Journal of Dentistry (2005) 33, 139–146

www.intl.elsevierhealth.com/journals/jden

Cuspal movement and microleakage in premolar

teeth restored with posterior filling materials of
varying reported volumetric shrinkage values
G.J.P. Fleminga,*, D.P. Halla, A.C.C. Shortalla, F.J.T. Burkeb
a
Biomaterials Unit, School of Dentistry, University of Birmingham,
St Chad’s Queensway, Birmingham B4 6NN, UK
b
Primary Dental Care Research Unit, School of Dentistry, University of Birmingham,
St Chad’s Queensway, Birmingham B4 6NN, UK

Received 29 October 2003; received in revised form 7 September 2004; accepted 16 September 2004

KEYWORDS Summary Objectives. To investigate the effect of polymerisation shrinkage stress

Cuspal movement; of various aesthetic posterior filling materials on cuspal movement and cervical
Gingival gingival microleakage of mesio-occlusal-distal (MOD) restorations placed in incre-
microleakage; ments in extracted maxillary premolar teeth.
Resin-based Methods. Forty sound extracted upper premolar teeth were subjected to
composite; standardised preparation of a large MOD cavity. One curing regimen was used and
Ormocer each posterior filling material was placed in eight increments with the appropriate
bonding system. A twin channel deflection-measuring gauge allowed a measurement
of individual cusp deflections at each stage of polymerisation. Restored teeth were
thermocycled before immersion in 0.2% basic fuchsin dye for 24 h. After sagittal
sectioning of the restored teeth in a mesio-distal plane, the sectioned restorations
were examined to assess cervical gingival microleakage.
Results. In general, cuspal deflection measurements were dependent upon the
constituent monomers and the associated shrinkage on curing, with significantly
increased cuspal movement (P!0.05) being recorded for Z100 (20.03G2.92 mm)
compared with Filtek Z250e (12.34G2.18 mm), P60 (13.41G4.43 mm) and Admira
(11.2G2.58 mm). No significant differences were identified between the posterior
filling materials when the cervical gingival microleakage scores were examined.
Conclusions. It would appear that a reduction in the manufacturers’ reported
volumetric polymerisation shrinkage for Z100 (4.0%) to below 3% for Filtek Z250e,
P60 and Admira, resulted in a significant reduction in the associated cuspal strain on
the MOD cavity. The diluent triethyleneglycol dimethacrylate (TEGDMA) increases
the polymerisation shrinkage of Z100 resin-based composite due to an increased
concentration of carbon-to-carbon double bonds (CaC). The replacement of TEGDMA
with urethane dimethacrylate (UDMA) and Bis-EMA in Filtek Z250e and P60,
decreases the polymerisation shrinkage stress by increasing the cross-linking of

* Corresponding author. Tel.: C44 121 237 2915; fax: C44 121 237 2932.
E-mail address: g.j.fleming@bham.ac.uk (G.J.P. Fleming).

