Dental Materials Journal 2020; : –

Influence of artificial saliva contamination on adhesion in class V restorations

Kisaki SHIMAZU1,2, Hiroyuki KARIBE1, Riyo OGUCHI1,2,3 and Kiyokazu OGATA1,3

1
Department of Pediatric Dentistry, The Nippon Dental University School of Life Dentistry at Tokyo, 1-9-20 Fujimi, Chiyoda-ku, Tokyo 102-8159,
Japan
2
Department of Pediatric Dentistry, Momose Dental Clinic, 6-25-16 Kinuta, Setagaya-ku, Tokyo 157-0073, Japan
3
Division of Dentistry, Tokyo Metropolitan Children’s Medical Center, 2-8-29 Musashidai, Fuchu-shi, Tokyo 183-8561, Japan
Corresponding author, Kisaki SHIMAZU; E-mail: kisaki-s@tky.ndu.ac.jp

This study aimed to evaluate the effects of artificial saliva contamination on a glass ionomer cement (GIC), a resin-modified GIC
(RMGIC), and a composite resin (CR) that was used with two different etching adhesive systems. Three surface conditions were
created on bovine teeth using artificial saliva: control (group I), mild saliva contamination (group II), and severe saliva contamination
(group III). The microtensile bond strength (μTBS) of CR with dentin was significantly lower in group III than in group I. However,
the μTBS of GIC and RMGIC with both enamel and dentin showed no significant intergroup differences. Moreover, CR exhibited
significantly greater microleakage on cementum in group III than in group I, whereas both GIC and RMGIC showed no significant
differences for both enamel and cementum. Thus, GIC and RMGIC may be suitable for preventing secondary caries after class V
restorative treatments when contamination by saliva cannot be avoided.

Keywords: Glass ionomer cements, Composite resins, Microtensile bond strength, Failure mode, Microleakage

strength and greater microleakage after artificial saliva

INTRODUCTION
contamination, whereas no significant differences were
Contamination of the dental operation area by saliva found for GIC and RMGIC15).
and other fluids is detrimental to positive restoration According to Siegward et al., microtensile bond
outcomes and should be controlled. The use of a rubber strength (μTBS) was evaluated to determine retention
dam enables the control of such contamination, allowing loss, marginal discoloration, and marginal integrity on
the dentist to focus on the actual clinical procedure1-4). cervical (class V) restorations16). However, few studies
The etiology of class V cervical lesions varies, including have compared μTBS and the degrees of microleakage
occlusal factors, brushing habits, dietary regimens, and exhibited by GICs, RMGICs, and CRs in a moist
psychological manifestations. Diagnosis of these lesions environment on class V restorations.
includes abrasion caused from incorrect brushing The purpose of this in vitro study was to evaluate the
techniques, erosion from improper dietary or chemical influence of artificial saliva contamination on the μTBS
occupational exposure, abfraction through malocclusion of three restorative materials (GIC, RMGIC, and CR)
and caries caused by dissolution of tooth structure from used for class V restorations and the effect on the degree
bacterial by-products5,6). However, class V restorations of microleakage exhibited by the restorative materials
(cervical caries) are difficult to perform under a rubber after thermocycling.
dam when the cavity approaches the attachment part of
the rubber dam clamp7,8). MATERIALS AND METHODS
Composite resins (CRs) are commonly used in
restoration procedures, even though they show reduced The materials used in this study, namely GIC (Fuji IX
bond strength when contaminated by saliva either before extra capsule [F9E], GC, Tokyo, Japan), RMGIC (Fuji
or after primer application9-12). In contrast, glass ionomer II LC capsule [2LC], GC), and CR (CLEARFIL AP-X,
cements (GICs) chemically adhere to mineralized dental Kuraray Noritake Dental, Tokyo, Japan), are listed
tissues, but incomplete chemical reactions and their in Table 1. For the CR, we used two different etching
sensitivity to water during the first stage of the GIC- adhesive systems: a total-etching adhesive (OptiBond
setting reaction can lead to softening and cracking Solo Plus [CR-OBS], Kerr, Orange, CA, USA) and a self-
of the cement surface, reducing wear resistance and etching primer (Scotchbond Universal Adhesive [CR-
fracture toughness13). According to Yamazaki et al., SUA], 3M ESPE, St Paul, MN, USA).
resin-modified GICs (RMGICs) have greater shear bond Various surface conditions were created on the
strength than conventional GICs in the case of luting enamel or dentin/cementum of bovine incisors using
by water immersion14). Our previous study suggested artificial saliva. The artificial saliva contained 20 mM
that CR exhibited significantly reduced shear bond 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid,
30 mM KCl, 4 mM KH2PO4, and 0.7 mM CaCl217). The
rationale for using artificial saliva was that the presence

