Q U I N T E S S E N C E I N T E R N AT I O N A L

Inhibition of dentin demineralization adjacent

to a glass-ionomer/composite sandwich
restoration
Daranee Tantbirojn, DDS, MS, PhD1/Richard P. Rusin, PhD2/
Hoa T. Bui, BS2/Sumita B. Mitra, PhD2

Objective: To evaluate, in vitro, dentin caries inhibition ability of a composite restoration

with glass-ionomer liners in an open-sandwich configuration. Method and Materials:
Rectangular dentin cavities (n = 5) were restored with a composite and glass-ionomer liner
in an open-sandwich configuration where the liner was applied up to the cavity margin.
Liners used were 3 resin-modified glass-ionomers (Vitrebond, 3M ESPE; Vitrebond Plus,
3M ESPE; Fuji Lining LC Paste Pak, GC) and a conventional glass-ionomer (Ketac Bond,
3M ESPE). The control group was a composite restoration (Filtek Z250, 3M ESPE) without
a liner. Specimens were immersed in lactic acid gel for 3 weeks to create a demineralized
lesion before being subjected to microradiographic analysis. The width of the area where
the demineralization was completely inhibited at the restoration interface was measured.
The total mineral loss (!Z) was determined at 0.25 and 1.0 mm from the cavity margin.
Results: An inhibition zone was observed at the interface of all open-sandwich restora-
tions but not in Filtek Z250. !Z at 0.25 mm of all the open-sandwich restorations was
significantly less than that of Filtek Z250 (analysis of variance, Scheffes S, P < .05). At
1.0 mm, only the open-sandwich restorations in Vitrebond and Vitrebond Plus groups had
significantly less !Z than Filtek Z250. Conclusion: Under an in vitro demineralization
challenge, glass-ionomer liners in an open-sandwich restoration exhibited pronounced
inhibition zones at the dentin margin and lowered the amount of mineral loss in the vicinity
of 0.25 mm from the restoration interface. (Quintessence Int 2009;40:287294)

Key words: demineralization, dentin, glass-ionomer liner, open-sandwich restoration

Secondary caries has been reported to be anticariogenic.5 However, clinical studies

the main reason for restoration failure and
replacement.14 Although risk factors for pri-
mary and secondary caries are well-known in
the dental community, patient compliance
and host factors are not always controllable.
Therefore, restorative materials with anticari-
ogenic properties that prevent secondary
caries would be advantageous.
Because of the release of fluoride and
other ions, glass-ionomers are considered
Assistant Professor, Department of Restorative Sciences,
1
University of Minnesota, Minneapolis, Minnesota, USA.
3M ESPE, St Paul, Minnesota, USA.
2
assessing this property have been inconclu-
sive. A systematic review of 28 studies from
1970 to 1996 showed no clear evidence for
or against inhibition of secondary caries by
glass-ionomer restorations.6 About half of
the glass-ionomer restorations in deciduous
teeth replaced in 1996 and 2000/2001 were
due to secondary caries.7 In addition, sec-
ondary caries was the reason for more than
half of failed glass-ionomer restorations
placed by general practitioners in 1993 to
1994.8 However, other studies reported that
glass-ionomer, both conventional and resin-
modified, provided protection against sec-

ondary caries under clinical conditions,

Correspondence: Dr Daranee Tantbirojn, 16-212 Moos Tower,
515 Delaware Street SE, Minneapolis, MN 55455. Fax: (612) 626-
especially in patients with high caries risk or
1484. Email: tant0002@umn.edu low compliance.912

2009 Quintessence Publishing Co, Inc.

All Rights Reserved
VOLUME 40 NUMBER 4 APRIL 2009 287
Q U I N T E S S E N C E I N T E R N AT I O N A L
Ta n t b i r o j n e t a l

Table 1 Study materials and application methods

Description Manufacturer Lot Application technique

Filtek Z250 Restorative Composite resin 3M ESPE 5JK Compressed into the cavity, light cured 40 s.
Scotchbond Etchant Phosphoric 3M ESPE 4CW Etched dentin/liner 15 s, rinsed with water, blot dried.
acid gel
Adper Single Bond Plus Dentin adhesive 3M ESPE 4AG Applied 2 coats on etched surface, agitated 15 s,
Adhesive gently air dried 5 s, and cured 10 s.
Ketac Bond Conditioner Polyacrylic acid 3M ESPE 14 Applied on dentin 10 s, rinsed with water, blot dried.
Ketac Bond Glass Conventional glass- 3M ESPE Powder: Weighed powder and liquid at ratio 4.2:1,
Ionomer Base Material ionomer liner 158562 mixed 1530 s, placed on conditioned dentin.
Liquid: Setting time 4 min.
209527

