Microleakage in Class II Composite Resin

Restorations: Total Bonding and Open

Sandwich Technique
Alessandro Dourado Loguercioa/Alessandra Reisb/Kelly Cristina
Mazzoccoc/André Lozano Diasd/Adair Luiz Stefanello Busatoe/
Julio da Motta Singerf/Patricia Rosag

Purpose: The objective of this in vitro study was to evaluate gingival microleakage in Class II total bond
resin restorations in comparison to open sandwich technique restorations using different materials.
Materials and Methods: Forty-eight human molar teeth were disinfected and stored in a 0.9% saline so-
lution. In each tooth, two standardized Class II cavities (3 mm x 6 mm x 2 mm) were prepared with the
gingival cavosurface margins located 1 mm below the cementoenamel junction. The teeth were divided
into 4 equally sized groups (n = 12), and the proximal boxes were treated as follows: in Group 1, no
base material was used and the cavity was restored using Syntac Sprint (SS) and Tetric Ceram (TC); in
Groups 2, 3, and 4 the gingival portion of the cavity was restored with different base materials (Group
2, Dyract; Group 3, Vitremer; Group 4, Chelon-fil) prior to the placement of the composite resin. After a
storage time of 7 days, the restorations were finished and polished. Then, specimens were submitted
to thermocycling (500 cycles, 5ºC to 55ºC, 15 s dwell time) and immersed in a 0.5% methylene blue
solution for 24 h. After washing, they were sectioned in a mesio-distal direction. Each restoration was
evaluated under a stereomicroscope at 20X by 2 examiners and scored on a 0 to 3 scale according to
the marginal leakage. Kappa statistics were used to evaluate the agreement between the examiners.
Given the ordinal nature of the scoring system, data were submitted to a nonparametric repeated mea-
sures ANOVA. The results were confirmed with a parametric repeated measures ANOVA.
Results: Significant differences (p < 0.001) among the four groups with respect to dye penetration were
detected, with the association Vitremer/Tetric showing the best results.
Conclusion: The use of Vitremer in the open sandwich technique presents the lowest degree of mi-
croleakage among the treatments considered in this study.

J Adhes Dent 2002; 4: 137–144. Submitted for publication: 11.07.00; accepted for publication: 10.02.02.

a PhD student, Department of Dental Materials, School of Dentistry, Reprint requests: Alessandro Dourado Loguercio, Departamento de
University of São Paulo, Brazil; Assistant professor, University of Materiais Dentários, FOUSP, Av. Prof. Lineu Prestes, 2227 CEP
Oeste de Santa Catarina, Joaçaba, SC, Brazil. 05554–900, São Paulo, SP, Brazil. Fax: +55-11-3091-7842-201.
b e-mail: aloguercio@hotmail.com
PhD student, Department of Dental Materials, School of Dentistry,
University of São Paulo, Brazil.
c Dentist in private practice, Concórdia, sc., Brazil. lthough composite resins are the most common-
d

e
Dentist in private practice, Pelotas, RS., Brazil.
Professor, School of Dentistry, University Luterana do Brasil, Rio
A ly used tooth-colored restorative material, they
still present difficulties when used directly in poste-
Grande do Sul, Brazil.
rior restorations.28 One of the most relevant con-
f Professor, Department of Statistics, Institute of Mathematics, Uni-
versity of São Paulo, Brazil. cerns is polymerization contraction, which may lead
g PhD student, Department of Statistics, Institute of Mathematics, to gap formation.7 This fact in conjunction with mas-
University of São Paulo, Brazil. ticatory stresses and thermal changes constitute

