed after removal of the temporomandibular ligaments or adaptation of the passive tensions of the masticatory muscles. While the masticatory muscles have a limited range of active shortening, it was hypothesized that they are the predominant factor for the shape of the envelope of border movement and that the influence of the passive constraints is much less.

J Dent Res 80(10):1908-1912, 2001

The sagittal and frontal active envelope of border movement is applied regularly as a clinical tool in functional examinations of the human masticatory system. In contrast, the three-dimensional movement area has hardly been examined. Furthermore, the determinants of this area are not established unambiguously. In the present study, the three-dimensional envelope of incisor movement was predicted with a three-dimensional mathematical model of the human masticatory system, which included the morphology of the system and the fine architecture of its muscles. With this model, the influence of the temporomandibular ligaments and the passive muscle tensions on the envelope were estimated. The predicted threedimensional active envelope of border movements was limited in horizontal directions, predominantly by the temporomandibular ligaments. The passive tensions of the masticatory muscles influenced, although marginally, its vertical extension. It appeared unlikely that, in a normal situation, active muscle tensions (casu quo muscle reflexes) contribute to the shape of the envelope.

KEY WORDS: masticatory muscles, jaw

movement, envelope, biomechanics.

Received October 11, 2000; Last revision September 10, 2001; Accepted September 20, 2001
A supplemental appendix to this article is published electronically only at


Downloaded from by guest on July 10, 2011 For personal use only. No other uses without permission.

J Dent Res 80(10) 2001

Jaw Border Movement Envelope



Border movemelnt simulationls were performied by mcans of a mathematical model ol the human masticatory system. This model enables one to analyze jawimovemilents with 6 degiees of freedom (Koolstra and van Eijden, 1995, 1996, 1997b, 1999). [For an extensive description, please refer to the Appendix (] The modcl consists of' a lower jaw /1 whichi is able to move freely in the thl-ee-dimilenisionial space. The jaw is accelerated by forces and accompanying torques with Sin: respect to its center o' gravity. applied by muscles (both active and passive muscIC tensions) and by joint surtfaccs, bite points, and ligaments througIl reaction forces. The muscles included were: :V00 superficial and deep masseter, anterior and posterior temporalis, 'P' medial ptcrygoid, superior and inferior lateral pterygoid, digastric, -wgeniohyoid, and mylohyoid (Fig. 2). The muscles weie assumed to :A run straight between their origins and insertions, except the superior lateral pterygoid, which was bent over the articular eminence when the condyle was located in the mandibular fossa. I MLuscle forces werc dependenit on the amouLit of activation, sarcomerc dynamics (forcc-length relationsliip, force-velocity relationshiip, passive muscle for-ce), and activation dyiiamics. Muscle fiber lengths, sarcomere lengths, and physiological crossFigure 1. Active three-dimensional envelope of border movement recorded from a 44-year-old male subject with no signs of sections weic obtained fromil van Eijden el ofl. ( 1995, 1996, 1997), temporomandibular disorders. The surfaces were fitted through the the dynamic muscle properties fromil van Ruijven and Weijs ( 1990), anteriormost and posteriormost locations of in total 48,000 incisor and activation dynamilics from Winter-s and Stark (1987). The positions obtained during various sagittal, mediolateral, and transverse ligaments were modeled as wires runlnilng from the lateral pole of border movements. (A) Cross-sectional frontal view. Bars: 1 cm. (B) Sagittal view. (C) Frontal view. the condyles to the lateral SUltface of' the articular eminlenices, which produced reaction forces wlien pulled taut. muscle model Recently, a method has been jaw model developed to predict adequatc musecle reernitellnlt patterns to movc the incisor point of the manaTEM_A dible to a desired position (KoolTEM P stra and van Eijden, 2001). I' this position was beyonid the active enLPT-S , velope of border movement, the LPT-1 , movement was stopped by, on the one side, passive conistraints and, .,.8b..s ' on the other side, by a tailule of' ..."..".. the muscles to move it any rirthcr. ,.'." The model was impleimented on an -M.10 IBM RS/6000 SP computcr sys-... AIX XL tem. It was written in FORTRAN, enabling parallelized .IEH / DIG computing to be done by means of' MPI (Message Passing Intcrface) Figure 2. Overview of the moc (Left) Jaw model in ventro-lateral view. Continuous lines: muscle lines of library routincs. action from origin (cross-bor) to insertion (circle). MAS_S, superficial masseter; MAS_P, deep masseter; In the present stLudy, we explored MPT, medial pterygoid; TEM_,4, anterior temporalis; TEM P, posterior temporalis; LPT S, superior lateral the limits of manldibular ioovemenit pterygoid; LPT_I, inferior laterral pterygoid; DIG, digastric; GEH, geniohyoid; MYH, mylohyoid. (Right) Muscle model. Total force is sL of sarcomere forces (Fsarcomeres). The active force (lF ,ve ) is dependent on by attempting incisor point movcactivation, instantaneous sar comere length, and contraction velocity. The passive tension (F pSSIve) is ments toward positions f'ar beyond dependent on instantoneous scarcomere length. the possible movemetit area. To cxplore the antei ior border of possiblc side was explored. TFle obtained border points wei-C irroied so that jaw movements, we located these positions about 1(0 cm in tront ol' we could obtain the complete three-dimilenisionial exivelope of the incisor point at rest and, for the posterior border, at about 10 Cmil movemenit. Three-dimensionial surfaces werc titted thioughl the bchinid it. Vertically and laterally, these positions were definied anterior and posterior border points by Matlab (the Math Works Inc., according to a 5 x 5 mm grid in fi-ontal planes ranging over 7 cm and Natick, MA, USA). We constructed three-dimensional eixvelopes ot 5 cm, rcspectively. EnougIl time (0.2 sec) was allowed to let the border movemeiit by combining the the subsequenit border points. Only guest onright 2011 For personal use only. No other uses without permission.these anterior and posterior surfaces. central incisors reach Downloaded from by July 10,

