Effect of abutment angulation on the strain on the bone around an implant

in the anterior maxilla: A finite element study

Xavier E. Saab, DDS,a Jason A. Griggs, PhD,b John M. Powers, PhD,c
and Robert L. Engelmeier, DMD, MSd
The University of Texas Health Science Center Dental Branch at Houston, Tex;
Baylor College of Dentistry, Texas A&M University System Health Science Center, Dallas, Tex
Statement of problem. Angled abutments are often used to restore dental implants placed in the anterior
maxilla due to esthetic or spatial needs. The effect of abutment angulation on bone strain is unknown.
Purpose. The purpose of the current study was to measure and compare the strain distribution on the bone
around an implant in the anterior maxilla using 2 different abutments by means of finite element analysis.
Material and methods. Two-dimensional finite element models were designed using software (ANSYS) for 2
situations: (1) an implant with a straight abutment in the anterior maxilla, and (2) an implant with an angled
abutment in the anterior maxilla. The implant used was 4 3 13 mm (MicroThread). The maxillary bone was
modeled as type 3 bone with a cortical layer thickness of 0.5 mm. Oblique loads of 178 N were applied on
the cingulum area of both models. Seven consecutive iterations of mesh refinement were performed in each
model to observe the convergence of the results.
Results. The greatest strain was found on the cancellous bone, adjacent to the 3 most apical microthreads on
the palatal side of the implant where tensile forces were created. The same strain distribution was observed
around both the straight and angled abutments. After several iterations, the results converged to a value for
the maximum first principal strain on the bone of both models, which was independent of element size.
Most of the deformation occurred in the cancellous bone and ranged between 1000 and 3500 microstrain.
Small areas of cancellous bone experienced strain above the physiologic limit (4000 microstrain).
Conclusions. The model predicted a 15% higher maximum bone strain for the straight abutment compared
with the angled abutment. The results converged after several iterations of mesh refinement, which confirmed
the lack of dependence of the maximum strain at the implant-bone interface on mesh density. Most of the strain
produced on the cancellous and cortical bone was within the range that has been reported to increase bone mass
and mineralization. (J Prosthet Dent 2007;97:85-92.)

CLINICAL IMPLICATIONS
The results of this finite element study suggest that using an angled abutment may decrease the
strain on bone when restoring implants in the anterior maxilla, as compared to using a straight
abutment.

S ince Branemark et al1 published the results of a 10-

year study on osseointegration in 1977, dental implants
years later. Lower survival rates were observed for im-
plants placed in the anterior maxilla. When teeth are
have increasingly been used to replace missing teeth. It lost in the anterior maxilla, the pattern of bone loss
was observed that implants placed in the anterior mandi- cannot be accurately predicted.3 This change in bone
ble of humans had a high success rate. Later, Adell et al2 morphology often dictates placement of implants with
reported that approximately 90% of the implants placed the long axis in different and exaggerated angulations to
in the anterior mandible were still in function 5 to 12 satisfy space and esthetic needs.
Regardless of the occlusal philosophy, the palatal sur-
a faces of the maxillary anterior teeth provide a vertical
Senior Resident, Department of Prosthodontics, The University of
Texas Health Science Center Dental Branch at Houston, Houston.
ramp for the mandibular anterior teeth to guide the
b
Associate Professor and Graduate Program Director, Department of mandible through protrusive and lateral excursions.4
Biomaterials Science, Baylor College of Dentistry, Texas A&M Thus, most occlusal loads applied to anterior teeth are
University System Health Science Center. at an angle to the long axis of the implants. Forces
c
Professor, Department of Restorative Dentistry and Biomaterials, applied off axis may be expected to overload the bone
and Director, Houston Biomaterials Research Center, The Univer-
sity of Texas Dental Branch at Houston.
surrounding single-tooth implants, as shown by Pa-
d
Graduate Program Director, Department of Prosthodontics, The pavasiliou et al5 by means of finite element (FE) analysis.
University of Texas Dental Branch at Houston. This creates a controversy when evaluating clinical

