Accepted Manuscript

Biomechanical behavior of mandibles reconstructed with fibular grafts

at different vertical positions using finite element method

Yun-feng Liu PhD , Kang-jie Cheng PhD candidate ,

Joanne H. Wang MD , Janice C. Jun PhD candidate ,
Xian-feng Jiang PhD , Russell Wang D.D.S., M.S.D. ,
Dale A. Baur DDS

PII: S1748-6815(18)30354-1
DOI: https://doi.org/10.1016/j.bjps.2018.10.002
Reference: PRAS 5813

To appear in: Journal of Plastic, Reconstructive & Aesthetic Surgery

Received date: 12 September 2017

Accepted date: 28 October 2018

Please cite this article as: Yun-feng Liu PhD , Kang-jie Cheng PhD candidate , Joanne H. Wang MD ,
Janice C. Jun PhD candidate , Xian-feng Jiang PhD , Russell Wang D.D.S., M.S.D. ,
Dale A. Baur DDS , Biomechanical behavior of mandibles reconstructed with fibular grafts at
different vertical positions using finite element method, Journal of Plastic, Reconstructive & Aesthetic
Surgery (2018), doi: https://doi.org/10.1016/j.bjps.2018.10.002

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service
to our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and
all legal disclaimers that apply to the journal pertain.
 ACCEPTED MANUSCRIPT

Biomechanical behavior of mandibles reconstructed with fibular grafts at

different vertical positions using finite element method
Yun-feng Liu, PhD*†
Key Laboratory of E&M (Zhejiang University of Technology), Ministry of
Education & Zhejiang Province
Hangzhou, Zhejiang Province, 310014, China

Kang-jie Cheng, PhD candidate†

T
Key Laboratory of E&M (Zhejiang University of Technology), Ministry of

IP
Education & Zhejiang Province
Hangzhou, Zhejiang Province, 310014, China

CR
Joanne H. Wang, MD
Department of Orthopedic Surgery,
University Hospitals of Cleveland, Case Medical Center
US
11100 Euclid Ave, Cleveland, OH 44016 U.S.A.

Janice C. Jun, PhD candidate

AN
Department of Oral Maxillary Surgery
Case Western Reserve University School of Dental Medicine
10900 Euclid Ave.
M

Cleveland, OH 44106-4905 U.S.A.

Xian-feng Jiang, PhD

ED

Key Laboratory of E&M (Zhejiang University of Technology), Ministry of

Education & Zhejiang Province
Hangzhou, Zhejiang Province, 310014, China
PT

Russell Wang, D.D.S., M.S.D.

CE

Department of Comprehensive Care

Case Western Reserve University School of Dental Medicine
10900 Euclid Ave.
AC

Cleveland, OH 44106-4905 U.S.A.

Dale A. Baur, DDS

Department of Oral Maxillary Surgery
Case Western Reserve University School of Dental Medicine
10900 Euclid Ave.
Cleveland, OH 44106-4905 U.S.A.

*Correspondent author:
Key Laboratory of E&M (Zhejiang University of Technology), Ministry of
 ACCEPTED MANUSCRIPT

Education & Zhejiang Province

Hangzhou, Zhejiang Province, 310014, China
E-mail: [email protected]
† These authors contributed equally.

T
IP
CR
US
AN
M
ED
PT
CE
AC
 ACCEPTED MANUSCRIPT

Summary
Background: For large mandibular defects, surgical reconstruction using
microvascular fibular grafts has advantages over other alternatives in terms of blood
supply and good quality of grafted bone. However, the fibular segment is usually
lower in height than that of the original mandible, meaning that the vertical
positioning of the fibular graft is variable, with different biomechanical consequences
on the reconstructed mandible.
Objectives: To use finite element method (FEM) to evaluate stress distribution and
displacement of a reconstructed mandible versus an intact mandible under occlusal
loads.