Vol 4, No 2, 2002 137

Loguercio et al
the principal causes of microleakage in Class II res-
torations and consequently of secondary or recur-
rent caries. In some clinical studies, these new le-
sions are considered as the most relevant reason
for replacement of these restorations.22
In spite of recent improvements in adhesive sys-
tems, several in vitro studies demonstrated consid-
erable marginal microleakage and poor marginal ad-
aptation of composite restorations, mainly when cer-
vical margins are located in the dentin.4,10,23,25,32
Several alternative clinical techniques have been
introduced to avoid the sealing problems in Class II
cavities. Among these, we mention the replace-
ment of a substantial part of the resin composite
in the proximal box with a glass-ionomer cement
base, first introduced by McLean et al.19 In fact, ini-
tially, conventional glass-ionomer cements were
completely covered with composite resins in the
so-called closed sandwich technique.16 Alternative-
ly, some authors left this material exposed in the
cervical margins of Class II restorations (“open
sandwich technique”). Nevertheless, this concept
failed mainly because of a continuous loss of the
material.16,35,36
Recently, some new materials less sensitive to
the application procedures, with improved mechan-
ical and physical properties, and with better control
during clinical handling17,20 were introduced on the
market. They are the so-called resin-modified glass
ionomers and polyacid modified composite resins
(compomers).18
The purpose of this study was to evaluate the mi-
croleakage at the cervical margins of Class II resto-
rations restored with different materials as a base
(“open sandwich technique”) compared to direct
composite resin restorations (total bonding tech-
nique).
MATERIALS AND METHODS
Sample Preparations
Forty-eight sound human molars were cleaned and
disinfected in 0.5% chloramine for 15 days and
stored for less than 6 months in a 0.9% saline so-
lution.9 On each tooth, two standardized Class II
cavities were prepared in the proximal surfaces
(3 mm wide, 6 mm high, and 2 mm deep) with gin-
gival margins located approximately 1 mm below
the cementoenamel junction (CEJ) by means of cy-
lindrical diamond burs (# 3145, Kg Sorensen, SP,
138
Brazil). The teeth were placed in a special device
that allowed simulation of a clinical situation (In-
odon, Porto Alegre, RS, Brazil). Metal matrix bands
and wood wedges were placed before each restor-
ative procedure.
Restorative Procedures
The teeth were randomly divided into 4 groups of
12 teeth each, as detailed in Tables 1 and 2. The
same material was placed both in the mesial and
distal cavities of each tooth, in order to verify the
effect of the proximal side. All materials were han-
dled according to the manufacturers’ instructions
(Table 1). Each Class II cavity was restored in three
increments (Fig 1).
Under the open sandwich technique, all base ma-
terials were inserted in the cavity by means of a Cen-
trix syringe (Centrix, Shelton, CT, USA) in a single
portion. Each resin increment (2 mm) was light
cured for 40 s with Optilux 400 (Demetron, Dan-
bury, CT, USA) with a light intensity of 650 mW/cm2.
The conventional glass-ionomer cement was pro-
tected with wet cotton rolls during the initial gelefi-
cation (4 min). The base materials in the proximal
surfaces of the restorations in groups 3 and 4 (Vit-
remer and Chelon-Fil) were protected with Finishing
Gloss (3M Dental Products, St Paul, MN, USA).
Assessment Procedure
The restored teeth were stored in distilled water at
37ºC for 7 days. After that, they were polished and
finished with aluminum oxide–coated flexible disks
(Soft Lex Pop-On, 3M Dental Products). The teeth
were subjected to thermocycling (500 cycles, 5ºC
to 55ºC, 15 s dwell time), and each tooth was cov-
ered with 2 coats of nail varnish except for the area
approximately 1 mm away from the gingival margin
of the restoration. Then, they were immersed in
0.5% methylene blue dye solution. After 24 h, they
were cleaned and sectioned in a mesio-distal direc-
tion using a diamond saw (LABCUT 1010 machine,
Extec, Enfield, CT, USA).
The severity of the dye penetration was analyzed
according to a 0 to 3 scale (0 = no dye penetration,
1 = dye penetration up to one half of the gingival
floor, 2 = dye penetration up to more than one half
of the gingival floor, and 3 = dye penetration up to
the axial wall). Both sections were evaluated at the
The Journal of Adhesive Dentistry
 Loguercio et al

Table 1 Materials, composition and manufacturers’ instructions

Materials Composition (Batch Number) Manufacturers’ instructions

Adhesive system 1 – Phosphoric Acid (37%) (908874) 1. Acid etching for 15 s.

