Keywords denture base thickness - dynamic loading - overdenture
Introduction
The treatment of edentulous patients with implant-supported overdenture is a common
practice in dentistry; however, denture base fracture is frequent. Fractures usually
occur after repeated flexing of the denture base under small loads. Repetitive forces
result in the development of microscopic cracks in the areas of biomechanical stress
concentration, and it is identified as the main reason for denture base fracture.[1 ]
[2 ] It is also reported that the thinnest acrylic resin areas which are around the copings
of overdentures are responsible for fracture.[3 ]
[4 ]
[5 ]
[6 ]
[7 ]
[8 ]
[9 ]
[10 ] Since biomechanical stresses are focused on these areas, it is suggested that where
the reinforcement is necessary.[7 ]
To understand fracture mechanism of overdentures, dynamic loading tests are important
and the number of these tests is limited.[11 ]
[12 ] Furthermore, there is no published literature about the effect of acrylic resin
thickness around the bar attachment. This study aimed to assess the effect of reinforcement,
denture base thickness, and acrylic resin types on dynamic and static fracture strength
in mandibular implant overdentures with bar attachments. The null hypothesis was that
reinforcement materials, denture base thickness, and acrylic resin types would have
no effect on the dynamic and static fracture strength of mandibular implant overdenture
with bar attachment.
Materials and Methods
A total of 108 experimental mandibular implant overdentures (EMOs) were fabricated
into three main groups, namely unreinforced (control: C), reinforced with unidirectional
glass fibers (FR), and Co–Cr cast metal (MR). Each of the main groups included 2-,
3-, and 4-mm denture base thicknesses and fabricated with conventional (CA) and high-impact
(HIA) acrylic resin. A power analysis of pilot data indicated a minimum specimen size
of 6 per group (α = 0.05, p = 0.80).
An edentulous mandibular model was constructed in epoxy resin (Epofix Kit, Struers
ApS, Ballerup, Denmark). Two bone-level implants with 60° conical–internal octagonal
connections (Trias, Servo-Dental GMBH, Hagen, Germany) were placed (4.1 × 12 mm) at
the positions of the canines. To simulate the viscoelasticity of the mucosa, the surface
of the residual ridge and the retromolar pad were ground out to a depth of 2 and 4
mm. An artificial mucosa was fabricated (Fit Checker, GC Corporation, Tokyo, Japan)
and adapted to epoxy resin model. A Hader-type bar attachment (Bredent, GmbH & Co.
KG, Senden, Germany) screwed onto the implants. The epoxy resin model was duplicated
for producing dental stone models.
A placeholder was fabricated at a thickness of 1.5 mm to provide sufficient space
for the reinforcing material (0.5 mm) and the acrylic resin (1 mm). It was length
to mesial sides of the first molars over the bar attachment surface ([Fig. 1 ]). Unidirectional glass FR (Stick Tech Ltd, Turku, Finland) and Co–Cr cast MR (Magnum
H60, Mesa, Travagliato, Italy) were used as a reinforcing material (0.5 mm in thickness
and 4 mm in width). Co–Cr cast MR were sandblasted with 110-μm grain-sized alumina
for 10 seconds at 0.3 MPa emission pressure, and a MR conditioner (Alloy Primer, Kuraray
Medical Inc., Tokyo, Japan) was applied.
Fig. 1 The space after the placeholder was removed.
For producing identical denture bases, thermoplastic sheets were used at 2-, 3-, and
4-mm thicknesses (Orthotechnology, Thermal Forming Splint, Florida, United States).
The denture bases were adapted onto the dental stone model and they were invested
in flasks ([Fig. 1 ]). Their surface was covered with a silicon impression material (Elite HD, Zhermack,
Italy) to be used again. Acrylic resins were mixed according to the manufacturer's
instructions. Polymerization was performed at 70°C water bath for 7 hours and at 100°C
water bath for 1 hour ([Fig. 2 ]).
Fig. 2 Thermoplastic denture bases at different thickness.