T
Methods: A three-dimensional intact edentulous mandibular bone (Model I) and a

IP
reconstructed mandible bone with fibular graft were created from CBCT images.
Calculation models were generated with fibular bone graft extracted from the

CR
reconstructed mandible of identical length placed into a mimicked defect area on the
right side of the mandible at three different vertical positions: superior (Model II),
intermediate (Model III), and inferior (Model IV). Forces were applied at lower left
first molar region and lower left central incisor area. Von Mises stresses and

US
mandibular displacement were calculated as outcome measurements during loadings.
Results: Maximum stress and strain within the reconstructed mandible were
identified at the posterior border of the graft and the contralateral condyle. Maximum
AN
displacement occurred near the interface of fibular graft and anterior segment of the
mandible. Stress distribution in the graft under functional loads is much higher than
that in the residual mandibular segments from Models II to IV. The combined average
M

maximum stress from anterior and posterior loads, there are 10.66 time higher stresses
in the mandible with inferior positioned graft (Model IV), 8.72 time of superior graft
(Model II) and 3.68 time of intermediate graft (Model III) than that of the control
ED

group (Model I). The worst displacement result during functional loadings was in the
group with fibular graft located at the inferior border of the mandible.
Conclusions: The position of fibular graft placed in the surgical resection site has
PT

significant effects on the mechanical behavior of the reconstructed mandible. The

fibular graft aligned with the inferior border of the mandible, the most common site
CE

designated location by clinicians, has the worst effects on the stress distribution and
displacement to the mandibular under functional loads. The fibular graft placed at the
intermediate location has the best biomechanics and provides favorable condition for
AC

subsequent prosthetic reconstruction.

Keywords:
Mandibular reconstruction,
Fibular graft,
Bone property,
Finite element method,
Von Mises stress
 ACCEPTED MANUSCRIPT

Introduction

Surgical resection of the mandible often is necessary in treatment of head and neck
tumors, infections, trauma or congenital deformities. 1 Facial disfigurement, impairment of
chewing, swallowing, speech, and psychologic well-being are challenges for patients who
require surgical and prosthetic reconstructions of mandible. 2-8 For large mandibular defects,
a fibular flap was first used by Hidalgo in 1989 to surgically reconstruct the mandible, 9 this
method has become the treatment of choice due to its many advantages, such as good
quality of bone from the fibula, less donor site mobility, and high graft survival based on

T
re-vascularization.10,11

IP
The mandible can experience five types of loading: tension, compression, shear,
torsion, and bending. Facture strength of the mandible depends on the direction, location,

CR
and type of stress. To be able to resist forces and bending and torsional moments, not only
the material properties of the mandible but also its geometrical design is of importance. The
mandible is subject to forces produced by the muscles of mastication and by reaction forces

US
acting through the teeth and temporomandibular joints during chewing and clenching. With
normal geometry of the mandible, in the longitudinal direction, the mandible is stiffer than
AN
in transverse directions, and the vertical cross-sectional dimension of the mandible is larger
than its transverse dimension. These features enhance the resistance of the mandible to the
relatively large vertical shear forces and bending moments that come into play in the
M

sagittal plane.
Biomechanics of intact mandible have been characterized in the literature;12-18 however,
biomechanical behavior of reconstructed mandibles with various grafting methods is not
ED

well understood. Physical models have been used such as animal bone, human cadaveric
bone, rapid prototyping, and bone substitutes to provide fracture strength and fatigue
PT

information by various mechanical tests at a gross level.19-21 Most of the literature on

mandibular reconstruction using microvascular free fibular flaps deal with surgical
techniques, clinical outcomes, and prosthesis designs. Biomechanical analyses of
CE

reconstructed mandible with fibular graft under functional loads are lacking because of the
complex geometry of the mandible, multidirectional muscle forces, heterogeneous bony
AC

structures, and difficulty in obtaining samples for in vivo study.

FEM is commonly used in industry for numerical mechanical analyses, such as the
aeronautical and automotive field. FEM is commonly used to assist design and development
of products. Due to its powerful adaptability in calculation, FEM is also suitable for
biomechanical analysis of bones, including the mandible.22,23 Given a high correlation
between the FEM and experiments per se, various data within the specimen can be obtained
using the FEM calculation. The accuracy of FEM describing the biomechanical behavior of
bony specimens has been investigated.24-27 FEM can provide useful insight into the complex
mechanical behavior of mandibles affected by functional loading that could be verified by
comparing the calculated data to experimental results on physical models or clinical
 ACCEPTED MANUSCRIPT

statistical data.28,29 Computational models using finite element analysis, they can predict
areas most likely to fail based on internal stress distribution and areas of maximum stress
concentration.
A fibula graft is commonly placed at the inferior border of the mandible due to surgeon
preference, although it makes a later dental implant placement and tooth restoration more
difficult. The diameter of a fibula is significantly less than that of a mandible. The vertical
distance between the reconstructed segment and the occlusal plane can be large. This often
is a difficult problem for subsequent rehabilitation by an implant-supported prosthesis. The
large leverage forces resulting from the high vertical dimension of the prosthetic
construction can lead to overloading of the osseointegrated implants and endanger the