FEBRUARY 2007 THE JOURNAL OF PROSTHETIC DENTISTRY 85

THE JOURNAL OF PROSTHETIC DENTISTRY SAAB ET AL

reports by Eger et al6 and Sethi et al.7 These authors

concluded that angled abutments may be considered a
suitable restorative option when implants are not placed
in ideal axial positions. Studies on the biomechanical
behavior of implants have concluded that the major
concentration of stresses at the implant-bone interface
usually occurs at the crestal bone level.5,8-18
Few investigators have studied the unavoidable situa-
tion of placing and loading implants at an angulation
in the anterior maxilla. Furthermore, few conclusions
have been drawn from the quantitative data obtained
Fig. 1. Maxillary implant with straight abutment with 0.5-
by most stress analysis studies, in terms of the criteria mm-thick cortical layer. Arrows indicate orthogonal compo-
for the elastic limit or failure limit of bone, such as nents of applied point load. Triangles mark boundary with
Frost’s ‘‘mechanostat,’’ Hill’s potential function, or the zero displacement constraint. Implant is cyan, cancellous
Tsai-Wu function.19-21 The purpose of the current study bone is red, and cortical bone is blue, purple, and green.
was to measure and compare the strain distribution on
the bone around an implant placed in the anterior maxilla
using 2 different abutments by means of FE analysis. An
attempt to correlate the results of the current study with
the physiologic threshold proposed by Frost was made.

MATERIAL AND METHODS

Two-dimensional (2-D) FE models were constructed
using software (ANSYS 7.0; DRD Technology Corp,
Tulsa, Okla) for 2 situations: (1) an implant with a
straight abutment in the anterior maxilla, and (2) an im-
plant with a 20-degree angled abutment in the anterior
maxilla. The maxilla was modeled as a sagital cut of the
Fig. 2. Maxillary implant with 20-degree angled abutment
palatine process of the maxilla, including the residual
with 0.5-mm-thick cortical layer. Arrows indicate orthogonal
alveolar process and the palatine bone. The anterior
components of applied point load. Triangles mark boundary
nasal spine–posterior nasal spine (ANS-PNS) line was with zero displacement constraint. Implant is cyan, cancel-
oriented horizontally. To create the digital image of lous bone is red, and cortical bone is pink, purple, and green.
the maxilla, a dried specimen obtained from the anat-
omy lab of Baylor College of Dentistry, Dallas, Texas,
was scanned (HP Scanjet 4570c; Hewlett-Packard outline of the maxilla, (x1 y1) to (x2 y2). Finally, spline
Development Co, Palo Alto, Calif). The scanned image curves were fitted through each set of keypoints, result-
was imported into image analysis software (Image Tool ing in the final image of the maxilla.
1.21; UTHSC, San Antonio, Tex). The outline of the The same procedure was used to create the implant
image was manually plotted and each point converted image. A 4 3 13-mm implant (MicroThread 4.0 ST;
into x and y coordinates. The coordinates were finally Astra Tech Inc, Molndal, Sweden) was used. Two im-
imported into the ANSYS software as keypoints of the plants were connected to a straight (20-degree Uni-
definitive image. The bone encountered in the anterior Abutment 3.5/4.0; Astra Tech Inc) and an angled
maxilla, classified as type 3 bone, was described by abutment (Angled Abutment 4.0, Astra Tech Inc),
Lekholm and Zarb22 as a thin layer of cortical bone sur- respectively. Cylinders (Semi-Burnout Cylinder; Astra
rounding a core of dense trabecular bone of favorable Tech Inc), which are part of the prosthetic components,
strength. A 0.5-mm cortical layer was modeled using were also attached to the implant-abutment complex.
the following formulas23: Each specimen was embedded in a thermoplastic resin
! " (Simplimet 3; Buehler, Lake Bluff, Ill), and then ground
cos 90 sin 90
ð x3 y3 Þ ¼ ð x2 2x1 y2 2y1 Þ and polished down 2 mm to a sagital cut using a pol-
2sin 90 cos 90
isher/grinder (Ecomet 3; Buehler). Digital images
ð x3 y 3 Þ were then created and imported into software, as previ-
ð x4 y4 Þ ¼ ð x1 y1 Þ 1 pffiffiffiffiffiffiffiffiffiffiffiffiffiffi
ffitcort ously described for the maxilla. Both implant-abutment
x32 1 y32
images were placed within the alveolar bone at an angle
which created a set of keypoints (x4 y4) at the predeter- of 113 degrees to the ANS-PNS line, which lies within
mined distance (tcort = 0.5 mm) parallel to the original most of the norms established in the literature.24-26