acbvanon dynDacs

..... .....r

ooz oon


ni-m (pm!





Koolstra et a/.
reference .....,..." ,,

J Dent Res 80(1 0) 2001





~ -i3wwm



W v









S dii .E


50% passive muscle tensions 200% passive muscle tensions



Figure 3. Border points as predicted by the model and the active threedimensional envelope of border movement constructed through these points. + = border points predicted for all maximally retruded positions. o = border points predicted for all maximally protruded positions. (A) Postero-lateral view. (B) Lateral view. (C) Antero-lateral view.
We explored the iniluence of the temporomilandibular ligaments on the active envelope of' border movxemeit by pertorminig a series of border movemenlt simulationis with a model lfrom whicih they werc rcmovcd. Thc iniluence ol' the passive elastic tension of the masticatory muscles was studicd by similar simulations while thc tension was dccreased or increased by a factor oft2.








RESULTS Border Movement Simulations

In Fig. 3, the simulated bordler points are displayed together with the surlaces fitted through them. The jaw could be opened over 3.4 cm. The maximum amount of' opening by rotation about an interconidylar axis only (deduced from the ridge about halfway fromil the posterior border of the envelope) was 2.3 cm. Maximum protrLision, retrusion, and laterodeviation were 9, 3, and 12.6 mim, respectively. The transverse cross-sections (Fig. 4A) of'the envelope have sharp edges at the right and left sides but are ilattenied anteriorly and posteriorly. Durinig the border movemilenits, the mandibular condyle maintained contact with the articular eminence ot'the skull. The largest antero-posterior excLirsions (1 .5 cm) were observed in the contralateral joint, and the largest medio-lateral excursions (0.7 cm) in the ipsilateral one (data not shown).