T
longevity of the prosthetic restoration. If the fibula is placed more superiorly, it will be

IP
easier to place and restore dental implants.
The purpose of this work was to investigate the mechanical behavior of reconstructed

CR
mandibles with a fibular graft placed at three vertical positions. Three-dimensional (3D)
models were constructed based on computerized tomography (CT) images. FEM was used

US
as an analytic tool to study the mechanical behaviors of the reconstructed mandibles.
Stress distributions and displacements are the outcome measurements of the reconstructed
mandibles during occlusal loadings.
AN
Material and Methods
M

Case information

All CBCT image data of two patients were obtained through a cone beam CT scanner at
ED

Case Western Reserve University Craniofacial Image Center, The parameters of the scanner
were set at 120 KV, 70 mA, with a field of view of 23 × 16 cm and voxel size of 0.39 mm.
PT

A total of 512 images of each patient were saved as DICOM data files. Three mandibular
3D virtual models were created via DICOM files. The first model was created from a
normal 50-year-old completely edentulous patient who had multiple extractions 5 months
CE

ago without fibular graft (Figure 1a). Virtual alveoloplasty was performed to clean-up thin
and sharp bony spikes on the first model. A triangular surface model was refined then
converted to a volume mesh FEA model. The 2nd model (Figure 1b) was from a partially
AC

edentulous patient who had mandibular reconstruction with fibular bone and fixation plates
resulting from resection of squamous cell carcinoma of the floor of the mouth with local
bony invasion. Noises from the metal plate of the 2nd model generated by forward and back
scatter radiation were manually removed. The mandibular reconstruction plate also was
segmented from the 2nd model. A 30 mm portion of the body of the mandible of the first
model was segmented and the resected area was replaced by the fibular graft from the 2nd
model. Therefore, a third master model (Figure 1c) with virtual mandibular reconstruction
using fibular graft was created by the combination of our first and 2nd models. In an ideal
clinical scenario, data from pre-operative and post-fibular graft of a same mandible would
 ACCEPTED MANUSCRIPT

be the best. Practically, it is hard to come by obtaining the same patient’s pre- and
post-operative mandibular DICOM files.

3D Reconstruction and meshing

A platform MIMICS (V16.0, Materialise, Belgium) was used to construct the two 3D
mandibles based on the DICOM data files. The reconstruction procedure was as follows:
first, a threshold for bony tissue segmentation was determined from the value and boundary.
The mandible was separated as a sole mask through ROI (region of interest) extraction.
Based on the mask, 3D models represented as triangular mesh (also known as STL file)

T
were created, as shown in figure 1.

IP
The triangular mesh is only a surface model (Figure 2a). FEM calculation requires
volume mesh (tetrahedron) model (Figure 2b). The MIMICS provides a mesh tool named

CR
3-matic (V8.0) for mesh reduction. 3D virtue models can be smoothed and re-meshed to
form volume meshes with high quality for numerical simulation. Geomagic (V12, 3D
system, Rock Hill, SC, USA) software was used to complete the final editing of the

US
triangular model. INP files in Geomagic program can be directly imported to Abaqus
(V6.13, Dassault Systèmes, Cedex , France) software which can create tetrahedron meshes
AN
of the models for subsequent simulation and calculation.

Material properties of jaw bone

M

Bone, consisting of cortical and cancellous parts, is a heterogeneous biomaterial with

various degrees of mineralization (various material mechanical properties) that can be
ED

detected and translated to different radio-densities from CT images. MIMICS software was
used to calculate the material properties of the models such as bone density and Young’s
modulus based on Hounsfield unit (HU) of CT images. Material properties such as bone
PT

density (ρ) and Young’s modulus E of each volume mesh of the 3D models were derived
from the following equations:
CE

ρ = 114 + 0.916 × HU (1)

1.37
E = 0.51 × ρ (2)
Figure 3a shows material properties of the master model with different colorations,
AC

which correspond to their mechanical properties. Figure 3b lists modulus and density, which
are derived from the equation 1 and 2. This model can be exported from MIMICS into
Abaqus via inp file for subsequent simulations and calculations. The material properties of
fibular graft were also calculated by the same method.