0300-5712/$ - see front matter q 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.jdent.2004.09.007
140
polymer networks. However, no group was identified as producing less gingival
microleakage at the cervical dentine cavosurface margin when the cavities were
sectioned and examined, regardless of the reported variations in cuspal strain and
the associated volumetric polymerisation shrinkage values.
q 2004 Elsevier Ltd. All rights reserved.
Introduction
Interest in aesthetic dentistry has resulted in tooth
coloured restorations being increasingly used, not
only as a replacement material for failed or un-
aesthetic amalgams, but as the first choice material
to restore previously ‘virgin’ posterior teeth.1
These restorations generally take the form of
directly placed, resin-based composite (RBC)
materials and at present there are a number of
RBCs with varying monomeric formulations avail-
able to the dental practitioner. Bowen2 revolutio-
nised resin-based technology following the
patenting of a novel RBC based on a highly viscous
dimethylmethacrylate monomer synthesised from
the reaction of bisphenol-A and glycidyl methacry-
late (BisGMA). However, in 1962 Bowen3 reported
the necessity for the addition of a co-monomer,
triethyleneglycol dimethacrylate (TEGDMA), to
decrease the viscosity of the mixture and aid
incorporation of the filler particles. Unfortunately,
light irradiation of RBCs involves free-radical
polymerisation of dimethacrylate monomers lead-
ing to bulk contraction4 and as a result, polymeris-
ation shrinkage of the composite material may be
manifested as shrinkage stress. Individual cuspal
movement on polymerisation may be perceived by
the patient as post-operative pain.5 Bacterial
microleakage will follow interfacial debonding at
the tooth-restoration interface and this may ulti-
mately lead to marginal staining, pulpal inflam-
mation or necrosis and possibly secondary caries.6
The elimination of TEGDMA and the incorporation
of urethane dimethacrylate (UDMA) and derivatives
of BisGMA, such as bisphenol-A ethoxylated
dimethacrylate (Bis-EMA), in an attempt to primar-
ily decrease polymerisation shrinkage, improve
handling and increase the cross-linking of polymer
networks.7 Despite a meta-analysis in 19948 which
highlighted satisfactory clinical performances of
RBCs in posterior teeth, researchers continue to try
and produce a RBC which will be a simple, reliable
and cost-effective alternative to dental amalgam.
Ormocers have been employed by different
scientific disciplines since 1978, however, they are
still under development for use in dental appli-
cations since the first dental restorative material
G.J.P. Fleming et al.
based on the ormocer technology was marketed in
1998.9 After incorporation of filler particles, the
ormocer composites can be manipulated by the
dentist like a hybrid composite. Ormocers (acronym
for organically modified ceramics) are character-
ized by inorganic–organic copolymers in the formu-
lation that modify the mechanical properties10 and
ormocers are reported to have increased fracture
and wear resistance compared with RBCs.9 The
alkoxysilyl groups of the silane form an inorganic Si–
O–Si network by hydrolysis and polycondensation
reactions whilst the (meth)acrylate groups photo-
chemically induce organic polymerisation11 which
may reduce polymerisation stress following light
irradiation.
The aims of the current study were to assess the
cuspal deflection at each stage of polymerisation
for the incremental restoration of standardised
large mesio-occlusal-distal (MOD) cavities with four
posterior filling materials (three RBCs and one
ormocer with variously reported volumetric shrink-
age values) using a twin channel deflection-measur-
ing gauge. The clinical performance of these
restorations was investigated by assessing the
cervical dentine cavosurface margin for gingival
microleakage following thermocycling and immer-
sion in 0.2% basic fuchsin dye prior to sagittal
sectioning.
Materials and methods
Cuspal deflection
Forty sound extracted upper premolar teeth that on
visual examination were free from hypoplastic
defects and cracks were selected for use. Surface
deposits were carefully removed using a hand
scaler. The teeth were stored for 24 h in buffered
formal saline post-extraction, and the teeth
were then stored in water at room temperature
(23G1 8C) except when aspects of the experimental
procedure required isolation from moisture. The
teeth were fixed, in accordance with the procedure
outlined previously12 so that the resin extended to
within 2 mm of the amelocemental junction (ACJ).
The maximum bucco-lingual width (BLW) for each
Cuspal movement and microleakage with resin-based composites 141

Table 1 Dimensions of the premolar teeth (mm) highlighting no statistical differences between groups connected
with the same superscript letter (a).
Z100 (group A) Filtek Z250e (group B) P60 (group C) Admira (group D)
Mean SD Mean SD Mean SD Mean SD
a a a a
9.07 0.37 9.13 0.30 9.21 0.38 9.19 0.48