86 VOLUME 97 NUMBER 2
SAAB ET AL THE JOURNAL OF PROSTHETIC DENTISTRY

Table I. Young’s modulus (E) and Poisson’s ratio (n) of bone used in previous FE studies
Study Cortical E (GPa) m Cancellous E (GPa) m
38
Weinstein et al 13.72 0.30 — —
Borchers and Reichart10 13.7 0.30 1.37 0.30
Siegele and Soltesz44 20.0 0.30 2.0 0.30
Matsushita et al45 10.67 0.30 0.18 0.30
Clelland et al30 15.0 0.30 0.056 0.30
0.25 0.30
1.4 0.30
Benzing et al11 15.0 0.25 2.0 0.495
Meijer et al33 13.7 0.30 1.37 0.30
van Zyl et al37 13.7 0.30 1.37 0.30
Akpinar et al46 10.0 0.30 0.5 0.30
Canay et al15 19.73 0.30 — —
Papavasiliou et al5 13.7 0.30 1.37 0.30
Holmes and Loftus47 14.8 0.30 1.85 0.30
0.231 0.30
Sertgoz35 13.7 0.30 1.37 0.30
Holmgren et al48 15.0 0.30 0.15 0.30
Menicucci et al34 13.7 0.30 1.37 0.30
Patra et al14 7.5 2.70 0.5 0.30
Stegaroiu et al12 15.0 0.30 1.5 0.30
Teixeira et al36 13.7 0.30 1.37 0.30
Van Oosterwyck et al13 13.7 0.30 1.37 0.30
Akca and Iplikcioglu32 13.7 0.30 1.85 0.30
Ciftci and Canay49 14.0 0.30 1.0 0.30
O’Mahony et al9 14.4 0.309 0.48 0.225
Geng et al17 13.4 0.30 1.37 0.31
10.0 0.30
7.5 0.30
5.0 0.30
1.37 0.30
Geng et al16 13.4 0.30 1.37 0.31
Geramy and Morgano18 3.40 0.26 1.37 0.38

Table II. Material properties used in this FE study that no strain occurs in the z direction, but some stress
Elastic modulus
will develop in the z direction. For the current study, a
Material (GPa) Poisson’s ratio plane-strain condition (ez = 0 and sz 6¼ 0) was given to
both models.
Ti alloy* 107.0 0.34
Cortical bone40 20.7 0.30
Zero displacement constraints must be placed on
Cancellous bone40 14.8 0.30 some boundaries of the model to ensure an equilibrium
solution. The constraints should be placed on nodes that
*Provided by the manufacturer.
are far away from the region of interest to prevent the
strain fields associated with reaction forces from overlap-
The 2 abutments used represented different restorative ping with the bone-implant interface. In the current
situations (Figs. 1 and 2). Since the primary goal of study, a zero displacement constraint was placed on all
this research was not to study the stress distribution at nodes lying along the external lines of the cortical
the implant-abutment or the abutment-prosthesis inter- bone on Figures 1 and 2. This line represents the bound-
faces, the implant-abutment complex was modeled as a ary of oral and nasopharyngeal cavities.
1-piece structure and the crown restoration omitted. Three different types of materials were assigned for
When a model is assumed to be 2-D, the z axis (third each model: cortical bone, cancellous bone, and tita-
dimension) must be specified to have either a plane- nium alloy. Each object was then subdivided in smaller
stress or a plane-strain condition. Plane stress assumes elements. The element shapes were 2-D quadralateral
the model to be thin enough that no stress occurs in with midside nodes (PLANE 82). The final models
the z direction, but it has some strain in the z direction. had a total number of 46,832 elements and 94,133
Plane strain assumes the model to be infinitely thick, so nodes for the model with the straight abutment and

FEBRUARY 2007 87
THE JOURNAL OF PROSTHETIC DENTISTRY SAAB ET AL

Fig. 3. Convergence of first principal strain values at location

of maximum strain for straight abutment (black square) and
angled abutment (black diamond) models. Fig. 4. Distribution of first principal strain on cancellous
bone. Close-up of image with straight abutment. Colors indi-
cate level of strain from dark blue (lowest) to red (highest).

