ORIGINAL ARTICLE

Moment-to-force ratio, center of rotation, and force level: A nite element study predicting their interdependency for simulated orthodontic loading regimens
Paolo M. Cattaneo,a Michel Dalstra,b and Birte Melsenc Aarhus, Denmark Introduction: Changes in the stress/strain distribution in the periodontium after the application of orthodontic forces trigger remodeling processes that make tooth movement possible. The type of orthodontic tooth movement is linked to the force system applied to the bracket. By combining moments and forces, often expressed as the moment-to-force (M/F) ratio, it is possible to determine the prescribed type of tooth movement. According to classical theory, xed values for M/F are associated with specic tooth movements. Methods: A segment of a mandible containing the canine and the rst premolar obtained from autopsy was scanned with microcomputed tomography, and a nite element model was generated. In a series of nite element analyses simulating teeth subjected to various orthodontic loading regimens, the inuences of the M/F ratio and the force magnitude were examined. Results: By applying a range of values of M/F, different types of tooth movement were generated, although the classic prescription of the M/F ratio suggested in the literature could not be conrmed. Due to the nonlinear behavior of the periodontal ligament, loading modes with a constant M/F ratio, yet varying the force magnitude, resulted in different types of tooth movement. Conclusions: The material properties of the periodontal ligament, the morphology of the root, and the alveolar bone are patient specic. Therefore, the M/F values generally advocated to obtain orthodontic tooth movement should be used only as guidelines. To be effective and accurate, the force system selected for a specic tooth movement must be monitored and the outcome compared with the predicted tooth movement. (Am J Orthod Dentofacial Orthop 2008;133:681-9)

fter an orthodontic loading regimen, the rst reaction is an alteration in the strain-stress distribution in the periodontal ligament (PDL) and the surrounding alveolar bone. This leads to intraalveolar displacement of the tooth and bending of the surrounding alveolar bone if the forces are large enough. The study of teeth displacement following the application of orthodontic loading regimes is not trivial as it depends on several and complex parameters such as material properties of periodontal tissues, shape and length of tooth, width of the PDL, and the marginal bone level.1 Moreover, since the decay rate of a force
From the Department of Orthodontics, School of Dentistry, University of Aarhus, Aarhus, Denmark. a Assistant professor. b Associate professor. c Professor and chairman. Reprint requests to: Paolo M. Cattaneo, Department of Orthodontics, School of Dentistry, University of Aarhus, Vennelyst Boulevard 9, DK-8000 Aarhus C, Denmark; e-mail, pcattaneo@odont.au.dk. Submitted, November 2005; revised and accepted, May 2006. 0889-5406/$34.00 Copyright 2008 by the American Association of Orthodontists. doi:10.1016/j.ajodo.2006.05.038

and the decay of the moment that this force generates are not equal, the actual force system can vary in experiment situations.2 As soon as the teeth start moving, the geometry characterizing the original force system also changes, and a new force system is generated, leading to a different displacement. The consequence of the undulating force system makes it nearly impossible to establish xed values describing the tooths displacement over time.3 Based on a mathematical model, Burstone4 established the localization of the center of resistance (CR) in a single-rooted tooth with a parabolic shape to be at 40% from the apex to the measured length between the alveolar crest and the apex of the root. Based on this assumption, the center of rotation (CRot) could be calculated with the so-called Burstone formula: moment force 0.068 h2 y

where h is the distance from the alveolar crest to the apex, and y is the distance between the CR and the CRot, assuming the PDL to be idealized as a 2-dimen681