tooth was measured with a micrometer screw gauge cavity preparation (Fig. 2). To standardise place-
accurate to 10 mm (Moore and Wright, Sheffield, ment the teeth were fixed with the palatal-
England). The specimens were distributed into four measuring gauge placed approximately 2.5 mm
groups of 10 teeth (A, B, C and D), and the mean from the palatal cusp tip (Fig. 3). A baseline
BLW of the teeth, between groups, differed by no measurement was taken and tooth restoration
more than 5% (Table 1). initiated with the mesial approximal box, packing
Standardised large MOD cavities were prepared the appropriate composite in increments against a
following a protocol previously used by the authors sectional matrix (3M sectional matrix system: 3M
to ensure consistency in cavity preparation ESPE, St Paul, MN) wedged firmly against the
(Fig. 1).12 Groups A, B and C were restored with approximal aspects of the teeth. The increment
Z100, Filtek Z250e and P60 (3M ESPE, St Paul, MN), sequence involved the placement of eight approxi-
respectively, in conjunction with the associated mately ‘triangular’ increments, three for each
bonding system (Scotch Bond 1; 3M ESPE, St Paul, approximal box and two for the occlusal surface,
MN) in accordance with the manufacturer’s guide- with each increment touching only one cavity wall.
lines. Following cavity preparation, the tooth Each increment was cured using an Optilux 501
surfaces were dried, and 37% phosphoric acid halogen light for a total duration of 40 s. Following
etching gel was applied for 15 s, before rinsing each stage of polymerisation a measurement of
with water for 10 s. The cavity surfaces were briefly
dried with compressed air and damp cotton wool
pledgets before two consecutive coats of the
adhesive were applied to the etched surfaces with
a fully saturated brush tip.13 Tooth surfaces were
lightly dried with compressed air for 2–5 s, light
cured for 10 s with an Optilux 501 halogen
light (Kerr Mfg. Co., Orange, CA) operating at a
light intensity of 740 mW/cm2 used in standard
mode. The teeth were isolated to avoid moisture
contamination.
Group D teeth were restored using an Ormocer
material (Admira:VOCO GmbH, Cuxhaven,
Germany) using its associated bonding agent
(Admira bond:VOCO GmbH, Cuxhaven, Germany)
in accordance with the manufacturer’s instruc-
tions.14 The tooth surfaces were dried, etched with
the supplied gel (Vococid TM Phosphoric acid) for
15–20 s rinsed with water and air dried with
compressed air. The cavity surfaces were coated
with Admira bond, for 30 s before being dispersed Figure 1 The tooth preparation of a large MOD cavity
with air. The bonding agent was then light cured for with the buccal–palatal width (BPW) of the approximal
20 s with the Optilux 501 halogen light at its boxes of the cavity being prepared to two-thirds of the
standard setting and the teeth were again isolated BLW of the tooth (B) and the occlusal isthmus (A) being
prepared to half the BPW. The cavity depth at the
avoid moisture contamination.
occlusal isthmus was standardised to 3.5 mm from
The buccal and lingual cusps of the extracted the tip of the palatal cusp. The cervical aspect of the
teeth were approximated to the receptors of a approximal boxes were kept 1 mm above the ACJ. The
twin channel deflection-measuring gauge (Twin facial and lingual walls of the cavity were also prepared
Channel Analogue Gauge Unit: Thomas Mercer parallel to each other to ensure consistency in cavity
Ltd, St Alban’s, UK) following the standardised preparation.
142
Figure 2 The twin channel deflection-measuring gauge
employed in the current study to assess cuspal
movement.
cuspal deflection was recorded after 3 min to allow
for stress relaxation, resulting in eight measure-
ments for each individual/tooth cusp. The data for
the buccal and palatal cusp deflections were also
combined to give the ‘overall’ cuspal deflection for
each increment of composite. The results were
subjected to statistical analysis by ANOVA tech-
niques and by Tukey’s HSD paired group post hoc
comparison procedure test.
Microleakage
The restored teeth were finished with Sof-Lex
Finishing discs (3M, St Paul, MN) in a slow hand-
piece and 15 mm grit Composhape finishing diamond
burs (Intensiv, Viganello-Lugano, Switzerland) used
in an air turbine hand-piece under water spray
Figure 3 The standardised placement the teeth fixed
with the palatal-measuring gauge placed approximately
2.5 mm from the palatal cusp tip where the buccal and
lingual cusps of the extracted teeth were approximated
to the receptors of a twin channel deflection-measuring
gauge.
G.J.P. Fleming et al.
coolant. Root apices were sealed with sticky wax
and all tooth surfaces were sealed with nail varnish,
with the exception of a 1 mm band around the
margins of each restoration, and the teeth replaced
in water when the varnish dried. The specimens
were thermocycled between two waterbaths main-
tained at 65G1 and 4G1 8C, respectively,15,16 so
that the restored teeth were submerged for 10 s
with a 25 s transfer from waterbath to waterbath
for the time equivalent of 500 cycles. The teeth
were then immersed in 0.2% basic fuchsin dye for
24 h and a vertical section was made through each
restored tooth mid-sagitally in a mesio-distal plane
using a diamond cutting saw (Struers, Glasgow,
Scotland) with a ceramic disc operating at a speed
of 125 rpm with an applied load of 100 g. Sectioned
restorations were examined under a stereo-micro-
scope (Wild M3C, Heerburg, Switzerland) at
25! magnification and the extent of the cervical
gingival microleakage was recorded. Accordingly,
the degree of cervical margin microleakage was
scored:12
† 0, no evidence of dye penetration.
† 1, superficial penetration not beyond the ame-
lodentinal junction (ADJ).
† 2, penetration beyond the ADJ but not the
cervico-axial line angle.
† 3, penetration along the axial wall.
† 4, penetration into the pulp chamber.
The resultant microleakage data was analysed
using a non-parametric one-way ANOVA (Kruskal–
Wallis) test at the 5% significance level.
Results
Cuspal deflection
The mean and standard deviations of the dimen-
sions of the teeth, used for the control groups did
not vary significantly within each group between
groups (Table 1). The palatal and buccal cuspal
deflection data were single dependent variables.
The overall mean palatal or mean buccal cusp
deflection per cure increment for each material
investigated are highlighted in Fig. 4. A one-way
ANOVA on the combined data of all individual
increment deflections with ‘cusp (buccal or pala-
tal)’ as the independent variable revealed no
significant difference (PZ0.11) between deflec-
tions according to cusp (buccal or palatal). Individ-
ual cuspal deflections for each tooth/increment
were combined for all subsequent data analysis.
A two-way factorial ANOVA (with product,
Cuspal movement and microleakage with resin-based composites 143