Figure 4. Active three-dimensional envelopes of border movement as predicted by the model. (A) Normal model. (B) Model without temporomandibular ligaments. (C) Model with reduced passive muscle tensions. (D) Model with increased muscle tensions. Left panels: crosssectional frontal view. Bars: 1 cm. Upper right panels: sagittal view. Lower right panels: frontal view.
(artilicially cxtended) temporal surtheCC, to retrude the inicisor as much as possible (data not shown). The applied chaniges in passive elastic properties of' the muscles had much less inIluelice on the cnxvclope of movexilemit. Reductioll ot' this property (Fig. 4C) led to a slightly larger rnaxilmti jaw openinig (3.6 cm). The moveminciit possibilities in the sagittal and fronital planes in the extended part of the envelope were small. The other maxilulIL exCursions wxere not aflfected. A doublinig of the passive elastic muscle forces (Fig. 4D) reduced maximlium jaw openinig to 3.1 cm. Also, maxilmlumIL protrusion, retrLisioni, and laterodeviation were slightly reduced to 7.7, 2.6, and II mim, respectively. Chaniges in this parameter did not affect the amount of'terimlinal hinige movemilenit. The condylar excursions were almost similar to those obtained with the norimial model. There was one exception: To reach the anterior border wheln the passive muscIe tensiolIs were reduced, the antero-posterior excursionl of' the ipsilateral condyle increased 6.6 mm, to about 19 mmll (data not shox n).

Influence of Passive Structures

The extension oh the active envelope of border movement became about 63% larger in the antero-posterior direction and about 27% in the rnedio-lateral direction after the temporomandibular ligaments were removed (Fig. 4B). The increasc was not found in the vertical direction; the maximum DISCUSSION amount of jaw opening remained the same. The amount of' Border Movement Simulations opening by rotation about the intercondylar axis was less (about x In the present study. we were able to explore the thiee1.7 cm), whereas the posterior prominence in the posterior dimensionial active envelope ol' border movemilenit. F urthermore, border of the envelope had almost disappeared. Maximum the inifluenice of the temporomanidibular ligamenits and the protrusion, retrusion, and laterodeviation were increased to 10, passive muscle tensionis On this envelope could be predicted. 9.4, and 16 mmr, respectively. The active envelope of' border movemilenit predicted by our Alter removal of the ligaments, the condyle was able to unmodified model (Fig. 4A) displayed featul-es similar- to those move beyond the natural borders of the area of articular obtained experimentally (Fig. I ). The vertical ranige ol' the the contralateral contact. The largest excursions were found for predicted movements, however, was smaller. A possible condyle, wliich moved 1.5 cm medially wheni the incisors explanation for this fact is an overestimationi of the passive moved to their most lateral position. Both condyles moved Downloaded from along the muscle only. No other (Langenbacih maximally about 3 cm in the posterior direction by guest on July 10, 2011 For personal use tensionisuses without permission. and llainiaii, 1999) (but Vide

J Dent Res 80(10) 2001

Jaw Border Movement Envelope


infra). Furthermore, the incorporated jaw-opening muscles could have been relatively weak. Their properties were estimated from cadaveric specimens. Although these specimens had a reasonable dentition, they were relatively old (van Eijden et al., 1995), and it can be assumed that their muscle strength was less than the average. One of the most likely explanations relates to head movements during jaw opening. It has been demonstrated that the upper jaw may move up to 50% of jaw opening (Eriksson et al., 1998; Kohno et al., 2001), herewith stretching the contracted jaw openers toward their optimum lengths. This mechanism presumably enables the jaw openers to produce enough force to counteract the stretching jaw closers to open the jaw much further than in the present simulations. Despite similar features between the simulated and measured envelopes of border movement, there were also differences. The differences were not larger than those observed in border movements from different subjects (Brown, 1975; Lewin, 1985; Nishigawa et al., 1991). The morphology of the relevant subject was not incorporated into the model (Koolstra and van Eijden, 1992), while such is possible only for macroscopic features such as overall muscle direction and cross-section. The fine muscle architecture incorporating sarcomere length and fiber length, which are assumed to play a dominant role for the active envelope of border movement, cannot be obtained in vivo without disturbing the integrity of the muscle tissue. The endpoints of the border movements were not distributed regularly over the simulated envelope of border movement (Fig. 3), although a 5 x 5 mm grid was applied for the putative destinations. This is a consequence of the applied procedure for generating muscle recruitment patterns (Koolstra and van Eijden, 2001), which attempted to move the incisors across, but not along, the border to get closer to a putative destination.