Fig. 6. Distribution of first principal strain on maxillary bone

with straight abutment. Colors indicate level of strain from
dark blue (lowest) to red (highest).
Fig. 5. Distribution of first principal strain on cancellous
bone. Close-up of image with 20-degree angled abutment.
Colors indicate level of strain from dark blue (lowest) to bone, such as tensile, compressive, bending, and torsion
orange (highest). testing, pure shear tests, micro- and nano-indentation
tests, acoustic tests, and back-calculation using FE
models (Table I).31 The values 13.7 GPa and 1.37 GPa
26,529 elements and 54,166 nodes for the model with have been frequently used for the Young’s moduli of
the angled abutment. After each load application, a cortical and cancellous bone, respectively.5,10,13,16-18,32-38
mesh refinement was performed by subdividing the The original source for those values is a compressive
elements adjacent to the maximum first principal strain test study on human vertebrae.39 However, compressive
on the bone. The loads were then reapplied and a new tests are subject to the confounding factors of proper
solution obtained. specimen alignment and compliance of the loading fix-
Bone is an anisotropic material, which means it has ture, which are not factors in ultrasonic pulse technique.
different physical properties when measured in different Consequently, in the current study, cortical bone and can-
directions.27-29 Most FE analysis studies in the literature cellous bone were given a Young’s modulus of 20.7 GPa
have modeled cortical and cancellous bone to be homo- and 14.8 GPa, respectively, according to the ultrasound
geneous and isotropic. Some authors have made at- study by Rho et al.40 Poisson’s ratios were assumed to
tempts to create models of the mandible and maxilla be 0.30 for both cortical and cancellous bone. Since the
with different degrees of anisotropy.9,30 There are sev- anisotropic characteristics of maxillary bone have not
eral methods to determine the physical properties of been described in the literature, cortical and cancellous

88 VOLUME 97 NUMBER 2
SAAB ET AL THE JOURNAL OF PROSTHETIC DENTISTRY

Fig. 7. Distribution of first principal strain on maxillary bone

with 20-degree angled abutment. Colors indicate level of
strain from dark blue (lowest) to red (highest).

Fig. 8. Distribution of first principal strain on maxillary bone

with straight abutment. Color scaled to observe strains within
1000 and 3000 microstrain (green). Colors indicate level of
strain from light blue (lowest) to red (highest) on nonlinear
scale.

also within the range of mean values reported in the

literature.41,42
After applying the loads on each model, a record of
the patterns of strain distribution and values were saved.
The mesh was then refined at the node of maximum
strain detected at the implant-bone interface by dividing
each element within 2 layers adjacent to the node into
new elements with side lengths half those of the original
element. The load was applied again to observe the con-
vergence of the maximum strain values to verify the
mesh independence of the test. Seven different itera-
tions of mesh refinement were calculated for each
model.
Fig. 9. Distribution of first principal strain on maxillary bone
with 20-degree angled abutment. Color scaled to observe
strains within 1000 and 3000 microstrain (green). Colors indi- RESULTS
cate level of strain from light blue (lowest) to red (highest) on
nonlinear scale. The values for the maximum first principal strain on
the bone for each of the 7 iterations with both models
are summarized in Figure 3. The maximum strain con-
tissues in the current study were assumed to be isotropic, verged toward a finite value as the mesh density in-
linear elastic, and homogeneous. The material properties creased. This indicates that further refinement of the
of the titanium implant and abutments were provided by mesh would not have affected the results. For the model
the manufacturer (Table II). The implant was modeled with the straight abutment, the maximum first principal
as completely integrated and bonded to the bone at strain was observed on cancellous bone at the implant-
the interface. bone interface, adjacent to the last microthread on the
According to a previous study, the interincisal mean palatal side of the implant, with a value of 4650 micro-
angle is approximately 130 degrees.24 It was assumed strain (Fig. 4). For the model with the angled abutment,
that the force applied to the palatal surface of the maxil- the maximum first principal strain was predicted to be
lary prosthesis would be parallel to the long axis of the at the same location, with a value of 4020 microstrain
mandibular incisor. Therefore, the load applied near (Fig. 5). The general patterns for strain distribution
the cingulum area of the prosthesis had an angle of were similar for both models (Figs. 6 and 7). Small areas
130 degrees in relationship with the long axis of the of strain higher than 4000 microstrain were distributed
implant. The magnitude of the force of 178 N was along the most apical microthreads on the palatal aspect