(Syntac Sprint-SS) 2 – Adhesive (maleic acid, HEMA, 2. Rinse for 15 s.
methacrylate-modified, polyacrylic acid, fluoride, 3. Blot dry (wet bonding technique)
initiators, stabilizers and acetone) (912453) 4. Application of two coats of adhesive (15 s)
5. Air dry for 3 s
6. Light cure for 10 s.

Composite resin Organic matrix (bis-GMA, UDMA and 1. Incremental placement (< 2mm)
(Tetric Ceram-TC) TEGDMA) with five types of inorganic 2. Light cure for 40 s.
Shade A3 particles (barium glass, fluorosilicate glass,
silica, oxides and yterbium trifluoride)
(911146)

Polyacid-modified Sr-F-Al silicate glass, polyacrylic acid, acidic poly- 1. Incremental placement (< 2mm)
resin (Dyract-DY) merizable monomers, UDMA and TCB 2. Light-cure for 40 s.
Shade A3 (970300371)

Resin-modified 1 – Primer (Vitrebond copolymer, HEMA, ethanol 1. Primer application for 30 s

glass-ionomer and photoinitiators) (3303) 2. Air thin for 15 s
(Vitremer-VT) 2 – Powder (Sr-Al-F silicate glass, potassium 3. Light cure for 30 s
Shade A3 persulfate and ascorbic acid) (466) 4. Manipulation (2, 5:1) and placement with Centrix
3 – Liquid (polycarboxylic acid with pendant syringe
methacrylate groups, Vitrebond copolymer, 5. Light cure for 40 s
HEMA, ethanol and photoiniciators) (3303) 6. Protection with Finishing Gloss
4 – Finishing Gloss (bis-GMA and photoinitiators)
(3303)

Traditional glass- 1 – Ketae conditioner (10% polyacrylic acid) (066) 1. Conditioner application for 10 s
ionomer 2 – Powder (silica, alumina, calcium fluoride and 2. Rinse for 30 s
(Chelon-Fil-CF) aluminium fluoride) (011) 3. Gentle air dry (blot dry)
Shade LY 3 – Liquid (aqueous solution of tartaric and 4. Manipulation (1:1) and placement with Centrix
polyacrylic acid) (066) syringe, wait for the initialgelefication.
5. Protection with Finishing Gloss

Table 2 Restorative procedures and sequence of materials

Groups Restorative technique

Total bonding 1 Acid etching and adhesive application (SS) + placement of

composite resin (TC)

2 Acid etching + adhesive application (SS) + placement of DY

+ composite resin (TC)

Open sandwich 3 Vitremer application (primer and placement of VT) + acid

technique etching and adhesive application (SS) + placement of
composite resin (TC)

4 Chelon-Fil application (Acid etching with polyacrylic acid

and placement of CF) + acid etching and adhesive
application (SS) + placement of composite resin (TC)