682 Cattaneo, Dalstra, and Melsen

American Journal of Orthodontics and Dentofacial Orthopedics May 2008

sional entity with a shape resembling a parabola, the stress distribution to be uniform, and the stress-strain ratio to be linear. Later, Christiansen and Burstone5 determined the CRot experimentally by applying different forces on maxillary central incisors, and they conrmed that the CRot is related to the moment-toforce (M/F) ratio, but the CRot seemed to be positioned more apically than theoretically calculated. Most clinical investigations conrmed these results. The discrepancy between the theoretically calculated and the experimentally established localization of the CR was explained by the fact that Burstones formula was developed on a 2-dimensional model. All considerations regarding the position of the CR were based on the assumption that the PDL exhibits linear behavior. Over the last decades, numeric methods to calculate the stress and strain elds in the periodontium have been extensively used, and the nite element (FE) method has frequently been the method of choice. The complexity of shape and tissue composition of the dental region does, however, set limitations on the validity of the results of some previous analyses.6-8 Although many attempts to relate the force system to tooth displacement and the reaction of the surrounding tissues have been made, a clear concept still has not been presented. One reason might be that the methods applied have suffered from deciencies in the methods and the properties of the tissues analyzed. The FE method was introduced into dental biomechanical research in 19739 and since then has been applied to analyze the stress and strain elds in the alveolar support structures.8,10-18 The load-transfer mechanism from the tooth through the PDL to the alveolar bone depends on the physical properties and the morphology of the periodontium. The choice of these parameters determines the outcome of tge FE analysis and, in orthodontics, the type of tooth movement generated when a specic force system is applied to the bracket. Even though the PDL is known to be a nonlinear viscoelastic material, most previous FE models incorporated homogeneous, linear elastic, isotropic, and continuous PDL properties. At the same time, the morphology of the alveolar structures has not been assigned any specic value in relation to the loadtransfer mechanism. Our aim in this study was to demonstrate by FE analyses the inuence of the material properties of the PDL on the type of tooth movement. Moreover, the inuence of the applied force level on the type of tooth movement, with a xed M/F ratio, was evaluated and the results interpreted in the light of existing prescriptions for orthodontic tooth movement.

MATERIAL AND METHODS

Using a microcomputed tomography scanner (model 40, Scanco Medical, Bassersdorf, Switzerland), a 3-dimensional (3D) data set of the left segment of a human mandible including a canine and a rst premolar obtained from autopsy was acquired. The 3D data set had a voxel dimension of 37 m. By using a procedure previously described,19 a 3D FE model comprising the mandibular segment, the PDL, the canine, and the rst premolar was generated. The model consisted of 197,186 ten-node tetrahedral elements, 253,309 nodes, and 754,347 degrees of freedom (Fig 1). The material properties of bone were assigned to each element individually based on the true morphology of the bone as obtained from the scans by using a slightly modied version of the procedure of Cattaneo et al.20 Three Youngs moduli were considered to represent full cortical bone (17,500 MPa, Poissons ratio of 0.3), partly cortical bone (5000 MPa, Poissons ratio of 0.3), and bone marrow (200 MPa, Poissons ratio of 0.3), respectively. A nonlinear and nonsymmetric approach was used to describe the material properties of the PDL when the behavior in compression and tension were not the same. In compression, the stiffness of the PDL was substantially lower than in tension. In compression, a Youngs modulus of 0.005 MPa up to the 93% strain level was used; then a Youngs modulus of 8.5 MPa was adopted to simulate precontact between the roots and the alveolar bone. In tension, the Youngs moduli was gradually increased from 0.044 MPa at zero strain level to 0.44 MPa at about 50% strain; then a smaller Youngs modulus of 0.032 MPa was used.19 A Poissons ratio of 0.3 was used. The elements representing the 2 teeth were assigned a Youngs modulus of 20,000 MPa and a Poissons ratio of 0.3.10,14,21 Different loading regimens were applied to the model to simulate distinct types of orthodontic tooth movements. In accordance with clinical practice, the various loading regimens were dened by the M/F ratio applied at the bracket. In the rst series of analyses, the force applied at the bracket was kept constant while the M/F ratio was varied. The force on the premolar acted in the buccolingual direction and in the opposite direction for the canine22; it had a magnitude of 100 cN for both teeth. This value was chosen because it corresponded to the force level used in clinical practice.2,5,22,23 The M/F ratios were varied from uncontrolled tipping (M/F 0) to 4, 6, 8, 9, and 10 for the premolar and from 0 to 5, 8, 10, 11, and 12 for the canine. In the second series of analyses, the force level was

American Journal of Orthodontics and Dentofacial Orthopedics Volume 133, Number 5

Cattaneo, Dalstra, and Melsen 683

gradually increased from 0.5 to 200 cN, with the M/F ratios kept constant at 10 for the premolar and at 12 for the canine. For each analysis, the CRot was determined, and its location was compared with that of the CR. For the boundary conditions, movement was suppressed in all directions for the nodes on the bottom edge of the bone segment.
RESULTS

Fig 1. Exploded view of the FE model of the mandibular segment with alveolar bone, PDLs, canine (left), and rst premolar (right). The PDL is the space between the alveolar bone and the roots of the teeth.

