A.I.I.M.B.

Associazione Italiana di Ingegneria Medica e Biologica

affiliated to
I.F.M.B.E.
International Federation for Medical and Biological Engineering

MEDICON and HEALTH TELEMATICS 2004

Health in the Information Society
X Mediterranean Conference on Medical and Biological Engineering

PROCEEDINGS

Hotel Continental Terme, Ischia Porto, Naples - Italy

31 July - 5 August, 2004

This document is available at http://www.ingbiomedica.unina.it/medicon04/

 COMPARISON BETWEEN TWO DIFFERENT IMPLANT SUPPORTS
FOR OVERDENTURE RETENTION
C. Bignardi*, G. Lorenzon**, A. Screti*, E.M. Zanetti***

*Dipartimento di Meccanica, Politecnico di Torino

** Centro Chirurgico s.r.l., Brandizzo, Torino
***Dipartimento di Ingegneria Industriale e Meccanica, Università di Catania

cristina.bignardi@polito.it

Abstract: In this work two different kinds of Brånemark implants inserted bicortically in rabbit
solutions regarding implant support for overdenture tibiae. The interpretation was that the bone loss had
retention were compared by means of finite element been caused by excessive stresses. Roberts et al. (1996)
method. Bone stress and strain fields were analysed [8] reported a high remodelling rate around the tops of
considering two different mandibular bone implant threads.
conditions: ‘normal’ and ‘resorbed’ one. All these researches confirm that the analysis of
The obtained results agreed with the clinical stress pattern can give important indications for the
experience: there were strain and stress peaks at the choice of the kind of implant to be used.
level of the insertion of implants in the cortical bone,
but the strain/stress distribution and the magnitude
of peaks was different in relation to the retention
configuration chosen for the overdenture.
Introduction
The implant-supported prosthesis is an alternative to
the conventional removable dentistry: while Figure 1: X-ray image (left) and picture of an implant
conventional denture may meet the needs of many support solution
patients, others require more retention, stability,
function and aesthetics. A review of recent literature [1] The biomechanical system to analyse is complex
revealed very high success rate of implants used to because of different structures involved (compact bone,
support a mandibular overdenture; for such a reason cancellous bone, gum, implant, prosthesis), which
their use will become more and more widespread. present complex geometry and different mechanical
Today there are already several prosthetic solutions for characteristics. For this reason, it is difficult to evaluate
same clinical situations: in particular, the implant load transmission from the teeth to the bone intuitively,
support can be different depending on the type of and Finite Element Method (FEM) is a necessary tool
implants used and their layout. The clinical comparison for comparative evaluations allowing the simulation of
of different surgical treatments is difficult because, as a different surgical treatments on the same bone situation.
matter of fact, each patient has its own specific In particular, in this work, two different kinds of
biomechanical situation (Figure 1) and the scientific implant supports for overdenture retention were
literature at the moment does not provide any clear compared by means of FEM: they differed for the
directives to claims of alleged benefits of specific number of implants, for their dimension, for their
morphological characteristics of dental implants [2]. location inside the mandible and, at last, for the
However, it is well known that the success or the presence/absence of a beam connecting all implants and
failure of implants interfaced with bone (orthopaedic making them all linked.
and dental implants) depends, counting on a favourable Materials and Methods
biological reaction, on the structural condition of the
biomechanical system constituted by the bone structure The first solution (Figure 2c) for overdenture
and the implant [3, 4]. The knowledge of strain/stress retention will be called ‘traditional’ in the following and
pattern can allow to establish if bone maintenance, simulates the insertion of two Brånemark implants,
resorption or addition is more likely to take place [5]: parallel to each other, in the chin area. A resinous saddle
Hoshaw et al. (1994) [6] applied a dynamic axial tensile is the basis for the prosthesis and is linked to these
load for 500 cycles per day for five consecutive days to implants.
Brånemark implants inserted in the tibia of rabbit. The The second solution (Figure 2d) will be called
result was bone loss around the implant neck; a finite ‘modified’ in the following and simulates the insertion
element analysis showed high strains in this region. of four screwed implants, anchored to the chin area with
Duyck et al. (2001) [7] found crater-like bone defects as bi-cortical fixation. These implants are differently
a result of a dynamic transversal load applied on oriented and are connected to each other by means of a
metallic wire. The resinous saddle used as a basis for Results and Discussion
the prosthesis is linked to this metallic wire; a plastic
The main attention of this analysis was focused on
layer takes place between the metallic wire and the
stress/strain pattern of cortical and trabecular bone in
resinous saddle and has a damping function.
order to assess if the structural condition was favourable
to bone remodelling. First of all, Von Mises stress were
considered to highlight the most stressed areas, after a
more detailed analysis was carried on to assess the
orientation of principal stress in that area.
The analysis of Von Mises stresses in cortical bone
pointed out that the peak stress occured in
correspondence of the implant nearest to the applied
a) b) load (Figure 3), for both implant support solutions. The
peak stress was located at the implant insertion into
cortical bone, on the distal side. The same pattern was
observed in the case of the atrophic mandible, but
surprisingly, the peak stress had slightly decreased
(Figure 4).
A more detailed stress analysis pointed out that the
peak stress is due to a notch effect: stress field is
c) d) typically three-axial, the stressed area is very small and
corresponds to the clinical evidence of conical
Figure 2: FEM models of a) ‘normal’ mandible, b) resorption [6]. The ‘resorbed’ mandible has a smaller
‘resorbed’ mandible, c) ‘traditional’ implant support, d) cross section but stresses remain nearly the same as the
‘modified’ implant support ‘normal’ mandible because, in both cases, only a small
portion of the total bone-implant interface area carries
These two solutions were applied to two different
the external load. Furthermore, the ‘resorbed’ mandible
types of mandibles: the first one, called ‘normal’, has a
is more flexible and therefore notch intensity factor is
physiological shape, while the second one, called
lower.
‘resorbed’, shows a remarkable resorption, as often
The analysis showed also that the most influent
encountered in the clinical practice (Figure 2a, 2b).
component of the force was that one along y (distal-
Finally, four different numerical models were
mesial) direction because the application point of the
created: both ‘normal’ and ‘resorbed’ mandible were
force was nearly aligned with one of the constrain
considered with both kinds of implant support for
points, along the z (vertical) direction.
overdenture retention (Figure 2).
However, the comparison between the ‘normal’ and
A full osteo-integrated condition was simulated
‘resorbed’ mandible results biased by the fact that the
(secondary stability). The numerical models consisted
same quality of bone was considered in both cases.
of about 33000 tetrahedral elements with four nodes; the
Having assessed the influence of pure morphological
constitutive materials are listed in Table 1, their
variation as done in this work, further study should be
mechanical properties agree with data found in literature
performed taking advantage of recent works [11]
[9][10].
describing how the mechanical properties of mandibular
Table 1: Mechanical properties of materials bone (Young’s modulus and ultimate stress, mainly)
vary for different degrees of bone resorption.
Materials Young Modulus ν
[MPa]
Cortical bone 13000 0.3
Spongious bone 300 0.3
Gum 20 0.3
Titanium 100000 0.3
Resin 2000 0.4
Damping layer 500 0.4