Figure 4 Mean (a) buccal and (b) palatal cuspal deflection measurements for MOD cavities restored with different
resin-based composites cured with a ‘turbo-boosted’ halogen curing light.

and increment number as the independent vari- the four test groups were subjected to statistical
ables) of the cusp strain data revealed that the analysis by a Kruskal Wallis non-parametric one-way
factors ‘product type’ (material chosen) and ANOVA procedure (Fig. 3) no significant difference
‘restoration increment’ (1–8) were both highly was detected between groups A–D (PO0.05).
significant (P!0.001 and PZ0.003, respectively)
but no significant interaction was revealed (PZ
0.53). Overall mean (standard deviation) cuspal Discussion
deflection was greatest for Z100 at 20.03 (2.92) mm
and least for Admira at 11.2 (2.58) mm with Filtek The preparations utilised in the current investi-
Z250e at 12.34 (2.18) mm and P60 at 13.41 gation during placement of the RBCs and ormocer
(4.42) mm ranked second and third of the four
products overall. A one-way ANOVA followed by
Table 2 Mean (standard deviation) cuspal deflection
Tukey’s HSD post hoc paired group comparison measurements for each posterior filling material
procedure revealed significant differences at the 5% examined in the current study.
significance level for Z100 compared with P60,
Filtek Z250e and Admira (Table 2). Product Mean overall cuspal
deflection (mm)
Z100 20.03 (2.92)a
Microleakage
Filtek Z250e 12.34 (2.18)b
P60 13.41 (4.43)b
No single test group revealed a totally leak-free Admira 11.2 (2.58)b
cervical dentine cavosurface margin when the
gingival microleakage results were examined Products with the same superscript letter coding do not differ
significantly (PO0.05).
(Table 3). When the results of the microleakage for
144 G.J.P. Fleming et al.