GC Fuji Lining LC Paste RMGI liner GC 505101 Clicker dispensed, mixed 1015 s, placed on dentin,
Pak Light-Cured Glass light cured 20 s.
Ionomer Lining Material
Vitrebond Plus Light RMGI liner 3M ESPE Lot 113 Clicker dispensed, mixed 1015 s, placed on dentin,
Cure Glass Ionomer light cured 20 s.
Liner Base
Vitrebond Light Cure RMGI liner 3M ESPE Powder: 3EW Weighed powder and liquid at ratio 1.4:1,
Glass Ionomer Liner Liquid: 3CU mixed 1015 s, placed on dentin, light cured 30 s.
Base

Glass-ionomer cement has been advocat-
ed as a base or liner under composite
restorations.13,14 The glass-ionomer/compos-
ite sandwich configuration is viewed as an
example of the biomimetic principle,15 in that
it emulates the structure and properties of
the natural tooth. The glass-ionomer, with the
benefits of fluoride release, adhesion, and
similar coefficient of thermal expansion as
tooth structures, serves as dentin replace-
ment, while the more durable and esthetic
resin-based composite serves as enamel
replacement. The sandwich configuration
with the glass-ionomer layer extending to the
external surface (open-sandwich) has been
recommended when the cavosurface mar-
gin is in dentin.13,14 Clinical studies have
reported improved seals at the dentin mar-
gin and fewer caries lesions developing adja-
cent to glass-ionomer/composite sandwich
restorations.1618
Early conventional glass-ionomer cements
had limited physical properties that con-
tributed to restoration failure when the
glass-ionomer component of the sandwich
restoration was exposed to the oral environ-
ment.19 At present, resin-modified glass-
ionomer (RMGI) liners have been improved
so that it can be used for nonstress-bearing
restorations in deciduous teeth.20 RMGIs
were developed to provide the advantages of
preferred handling and setting characteristics
while maintaining the sustained fluoride
release of conventional glass-ionomer
cements.21
The cariostatic effect of glass-ionomers is
attributed mostly to their ability to inhibit
demineralization and enhance remineraliza-
tion through the release of fluoride ions to
surrounding fluid phases and adjacent tooth
structures.21,22 Indeed, glass-ionomer/com-
posite sandwich restorations with substantial
thickness of restorative-type glass-ionomer
exhibited some caries protection clinically.17,18
In contrast, a thin layer of glass-ionomer liner
exposed to the external environment in a
glass-ionomer/composite sandwich configu-
ration might provide a lesser amount of
fluoride, raising the question whether it will
have an effect on secondary caries inhibi-
tion.
The purpose of this in vitro study was to
evaluate the ability of various glass-ionomer
liners to inhibit demineralization of adjacent

2009 Quintessence Publishing Co, Inc.

All Rights Reserved
288 VOLUME 40 NUMBER 4 APRIL 2009
 Q U I N T E S S E N C E I N T E R N AT I O N A L
Ta n t b i r o j n e t a l

Lesion

Z250

Inhibition Vitrebond
zone Vitrebond Plus

a b c
Glass-ionomer liner

Fuji Ketac Bond Z250

Paste Pak without liner

d e f

Figs 1a to 1f (a) Diagram of the open-sandwich restoration and adjacent dentin demineralized lesion with the inhibition zone.
(b to f) Microradiographs of the restoration and adjacent dentin with demineralized lesion. Composite restoration without liner
(f) did not exhibit the inhibition zone, while the inhibition zone was observed at the dentin interface in all open-sandwich restora-
tions (b to e).