FEBRUARY 2007 89
THE JOURNAL OF PROSTHETIC DENTISTRY SAAB ET AL

of the implant for both abutments. Large areas with results may be overestimated. Applying the load near
values between 1000 and 3000 microstrain were ob- the cingulum area with a buccal apical direction simu-
served surrounding the implant body in both models lates a clinical situation in which the mandibular incisors
on the buccal and palatal aspects and at the apex (Figs. close on the lingual surfaces of the maxillary incisors,
8 and 9). with the mandible near centric occlusion. It also creates
a better condition to assess the behavior of implants in
the anterior maxilla. Although the load was directed at
DISCUSSION
a greater angle with relation to the direction of the im-
In FE analysis studies, the assumptions made regard- plant, it was observed that most deformation on the
ing the geometry, mechanical properties of the mate- bone was still within the physiologic limits proposed in
rials, and loads and constraints applied to the model the literature.21 The theory proposed by Frost21 indi-
have a key role in the accuracy of the experiment. cates that mechanically induced bone remodeling re-
Several authors have reviewed the literature on FE anal- mains inactive during loads that cause maximum bone
ysis of dental implants27-29 and stressed the importance strains smaller than approximately 1500 microstrain.
of modeling bone as an anisotropic material. O’Mahony However, it is enabled when strains reach the 1500 to
et al9 compared a completely isotropic model of the 3000 range, at which point the bone mass is increased.
mandible with a transversely isotropic model and found Bone deformation higher than 4000 microstrain is con-
a 20% higher level of stress at the crestal level for the sidered to be pathologic.21 The increase in bone mass
transversely isotropic model. Clelland et al43 created a caused by strains within the physiologic range would
3-dimensional (3-D) model of the anterior maxilla tend to reduce the deformation back to the threshold
with a 1.5- and 3.0-mm-thick cortical layer with isotro- strain.
pic characteristics, which does not represent type 3 bone Future studies on the biomechanics of implants
with a thin cortical layer. In the present study, the corti- placed in the maxilla are indicated. The higher failure
cal bone for the maxilla was modeled as a 0.5-mm layer, rate of implants restored in the maxilla requires the clini-
which better represents a clinical situation. cian to be more aware of the mechanical behavior of im-
It is important to mention that similar conclusions of plant-supported prosthesis in type 3 bone. Research to
several studies regarding the location of the maximum explain the anisotropic behavior of the maxilla and man-
stresses exactly on the cortical layer5,8-15 may be closely dible and to accurately calculate the material properties
related to the material properties assigned to the bone for cortical and cancellous bone is needed, and then
model.17 Models of the mandible or maxilla that have conclusions drawn from future FE analysis studies may
a cancellous core with a Young’s modulus an order of be more clinically relevant.
magnitude less than that of the cortical layer may behave The model predicted a 15% higher maximum strain in
as if the implants were only supported by cortical the bone for the straight abutment compared with the
bone.5,9-14,16,30,32-37,44-49 Cortical bone would absorb angled abutment. The software automatic mesh genera-
most of the stresses, while the reaction forces of the can- tion resulted in approximately twice as many elements
cellous bone upon the loaded implant would be under- for the straight abutment model as for the angled abut-
estimated. By assigning to the cortical and cancellous ment model. This is not responsible for the difference
bones properties that are not so different, it can be in maximum strain between the 2 models because the
seen that the highest strains observed on the bone are solution was independent of mesh density by the final
at the coronal third of the implant-bone interface. iteration of mesh refinement (Fig. 3).
However, they are not limited to the cortical layer; Although the results of the current study can be con-
they are also shared with the cancellous core. This may sidered to be independent of mesh density, the use of a
motivate future investigators to be more thorough in 2-D model may cause its clinical application to be lim-
the research of the mechanical properties assigned to ited. The condition of plane-strain condition given to
the materials involved in FE analysis experiments. this 2-D model assumes it to be infinitely thick, disre-
Designing models that simulate clinical situations is garding the cylindrical shape of the implant and its con-
also essential. For the present study, a careful review of tact with the bone around it. Therefore, the axial forces
cephalometric norms was done to create a 2-D model. that would have been absorbed by the bone surrounding
Clelland et al43 loaded differently angled abutments in the implant were not considered, and the maximum
the anterior maxilla. The authors applied a masticatory strains reported in the current study might have been
force along the long axis of the abutments tested. That lower if a 3-D model were used. The use of a 3-D model
force direction would simulate a clinical situation with anisotropic characteristics would increase the clin-
in which the incisors are in an edge-to-edge position. ical relevance of these results. Also, in the current study
Since masticatory forces decrease significantly when it was assumed that the bone-implant interface was
the mandible is in an eccentric position and there is com- continuously bonded, which may not accurately reflect
plete disclusion of all molars, the magnitude of the all clinical situations.