a long period of time. Causton et al.18 reported cusp

Table 3 Gingival microleakage scores highlighting
movement suggesting that stress relief occurred
no significant differences between groups.
due to fractures within the tooth structure. The
Tooth Degree of microleakage cusp movements detected in the present work
Z100 Z250 P60 Admira (Table 2) are substantially less than the 2% in total
1 2 2 1 1 over a one week period seen by Causton et al.18 on
2 1 2 1 1 polymerisation of P30 (3M, St Paul, MN) but may
3 1 3 3 3 possibly be considered of a magnitude to cause
4 3 3 0 1 post-placement pain in some patients. Medige
5 0 0 0 0 et al.19 considered that two factors could be
6 0 3 2 3 involved in cusp movement, namely the bonding
7 3 1 1 3 system and the composite itself when linear strain
8 3 1 0 3 gauges were utilised to assess cuspal flexure.
9 1 0 1 1
A recent study by Molinaro et al.20 examined the
10 0 2 1 2
influence of a packable resin composite on molar
cuspal stiffness using vertical loading in a materials
testing machine. The authors reported that an MOD
material were large MOD cavities, with the prep-
cavity preparation reduced cuspal stiffness by in
arations being designed to weaken remaining tooth
excess of 60% compared with intact teeth whilst the
structure, to favour possible cuspal movement.
mean cuspal stiffness of teeth restored with the
However, these cavities could be considered to be
packable RBC was 73% of the intact unprepared
typical of amalgam replacement cavities and the
teeth at 12 months post-restoration. The gelation
number of such restorations currently placed in
point of an RBC occurs when the viscous flow of the
clinical practice is increasing since, for example,
curing monomer is unable to keep pace with
the advent of improved matrix and bonding systems
the curing contraction and post-gel contraction of
have made the use of these restorations more
the RBC is constricted by the strength of the
viable.1 Their popularity is also high with dentists adhesive bond at the tooth-restoration interface.21
and patients alike on account of their aesthetics Therefore, polymerisation shrinkage of the RBC
and, given that the cost of a direct-placement material may be manifested as shrinkage stress.4
restoration is substantially less than an indirect The principal monomers utilised in the Z100
restoration in a similar material.5 methacrylate RBC include BisGMA which is a highly
Previously, a direct current differential transfor- rigid and viscous material which requires the
mer (DCDT) was used to measure the linear addition of a co-monomer, such as TEGDMA to
displacement of the cusps, however, Jantarat decrease the viscosity of the mixture and aid
et al.17 compared the use of DCDTs with a linear incorporation of the filler particles.3 However, the
variable differential transformer (LVDT), and addition of TEGDMA increases polymerisation
reported considerable variation in results with shrinkage of the composite due to an increased
both measurement techniques. Their result was concentration of carbon-to-carbon double bonds
perhaps to be expected given the potential differ- (CaC) and therefore an increased degree of con-
ences in morphology among teeth. In the present version of the methacrylate bond. The replacement
work, the size of the teeth used in the study was of the TEGDMA with UDMA provides RBCs with
closely controlled, with the differences in mean improved mechanical properties7,22 and derivatives
bucco-palatal width being kept less than 5% of Bis-GMA, namely Bis-EMA, were also developed
between groups. Jantarat et al.17 also reported to primarily decrease polymerisation shrinkage,
that reproducable and correct vertical orientation improve handling and increase the cross-linking of
of teeth in the experimental apparatus was critical polymer networks as is the case for Filtek Z250e
to reducing the scatter in the experimental results. and P60 which contain 60 and 61% by volume of