Calculation models with graft at various positions

A defect on the right side of the mandibular was created with MIMICS, and replaced with
either the original bone segment as a control, or a fibular graft at various vertical positions.
Four tested models were created: Model I, the original segment was used (Figure 4a);
 ACCEPTED MANUSCRIPT

Model II, a fibular graft was placed at the superior position between two residual
mandibular segments (Figure 4b); Model III, a fibular graft was placed at the intermediate
vertical position (Figure 4c); Model IV, a fibular graft was placed at the inferior border of
the mandible (Figure 4d).
Loading and constraints
To simulate a static status with loading forces during mastication, two TMJ condyles could
be fixed on all six freedoms, as references 30-32. Figure 5 shows occlusal loadings applied
to the control and the experimental groups at two locations of the mandibles: Location
①-lower left first molar; Location ②-lower left central incisor. Forces were perpendicular

T
to the occlusal table (Z axis). Mandibular condyles were fixated to prevent movement in

IP
any direction (Figure 5, black).30 In order to create more accurate models, jaw opening and
closing muscles were added to the model design. Muscles origins are labeled with blue in

CR
Figure 5. The direction and magnitude of muscle forces were based on data from Nelson33
for FEM calculations. Tables 1 and 2 are the values of muscle forces and directions of
forces in x, y, z directions. Comparing to closing muscles, reactions from the opening

US
masticatory muscles, such as, digastric and mylohyoid muscle are too small that would not
significantly affect the calculation results in the static status. Therefore, there was no need
to integrate the opening muscles in the calculation model.30-32
AN

Results
M

The Outcome measurements from FEM results are stresses, strains and displacements
within the mandibles and the fibular bone graft. Figure 6 shows the distributions of Von
ED

Mises stress (MPa), strains (%) and the amount of mandibular displacement (mm) of the
mandible of Model III under occlusal loading at left molar area. The maximum stress and
strain are identified at both the posterior border of the graft and the contralateral condyle
PT

area. Maximum displacement occurs near the interface of fibular graft and anterior segment
of the mandible.
CE

Figure 7 shows the Von Mises stress distributions in the mandible and the fibular bone
under occlusal loadings at the left molar area as well as at the left central incisor. Under
both loading conditions, maximum stresses among the four groups show the same pattern.
AC

Grafts placed at the inferior border of the mandible always yield the worst results. The
superior location, middle location, and the control group are progressively better, in that
order.
Tables 3 and 4 summarize the results of maximum Von Mises stress and the amount of
mandibular displacement under loading ① and loading ② conditions.

Discussion

Common method in the literature using FEM have tended to assume that the mandible is
isotropic rather anisotropic to simplify the calculation. In this study, we derive various bone
 ACCEPTED MANUSCRIPT

properties throughout the mandible (Figure 3) and integrate force directions of 7

masticatory muscles (Figure 5) in the modeling system. We intentional choose a completely
edentulous model for the FEM analysis of fibular graft because of the difficulty of modeling
tooth structures (three layers: enamel, dentine and pulp chamber) and periodontal ligament.
Our goal was to evaluate the mechanical behavior of the graft and native bone without the
compounding factor from the teeth. The model provides more accurate FEM data to analyze
the internal stress patterns throughout the mandible, the effect of changes in loading and
prediction of areas of likely failure.
Our results showed that stresses always were concentrated at the posterior aspect of the
graft and the opposite side of the condyle from the point of loading and while maximum

T
displacement occurred at the bone/graft interface. Our previous publication reveals the

IP
similar maximum stress and displacement patterns with treatment of mandibular angel
fracture.18 During functional loading to a fractured or reconstructed mandible, the weakest

CR
link of the mechanical system of the mandible is the interfragmental gap areas not the TMJ,
Therefore, the fixation points of the TMJ was not measured in the study. The measurement

US
of stress and displacement TMJ during functional loading was discussed in our previous
study that would not be the key factor for the outcome assessment of the scenario for
mandibular reconstruction with fibular graft.34 Fibula positioned along the inferior border of
AN
the mandible always leads to the least desirable mechanical results. Superior positioned
graft, intermediate positioned group and the control group have significantly better results.
The same pattern was observed with maximum displacement. Our data show that placing a
M

fibular graft at different vertical positions results in differences in the distribution of stress
throughout the reconstructed mandible, and also within the graft itself. The results from this
ED

FEM study contribute to the biomechanical understanding of a reconstructed body of the

mandible with a fibular graft.
Stress is an essential factor in evaluating the mechanical behavior of mandibular bone.
PT