Color figures can be viewed in the online issue, which is avail-

able at J-STAGE.
Received Feb 1, 2019: Accepted May 30, 2019
doi:10.4012/dmj.2019-032 JOI JST.JSTAGE/dmj/2019-032
2 Dent Mater J 2020; : –

Table 1 Materials used in the study

Material Composition (batch no.) Application conditions

Fuji IX extra Powder: fluoroaluminosilicate glass, polycarboxylic acid.

capsule Liquid: polycarboxylic acid, water, polybasic carboxylic acid. —
(F9E) Shade: A2 (1504161).

Fuji II LC Powder: fluoroaluminosilicate glass.

capsule Liquid: methacrylic acid ester, polycarboxylic acid, water. —
(2LC) Shade: A2 (1501151).

Cavity Water, polycarboxylic acid, aluminum chloride Apply the conditioner for 10 s,
conditioner (140129). rinse for 10 s, and gently air dry.

Monomer: Bis-GMA, TEGDMA.

Fillers: surface-treatment glass powder, surface-treatment
CLEARFIL
silica-based micro filler. —
AP-X
Additional contents: photo-initiator, colorant.
Shade: A3 (CG0056).

Bond: Bis-GMA, hydroxyethyl methacrylate, ethyl alcohol, Apply the etchant for 15 s,
OptiBond
fillers (5450989). rinse for 15 s, and gently air dry.
Solo Plus
Gel etchant: 37.5% phosphoric Bond for 15 s, gently air dry for 10 s,
(CR-OBS)
acid (5225747). and cure under light for 20 s.

Monomer: Bis-GMA, MDP, HEMA, Decamethylene

Scotchbond Apply the adhesive to the prepared
dimethacrylate.
Universal tooth and rub it in 20 s. Gently air
Filler: Silane treated silica.
Adhesive dry the adhesive for approximately 5 s.
Additional contents: ethyl alcohol, photo-initiator, water
(CR-SUA) Light cure for 10 s.
(587216).

Bis-GMA: bisphenol A-diglycidyl methacrylate; TEGDMA: triethyleneglycol dimethacrylate; MDP: methacryloyloxydecyl

dihydrogen phosphate; HEMA: 2-hydroxyethyl methacrylate; DMA: dimethacrylate.

of calcium and phosphate ions would prevent additional manuals; a cavity conditioner was used for F9E and
demineralization, which can alter the etching depth for 2LC, and a gel etchant was used for CR-OBS. Artificial
enamel and dentin/cementum surfaces treated with acid saliva was applied to the group II and III blocks, after
or adhesives. The artificial saliva did not contain sodium which the appropriate adhesives were applied to the
azide because there was no need to store the specimens CR-OBS and CR-SUA blocks, in accordance with their
in sodium azide. The test samples were divided into respective manufacturers’ instructions. The specimens
the following groups: group I (control), in which the were prepared using a silicone mold. The 2LC, CR-
bonding surface remained dry; group II (mild saliva OBS, and CR-SUA blocks were light-cured for 20 s
contamination), in which 0.1 mL of the artificial saliva using a visible-light curing unit (G-Light Prima-II; GC),
was placed on the bonding surface and dried slightly; whereas the F9E blocks were stored for 5 min at 37°C
and group III (severe saliva contamination), in which 0.1 and 100% relative humidity. After the silicone molds
mL of the artificial saliva was used as is. were removed, all specimens were stored in water at
37°C for 24 h. Then, eight specimens were prepared for
μTBS test each of the four materials in each of the three surface
After confirming the absence of abnormalities such as condition groups (approximate dimensions, 1×1×3 mm)
discoloration or hypoplasia on the labial side, the enamel (i.e., a total of 96 specimens each for enamel and dentin).
and dentin on the cervical area of bovine incisors were The specimens were loaded to failure under tension at a
cut into blocks using a low-speed diamond blade (IsoMet, crosshead speed of 1.0 mm/min using a universal testing
Buehler, Lake Bluff, IL, USA). These blocks were machine (Autograph EZ-S; Shimadzu, Kyoto, Japan).
embedded in an acrylic resin (Unifast II, GC). Next, the The enamel/dentin side of all fractured specimens
enamel or dentin surfaces of the blocks were polished from each group was air-dried and examined with a
using 600-grit sandpaper. The blocks were then divided scanning electron microscope (Miniscope TM3000;
into the three groups. Blocks in each of the three groups Hitachi High-Technologies, Tokyo, Japan) operating at
were then further divided into four material categories: 15 kV. The fracture modes were classified as follows: (1)
F9E, 2LC, CR-OBS, and CR-SUA. The surfaces of cohesive failure in the restorative material; (2) mixed
the F9E, 2LC, and CR-OBS blocks were treated in fractures; (3) adhesive failure at the enamel/dentin-
accordance with the instructions in their respective restorative materials interface; (4) cohesive failure in
 Dent Mater J 2020; : – 3