90 VOLUME 97 NUMBER 2
SAAB ET AL THE JOURNAL OF PROSTHETIC DENTISTRY

CONCLUSIONS 16. Geng JP, Ma QS, Xu W, Tan KB, Liu GR. Finite element analysis of four
thread-form configurations in a stepped screw implant. J Oral Rehabil
Within the limitations of the current study, and con- 2004;31:233-9.
17. Geng JP, Xu W, Tan KB, Liu GR. Finite element analysis of an osseointe-
sidering all of the assumptions made while creating the grated stepped screw dental implant. J Oral Implantol 2004;30:223-33.
FE models, there was no difference in the pattern of 18. Geramy A, Morgano SM. Finite element analysis of three designs of an
strain distribution predicted using a straight or an implant-supported molar crown. J Prosthet Dent 2004;92:434-40.
19. Ellis DK, Natali A. Dental biomechanics. New York: Taylor & Francis;
angled abutment in the anterior maxilla. The results con- 2003. p. 1-19.
verged after several iterations of mesh refinement, which 20. Tsai SW, Wu EM. General theory of strength for anisotropic materials.
confirmed the lack of dependence of the maximum strain J Composite Materials 1971;5:58-80.
21. Frost HM. Bone ‘‘mass’’ and the ‘‘mechanostat‘‘: a proposal. Anat Rec
at the implant-bone interface on mesh density. 1987;219:1-9.
Small areas of strain higher than 4000 microstrain 22. Lekholm U, Zarb GA. Patient selection and preparation. In: Branemark
were found within the cancellous bone at the implant- P-I, Zarb GA, Albrektsson T, editors. Tissue-integrated prosthesis: osseoin-
tegration in clinical dentistry. Chicago: Quintessence; 1985. p. 199-209.
bone interface adjacent to the 3 most apical micro- 23. Boas ML. Mathematical methods in the physical sciences. 3rd ed. New
threads on the lingual aspect of both models. The model York: John Wiley; 2005. p. 410-3.
with the straight abutment had slightly higher values of 24. Schaeffer A. Behavior of the axis of human incisor teeth during growth.
Angle Orthodont 1949;19:254-75.
microstrain than the model with the angled abutment. 25. Bjork A, Palling M. Adolescent age changes in sagittal jaw relation, alve-
Most of the strain produced in the cancellous and corti- olar prognathy, and incisal inclination. Acta Odontol Scand 1955;12:
cal bone was within the range reported to increase bone 201-32.
26. Noyes HJ, Rushing CH, Sims HA. The angle of axial inclination of human
mass and mineralization.
central incisor teeth. Angle Orthod 1943;13:60-8.
27. Geng JP, Tan K, Liu GR. Application of finite element analysis in
implant dentistry: a review of the literature. J Prosthet Dent 2001;85:
585-98.
REFERENCES 28. Korioth TW, Versluis A. Modeling the mechanical behavior of the jaws
1. Branemark PI, Hansson BO, Adell R, Breine U, Lindstrom J, Hallen O, and their related structures by finite element (FE) analysis. Crit Rev Oral
et al. Osseointegrated implants in the treatment of edentulous jaw. Expe- Biol Med 1997;8:90-104.
rience from a 10-year period. Scand J Plastic Reconstr Surg Suppl 1977; 29. DeTolla DH, Adreana S, Patra A, Buhite R, Comella B. Role of the finite
16:1-132. element model in dental implants. J Oral Implantol 2000;26:77-81.
2. Adell R, Lekholm U, Rokler B, Branemark PI. A 15-year study of osseoin- 30. Clelland NL, Lee JK, Bimbenet OC, Gilat A. Use of an axisymmetric finite
tegrated implants in the treatment of the edentulous jaw. Int J Oral Surg element method to compare maxillary bone variables for a loaded
1981;10:387-416. implant. J Prosthodont 1993;2:183-9.
3. Atwood DA. Some clinical factors related to rate of resorption of residual 31. Cowin SC. Bone mechanics handbook. 2nd ed. Boca Raton (FL): CRC
ridges. J Prosthet Dent 1962;12:441-50. Press; 2001. p. 1.1-1.23.
4. McHorris WH. The importance of anterior teeth. J Gnath 1982;1:19-36. 32. Akca K, Iplikcioglu H. Finite element stress analysis of the influence of
5. Papavasiliou G, Kamposiora P, Bayne SC, Felton DA. Three-dimensional staggered versus straight placement of dental implants. Int J Oral Maxillo-
finite element analysis of stress-distribution around single tooth implants fac Implants 2001;16:722-30.