35,36
Stress distribution in the graft under functional loads is much higher than that in the
residual mandibular segments from Models II to IV. The combined average maximum stress
CE

from anterior and posterior loads, there are 10.66 time higher stresses in the mandible with
inferior positioned graft (model IV), 8.72 time of superior graft (model II) and 3.68 time of
intermediate graft (model III) than that of the control group (model I). Our results show that
AC

with combined average maximum stress from anterior and posterior loads, there are 2.85
time higher stresses in the mandible with inferior positioned graft (model IV), 2.33 time of
superior graft (model II) and 1.16 time of intermediate graft (model III) than that of the
control group (model I). Our results agree with the mechanical principle of one beam theory
that stress concentration increases in a small diameter area when forces are applied to a
beam system that has one end immobile. A significant discrepancy of stress concentration
between the graft and remaining mandibular segments may contribute to the faster
resorption rate of a grafted bone (i.e. iliac or others) versus a pristine bone. Fibular grafts
are commonly placed at the inferior position because the mandible has the largest diameter
at the bottom which allows the most bone-to-bone contact. Fibular grafts are used for large
 ACCEPTED MANUSCRIPT

mandibular defects, therefore, positioning of a reconstruction plate away from dentition

along the inferior border rather than using the Champy method is frequently chosen for
mandibular reconstruction. Our results show this common practice produces the least
desirable mechanical effectiveness.
Cross-sections of fibular bone exhibit a structural pattern that is the opposite of that
present in the mandible. The diameter of the fibula has larger width than its vertical height.
The fibular segment is usually significantly lower in height than that of the remaining
mandible which creates problems for a later prosthetic restoration of a patient’s dentition.
The vertical distance between the reconstructed segment and the occlusal plane can be
substantially large. This often is a difficult problem for subsequent rehabilitation by an

T
implant-supported prosthesis. The large leverage forces resulting from the high vertical

IP
dimension of the prosthetic construction can lead to overloading of the osseointegrated
implants and endanger the longevity of the prosthetic restoration.

CR
Based on our results, the height, volume, and shapes of the grafted bone matter,
suggest that increasing graft volume by methods such as the double-barrel grafts may

US
improve biomechanical properties of the reconstructed mandible for patients who are
candidates for later implant-supported dental prostheses. “Double-barrel” grafts are not
routinely done because they require more osteotomies and are technically difficult.
AN
One reason for the optimal biomechanics of the graft placed in the intermediate
vertical position (Model III) may be that the overall structure of the reconstructed mandible
in this case is the most similar to that of the intact mandible. Another reason may be that
M

during chewing, the superior part of mandible suffers tensile stress and the inferior part
suffers compressive stress, so a balance of these forces at an intermediate zone may
ED

minimize stress within the graft at this position. A graft placed inferiorly is subjected to
great moments stress generated by chewing forces. Thus, a graft placed in the intermediate
zone may offer the best biomechanical compromise for the reconstructed mandible that
PT

facilitates ease of dental rehabilitation including placement of dental implants and

prostheses. The relationship of our results to double-barrel grafts is that the first barrel of
CE

full length should be placed at an intermediate to superior position, with the second
non-vascularized barrel of shorter length placed either above or below the first barrel as
would be most convenient for restorative purposes, as determined by the surgeon.
AC

Even more compellingly, our study supports efforts in developing novel ways of
generating bone graft segments of the same size and shape of the original segment, such as
by 3D printing of biocompatible metals or polymers as bone analogs would be the new
direction for mandibular reconstruction. Future research using FEM and statistical analysis
of clinical data, as well as experimental tests, may help to quantify and understand the long
term outcome of reconstruction and the restoration of biomechanical function of the
reconstructed mandible.

Conclusions
 ACCEPTED MANUSCRIPT

The position of fibular graft placed in the surgical resection site has significant effects on
the mechanical behavior of the reconstructed mandible. The fibular graft aligned with the
inferior border of the mandible, the most common site designated location by clinicians, has
the worst effects on the stress distribution and displacement to the mandibular under
functional loads. The fibular graft placed at the intermediate location has the best
biomechanics and provides favorable condition for subsequent prosthetic reconstruction.
This work provides an important basis for future improvement of the surgical and prosthetic
rational for mandibular reconstruction and ultimately the benefits to those patients.

Conflict of interest

T
IP
All authors declare that there is no conflict of interest.

CR
Acknowledgement

This project is supported by the grants from the National Natural Science Foundation of

US
China (grants no. 51775506), the Natural Science Foundation of Zhejiang Province (Grant
No. LY18E050022) and the James Hayward Research Fund.
AN
References

1. Maciejewski A, Szymczyk C. Fibula free flap for mandible reconstruction: analysis of 30 consecutive
M

cases and quality of life evaluation. J Reconstr Microsurg 2007; 23:1-10.