Influence of Temporomandibular Ligaments

The envelope of border movement is determined by both the possibilities of the masticatory muscles to move the mandible and the constraints that are enforced by the passive structures such as joint surfaces, ligaments, and passive muscle tensions. The influence of the joint surfaces was not examined separately in the present study, while changes within more or less natural limits (a few mm) were expected to alter the envelope of border movement only slightly. Furthermore, the shapes of the joints are not defined by a limited number of parameters which can be adapted to determine an unambiguous relationship between border movement envelope and joint shape. Finally, the contact area of the joint is presumably bounded by the articular capsule rather than by its shape. The surrounding soft structures, such as skin and glands, also may limit condylar excursions. These were not implemented in the model. Therefore, the mandibular condyle was able to move far beyond its normal movement area, and the resulting envelope was in the medio-lateral and anteroposterior direction, limited only by the passive tensions of the muscles and their possibilities for force generation.

of the envelope, however, was much smaller than generally assumed. By postulation of optimum length or 150% optimum length for all muscles at a jaw opening of 12 mm or 50 mm, respectively, wide-open jaw positions could be simulated (Langenbach and Hannam, 1999; Peck et al., 2000). However, human masticatory muscles do not reach optimum sarcomere length simultaneously (Koolstra and van Eijden, 1997a). The model applied optimum sarcomere length obtained from related animal masticatory muscles (Muhl et al., 1978), while this parameter has not been experimentally established in humans. We were reluctant to adapt optimum sarcomere length, without a physiological or experimental basis, only to reach a larger jaw opening. At the present maximum amount of jaw opening, the jaw closers apparently do not become stretched beyond the length where their passive forces become dominant. In the present study, the influence of active muscle forces on the active envelope of border movement was not examined separately. Since the simulated envelope was not larger than the measured one, it is unlikely that the maximum excursions of the lower jaw are limited actively by muscle reflexes (Posselt, 1962; Lauer, 1985). The differences in the active envelope of border movement predicted by the models with and without the presence of the temporomandibular ligament suggest that the articular capsule is loaded during extreme laterodeviation, protrusion, or retrusion. Extreme wide jaw opening without head movement presumably does not load these capsules. The difference between the model predictions as a function of the implemented amount of passive muscle tensions suggests that if the position of the hyoid bone with respect to the skull remains stationary, the jaw muscles are not stretched to the point where their passive forces become the dominant limiting factor for movement possibilities. The mechanism that renders the muscles insufficient for force production is suggested to be the dominant factor for vertical movement limitations.

We gratefully thank Academic Computing Services Amsterdam (SARA) for technical support. This research was institutionally supported by the Interuniversity Research School of Dentistry, through the Academic Centre of Dentistry Amsterdam.