as a function of bony support, prosthesis type, and loading during func- 33. Meijer GJ, Starmans FJ, de Putter C, van Blitterswijk CA. The influence of a
tion. J Prosthet Dent 1996;76:633-40. flexible coating on the bone stress around dental implants. J Oral Rehabil
6. Eger D, Gunsolley JC, Feldman S. Comparison of angled and standard 1995;22:105-11.
abutments and their effect on clinical outcomes: a preliminary report. 34. Menicucci G, Lorenzetti M, Pera P, Preti G. Mandibular implant-retained
Int J Oral Maxillofac Implants 2000;15:819-23. overdenture: finite element analysis of two anchorage systems. Int J Oral
7. Sethi A, Kaus T, Sochor P. The use of angulated abutments in implant den- Maxillofac Implants 1998;13:369-76.
tistry: five-year clinical results of an ongoing prospective study. Int J Oral 35. Sertgoz A. Finite element analysis study of the effect of superstructure
Maxillofac Implants 2000;15:801-10. material on stress distribution in an implant-supported fixed prosthesis.
8. Kenney R, Richards MW. Photoelastic stress patterns produced by implant- Int J Prosthodont 1997;10:19-27.
retained overdentures. J Prosthet Dent 1998;80:559-64. 36. Teixeira ER, Sato Y, Akagawa Y, Shindoi N. A comparative evaluation of
9. O’Mahony AM, Williams JL, Spencer P. Anisotropic elasticity of cortical mandibular finite element models with different lengths and elements
and cancellous bone in the posterior mandible increases peri-implant for implant biomechanics. J Oral Rehabil 1998;25:299-303.
stress and strain under oblique loading. Clin Oral Implants Res 2001;12: 37. van Zyl PP, Grundling NL, Jooste CH, Terblanche E. Three-dimensional fi-
648-57. nite element model of a human mandible incorporating six osseointe-
10. Borchers L, Reichart P. Three-dimensional stress distribution around a grated implants for stress analysis of mandibular cantilever prostheses.
dental implant at different stages of interface development. J Dent Res Int J Oral Maxillofac Implants 1995;10:51-7.
1983;62:155-9. 38. Weinstein AM, Klawitter JJ, Anand SC, Schuessler R. Stress analysis of
11. Benzing UR, Gall H, Weber H. Biomechanical aspects of two different porous rooted dental implants. J Dent Res 1976;55:772-7.
implant-prosthetic concepts for edentulous maxillae. Int J Oral Maxillofac 39. Bell GH, Dunbar O, Beck JS, Gibb A. Variations in strength of vertebrae with
Implants 1995;10:188-98. age and their relation to osteoporosis. Calcif Tissue Res 1967;1:75-86.
12. Stegaroiu R, Sato T, Kusakari H, Miyakawa O. Influence of restoration type 40. Rho JY, Ashman RB, Turner CH. Young’s modulus of trabecular and corti-
on stress distribution in bone around implants: a three-dimensional finite cal bone material: ultrasonic and microtensile measurements. J Biomech
element analysis. Int J Oral Maxillofac Implants 1998;13:82-90. 1993;26:111-9.
13. Van Oosterwyck H, Duyck J, Vander Sloten J, Van der Perre G, De Coo- 41. Hellsing G. On the regulation of interincisor bite force in man. J Oral
man M, Lievens S, et al. The influence of bone mechanical properties Rehabil 1980;7:403-11.
and implant fixation upon bone loading around oral implants. Clin Oral 42. Helkimo E, Carlsson GE, Helkimo M. Bite force and state of dentition.
Implants Res 1998;9:407-18. Acta Odontol Scand 1977;35:297-303.
14. Patra AK, DePaolo JM, D’Souza KS, DeTolla D, Meenaghan MA. Guide- 43. Clelland NL, Lee JK, Bimbenet OC, Brantley WA. A three-dimensional
lines for analysis and redesign of dental implants. Implant Dent 1998;7: finite element analysis of angled abutments for an implant placed in the
355-68. anterior maxilla. J Prosthodont 1995;4:95-100.
15. Canay S, Hersek N, Akpinar I, Asik Z. Comparison of stress distribution 44. Siegele D, Soltesz U. Numerical investigations of the influence of implant
around vertical and angled implants with finite element analysis. Quintes- shape distribution in the jaw bone. Int J Oral Maxillofac Implants 1989;4:
sence Int 1996;27:591-8. 333-40.