2. Hidalgo DA. Fibular free flap: a new method of mandible reconstruction. Plast Reconstr Surg 1989;
84:71-9.
ED

3. Wei FC, Seah CS, Tsai, YC, et al. Fibular osteoseptocutaneous flap for reconstruction of composite
mandibular defects. Plast Reconstr Surg 1994; 93:294-304.
4. Smolka K, Kraehenbuehl M, Eggensperger N. Fibula free flap reconstruction of the mandible in cancer
PT

patients: evalution of a combined surgical and prosthodontic treatment concept. Oral Oncol 2008;
44:571-81.
5. Klesper B, Lazar F, Siessegger M, et al. Vertical distraction osteogenesis of fibular transplants for
CE

mandibular reconstruction-a preliminary study. J Craniomaxillofac Surg 2002; 30:280-5.

6. Militsakh ON, Werle A, Mohyuddin N, et al. Comparison of radial forearm with fibula and scapula
osteocutaneous free flaps for oromandibular reconstruction. Arch Otolaryngol Head Neck Surg 2005;
AC

131:571-5.
7. Lópezarcas JM, Arias J, Del Castillo JL, et al. The Fibula Osteomyocutaneous Flap for Mandible
Reconstruction: A 15-Year Experience. J Oral Maxillofac Surg 2010; 68:2377-84.
8. Wallace CG, Chang YM, Tsai CY. Harnessing the potential of the free fibula osteoseptocutaneous flap
in mandible reconstruction. Plast Reconstr Surg 2010; 125:305-14.
9. Hidalgo DA. Fibula free flap: a new method of mandible reconstruction. Plast Reconstr Surg 1989;
84:71-9.
10. Fujiki M, Miyamoto S, Sakuraba M, et al. A comparison of perioperative complications following
transfer of fibular and scapular flaps for immediate mandibular reconstruction. J Plast Reconstr Aesthet
Surg 2013; 66:372–5.
11. Wijbenga JG, Schepers RH, Werker PM, et al. A systematic review of functional outcome and quality of
 ACCEPTED MANUSCRIPT
life following reconstruction of maxillofacial defects using vascularized free fibula flaps and dental
rehabilitation reveals poor data quality. J Plast Reconstr Aesthetic Surg 2016; 69: 1024-36.
12. Bezerra TP, Junior S, Scarparo HC, et al. Do erupted third molars weaken the mandibular angle after
trauma to the chin region? A 3D finite element study. Int Illofac Surg 2013; 42:474-80.
13. Meyer C, Kahn JL, Boutemi P, et al. Photoelastic analysis of bone deformation in the region of the
mandibular condyle during mastication. J Craniomaxillofac Surg 2002; 30:160-9.
14. Kober C, Erdmann B, Lang J, et al. Adaptive Finite Element Simulation of the Human Mandible Using
a New Physiological Model of the Masticatory Muscles. Pamm 2004; 4:332–3.
15. Antic S, Vukicevic A, Milasinovic M, et al. Impact of the lower third molar presence and position on
the fragility of mandibular angle and condyle: A Three-dimensional finite element study. J
Craniomaxillofac Surg 2015; 43:870-8.

T
16. Bujtár P, Sándor GK, Bojtos A, et al. Finite element analysis of the human mandible at 3 different

IP
stages of life. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2010; 110:301-9.
17. Murakami K, Yamamoto K, Sugiura T, et al. Biomechanical analysis of poly-L-lactic acid and titanium
plates fixated for mandibular symphyseal fracture with a conservatively treated unilateral condylar

CR
fracture using the three-dimensional finite element method. Dent Traumatol 2015; 31:396-402.
18. Wang R, Liu Y, Wang JH, et al. Effect of interfragmentary gap on the mechanical behavior of
mandibular angle fracture with three fixation designs: A finite element analysis. J Plast Reconstr
Aesthetic Surg 2017; 70:3360-9.
US
19. Vollmer D, Meyer U, Joos U, et al. Experimental and finite element study of a human mandible. J
Craniomaxillofac Surg 2000; 28:91-6.
AN
20. Meyer C, Kahn JL, Lambert A, et al. Development of a static simulator of the mandible. J
Craniomaxillofac Surg 2000; 28:278-86.
21. Meyer C, Serhir L, Boutemi P. Experimental evaluation of three osteosynthesis devices used for
M

stabilizing condylar fractures of the mandible. J Craniomaxillofac Surg 2006; 34:173-81.