Specimens were stored in distilled water for 7 days before being thermocycled at 5
to 55°C for 5,000 cycles (30 seconds dwell time). Specimens were then subjected to
a 400,000 load regime which is equivalent to using a denture for 20 months.[13 ] The average fatigue load has been cited as 43 to 100 N.[14 ] In line with this, vertical loading of 98 N was used in this study and hence the
loading delivered to the first molar on the one side was approximately 49 N. Dynamic
loading was performed (CS-4.8; SD Mechatronik, Feldkirchen-Westerham, Germany) in
a chewing simulator ([Fig. 3 ]) with 1.2 Hz in distilled water. A concentric cyclic loading was performed. Dynamic
loading performed vertically 2 mm away from the central fossa with 2-mm horizontal
sliding motion for imitating chewing procedure. Unbroken specimens were then loaded
until fracture by a universal testing machine (MTS, 858 mini Bionix II, Eden Prairie,
Minneapolis, United States) at a crosshead speed of 2 mm/min ([Fig. 4 ]). During the natural chewing function, especially molar clench, mandible shows horizontal
and vertical bending and this function causes torsion of the mandible horizontally.[15 ] To mimic horizontal and vertical bending of the mandible, a stainless steel trapezoid
tip with a 45° base angle was designed. The static load applied through this trapezoid
tip. This trapezoid tip applied a simultaneous and equal static load to the lingual
surfaces of the first molar teeth at an angle of 45° to imitate static loads occurring
in a natural molar clench ([Fig. 5 ]).
Fig. 3 Test specimens in chewing simulator.
Fig. 4 Fracture apparatus which mimics the mandibular bending in horizontally and vertically.
Fig. 5 Sample of fractured fiber specimens.
All tests were performed using statistical software (IBM SPSS, Version 22.0, Armonk,
New York, United States). The distribution of the test results was evaluated using
the Shapiro–Wilk test. Differences in mean fracture resistance among the groups were
compared using the one-way analysis of variance (with post hoc Tukey's honestly significant
difference test) and Student's t -tests (α = 0.05).
Results
All specimens survived from the dynamic loading. C groups showed the highest fracture
strength against the reinforced groups in 2-mm (CA = 355.99 ± 28.16 and HIA = 469.78
± 30.4) thicknesses. The 2-mm denture base thickness was found insufficient for any
of the reinforcing methods. Fracture strength increased significantly when the denture
base thickness was increased (p = 0.001). The fracture strength of HIA resin was significantly higher than CA resin
in FR groups for all thicknesses (p = 0.001, p = 0.029, and p = 0.001). No significant difference was found in HIA and CA resins in C (p = 0.879 and p = 0.564) and MR groups (p = 0.627 and p = 0.804) at 3- and 4-mm thicknesses. MR groups appeared to be the weakest groups
([Table 1 ]).
Table 1
Comparisons of fracture strength means and SD among (n) in specimen groups
Reinforced and unreinforced Groups
Thickness (mm)
Mean ± SD
p -Value
CA
HIA
Abbreviations: ANOVA, analysis of variance; C, control; CA, conventional; FR, fibers;
HIA, high impact; MR, metal; SD, standard deviation.
a Student's t -test.
b One-way ANOVA test p < 0.05.
C
2
355.99 ± 28.16
469.78 ± 30.4
0.001a
MR
2
211.66 ± 44.49
404.64 ± 57.05
0.001a
FR
2
311.6 ± 41.15
519.54 ± 41.64
0.001a
P
0.001b
0.001
C
3
467.09 ± 29.26
534.50 ± 62.06
0.879
MR
3
452.4 ± 21.8
457.56 ± 12.63
0.627
FR
3
620.86 ± 67.85
704.6 ± 13.5
0.029a
P
0.001†
0.001b
C
4
803.48 ± 40.34
780.35 ± 85.92
0.564
MR
4
762.19 ± 14.63
765.53 ± 28.47
0.804
FR
4
865.54 ± 14.5
917.7 ± 20.65
0.001a
P
0.001
0.001b
Discussion
It was hypothesized that the reinforcement materials, denture base thickness, and
acrylic resin types would have no effect on the dynamic and static fracture strength
of mandibular implant overdenture with bar attachment. The statistical analysis showed
that the null hypothesis for dynamic strength should not be rejected but for static
strength should be rejected.