dentin when used under a resin-based com-
posite in an open-sandwich restoration. The
effect was expected to be visible as an inhi-
bition zone where demineralization had not
taken place at the dentin interface and as a
reduction in the amount of mineral loss in
adjacent dentin. The null hypothesis was
that there was no difference in the width of
the inhibition zone or in the amount of miner-
al loss between the composite restoration
and glass-ionomer/composite sandwich
restorations.
METHOD AND MATERIALS
Rectangular cavities (6.0 " 2.0 " 1.5 mm3)
were prepared in 25 dentin specimens cut
from 5 bovine roots using a carbide bur in a
high-speed handpiece with a copious
amount of water (n = 5). The cavities were
restored with a composite (Filtek Z250
Restorative, 3M ESPE) and glass-ionomer
liners in an open-sandwich configuration in
which a thin layer of glass-ionomer liner was
extended up to the external surface along
the cavity margin (Fig 1a). The glass-ionomer
liner materials used were 3 RMGIs
(Vitrebond Light Cure Glass Ionomer Liner
Base, 3M ESPE; Vitrebond Plus Light Cure
Glass Ionomer Liner Base, 3M ESPE; GC
Fuji Lining LC Paste Pak Light-Cured Glass
Ionomer Lining Material, GC) and a conven-
tional glass ionomer (Ketac Bond Glass
Ionomer Base Material, 3M ESPE). The con-
trol group was a composite restoration with-
out glass-ionomer liner (Filtek Z250
Restorative).
A total-etch bonding technique (Scotch-
bond Etchant and Adper Single Bond Plus
Adhesive, 3M ESPE) was performed before
composite application. Table 1 shows the
details of materials used and application
methods. After being restored, the speci-
mens were kept moist at 37C for at least 1
hour and polished with a series of Sof-Lex
disks (3M ESPE). The surrounding dentin
was coated with acid-resistant nail polish
except for a window area (about 6 " 3 mm2)
adjacent to the restoration. The specimens
were separately immersed in 20 mL acid gel
at 37C for 3 weeks to create a demineral-
ized lesion. The acid gel consisted of 6% by

2009 Quintessence Publishing Co, Inc.

All Rights Reserved
VOLUME 40 NUMBER 4 APRIL 2009 289
Q U I N T E S S E N C E I N T E R N AT I O N A L
Ta n t b i r o j n e t a l

Width of inhibition zone (m)

Fig 2 Widths of the inhibition zone (m) at the dentin interface were
significantly different among the 5 groups (P = .0001). Same letters represent
mean values that were not significantly different (ANOVA and Scheffes S;
significance level, .05).

weight hydroxyethylcellulose in a 0.1 mol/L
lactic acid solution adjusted to pH 5.1 with
1.0 mol/L sodium hydroxide.23
After 3 weeks, the specimens were
removed from the acid gel, washed with
deionized water, and embedded in acrylic
resin. The specimens were cut perpendicular
to the tooth-restoration interface into thin sec-
tions using a diamond blade (Isomet Low
Speed Saw, Buehler) with water coolant.
Each section was ground by hand on wet
600-grit silicon carbide paper (Buehler) to
achieve a thickness of approximately 0.4 mm
for microradiography.
The microradiograph was obtained with a
Picker Hotshot x-ray machine (Picker
Industrial Products) operating at 12 kV and 1
mA for an exposure time of 54 seconds,
using Ultra-speed Dental Films (Kodak DF-
58, Eastman Kodak).24 The image of the
microradiograph was captured at 100"
magnification under a stereomicroscope
(Nikon SMZ-2T, Nikon) with a charge-cou-
pled device camera (Spot Insight QE,
Diagnostic Instruments). The light intensity
was standardized by adjusting the light
source so that an unexposed x-ray film
reached a particular luminance value, which
allowed a constant light incident.
Microradiographic analysis was per-
formed on 2 sections from each specimen,
using an image analysis software Image-Pro
Plus 4.5 (Media Cybernetics). Parameters
assessed were the width of the inhibition
zone and the total mineral loss (!Z). The
width of the inhibition zone is the width (in
m) of the area where the demineralization
was completely inhibited at the restoration
interface. The severity of demineralization,
represented as !Z, was computed from the
demineralized area at 0.25 and 1.0 mm from
the cavity margin by the following proce-
dures. First, the grayscale profile across the
area of interest (250 m width " 500 m
length from lesion surface to underlying
sound dentin) was obtained. The gray value
(radiopacity) along the profile was converted
into mineral content (volume % mineral) by
calibrating with the gray values from an
aluminum step wedge.25,26 Then, a mineral
profile was constructed from the mineral
content as a function of distance from the
lesion surface to the underlying sound
dentin. Finally, !Z (volume % mineralm)
was integrated from the area between the
mineral profile of the lesion and the average
volume % mineral that had been extrapolat-
ed from the underlying sound dentin, which
was defined to contain 45 volume % mineral
as an internal standard.24
The width of the inhibition zone and !Z at
0.25 mm and !Z at 1.0 mm from the cavity
margin were compared using 1-way analysis
of variance (ANOVA) followed by Scheffes S
test (SuperANOVA, Abacus Concepts) (level
of significance, .05).