22. Hart RT, Hennebel VV, Thongpreda N, et al. Modeling the biomechanics of the mandible: a
three-dimensional finite element study. J Biomech 1992; 25:261-86.
ED

23. Parashar SK, Sharma JK. A review on application of finite element modeling in bone biomechanics.
Perspect Sci 2016; 8: 696-8.
24. Tsouknidas A, Anagnostidis K, Maliaris G, et al. Fracture risk in the femoral hip region: a finite element
PT

analysis supported experimental approach. J Biomech 2012; 45:1959-64.

25. Khayyeri H, Checa S, Tägil M, et al. Corroboration of mechanobiological simulations of tissue
differentiation in an in vivo bone chamber using a lattice-modeling approach. J Orthop Res 2009;
CE

27:1659-66.
26. Rho JY, Hobatho MC, Ashman RB. Relations of mechanical properties to density and CT numbers in
human bone. Med Eng Phys 1995; 17:347-55.
AC

27. Rice JC. On the dependence of the elasticity and strength of cancellous bone on the apparent density. J
Biomech 1998; 2:155-68.
28. Meyer C, Martin E, Kahn JL, et al. Development and biomechanical testing of a new osteosynthesis
plate (TCP) designed to stabilize mandibular condyle fractures. J Craniomaxillofac Surg 2007; 35:
84-90.
29. Ramos A, Ballu A, Mesnard M, et al. Numerical and experimental models of the mandible. Exp Mech
2011; 51:1053-59.
30. Vajgel A, Camargo IB, Willmersdorf RB, et al. Comparative finite element analysis of the
biomechanical stability of 2.0 fixation plates in atrophic mandibular fractures. J Oral Maxillofac Surg
2013; 71:335-42.
31. Narra N, Valášek J, Hannula M, et al. Finite element analysis of customized reconstruction plates for
 ACCEPTED MANUSCRIPT

mandibular continuity defect therapy. J Biomech 2014; 47:264-8.

32. Gregolin RF, Tokimatsu RC, Pereira JA. Biomechanical Stress and Strain Analysis of Mandibular
Human Region from Computed Tomography to Custom Implant Development. Adv Mater Sci Eng
2017; 2017:1-9.
33. Nelson GJ. Three dimensional computer modeling of human mandibular biomechanics. D.M.D. Thesis.
The University of British Columbia, Vancouver. 1986.
34. Liu Y, Wang R, Baur DA, et al. A finite element analysis of the stress distribution to the mandible from
impact forces with various orientations of third molars. J Zhejiang Univ-SC B 2018; 19:38-48.
35. Sevimay M, Turhan F, Kiliçarslan MA, et al. Three-dimensional finite element analysis of the effect of
different bone quality on stress distribution in an implant-supported crown. J Prosthet Dent 2005;
93:227-34.

T
36. Shigemitsu R, Yoda N, Ogawa T, et al. Biological-data-based finite-element stress analysis of

IP
mandibular bone with implant-supported overdenture. Comput Bio Med 2014; 54:44-52.

CR
US
AN
M
ED
PT
CE
AC

Figure legends
 ACCEPTED MANUSCRIPT

Figure 1. 3D models for simulation calculation. (a) master model of an edentulous mandible;
(b) a reconstructed mandible with a fibular graft which is located in the middle of the bone

T
segments along vertical direction; (c) the two mandibles are registered and a bone defect on

IP
the master model was mimicked by resecting and removing a bone segment with two planes
and substituted by the fibular graft extracted from the reconstructed mandible by resecting

CR
with the same two planes.

US
AN
M
ED

Figure 2. Triangular mesh model and volume mesh model. (a) A triangular mesh model of
PT

an intact mandible, a total of 25,940 triangles included; (b) a tetrahedron volume mesh
model created from a triangular mesh model.
CE
AC
 ACCEPTED MANUSCRIPT

T
IP
CR
Figure 3. Modeling of material properties. (a) Color-coded model, different colors of the
meshes represent different material properties. (b) Material properties were calculated by

US
the CT number, and the color index of the properties was listed.
AN
M
ED

Figure 4. Model groups. (a) Model I: original segment from the intact mandible; (b) Model
II: fibular graft is placed at the superior position flush with the adjacent mandibular
PT

segments; (c) Model III: fibular graft is placed at the intermediate vertical position; (d)
Model IV: fibular graft is placed at the inferior position flush with the adjacent mandibular
CE

segments.
AC
 ACCEPTED MANUSCRIPT

T
IP
CR
US
AN

Figure 5. Loading and boundary constraints on mandible. Condyles fixated in all directions
M

(black), origins of mandibular muscles (blue), and occlusal loading (green).