enamel/dentin. among groups, Tukey’s honestly significant difference

(HSD) post-hoc test was used to identify group differences
Microleakage analysis accounting for the significant p-value. All analyses were
Cavities (1.5 mm in depth and 3.0 mm in diameter) were performed using IBM SPSS Statistics for Windows,
created at the occlusal (enamel) and gingival (cementum) version 21.0 (IBM Japan, Tokyo, Japan).
margins of the labial aspect of the bovine incisor enamel.
The specimens were randomly divided into one of the RESULTS
three surface condition groups, namely, group I, group
II, and group III. After application of artificial saliva, the Figure 2 shows the enamel and dentin μTBS values of
cavities were filled with the respective materials (F9E, the tested materials at the different surface conditions.
2LC, CR-OBS, and CR-SUA) as described above (n=5). F9E and 2LC exhibited no significant intergroup
The filled cavities were covered with a polyester film. The differences in both enamel and dentin μTBS (Figs. 2a
2LC, CR-OBS, and CR-SUA blocks were irradiated for 20 and b). Conversely, for CR-OBS and CR-SUA, the dentin
s, whereas the F9E blocks were stored for 5 min at 37°C μTBS was significantly lower in group III than in group I
and 100% relative humidity. Next, all specimens were (control) (CR-OBS: p=0.009, CR-SUB: p<0.001), but there
stored for 24 h at 37°C in distilled water. After removing were no significant differences in the enamel μTBS (Fig.
the polyester film, the specimens were gently polished 2a). Figure 3 illustrates the percentage of each fracture
using #600 silicon carbide paper under water irrigation. type in each group, and Fig. 4 shows the scanning
All specimens were then thermocycled for 10,000 cycles electron microscopy micrographs of the enamel/dentin
at 5 and 55°C; the dwell time was 30 s18,19). They were side of a representative fractured beam in each material.
then soaked in a 0.1% methylene blue solution at 37°C Most (46–83%) of the F9E and 2LC post-test specimens
for 20 h. The specimens were subsequently rinsed in showed cohesive failure in restorative material, while
distilled water, embedded in acrylic resin (Unifast II; most (67–100%) of the CR post-test specimens showed
GC), and sectioned longitudinally on either side of the mixed fractures (Figs. 3a and b). For CR with dentin,
cavity midline using a low-speed diamond blade (IsoMet, the percentage of specimens showing interfacial failure
Buehler). The microleakage distances were assessed
using a digital microscope (One-shot 3D Measurement
Microscope VR-3000; Keyence, Osaka, Japan, Fig. 1).

Statistical analysis
Before performing any analyses for multiple group
comparisons, the homogeneity of variance was
assessed with Levene’s test. As Levene’s test revealed
no significant differences among the groups, one-way
analysis of variance (ANOVA) was used to analyze
group differences in enamel/dentin μTBS and enamel/
cementum microleakage distances in each material. A
p-value <0.05 was considered statistically significant.
When one-way ANOVA showed a significant difference

Fig. 2 (a) Enamel μTBSs and (b) dentin μTBSs.

*ANOVA with Tukey HSD (compare with group I
Fig. 1 Measurement of the dye penetration distances. of each material), p<0.05.
4 Dent Mater J 2020; : –

Fig. 3 The distribution of failure modes in each material for the three groups. The percentage
surface area of a particular failure mode in each group represented the ratio of the
fracture surface area exhibited by that failure mode to the total surface area in all the
fracture specimens. (a) Enamel surface area and (b) dentin surface area.