2009 Quintessence Publishing Co, Inc.

All Rights Reserved
290 VOLUME 40 NUMBER 4 APRIL 2009

Table 2 Mean (SD) Z (volume % mineralm) at 0.25 and 1.0 mm from the restoration-tooth
interface*
Composite Ketac Bond/
without liner Composite
Z 0.25 mm 6,696 (881)a 5,099 (1,031)b
Z 1.00 mm 6,107 (1130)a 5,652 (772)a,b
*The values were significantly different among the 5 groups at 0.25 and 1.0 mm from the interface (P = .0001 and .0006, respectively). Same letters
represent mean values that were not statistically different at the same distance (ANOVA and Scheffes S; significance level, .05).
RESULTS
An inhibition zone was observed next to all
open-sandwich restorations (Figs 1b to 1e).
No inhibition zone was found next to the
composite restoration without glass-ionomer
liner (Fig 1f). Figure 2 shows the mean width
of inhibition zones in each group. There
was a significant difference in the width of
inhibition zone among the 5 groups
(ANOVA; P = .0001). The inhibition zone
adjacent to Vitrebond was significantly wider
than the zones adjacent to Fuji Paste Pak
and Ketac Bond, but not significantly differ-
ent from the width of the inhibition zone adja-
cent to Vitrebond Plus (Scheffes S test; sig-
nificance level, .05).
Means of !Z, standard deviations, and
statistical results are shown in Table 2.
There was a significant difference in !Z at
0.25 mm from the cavity margin among the
5 groups (ANOVA; P = .0001). The open-
sandwich restorations exhibited significantly
lower !Z at 0.25 mm than the composite
restorations without glass-ionomer liner
(Scheffes S test; significance level, .05). The
restorations with Vitrebond had significantly
lower !Z than those with Fuji Paste Pak or
Ketac Bond. !Zs at 0.25 mm in the groups
restored with Vitrebond and Vitrebond Plus
were not significantly different (Scheffes S
test; significance level, .05). There was a sig-
nificant difference in !Z at 1.0 mm from the
cavity margin among the 5 groups (ANOVA;
P = .0006). !Zs at 1.0 mm in the groups
restored with Vitrebond and Vitrebond Plus
were significantly different from the compos-
ite restorations without a glass-ionomer liner
(Scheffes S test; significance level, .05).
2009 Quintessence Publishing Co, Inc.
All Rights Reserved
VOLUME 40 NUMBER 4 APRIL 2009
Q U I N T E S S E N C E I N T E R N AT I O N A L
Ta n t b i r o j n e t a l
Fuji Paste Pak/ Vitrebond Plus/ Vitrebond/
Composite Composite Composite P
4,551 (534)b 4,237 (473)b,c 3,236 (694)c .0001
5,106 (484)a,b 4,931 (574)b 4,576 (740)b .0006
DISCUSSION
Under laboratory conditions, it has been
reported that glass-ionomers were able to
protect surrounding tooth structures from a
demineralization challenge.22,23,27 However,
in situ and clinical studies have been incon-
clusive,6 reporting positive,912,17,18,28,29 nega-
tive,7,8 as well as neutral results.30 It is
possible that inferior physical properties of
glass-ionomers, especially the earlier prod-
ucts, played an important role in the success
or failure of the restorations. More important-
ly, glass-ionomer restorations were shown to
be more effective in patients with high caries
risk or low compliance.912
In two of the referenced clinical studies
with positive effects, sandwich restorations
with substantial restorative glass-ionomer
thickness exhibited some caries protec-
tion.17,18 This in vitro study examined whether
a thin layer of glass-ionomer liner placed
under a resin-based composite in an open-
sandwich restoration was able to inhibit
demineralization of adjacent dentin. The null
hypothesis was that there was no difference
in the width of the inhibition zone and in the
amount of mineral loss between the com-
posite restoration and glass-ionomer/com-
posite sandwich restorations.
The results of this study rejected the null
hypothesis and showed that a glass-
ionomer/composite sandwich configuration
with the glass-ionomer layer extending to the
external surface affected the development of
a demineralized lesion in the adjacent dentin.
A thin area where the demineralization was
completely inhibited was observed at the cav-
ity margin in all glass-ionomer/composite
291
Q U I N T E S S E N C E I N T E R N AT I O N A L
Ta n t b i r o j n e t a l
sandwich restorations in contrast to the con-
trol group without the glass-ionomer liner (see
Figs 1 and 2). The in vitro results support clin-
ical evidence that found fewer caries lesions