The 2-mm thick specimens were found too thin to pack the amount of acrylic resin necessary
for efficient impregnation of the reinforcements. C groups showed significantly higher
fracture strength than reinforced groups at 2-mm denture base thickness. This finding
in combination with the available literature implies that care should be given to
keep the denture base thickness at a minimum of 2 mm even if there is limited space
for the resin. In case of limited interalveolar distance, the artificial tooth can
be adjusted to provide a 2-mm space for the acrylic resin to avoid any fracture failures.
Furthermore, the design of both the bar attachment and the denture base can be made
with one of the indexing techniques to keep the resin thickness at 2 mm around the
attachment. According to the current study result, denture base thickness significantly
affects the fracture strength of the EMOs. The positive relationship between acrylic
resin thickness and fracture strength was confirmed in accordance with the previous
studies.[3 ]
[6 ]
[16 ]
MR groups showed the lowest fracture strength in all groups. The result of the current
study is similar to those reported by others.[17 ]
[18 ] This finding may be attributed to the foreign body effects of MR reinforcement on
acrylic resin.[12 ] Cast MR inserts, even if they are physically and/or chemically treated, do not provide
reinforcement to the denture base; instead, they weaken the overall structure by reducing
the thickness of the resin. FR groups showed the highest fracture strength at 3- and
4-mm thick EMOs. These findings were similarly with previous studies.[19 ]
[20 ]
[21 ]
[22 ]
[23 ]
[24 ]
According to previous studies, the performance of HIA resins has not been investigated
comparatively below 2.5-mm thickness, where it is actually critical. The specimen
geometry and loading protocols up-to-date were far from being able to generate similar
strains as an overdenture under function.[21 ]
[25 ]
[26 ]
[27 ]
[28 ] Thus, test specimens and loading protocol were designed in a different methodology.
In the current study, HIA resin showed significantly higher fracture strength than
CA resin in 2-mm thickness. Furthermore, HIA resin showed significantly higher fracture
strength than CA resin in 3- and 4-mm thicknesses for FR groups. Hamouda and Beyari
reported in a study that HIA resins with FR reinforcement showed better flexural strength
than CA resin.[21 ] FR reinforcement may affect the fracture strength of HIA resin, but further studies
are needed. The fracture strength was not significantly different between HIA and
CA resin for MR and C groups at 3- and 4-mm thicknesses, respectively. These results
were similar with previous studies.[25 ]
[26 ]
[28 ]
There are several limitations in this study. There is no standard protocol or information
exists about when a dynamic loading should be performed after the thermocycling in
previous studies or when a static test should be performed after the dynamic loading
test.[11 ]
[12 ] For the current study, dynamic loading test was performed within 24 hours, after
the thermocycling and static test was performed within 48 hours after the dynamic
loading test performed. All specimens were stored in distilled water at room temperature
until the dynamic loading began and over. The oral conditions cannot be accurately
simulated, although the mandibular models were fabricated in epoxy resin, which has
the similar young modulus with the mandibular jaw. EMOs were supported by epoxy models
during the dynamic test, which may not show the same flexibility as that of the mandible
in function. Although the dynamic test was performed in wet conditions, the Ph of
the oral environment could not mimic exactly. This is the first study of this kind;
hence, further in vitro studies needed to validate the result of the present study.
However, the findings as well as the procedures used in this study may aid researchers
in conducting more studies that mimic the clinical conditions more closely in the
future.
Conclusions
Within the limitations of this study, 2-mm denture base thickness had sufficient fracture
strength without reinforcement and a positive relationship between acrylic resin thickness
and fracture resistance was found.
Financial Support and Sponsorship
This work was supported by Scientific Research Projects Coordination Unit of Istanbul
University (Project No: 2097–47444 and ONAP 1509–42829). Finally, the author, Selen
Tokgöz, thanks TUBITAK-2211 for supporting the PhD education.