Fig. 4 Scanning electron microscopy micrographs of the enamel/dentin side of a

representative fractured beam in each material.
(1) Cohesive failure in restorative material; (2) mixed fractures; (3) adhesive
failure at the dentin–restorative material interface; and (4) cohesive failure in
enamel.
 Dent Mater J 2020; : – 5

The results of the statistical analyses showed that

artificial saliva contamination decreased the dentin
μTBSs of CR-SUA and CR-OBS; however, their enamel
μTBSs were not significantly affected. Similarly, the
degree of microleakage for cementum with CR-OBS and
CR-TSB after thermocycling increased significantly with
artificial saliva contamination.
The micromechanical bonds formed on enamel
surfaces demineralized by acid etching may explain
the differences between the dentin and enamel bond
strengths. However, artificial saliva contamination did
not affect the μTBSs for F9E and 2LC; this was true for
both enamel and dentin. This result may be explained
by the chemical self-adhesion of enamel and dentin even
in the absence of applied adhesives, which are affected
by saliva.
The degree of microleakage after thermocycling
in F9E and 2LC was not affected by artificial saliva
contamination for both enamel and dentin. However, for
CR-SUA and CR-OBS, the dentin bond strengths were
significantly lower in severe saliva contamination than
in the control. The resulting microleakage may engender
staining, marginal breakdown, hypersensitivity,
secondary caries, and the development of pulpal
pathology21). Thus, these results indicate that CR with
a high bond strength was more effective for preventing
secondary caries in environments where a rubber dam
can be utilized, while GIC and RMGIC are suitable in
Fig. 5 (a) Enamel microleakage distances and (b)
environments in which the use of a rubber dam is not
cementum microleakage distances.
possible.
*ANOVA with Tukey HSD (compare with group I
of each material), p<0.05.
Assessments of the fracture surface area indicated
cohesive failure in restorative material in most (46–
83%) of the F9E and 2LC post-test specimens. This may
be attributable to the fact that the strength of F9E and
2LC is lower than that of enamel/dentin, and adhesion is
increased with the level of artificial saliva contamination tight. Under clinical conditions, it is possible to restrict
(Fig. 3b). the progression of caries, since some material remains on
The microleakage distances of the four materials and the dental surface even after removal of the restorative
the significance of the three different surface conditions materials. The CR specimens showed mixed fractures
for each material are presented in Fig. 5. F9E and 2LC or adhesive failure at the dentin-restorative material
showed no significant differences with both enamel interface, while cohesive failure in enamel was noted
and cementum (Figs. 5a and b), whereas CR exhibited in 13–25% of the specimens, which may be because the
significantly greater microleakage distances in group III strength of CR is higher than that of the enamel surface,
than in group I with the cementum (CR-OBS: p=0.027, and adhesion is tight. Therefore, when the adhesive
CR-SUB: p=0.025, Fig. 5b). force decreases as a result of the saliva mixture, severe
secondary caries occurs due to exposure of the enamel/
DISCUSSION dentin.
Clinically, class V cavities in the vicinity of the
Saliva and gingival crevice fluid can often cause cervical region of the tooth are located at the site where
contamination of a filling even when the caries is present the clamp is placed; therefore, isolation using a rubber
under the edge of the gingiva. Therefore, for successful dam is difficult. Gingival crevice fluid contamination is a
prevention of secondary caries in restored teeth, it is common occurrence during restoration procedures. This
critical to maintain the restorative material intact over suggests that GIC or RMGICs are suitable for the repair
time. Achievement of good moisture control is a common of class V cavities when isolation is not feasible.
challenge in restorative dentistry, especially when There are some limitations in the present study.
isolation using a rubber dam is not feasible18,20). Since we utilized artificial saliva, it is necessary to
In this study, we evaluated the enamel and dentin conduct experiments using solutions that replicate blood
μTBS values and degrees of microleakage for three and gingival crevice fluid. Moreover, we used bovine
restorative materials (GIC, RMGIC, and CR) exposed to incisors for the experiment. Human and bovine teeth
artificial saliva contamination for a class V restoration. are regarded as essentially homogeneous structures22,23),
6 Dent Mater J 2020; : –