The results from the rst series of analyses (force level of 100 cN and M/F ratio varying from 0 to 10 for the premolar and from 0 to 12 for the canine) are shown in Figure 2. From these analyses, it can be seen that, with an M/F ratio of 0, uncontrolled tipping movement is obtained. With an M/F ratio of 10 for the premolar and 12 for the canine, the CRot was situated far from the innite, where it theoretically should have been located. M/F ratio was shown to be the best approximation for pure translation for the premolar, and a value of 11 for the canine would be the best approximation. Furthermore, it could be seen that M/F ratios of 6 for the premolar and about 8 for the canine produced controlled tipping because the CRot coincided with the apex of the root in that case. During uncontrolled tipping (M/F 0) at the bracket level, the premolar moved lingually 0.163 mm, and the canine moved buccally 0.181 mm (Fig 3). As expected, when M/F ratios of 9 for the premolar and 11 for the canine are used, the displacements of both teeth were smaller than in the case of uncontrolled tipping. The crowns of the premolar and the canine were displaced 0.024 mm lingually and 0.018 mm buccally, respectively, whereas the apexes of both teeth were displaced 0.021 mm (lingually for the premolar and buccally for the canine). This indicates that, although aiming at translation, moderate tipping occurred for the premolar whereas the canine exhibited root movement (Fig 3). By looking at the displacements of the crown and the apex (Fig 3), one can notice that the canine had a peculiar movement: when low forces were applied with an M/F ratio of 12, the crown and the apex started moving in opposite directions, and, instead of pure translation, reverse tipping occurred. However, above 30 cN, the movement of the crown tended to be zero, and root movement occurred. During uncontrolled tipping (M/F 0) with a force level at the brackets of 100 cN, regions of compressive and tensile normal stresses could be identied in the PDL, although the stress levels in the compression areas were signicantly lower than the tensile stresses. This was evident at the PDL-bone interface of both the

684 Cattaneo, Dalstra, and Melsen

American Journal of Orthodontics and Dentofacial Orthopedics May 2008

Fig 2. A, Location of the CRot when the force applied at the bracket of the premolar is kept constant at 100 cN while the M/F ratio is varied from 0 to 10. B, Location of the CRot when the force applied at the bracket of the canine is kept constant at 100 cN while the M/F ratio is varied from 0 to 12.

premolar and the canine, particularly in the cervical areas opposite the direction of the force (Fig 4, top). The strains in the PDL were in the range of 20% to 14%. Based on the previous results, the stresses in the PDL were evaluated when translation was simulated by using a force of 100 cN and M/F ratios of 9 for the premolar and 11 for the canine. In this case, again the stress values in the PDL were much higher in the tension side that in the compression side (Fig 4, bottom).

When the force level was gradually increased from 0 to 200 cN, the canine and the premolar displayed a remarkable phenomenon, especially when M/F of 10 for the premolar and M/F of 12 for the canine were applied: the CRot is not in a xed position, but it is displaced after the force level that is applied to the tooth. To depict these changes, the distance between the actual CRot and the CR was calculated (Fig 5). This distance becomes maximal at a force level of about 50 cN for the canine and about 40 cN for the premolar. After these force levels, the CR-CRot distance starts to

American Journal of Orthodontics and Dentofacial Orthopedics Volume 133, Number 5

Cattaneo, Dalstra, and Melsen 685

Fig 3. Lingual-buccal displacements of the crowns and the apices of the premolar and the canine when the force is increased from 0 to 100 cN. Three values of M/F ratio are simulated.

decrease for both teeth, approaching an asymptotic value. In the case of both the canine and the premolar, the initial displacement will vary according to the force level, and this phenomenon is particularly evident in the range of forces frequently used in daily practice (20-60 cN).
DISCUSSION

The relationship between the type of tooth movement generated by an orthodontic appliance and the M/F ratio has been based on inadequate methods. In our study, the importance of the morphology and the physical properties of the tissues involved was studied. An FE model was used to determine the relationship between the M/F ratio applied at the brackets and the CRot, and the inuence of the force level on the CRot.