Note. SM is Superficial Masseter, DM is Deep Masseter, MP is Medial Pterygoid, AT is
ED

Anterior Temporal, MT is Middle Temporal, PT is Posterior Temporal, LIP is Lateral

Inferior Pterygoid.
PT
CE
AC

Figure 6. FEM results of Model III with left molar loading. (a) Von Mises stress distribution,
(b) principal strain distribution, and (c) displacement of mandible and graft.
 ACCEPTED MANUSCRIPT

T
IP
CR
US
AN

Figure 7. Von Mises stress distribution to the mandible as well as the graft with occlusal
loadings. First row: Model I - control group - a and b are loading applied at lower left molar
M

(loading ①), c and d are loading at lower left central incisor(loading ②); second row is
Model II; third row is Model III; and fourth row is Model IV. Two left columns are results
under loading ① and two right columns are results from loading ②.
ED
PT
CE
AC
 ACCEPTED MANUSCRIPT

Table 1 Muscle force with constraint, loading ①- at first molar area.

Scaling Factors Unit Vector Coordinates
Balancing
Muscle Muscle Weight(N) Working Side X Y Z
Side
Superficial masseter 190.40 0.72 0.60 -0.21 -0.42 +0.89
Deep masseter 81.60 0.72 0.60 -0.55 +0.36 +0.76
Medial pterygoid 174.80 0.84 0.60 +0.49 -0.37 +0.79
Anterior temporal 158.00 0.73 0.58 -0.15 -0.04 +0.99
Middle temporal 95.60 0.66 0.67 -0.22 +0.50 +0.83
Posterior temporal 75.60 0.59 0.39 -0.21 +0.86 +0.47
Lateral inferior pterygoid 66.90 0.30 0.65 +0.63 -0.76 -0.17

T
IP
CR
US
AN
M
ED
PT
CE
AC
 ACCEPTED MANUSCRIPT

Table 2 Muscle force with constraint, loading ②-at central incisor.

Scaling Factors Unit Vector Coordinates
Balancing
Muscle Muscle Weight(N) Working Side X Y Z
Side
Superficial masseter 190.40 1.00 1.00 -0.21 -0.42 +0.89
Deep masseter 81.60 1.00 1.00 -0.55 +0.36 +0.76
Medial pterygoid 174.80 0.76 0.76 +0.49 -0.37 +0.79
Anterior temporal 158.00 0.98 0.98 -0.15 -0.04 +0.99
Middle temporal 95.60 0.96 0.96 -0.22 +0.50 +0.83
Posterior temporal 75.60 0.94 0.94 -0.21 +0.86 +0.47
Lateral inferior pterygoid 66.90 0.27 0.27 +0.63 -0.76 -0.17

T
NOTE. The force of each muscle was determined by the muscle weight multiplied by the

IP
scaling factors and the three-unit vector coordinates across the area of attachment. All
coordinates are referenced to a global Cartesian coordinate system, where the XY-plane is

CR
the frontal plane, XZ represents the orientation of the occlusal plane, and the YZ-plane is
orthogonal to both XY and XZ.

US
AN
M
ED
PT
CE
AC
 ACCEPTED MANUSCRIPT

Table 3 Maximum Von Mises stress (MPa).

Loading by Constraints loading ① loading ②
Reconstructed mandible Model I 85.78 93.79
Model II 164.50 253.80
Model III 97.18 112.00
Model IV 199.50 312.00
Fibular graft Model I 18.23 29.75
Model II 164.50 253.80
Model III 64.36 112.00

T
Model IV 199.50 312.00

IP
CR
US
AN
M
ED
PT
CE
AC
 ACCEPTED MANUSCRIPT

Table 4 Maximum displacement (mm).

Loading by Constraints loading ① loading ②
Reconstructed mandible Model I 0.704 0.773
Model II 0.795 0.792
Model III 0.817 0.801
Model IV 1.085 1.076
Fibular graft Model I 0.704 0.529
Model II 0.718 0.628
Model III 0.765 0.672

T
Model IV 1.035 1.021

IP
CR
US
AN
M
ED
PT
CE
AC

20