Vol 4, No 2, 2002 139

Loguercio et al
Fig 1 Flowchart of the investigation design.
gingival margin and, to stay on the conservative
side, only one score (the worst one) was recorded
for each restoration. The evaluation was made by
two previously calibrated examiners using a stere-
omicroscope at 20X (Olympus SZPT, Tokyo, Japan).
Kappa statistics were calculated by section to mea-
sure the agreement between the scores assigned
by the examiners. Due to the ordinal nature of the
data, nonparametric repeated measures analysis
of variance was used5 to evaluate the effects of
treatment groups, proximal surfaces, and their in-
teraction. This technique is based on the compari-
son of the so-called relative treatment effects, cal-
culated from the mid-ranks assigned to the ordinal
response variable (scores), for each combination of
the levels of the factors group and section. This
measure may be interpreted as the probability that
an observation randomly selected from a certain
140
treatment has a higher classification than an obser-
vation randomly selected from the whole sample.
Therefore, higher values of this measure are asso-
ciated to treatments with higher scores taking only
the relative positions and not the actual score val-
ues into account.
RESULTS
The Kappa statistics ([0.846 0.062] for the mesial
section and [0.967 0.032] for the distal section)
indicate strong agreement between the examiners.
The observed frequency of microleakage scores for
each section and groups with the estimated rela-
tive effects are presented in Table 3. Examples of
restorations exhibiting different microleakage
scores are shown in Figs 2 and 5 to 7.
The Journal of Adhesive Dentistry
 Loguercio et al

Table 3 Observed frequency of microleakage scores and astimated

treatment effects

Groups Scores M Scores D

0 1 2 3 (*) 0 1 2 3 (*)

Total bonding 1 3 1 2 6 0.62 4 0 2 6 0.60

Open sandwich 2 6 0 0 6 0.53 5 2 0 5 0.53

technique 3 10 0 0 2 0.33 10 1 0 1 0.31
4 1 6 2 3 0.59 5 2 2 3 0.49

(*) - Estimated treatment effects for each section and group.

Fig 2 Composite resin restoration Fig 3 Sandwich restoration Fig 4 Sandwich restoration using
(Tetric Ceram plus Syntac Sprint) show- using Vitremer/Tetric Ceram Chelon-Fil/Tetric Ceram showing
ing leakage score 0. showing leakage score 0. leakage score 3.

The results suggest no significant interaction be-
tween group and section (p = 0.853) and no signif-
icant main effect of section (p = 0.500). Significant
differences were detected among groups (p =
0.004). Paired comparisons of the treatment rela-
tive effects suggest that only Group 3 differs from
the remaining ones with respect to microleakage.
The corresponding p-values are presented in
Table 4. These results were confirmed by a re-anal-
ysis of the data with a parametric repeated mea-
sures ANOVA.24
Some internal porosity may be noted in the cas-
es where Chelon-fil and Vitremer were used as
base (Figs 3 and 4).
DISCUSSION
The results suggest that direct composite restora-
tions in Class II cavities (Group 1) do not provide
good marginal sealing at the cervical margins of the
restorations. This fact has been previously de-

Vol 4, No 2, 2002 141

Loguercio et al
Fig 5 Sandwich restoration using Fig 6 Sandwich restoration us-
Chelon-Fil/Tetric Ceram showing leak- ing Dyract/Tetric Ceram showing
age score 2. leakage score 3.
Table 4 P-values for paired comparison between
treatment groups (nonparametric repeated mea-
sures analysis of variance)
Groups 2 3 4 In our study, the conventional glass-ionomer was
1 0.271 0.000 0.227
2 0.001 0.912
3 0.014
scribed with different composite resins and adhe-
sive systems,11,12,14,21,28 mainly when total bond-
ing restorations are compared to those obtained
under open sandwich techniques.
Both polymerization shrinkage stresses induced
by composite resins and the insufficient penetra-
tion of the bonding agent into the demineralized
dentin may reasonably explain these results.32 The
latter may occur either as a consequence of the col-
lapse of the collagen structure when dentin is unin-
tentionally desiccated or by an inadequate satura-
tion with resin monomers (primer dilution) in the
presence of excess water.26,31 All these events
142
Fig 7 Composite resin restoration
(Tetric Ceram plus Syntac Sprint)
showing leakage score 3.
may occur simultaneously in Class II cavities after
acid conditioning, where some of the cavity walls
may be extensively dried, but the internal angles of
the preparation may present an excess of water.
used with the open sandwich technique, since Reid
et al27 demonstrated a better behavior of this ma-
terial using this technique compared to its use in a
closed sandwich technique. Several other studies
have also shown bad results when this material
was used as a base with a closed sandwich tech-
nique.10,12,23,25,29,32 The idea that the contraction
forces which occur within a polymerizing composite
resin are sufficiently strong to disrupt the bond be-
tween glass ionomer and dentin, mainly in the ini-
tial stages of the glass-ionomer cement matura-
tion,15,27,30 may be to blame for the poor results
observed in this group.
The best marginal sealing was obtained when a
resin-modified glass ionomer (Vitremer) was used
with the open sandwich technique. Miller et al,21
testing the marginal microleakage of different ma-
terials, observed that the lowest degree of mi-
croleakage occured when Vitremer was associated
with Heliomolar RO or Tetric Ceram, in contrast to
what they observed when the same composite res-
The Journal of Adhesive Dentistry