Although we used correct morphology and approximated the true physical properties of the PDL and the surrounding bone, certain limitations of this study should be mentioned. This study was based on 1 human sample. The material property of the PDL was assumed to be nonlinear and nonsymmetric, but viscoelasticity was not considered.19 Moreover, the results were not directly compared with in-vivo loading, although the calculated amount of deection of the teeth reects the results of previous experimental studies.5,24 In previously published studies, the material properties of the PDL were modeled as linear elastic with elastic moduli ranging from 0.07 to 100 MPa.14,24,25 In other studies, the PDL was modeled as nonlinear, ber-reinforced, or with viscoelastic properties.10,15-17,26-28 In a recent study,19 the PDL

686 Cattaneo, Dalstra, and Melsen

American Journal of Orthodontics and Dentofacial Orthopedics May 2008

Fig 4. Bucco-lingual stress at the bone-PDL interface around the root of the premolar (left) and the canine (right). Negative values represent compression, and positive values represent tension. Uncontrolled tipping is simulated by applying a force of 100 cN at the bracket level and an M/F ratio of 0 (top), and translation is obtained by applying again a force of 100 cN plus M/F ratios of 9 for the premolar and 11 for the canine (bottom).

was given a nonlinear stress-strain relationship based on experimental results.26,27 Its basic shape with a low-stiffness toe region and a high-stiffness slope closely resembles both experimentally29 and mathematically determined relationships.28 Although teeth comprise various tissues such as dentin, enamel, and cementum, only the dentin was modeled because our main goal was to study the reaction in the PDL and the surrounding alveolar bone. The material property of the dentin was adapted from several values in the literature.8,10,21,30

Unlike most previous FE analyses, our model comprised 2 teeth loaded simultaneously; thus, it was possible to counteract the force applied on the canine with another force of the same magnitude, but in the opposite direction, applied on the premolar. This way of loading limited the force transfer through the constrained nodes because the forces applied to the teeth counteracted each other. The inuence of boundary conditions on the nal outcome was therefore smaller, as opposed to the situation when only 1 tooth would have been modeled. In addition to this, it also reects

American Journal of Orthodontics and Dentofacial Orthopedics Volume 133, Number 5

Cattaneo, Dalstra, and Melsen 687

Fig 5. Distance between the CRot and the CR when the force level is gradually increased from 0 to 200 cN while keeping the M/F ratios constant at 10 for the premolar and 12 for the canine.

the clinical situation better: a tooth or a group of teeth is used to load the others. We simulated various tooth movements from uncontrolled tipping to pure translation. This was achieved by using the values of the M/F ratios recommended by Burstone and Pryputniewicz.23 The CRot values of the resulting tooth movements are located in a straight line passing through the CR. The actual location of the CRot is determined by the following: 1. The value of the M/F ratio. This dependency is evident, yet our results show that the traditionally assumed M/F ratios for translation do not perfectly

apply. An M/F ratio of 10, which is generally accepted for translation of incisors and premolars, was too large for the premolar in this study; it seemed to approach translation with an M/F ratio of 9. The lack of validity of the assumed values of an M/F ratio of 12 for the canine was also demonstrated; the correct value seemed to be between 10 and 11. However, these results were based on 1 human sample; therefore, they should be considered valid only for this canine and this premolar. In accordance with the report of Burstone and Pryputniewicz,23 small changes in the M/F ratio would produce large changes in the position of the CRot as