Accordingly, considerable care and attention was zirconia/silica filler. Moderate improvements in the
taken in the present work to ensure that all teeth mechanical and physical properties of modified
were placed consistently vertically in a precisely dimethacrylate-based resins have been identified,
standardised orientation to the experimental however, a significant decrease in polymerisation
apparatus. shrinkage stress has been achieved in the current
The present study examined the effect of study compared with Z100. The manufacturers of
polymerisation shrinkage on cuspal movement Filtek Z250e report that it shrinks 18% less than
during the polymerisation procedure itself. It is Z100 and that P60 has a 25% decrease in polymeris-
recognised that cusp movement may occur over ation shrinkage compared with Z100 due to
Cuspal movement and microleakage with resin-based composites
differences in monomer constituent and volume of
reinforcing filler.
The present work focused on three RBC systems
and one ormocer and the appropriate bonding
systems. The results of the present study high-
lighted that cuspal movement was greatest with
group A (Z100) and least with group D (Admira)
although there was no significant differences
between groups B–D (Filtek Z250e, P60 and Admira)
at the 95% significance level. Total cuspal move-
ment was greatest with the dimethacrylate-based
system that utilised TEGDMA (Z100) as a diluent,
which appeared to maximise cuspal movement and
possibly the potential for post-operative pain on
biting which patients may experience as a result.
The free-radical polymerisation of dimethacrylate
monomers is accompanied by the closer packing of
molecules leading to bulk contraction.23 Ormocers
are reported to have increased fracture and wear
resistance compared with resin-based composites4
and are advertised as having a total polymerisation
shrinkage of 1.97% due to the constituent com-
ponents of anorganic–organic copolymers and addi-
tive aliphatic and aromatic dimethacrylates.
However, whilst decreased polymerisation shrink-
age and the associated shrinkage stress was evident
between groups A and D no significant reduction in
shrinkage stress was evident between groups B–D. It
would appear that non-shrink polymerisation will
possibly only be realised if novel monomer systems
are developed which do not employ dimethacrylate
matrices.
No groups were identified as producing less
gingival microleakage at the cervical dentine
cavosurface margin when the results were exam-
ined. All groups experienced severe (code 3) levels
of microleakage (Fig. 5), which would appear to
indicate that the curing regimes utilised for these
groups and/or the subsequent in vitro thermal
stressing regime employed caused the bond to
Figure 5 Boxplot display of median leakage scores for
each group showing inter-quartile ranges.
145
fail. As a result, it is proposed that whilst a
reduction in polymerisation shrinkage below 4% is
advantageous and can be achieved by modification
of the constituent monomers present, any effective
reduction in polymerisation shrinkage would appear
not to have taken place as evidenced by the degree
of microleakage seen at the cervical dentine
cavosurface margin across all groups. However, it
should be remembered that the volumetric con-
traction of a restorative resin measured under
‘free’ or ‘unrestrained’ test conditions does not
correlate directly with measurements of restor-
ation/cavity wall adaptation measurements.24 The
latter ‘wall-to-wall’ shrinkage is the clinically
important parameter in determining initial cavity
seal.
Conclusions
It would appear that a reduction in the reported
volumetric polymerisation shrinkage from 4.0 to
1.97% for Z100 and Admira, respectively, resulted in
a significant reduction in the associated cuspal
strain on the MOD cavity. However, no group was
identified as producing less gingival microleakage at
the cervical dentine cavosurface margin when the
cavities were sectioned and examined regardless of
the reported volumetric polymerisation shrinkage