Brown T (1975). Mandibular movements. Monogr Oral Sci 4:126-150. Eriksson PO, Zafar H, Nordh E (1998). Concomitant mandibular and head-neck movements during jaw opening-closing in man. J Oral Rehabil 25:859-870. Heners M (1977). Elektronische Untersuchungen zur dreidimensionalen Bewegung des Unterkiefers. Dtsch Zahndrztl Z 32:104-107. Kohno S, Matsuyama T, Medina RU, Arai Y (2001). Functionalrhythmical coupling of head and mandibular movements. J Oral Rehabil 28:161-167. Koolstra JH, van Eijden TMGJ (1992). Application and validation of a three-dimensional mathematical model of the human masticatory system in vivo. JBiomech 25:175-187. Koolstra JH, van Eijden TMGJ (1995). Biomechanical analysis of jaw closing movements. JDent Res 74:1564-1570. Koolstra JH, van Eijden TMGJ (1996). Influence of the dynamical properties of the human masticatory muscles on jaw closing movements. Eur J Morphol 34:11-18.

Influence of Passive Muscle Tensions

The passive muscle tensions influenced primarily the vertical dimension of the envelope of border movement. This was most clearly seen by the reduction or enlargement of the inferiormost part of the envelope, where the transverse cross-sectional area became very small. The influence on the maximum open part

Downloaded from by guest on July 10, 2011 For personal use only. No other uses without permission.


Koolstra et al.

J Dent Res 80(10) 2001

Koolstra JH, van Eijden TMGJ (1997a). Dynamics of the human masticatory muscles during a jaw open-close movement. J Biomech 30:883-889. Koolstra JH, van Eijden TMGJ (1997b). The jaw open-close movements predicted by biomechanical modelling. J Biomech 30:943950. Koolstra JH, van Eijden TMGJ (1999). Three-dimensional dynamical capabilities of the human masticatory muscles. J Biomech 32:145-152. Koolstra JH, van Eijden TMGJ (2001). A method to predict muscle control in the kinematically and dynamically indeterminate human masticatory system. JBiomech 34:1179-1188. Langenbach GEJ, Hannam AG (1999). The role of passive muscle tensions in a three-dimensional dynamic model of the human jaw. Arch Oral Biol 44:557-573. Lauer HC (1985). Uber die Abhangigkeit der Unterkiefergrenzbewegung von der Okklusionshohe. Dtsch Zahndrztl Z40:1016-1019. Lewin A (1985). Electrognathographics. Atlas of diagnostic procedures and interpretation. Chicago: Quintessence Publishing Co., Inc. Muhl ZF, Grimm AF, Glick PL (1978). Physiologic and histologic measurements of the rabbit digastric muscle. Arch Oral Biol 23:1051-1059. Naeije M, van der Weijden JJ, Megens CCEJ (1995). OKAS-3D: opto-

electronic movement recording system with six degrees of freedom. Med Biol Eng Comput 33:683-688. Nishigawa K, Nakano M, Bando E, Clark GT (1991). The relationship between lateral border movements of the mandible and the determinants of occlusion. JProsthet Dent 66:486-492. Peck CC, Langenbach GEJ, Hannam AG (2000). Dynamic simulation of muscle and articular properties during human wide jaw opening. Arch Oral Biol 45:963-982. Posselt U (1952). Studies in the mobility of the human mandible. Acta Odontol Scand 10(Suppl 10): 19-160. Posselt U (1962). Physiology of occlusion and rehabilitation. Oxford: Blackwell Scientific Publications. van Eijden TMGJ, Koolstra JH, Brugman P (1995). Architecture of the human pterygoid muscles. JDent Res 74:1489-1495. van Eijden TMGJ, Koolstra JH, Brugman P (1996). Three-dimensional structure of the human temporalis muscle. Anat Rec 246:565-572. van Eijden TMGJ, Korfage JAM, Brugman P (1997). Architecture of the human jaw-closing and jaw-opening muscles. Anat Rec 248:464-474. van Ruijven LJ, Weijs WA (1990). A new model for calculating muscle forces from electromyograms. Eur JAppl Physiol 61:479-485. Winters JM, Stark L (1987). Muscle models: what is gained and what is lost by varying model complexity. Biol Cybern 55:403-420.

Downloaded from by guest on July 10, 2011 For personal use only. No other uses without permission.