ins were used without any base materials. In anoth-
er study, Dietrich et al12 evaluated the marginal ad-
aptation of a composite resin in large mesio-occlus-
al-distal cavities by using only adhesive systems (to-
tal bonding restoration), conventional or resin-mod-
ified glass ionomers with closed or open sandwich
technique. The authors showed that the use of res-
in-modified glass ionomers or compomers under
the open sandwich technique may improve the mar-
ginal adaptation of the restorations.
Our results did not show good cervical sealing
when the compomer (Dyract) was used as a base in
the open sandwich technique. Despite the fact that
studies with compomers are very rare, they have in-
dicated that such products are as good as res-
in-modified glass ionomers in terms of microleak-
age, and that they provide a better seal when
compared to total-bonding restorations.11,12,14,28
These results contradict those obtained in our study.
Comparisons with other studies are often quite
difficult, because some conditions – eg, size of
preparations, techniques, and materials – are sel-
dom alike. Despite the methodological differences,
Dietrich et al12 and Friedl et al14 suggested that the
resin-modified glass ionomer cements have a high-
er percentage of excellent margins12 or fewer mar-
ginal gaps14 than compomers; however, the differ-
ences they observed were not statistically signifi-
cant. Dietrich et al12 emphasized that the good re-
sults obtained with compomers using open sand-
wich techniques depend on the kind of association
between the compomer and composite resins.
The good results obtained with Vitremer in our
study may be attributed to some features of this
material. Tolidis et al33 indicated that the underly-
ing resin-modified glass ionomer appeared to be
able to absorb some of the polymerization stresses
of the composite resin setting, reducing the stress
accumulation in the dentin-restoration interface.8
Other authors suggest that the use of resin-modi-
fied glass ionomer could change the configuration
factor to a more favorable internal shape, minimiz-
ing the polymerization contraction effects.7,8 The
intrinsic porosity of this material, introduced by
hand mixing, can increase the “within-material”
free surface area, which also contributes to stress
relief.2,8 Furthermore, the higher water sorption
and hygroscopic expansion that occur with this kind
of material may decrease the gaps developed in
the bonding interface.3,13
The low elasticity modulus of this material is also
considered by some authors as another reason for
Vol 4, No 2, 2002
Loguercio et al
the good seal it provides. Its relative “flexibility” can
compensate the internal stress and the high stiff-
ness of the composite resins after cure,8,33 pre-
venting the adhesive interface from debonding.8
This fact has been correlated with a better marginal
adaptation. However, based on the methodology
considered in our study, we cannot identify this fea-
ture as one of the reasons for the good results pre-
sented by the resin-modified glass ionomer. This
could only be assessed if the effect of mechanical
loading were evaluated. Studies testing different
materials under load cycles have also shown con-
troversial results.4,14
It seems that these resin-modified glass iono-
mers have excellent behavior when used in con-
junction with composite resins in posterior restora-
tions without enamel margins. Recent 3-year clini-