values and associated cuspal deflection on
irradiation.
Acknowledgements
The authors wish to acknowledge the financial
support received from The Nuffield Foundation—
Undergraduate Research Bursaries 2002 (No.
URB/00496/G) and thank 3M ESPE and Voco UK for
supplying the materials used in the study.
References
1. Christensen GJ. Alternatives for the restoration of posterior
teeth. International Dental Journal 1989;39:155–61.
2. Bowen RL. Synthesis of a silica-resin direct filling material:
progress report. Journal of Dental Research 1958;37:90.
3. Bowen RL. Dental filling materials comprising of vinyl-silane
treated fused silica and binder consisting of the reaction
product of bisphenol-A and glycidyl methacrylate. US Patent
3,066,112; 1962.
4. Davidson CL, DeGee AJ, Feilzer AJ. The competition
between the composite–dentin bond strength and the
polymerisation contraction stress. Journal of Dental
Research 1984;63:1396–9.
146
5. Christensen GJ. Curing restorative resin. A significant
controversy. Journal of the American Dental Association
2000;131:1067–9.
6. Lutz F, Kreici I, Barbakow F. Quality and durability of
marginal adaptation in bonded composite restorations.
Dental Materials 1991;7:107–13.
7. Asmussen E, Peutzfeldt A. Influence of UEDMA, BisGMA and
TEGDMA on selected mechanical properties of experimental
resin composites. Dental Materials 1998;14:51–6.
8. El-Mowafy OM, Lewis DW, Benmerui C, Levinton C.
Meta-analysis on long-term clinical performance of
posterior composite restorations. Journal of Dentistry
1994;22:33–43.
9. Manhart J, Kunzelmann K-H, Chen HY, Hickel R. Mechanical
properties and wear behaviour of light-cured packable
composite resins. Dental Materials 2000;16:33–40.
10. Wolter H, Storch W, Ott H. New inorganic/organic
copolymers (ORMOCERs) for dental applications.
Materials Research Society Symposium Proceedings 1994;
346:143–9.
11. Wolter H, Storch W, Ott H. Dental filling materials (posterior
composites) based on inorganic/organic copolymers (ORMO-
CERs). Macro Akron 1994;503.
12. Abbas G, Fleming GJP, Harrington E, Shorthall ACC,
Burke FJT. Cuspal movement in premolar teeth restored
with a packable composite cured in bulk or incrementally.
Journal of Dentistry 2003;31:437–44.
13. Data Sheet for Scotch bond 1 supplied by 3M ESPE (St Paul,
MN).
14. Data Sheet for Admira supplied by VOCO (GmbH, Germany).
G.J.P. Fleming et al.
15. Palmer DS, Barco MT, Billy EJ. Temperature extremes
produced orally by hot and cold liquids. Journal of Prosthe-
tic Dentistry 1992;67:325–7.
16. Spierings TM, Peters MB, Bosman F, Plasschaert AM. Ver-
ification of theoretical modelling of heat transmission in
teeth by in vivo experiments. Journal of Dental Research
1987;66:1336–9.
17. Jantarat J, Panitvisai P, Palamara JEA, Messer HH. Com-
parison of methods for measuring cuspal deformation of
teeth. Journal of Dentistry 2001;29:75–82.
18. Causton BE, Miller B, Sefton J. The deformation of cusps by
bonded posterior composite restorations: an in vitro study.
British Dental Journal 1985;159:397–400.
19. Medige J, Deng Y, Yu X, Davis EL, Joynt RB. Effect of
restorative materials on cuspal flexure. Quintessence
International 1995;26:571–6.
20. Molinaro JD, Diefenderfer KE, Strother JM. The influence of a
packable resin composite, conventional resin composite and
amalgam on molar cuspal stiffness. Operative Dentistry
2002;27:516–24.
21. Palin WM, Fleming GJP. Low-shrink monomers for dental
restorations. Dental Update 2003;30:118–22.
22. Indrani DJ, Cook WD, Televantos F, Tyas MJ, Harcourt JK.
Fracture toughness of water-aged resin composite restora-
tive materials. Dental Materials 1995;11:201–7.
23. Davidson CL, Feilzer AJ. Polymerisation shrinkage and
polymerisation shrinkage stress in polymer-based restora-
tives. Journal of Dentistry 1997;25:435–40.
24. Munksgaard EC, Hansen EK, Kato H. Wall-to-wall polymeris-
ation contraction of composite resins versus filler content.
Scandinavian Journal of Dental Research 1987;95:526–31.