688 Cattaneo, Dalstra, and Melsen

American Journal of Orthodontics and Dentofacial Orthopedics May 2008

the M/F ratio required to generate translation is approached. 2. The level of the applied force. This was initially not anticipated but could be explained by the nonlinear behavior of the PDL. At different force levels, the actual load transfer across the PDL differs, leading to different types of tooth movement. Christiansen and Burstone4 observed this phenomenon in their clinical experiments with forces below 50 cN. Yet they failed to note that it is related to the mechanics of the PDL, and they instead ascribed the deviation between the expected behavior and their ndings to inaccuracies in measuring technique and instrumentation. This means that, with a nonlinear model to describe the PDL, the location of the CRot is not constant but depends not only on the M/F ratio, but also on the force level. In the deection curves depicted in Figure 3, it can be seen that the amount of tooth movement is not linearly correlated to the applied load. In the rst part of the curves, a relatively small increment of the force is followed by a relatively large movement of the teeth; on the contrary, after a force level of about 15 cN, the slope of the curve diminishes as the force magnitude increases. Therefore, the application of higher forces will generate only a moderate increment in tooth movement. This phenomenon could explain why, as reported by some authors, increasing the force magnitude from 10 to 25 cN increases the mean rate of tooth movement,31 whereas the mean rate of tooth movement is independent of the force magnitude when the force level is increased from 50 to 200 cN or more.32,33 Our results indicate that the M/F values generally advocated for various movements are too high, and a reduction of the M/F ratio could be suggested. However, tooth movement should always be monitored and the outcome compared with the expected tooth movement. In the case of discrepancy between the predicted and actual tooth movements, the force system should be adjusted.

REFERENCES 1. Melsen B. Tissue reaction to orthodontic tooth movementa new paradigm. Eur J Orthod 2001;23:671-81. 2. Burstone CJ. The segmented arch approach to space closure. Am J Orthod 1982;82:361-78. 3. Isaacson RJ, Lindauer SJ, Davidovitch M. On tooth movement. Angle Orthod 1993;63:305-9. 4. Burstone CJ. The biomechanics of tooth movement. In: Kraus BS, Ripamonti A, editors. Vistas in orthodontics. Philadelphia: Lea and Febinger; 1962. p. 197-213. 5. Christiansen RL, Burstone CJ. Centers of rotation within the periodontal space. Am J Orthod 1969;55:353-69.