cal evaluations published by Aboush et al,1 Bur-
gess et al,6 and Van Dijken et al34 also confirmed
the good laboratory results. However, further clini-
cal studies must be conducted in order to validate
all the conclusions presented.
CONCLUSION
Within the limits of this study, the resin-modified
glass ionomer (Vitremer) placed in the gingival por-
tion of Class II composite resin restorations (open
sandwich technique) may be a practical method to
reduce microleakage.
ACKNOWLEDGMENTS
The authors of this study are grateful for the financial support
provided by PRONEX - Federal University of Pelotas, School of
Dentistry, as well as FAPESP (99/05124-0 and 99/10611-8)
and the Department of Dental Materials, School of Dentistry,
University of São Paulo. The authors are also grateful to Paulo
Eduardo dos Santos for graphic illustration support.
REFERENCES
1. Aboush YEY, Torabzadeh H. Clinical performance of Class II
restorations in which resin composite is laminated over res-
in-modified glass-ionomer. Oper Dent 2000;25:367-373.
2. Alster D, Feilzer AJ, De Gee AJ, Mol A, Davidson CL. The de-
pendence of shrinkage stress reduction on porosity concen-
tration in thin resin layers. J Dent Res 1992;71:1619-1622.
3. Attin T, Buchalla W, Kielbassa AM, Helwig E. Curing shrinkage
and volumetric changes of resin-modified glass ionomer re-
storative materials. Dent Mater 1995;11:359-62.
143
Loguercio et al
4. Ausiello P, Davidson CL, Cascone P, DeGee AJ, Rengo S. Deb-
onding of adhesively restored deep Class II MOD restorations
after functional loading. Am J Dent 1999;12:84-88.
5. Brunner E, Langer F. Nonparametric analysis of ordered cat-
egorical data in designs with longitudinal observations and
small sample sizes. Biom J 2000;42:663-675.
6. Burgess JO, Summitt JB, Robbins JW, Haveman CW, Nimmi-
koski P, Gardiner D. Clinical evaluation of base, sandwich and
bonded Class 2 composite restorations. J Dent Res
2000;79:163.
7. Carvalho RM, Pereira JC, Yoshiyama M, Pashley DH. A review
of polymerization contraction: the influence of stress devel-
opment versus stress relief. Oper Dent 1996;21:17-24.
8. Davidson CL. Glass-ionomer bases under posterior compos-
ites. J Esthet Dent 1994;6:223-236.
9. DeWald JP. The use of extracted teeth for in vitro bonding stud-
ies: A review of injection control considerations Dent Mater
1997;13:74-81.
10. Dietschi D, De Siebenthal G, Neveu-Rosenstand L, Holz J.
Influence of the restorative technique and new adhesives on
the dentin marginal seal and adaptation of resin composite
Class II restorations: an in vitro evaluation. Quintessence Int
1995;26:717-727.
11. Dietrich T, Kraemer M, Losche GM, Roulet JF. Marginal integ-
rity of large compomer Class II restorations with cervical mar-
gins in dentine. J Dent 2000;28:399-405.
12. Dietrich T, Losche AC, Losche GM, Roulet JF. Marginal adap-
tation of direct composite and sandwich restorations in Class
II cavities with cervical margins in dentine. J Dent 1999;
27:119-128 (Erratum J Dent 1999;27:463-464).
13. Feilzer AJ, Kakaboura AI, de Gee AJ, Davidson CL. The influ-
ence of water sorption on the development of setting shrink-
age stress in traditional and resin-modified glass ionomer ce-
ments. Dent Mater 1995;11:186-190.
14. Friedl KH, Schmalz G, Hiller KA, Mortazavi F. Marginal adap-
tation of composite restorations versus hybrid ionomer/com-
posite sandwich restorations. Oper Dent 1997;22:21-29.