All models were asymmetrically loaded in

correspondence of the second pre-molar. Loads were
distributed, simulating the contact with the
corresponding upper tooth; the vertical component of
the load is equal to 50 N and the distal-mesial
component is equal to 50 N [7]. Constrains simulate the
action of muscles during mastication.
Figure 3: Von Mises stresses in cortical bone:
'traditional' solution applied on a 'normal' mandible
 On the whole, the ‘modified’ overdenture produced stiffness was moderate because, considering secondary
smaller stresses than the ‘traditional’ one (-34% of Von stability, the implants were linked to each other by
Mises stress, figure 5). means of cortical bone which has a lower Young’s
This result can be explained on the basis of different modulus than the metallic wire, but shows a definitively
observations: first, the load is distributed on a larger more favourable geometry due to its larger dimensions.
number of implants, second, the notch effect is reduced Different results should be expected if primary
whenever more than one discontinuities are present: stability had been studied, because the implant would
stress has a more homogeneous distribution, even if the not be osteointegrated yet and consequently the
average stress level raises. constrain given by cortical bone would rely only on
The numerical finding is corroborated by the clinical contact forces.
experience of the second author and by radiographic The stresses in the trabecular bone for ‘traditional’
findings where larger alveolar bone losses (typical implant pointed out how the most stressed area, this
resorbed cones) are visible in correspondence of time, was located in correspondence of the distal tip of
Brånemark implant insertion into the bone. Other works the implant opposite to the loaded area (Figure 6); these
in literature agree with this observations [12]. stresses could be disregarded for two main reasons: their
magnitude was low [11] and their location was far away
from the proximal implant area which is the most
critical for what concern bone remodelling.

Figure 6: Von Mises stresses in trabecular bone:

Figure 4: Von Mises stresses in cortical bone:
'traditional' solution applied on a 'normal' mandible
'traditional' solution applied on a ‘resorbed’ mandible

Figure 5: Von Mises stresses in cortical bone:

‘modified’ solution applied on a 'normal' mandible Figure 7: Von Mises stresses in trabecular bone:
‘modified’ solution applied on a 'normal' mandible
A more detailed analysis was performed in order to
assess the structural importance of the metallic wire In the case of ‘modified’ implant, the most stressed
connecting all implants in the ‘modified’ solution; for area was near to implant insertion (Figure 7), the
this aim an hypothetical model without wire was location changed moving form the ‘normal’ mandible to
developed. The numerical analysis demonstrated that the atrophic one (Figure 8), on the whole, they were
the removal of a 2 mm diameter wire, produces a peak quite well distributed on the entire implant area and they
stress 5% higher: the reduction of implant-bone system never reached critical values [11].
 Dental Implants’, Int. Dent. J., 53 (6 Suppl 2), pp.
409-43
[3] LORENZI G.L., CALDERALE P.M. (1974): ‘Problemi
Meccanici delle Artroprotesi’, Proc. of LIX Congr.
SIOT Cagliari Italy, 1974, pp. 1-107
[4] CALDERALE P.M., FASOLIO G., MONGINI F. (1980):
‘Experimental Analysis of Strains Influencing
Mandibular Remodelling’, Acta Orthopaedica
Belgica, 46(5), pp. 601-610
[5] FROST H.M. (1990) ‘Skeletal Structural Adaptions
to Mechanical Usage (SATMU): 1. Redefining
Wolff’s law: the bone remodelling problem’,
Anatomic Records, 226, pp. 403-413
[6] HOSHAW S.J., BRUNSKI J.B.,COCHRAN G.V.B.
Figure 8: Von Mises stresses in trabecular bone: (1994): ‘Mechanical Loading of Brånemark
‘modified’ solution applied on a ‘resorbed’ mandible Implants affects Interfacial Bone Modeling and
Remodeling’, Int. J. Oral and Maxillofacial
Conclusions implants, 9, pp.345-360
The purpose of this study was to compare the effect [7] DUYCK J., RONOLD H.J., VAN OOSTERWYCK H.,
of different implant supports for overdenture retention NAERT I., VANDER SLOTEN J., ELLINGSEN J.E.
in order to identify the solution with better (2001): ‘The Influence of Static and Dynamic
biomechanical behaviour of bone-implant system. This Loading on Marginal Bone Reactions around
comparison was conducted numerically, by means of a Osseointegrated Implants: an Animal Experimental
three-dimensional finite elements model, loaded Study’, Clin. Oral Implants Res., 12(3), pp. 207-218
asymmetrically in correspondence of the second pre- [8] ROBERTS W.E., HOHLT W.F., ANALOUI M. (1996):
molar. ‘Implant-anchored Space Closure as a Viable
The implant support consisting of four bicortical Alternative to Fixed Prostheses’, in DAVIDOVITCH
screws resulted in a peak principal stress 34% lower Z., NORTON L.A. (ED): ‘Biological Mechanisms of
than implant support consisting of two Brånemark Tooth Movement and Craniofacial Adaptation’,
implants. (Harvard Society for the Advancements of
The difference between the structural behaviour of Orthodontics, Boston), pp. 617-621
prosthetized ‘resorbed’ and ‘normal’ mandibles was [9] HARIBHAKTI V.V. (1996):’The Dentate Adult
negligible for the considered load, however ultimate Human Mandible: an Anatomic Basis for Surgical
stress was different between these two situations and Decision Making’, Plast. Reconstr. Surg., 97(3),
further studies should take into account the different pp.536-541
mechanical properties of bone between a ‘normal’ and a [10] TERHEYDEN H., MUHLENDYCK C., SPRENGEL M.,
‘resorbed’ mandible. LUDWIG K., HARLE F. (1999): ’Self-Adapting
The application of a metal wire which links all bi- Washer System for Lag Screw Fixation of
cortical screws produced a stress reduction equal to Mandibular Fractures. Part II: In vitro Mechanical
about 5%, but also primary stability should be Characterization of 2.3 and 2.7 mm Lag Screw
investigated because, in this case, its role of limiting Prototypes and in vivo Removal Torque after
implant bending would be surely emphasised. Healing’, J Craniomaxillofac. Surg., 27(4), pp.243-
Finally, the finite element method has demonstrated 251
to be suitable for simulating complex biomechanical [11] GIESEN E.B., DING M., DALSTRA M., VAN EIJDEN
systems in the maxillofacial area, as also confirmed by T.M. (2004): ‘Changed Morphology and Mechanical
the agreement between obtained data and clinical Properties of Cancellous Bone in the Mandibular
experience. It should enable considerable savings to be Condyles of Edentate People’, J. Dent. Res., 83(3),
made in terms of time, material and animal experiments pp. 255-259
in the future development of different implant [12] LUO X., OUYANG G., M A X. (1998): ‘Three
techniques. Dimensional Finite Element Analysis on the
Mandibular Complete Overdenture Supported by
References Nature Roots or Implants’, Zhonghua Kou Qiang Yi
[1] DOUNDOULAKIS J.H., ECKERT S.E., LINDQUIST C.C., Xue Za Zhi, 33(5), pp.303-305.
JEFFCOAT M.K. (2003): ‘The Implant-supported
Overdenture as an Alternative to the Complete
Mandibular Denture’, J Am. Dent. Assoc., 134(11),
pp. 1455-1458
[2] JOKSTAD A., BRAEGGER U., BRUNSKI J.B., CARR
A.B., NAERT I., WENNERBERG A. (2003): ‘Quality of