and the teeth used in this study were less affected by 31: 277-280.
environmental conditions, such as the frequent use of 7) Iwatani K, Matsuo K, Kawase S, Wakimoto N, Taguchi A,
Ogasawara T. Effects of open mouth and rubber dam on
fluorine.
upper airway patency and breathing. Clin Oral Investig 2013;
17: 1295-1299.
CONCLUSIONS 8) Yoon RK, Chussid S. Topical anesthesia for rubber dam clamp
placement in sealant placement: comparison of lidocaine/
Based on the results of this study, the following prilocaine gel and benzocaine. Pediatr Dent 2009; 31: 377-
conclusions could be drawn: 381.
1. Contamination with artificial saliva does not 9) Cobanoglu N, Unlu N, Ozer F, Blatz M. Bond strength of
self-etch adhesives after saliva contamination at different
affect the enamel/dentin μTBSs of GICs and
application steps. Oper Dent 2013; 38: 505-511.
RMGICs. 10) van Meerbeek B, De Munck J, Yoshida Y, Inoue S, Vargas
2. The μTBS of CRs with dentin is reduced M, Vijay P, et al. Buonocore memorial lecture. Adhesion to
significantly when a total-etching adhesive or a enamel and dentin: Current status and future challenges.
self-etching primer is used under severe saliva Oper Dent 2003; 28: 215-235.
contamination. 11) Duke ES. The science and practice of dental adhesive systems.
Compend Contin Educ Dent 2003; 24: 417-419.
3. The microleakages in cavities with enamel
12) Leinfelder KF, Kurdziolek SM. Self-etching bonding agents.
and cementum margins filled with a GIC or an Compend Contin Educ Dent 2003; 24: 447-454.
RMGIC do not increase under artificial saliva 13) Molina GF, Cabral RJ, Mazzola I, Lascano LB, Frencken
contamination after thermocycling. JE. Biaxial flexural strength of high-viscosity glass-ionomer
4. For CR with dentin, the percentage of interfacial cements heat-cured with an LED lamp during setting. Biomed
failure increased with the level of artificial saliva Res Int 2013; 2013: 838460.
14) Yamazaki A, Hibino Y, Honda M, Nagasawa Y, Hasegawa
contamination.
Y, Omatsu J, et al. Effect of water on shear strength of glass
The presence of microleakage and the subsequent ionomer cements for luting. Dent Mater J 2007; 26: 708-712.
decrease in bond strength may result in the formation 15) Shimazu K, Karibe H, Ogata K. Effect of artificial saliva
of secondary caries around the restorative material. contamination on adhesion of dental restorative materials.
Therefore, GIC or RMGIC restorative materials are Dent Mater J 2014; 33: 545-550.
suitable for positive class V restorations. 16) Heintze SD, Thunpithayakul C, Armstrong SR, Rousson V.
Correlation between microtensile bond strength data and
clinical outcome of Class V restorations. Dent Mater 2011;
REFERENCES 27: 114-125.
17) Pashley DH, Tay FR, Yiu C, Hashimoto M, Breschi L,
1) Ammann P, Kolb A, Lussi A, Seemann R. Influence of rubber Carvalho RM, et al. Collagen degradation by host-derived
dam on objective and subjective parameters of stress during enzymes during aging. J Dent Res 2004; 83: 216-221.
dental treatment of children and adolescents a randomized 18) Eiriksson SO, Pereira PN, Swift EJ Jr, Heymann HO,
controlled clinical pilot study. Int J Paediatr Dent 2013; 23: Sigurdsson A. Effects of saliva contamination on resin-resin
110-115. bond strength. Dent Mater 2004; 20: 37-44.
2) Kameyama A, Asami M, Noro A, Abo H, Hirai Y, Tsunoda 19) Ibarra G, Johnson GH, Geurtsen W, Vargas MA. Microleakage
M. The effects of three dry-field techniques on intraoral of porcelain veneer restorations bonded to enamel and dentin
temperature and relative humidity. J Am Dent Assoc 2011; with a new self-adhesive resin-based dental cement. Dent
142: 274-280. Mater 2007; 23: 218-225.
3) Slawinski D, Wilson S. Rubber dam use: a survey of pediatric 20) Yoo HM, Pereira PN. Effect of blood contamination with
dentistry training programs and private practitioners. 1-step self-etching adhesives on microtensile bond strength
Pediatr Dent 2010; 32: 64-68. to dentin. Oper Dent 2006; 31: 660-665.
4) van Dijken JW, Horstedt P. Effect of the use of rubber dam 21) Mali P, Deshpande S, Singh A. Microleakage of restorative
versus cotton rolls on marginal adaptation of composite resin materials: an in vitro study. J Indian Soc Pedod Prev Dent
fillings to acid-etched enamel. Acta Odontol Scand 1987; 45: 2006; 24: 15-18.
303-308. 22) Sognnaes RF. Calcification in biological systems. Amer Assoc
5) Vandewalle KS, Vigil G. Guidelines for the restoration of Advance Sci 1960; 395: 411.
Class V lesions. Gen Dent 1997; 45: 254-260. 23) Eastoe JE. The amino acid composition of proteins from the
6) Owens BM. Alternative rubber dam isolation technique for oral tissues. Arch Oral Biol 1963; 8: 633-652.
the restoration of Class V cervical lesions. Oper Dent 2006;