15. Garcia-Godoy F. Glass ionomer materials in Class II compos-
ite resin restorations: to etch or not to etch? Quintessence
Int 1988;19:241-242.
16. Knibbs PJ. The clinical performance of a glass polyalkenoate
(glass ionomer) cement used in a “sandwich” technique with
a composite resin to restore Class II cavities. Br Dent J
1992;172:103-107.
17. McCabe JF. Resin-modified glass-ionomers. Biomaterials
1998;19:521-527.
18. McLean JW, Nicholson JW, Wilson AD. Proposed nomencla-
ture for glass-ionomer dental cements and related materials
[letter to the Editor]. Quintessence Int 1994;25:587-589.
19. McLean JW, Powis DR, Prosser HA. The use of glass-ionomer
cements in bonding composite resins to dentine. Br Dent J
1985;185:410-414.
144
20. Meyer JM, Cattani-Lorente MA, Dupuis V. Compomers: be-
tween glass-ionomer cements and composites. Biomaterials
1998;19:529-539.
21. Miller MB, Castellanos IR, Vargas MA, Denehy GE. Effect of
restorative materials on microleakage of Class II compos-
ites. J Esthet Dent 1996;8:107-113.
22. Mjör IA. The location of clinically diagnosed secondary caries.
Quintessence Int 1998;29:313-317.
23. Neiva IF, de Andrada MA, Baratieri LN, Monteiro Junior S, Rit-
ter AV. An in vitro study of the effect of restorative technique
on marginal leakage in posterior composites. Oper Dent
1998;23:282-289.
24. Neter J, Wasserman W, Nachtstein CJ, Kutner MH. Applied
Linear Statistical Models. Homewood, IL: Richard D. Irving,
1996.
25. Opdam NJM, Roeters JJM, Burgersdijk RCW. Microleakage
of Class II box-type composite restorations. Am J Dent 1998;
11:160-164.
26. Pashley DH, Carvalho RM. Dentine permeability and dentin
adhesion. J Dent 1997;25:355-372.
27. Reid JS, Saunders WP, Sharkey SW, Williams CE. An in-vitro
investigation of microleakage and gap size of glass iono-
mer/composite resin “sandwich” restorations in primary
teeth. ASCD J Dent Child 1994;61:255-258.
28. Roulet J-F. Benefits and disadvantages of tooth-coloured al-
ternatives to amalgam. J Dent 1997;25:459-473.
29. Sjödin L, Uusitalo M, Van Dijken J. Resin modified glass ion-
omer cements. In vitro microleakage in direct Class V and
Class II sandwich restorations. Swed Dent J 1996;20:77-86.
30. Subrata G, Davidson CL. The effect of various surface treat-
ments on the shear strength between composite resin and
glass ionomer cement. J Dent 1989;17:28-32.
31. Tay FR, Gwinnett JA, Wei SHY. Relation between water content
in acetone/alcohol-based primer and interfacial ultrastruc-
ture. J Dent 1998;26:147-156.
32. Thonemann B, Federlin M, Schmalz G, Glunder W. Total bond-
ing vs selective bonding: Marginal adaptation of Class 2 com-
posite restorations. Oper Dent 1999;24:261-271.
33. Tolidis K, Nobecourt A, Randall RC. Effect of a resin-modified
glass ionomer liner on volumetric polymerization shrinkage
of various composite. Dent Mater 1998;14:417-423.
34. Van Dijken JW, Kieri C, Carlén M. Longevity of extensive Class
II open-sandwich restorations with a resin-modified glass-ion-
omer cement. J Dent Res 1999;78:1319-1325.
35. Welbury RR, Murray JJ. A clinical trial of the glass-ionomer
cement composite resin “sandwich” technique in Class II
cavities in permanent premolar and molar teeth. Quintes-
sence Int 1990;21:507-512.
36. Yap AUJ, Mok BYY, Pearson G. An in vitro microleakage study
of the “bonded-base” restorative technique. J Oral Rehabil
1997;24:230-236.
The Journal of Adhesive Dentistry