FEBRUARY 2007 91
THE JOURNAL OF PROSTHETIC DENTISTRY SAAB ET AL

45. Matsushita Y, Kitoh M, Mizuta K, Ikeda H, Suetsugu T. Two-dimensional Reprint requests to:
FEM analysis of hydroxyapatite implants: diameter effects on stress distri- DR JASON A. GRIGGS
bution. J Oral Implantol 1990;16:6-11. BAYLOR COLLEGE OF DENTISTRY
46. Akpinar I, Demirel F, Parnas L, Sahin S. A comparison of stress and strain TEXAS A&M UNIVERSITY SYSTEM HEALTH SCIENCE CENTER
distribution characteristics of two different rigid implant designs for distal- DEPARTMENT OF BIOMATERIALS SCIENCE
extension fixed prosthesis. Quintessence Int 1996;27:11-7. 3302 GASTON AVE
47. Holmes DC, Loftus JT. Influence of bone quality on stress distribution for DALLAS, TX 75246
endosseous implants. J Oral Implantol 1997;23:104-11. FAX: 214-370-7001
48. Holmgren EP, Seckinger RJ, Kilgren LM, Mante F. Evaluating parameters E-MAIL: [email protected]
of osseointegrated dental implants using finite element analysis — a
two-dimensional comparative study examining the effects of implant 0022-3913/$32.00
diameter, implant shape, and load direction. J Oral Implantol 1998;24: Copyright ! 2007 by The Editorial Council of The Journal of Prosthetic
80-8. Dentistry.
49. Ciftci Y, Canay S. Stress distribution on the metal framework of the im-
plant-supported fixed prosthesis using different veneering materials. Int J
Prosthodont 2001;14:406-11. doi:10.1016/j.prosdent.2006.12.002

The effects of common errors on sealing ability of total-etch adhesives

Noteworthy Abstracts Hashimoto M, Tay FR, Svizero NR, de Gee AJ, Feilzer AJ, Sano H,
of the Kaga M, Pashley DH. Dent Mater 2006;22:560-8.
Current Literature

Objectives: This study evaluated the effect of errors commonly made in using total-etch adhesives, on the
resulting bond strength, fluid movement and nanoleakage of resin dentin bonds.
Methods: Two total-etch adhesives were used for bonding to dentin according to the manufacturers’
recommendations, with meticulous solvent evaporation (control), or with the introduction of common bond-
ing errors—wet bonding without solvent evaporation (no evaporation), and dry bonding.
Results: The 24-hour bond strength of the control was significantly higher than the other groups (P,0.05).
For all groups, the higher initial permeability declined significantly after 24 h. The fluid movement across bonded
dentin was similar in the control and dry bonding for both adhesives, whereas significantly higher permeability
(P,0.05) was recorded for the no evaporation groups even after 24 h. Extensive silver impregnation within
hybrid layers was seen by TEM in the no evaporation and dry bonding specimens after 24 h. Dry bonding caused
collapse of the collagen matrix and interfered with resin infiltration. In contrast, inadequate solvent evaporation
and/or residual water during dentin bonding results in dilution or incomplete polymerization of the resin,
leading to severe nanoleakage formation.
Significance: Increased permeability associated with incomplete solvent evaporation in total-etch adhesives
may lead to poor bond strength.—Reprinted with permission of Elsevier Publishing.

92 VOLUME 97 NUMBER 2