6. Williams KR, Edmundson JT. Orthodontic tooth movement analyzed by the nite element method. Biomaterials 1984;5:34751. 7. Tanne K, Bantleon HP. Stress distribution in the periodontal ligament induced by orthodontic forces. Use of nite-element method. Inf Orthod Kieferorthop 1989;21:185-94. 8. Cobo J, Sicilia A, Arguelles J, Suarez D, Vijande M. Initial stress induced in periodontal tissue with diverse degrees of bone loss by an orthodontic force: tridimensional analysis by means of the nite element method. Am J Orthod Dentofacial Orthop 1993; 104:448-54. 9. Farah JW, Craig RG, Sikarskie DL. Photoelastic and nite element stress analysis of a restored axisymmetric rst molar. J Biomech 1973;6:511-20. 10. Bourauel C, Freudenreich D, Vollmer D, Kobe D, Drescher D, Jager A. Simulation of orthodontic tooth movements. A comparison of numerical models. J Orofac Orthop 1999;60:136-51. 11. Middleton J, Jones ML, Wilson AN. Three-dimensional analysis of orthodontic tooth movement. J Biomed Eng 1990;12:319-27. 12. Middleton J, Jones M, Wilson A. The role of the periodontal ligament in bone modeling: the initial development of a timedependent nite element model. Am J Orthod Dentofacial Orthop 1996;109:155-62. 13. Tanne K, Yoshida S, Kawata T, Sasaki A, Knox J, Jones ML. An evaluation of the biomechanical response of the tooth and periodontium to orthodontic forces in adolescent and adult subjects. Br J Orthod 1998;25:109-15. 14. Tanne K, Sakuda M, Burstone CJ. Three-dimensional niteelement analysis for stress in the periodontal tissue by orthodontic forces. Am J Orthod Dentofacial Orthop 1987;92:499-505. 15. van Driel WD, van Leeuwen EJ, Von den Hoff JW, Maltha JC, Kuijpers-Jagtman AM. Time-dependent mechanical behaviour of the periodontal ligament. Proc Inst Mech Eng [H] 2000;214: 497-504. 16. Provatidis CG. A comparative FEM-study of tooth mobility using isotropic and anisotropic models of the periodontal ligament. Med Eng Phys 2000;22:359-70. 17. Qian H, Chen J, Katona TR. The inuence of PDL principal bers in a 3-dimensional analysis of orthodontic tooth movement. Am J Orthod Dentofacial Orthop 2001;120:272-9. 18. Toms SR, Eberhardt AW. A nonlinear nite element analysis of the periodontal ligament under orthodontic tooth loading. Am J Orthod Dentofacial Orthop 2003;123:657-65. 19. Cattaneo PM, Dalstra M, Melsen B. The nite element method: a tool to study orthodontic tooth movement. J Dent Res 2005; 84:428-33. 20. Cattaneo PM, Dalstra M, Frich LH. A three-dimensional nite element model from computed tomography data: a semi-automated method. Proc Inst Mech Eng [H]. 2001;215:203-13. 21. Verdonschot N, Fennis WM, Kuijs RH, Stolk J, Kreulen CM, Creugers NH. Generation of 3-D nite element models of restored human teeth using micro-CT techniques. Int J Prosthodont 2001;14:310-5. 22. Proft WR, Fields HW. The biological basis of orthodontic therapy. In: Rudolph P, editor. Contemporary orthodontics. 3rd ed. St Louis: Mosby; 2000. p. 296-325. 23. Burstone CJ, Pryputniewicz RJ. Holographic determination of centers of rotation produced by orthodontic forces. Am J Orthod 1980;77:396-409. 24. Jones ML, Hickman J, Middleton J, Knox J, Volp C. A validated nite element method study of orthodontic tooth movement in the human subject. J Orthod 2001;28:29-38.

American Journal of Orthodontics and Dentofacial Orthopedics Volume 133, Number 5

Cattaneo, Dalstra, and Melsen 689

25. Andersen KL, Pedersen EH, Melsen B. Material parameters and stress proles within the periodontal ligament. Am J Orthod Dentofacial Orthop 1991;99:427-40. 26. Poppe M, Bourauel C, Jager A. Determination of the elasticity parameters of the human periodontal ligament and the location of the center of resistance of single-rooted teeth. A study of autopsy specimens and their conversion into nite element models. J Orofac Orthop 2002;63:358-70. 27. Vollmer D, Bourauel C, Maier K, Jager A. Determination of the centre of resistance in an upper human canine and idealized tooth model. Eur J Orthod 1999;21:633-48. 28. Pietrzak G, Curnier A, Botsis J, Scherrer S, Wiskott A, Belser U. A nonlinear elastic model of the periodontal ligament and its numerical calibration for the study of tooth mobility. Comput Methods Biomech Biomed Engin 2002;5:91-100. 29. Toms SR, Lemons JE, Bartolucci AA, Eberhardt AW. Nonlinear

30.

31.

32.

33.

stress-strain behavior of periodontal ligament under orthodontic loading. Am J Orthod Dentofacial Orthop 2002;122:174-9. Jeon PD, Turley PK, Ting K. Three-dimensional nite element analysis of stress in the periodontal ligament of the maxillary rst molar with simulated bone loss. Am J Orthod Dentofacial Orthop 2001;119:498-504. van Leeuwen EJ, Maltha JC, Kuijpers-Jagtman AM. Tooth movement with light continuous and discontinuous forces in beagle dogs. Eur J Oral Sci 1999;107:468-74. Pilon JJGM, Kuijpers-Jagtman AM, Maltha JC. Magnitude of orthodontic forces and rate of bodily tooth movement. An experimental study. Am J Orthod Dentofacial Orthop 1996;110:16-23. Ren Y, Maltha JC, Vant Hof MA, Kuijpers-Jagtman AM. Optimum force magnitude for orthodontic tooth movement: a mathematic model. Am J Orthod Dentofacial Orthop 2004; 125:71-7.