developed adjacent to glass-ionomer/com-
posite sandwich restorations.17,18
The cariostatic effect of glass-ionomers
can be attributed to the ability of the cement
to inhibit demineralization and enhance rem-
ineralization through the release of fluoride
to adjacent tissue and surrounding fluid.2024
RMGI liner/base Vitrebond was shown to
have fluoride release comparable to conven-
tional glass-ionomer cements.21 Among the
glass-ionomer liners used in this study, Ketac
Bond is a conventional glass-ionomer
cement, whereas Vitrebond, Vitrebond Plus,
and Fuji Paste Pak are RMGIs. Fluoride
released from Vitrebond, Vitrebond Plus,
and Fuji Paste Pak is higher than from Ketac
Bond,21,31 hence Ketac Bond had the small-
est inhibition zone and the highest !Z value
among the glass-ionomer liners tested. The
amount of fluoride released from a glass-
ionomer cement in an acidic environment
was higher than in a neutral environment,32
which may have been another contributing
factor to the result of this study.
The thickness of a glass-ionomer liner
exposed to the surrounding fluid phase at
the tooth-restoration interface may affect the
ion-exchange mechanism, and thus the car-
iostatic effect of the liner. We measured the
mean thickness of the liner layers at the cav-
ity margin to be 117, 168, 215, and 99 m for
Ketac Bond, Fuji Paste Pak, Vitrebond Plus,
and Vitrebond, respectively. Linear regres-
sion was performed between the thickness
of the liner layer at the cavity margin of each
specimen and the width of the inhibition
zone or the amount of mineral loss at 0.25 or
1.0 mm, using the Regression Data Analysis
Tool in Microsoft Excel 97 (Microsoft). There
was no correlation between the thickness of
liner layer and the width of the inhibition
zone (r = 0.16; P = .33) nor the amount of
mineral loss at 0.25 and 1.0 mm (r = 0.005
and 0.05; P = .98 and .75, respectively).
Consequently, the cariostatic effect depend-
ed on the presence of the glass-ionomer
liner rather than the thickness of the
exposed liner.
2009 Quintessence Publishing Co, Inc.
All Rights Reserved
292
In the present study, periapical
radiograph films were used for microradiog-
raphy.24 The microradiograph was done with
low energy x-ray (12 kV) to permit high
absorption by the specimen; thus, regions
with slight differences in mineral content
would be distinguishable.26 Mineral content
was calculated by averaging the grayscale
value across an area of 250 m to compen-
sate for the relatively coarse grain of the peri-
apical films. Hence, the mineral profile at
0.25 mm represented the measurement
from 0.125 to 0.375 mm, and the mineral
profile at 1.0 mm represented the measure-
ment from 0.875 to 1.125 mm from the cavi-
ty margin. The average mineral content of
sound dentin computed by this method was
45.5 volume % mineral,33 which is compara-
ble with the established value of 47.0 volume
% mineral.34
Considering the high prevalence of
restoration failure and replacement due to
secondary caries, restorative materials with
anticariogenic potential are valuable for the
practice of dentistry.2 Apart from the effect of
fluoride on the dynamics of de/remineraliza-
tion, glass-ionomer cements have been
shown to inhibit the growth of cariogenic
bacteria in approximal dental plaque and
remaining carious dentin underneath the
material.35,36 Whether the incidence of sec-
ondary caries is reduced by glass-ionomers,
however, still remains to be determined. The
anticariogenicity of glass-ionomers is per-
haps the most effective in high-risk and less
compliant patients.912 An in situ study
showed less demineralization of enamel
adjacent to RMGI restorative materials than
a composite resin after a month without
fluoride dentifrice.29 Secondary caries reduc-
tions around cervical restorations for conven-
tional and RMGIs relative to composite were
greater than 80% in patients with xerostomia
with less compliance to fluoride supplemen-
tation.11 Similarly, less caries developed at
margins of glass-ionomer restorations than
at amalgam margins after 2 years in patients
with xerostomia who did not routinely use
topical fluoride.12 The present study has
demonstrated the potential of glass-ionomer
liner/composite open-sandwich restorations
to intervene with the development of dentin
VOLUME 40 NUMBER 4 APRIL 2009

caries, which conforms to the contemporary
concept of minimally invasive dentistry.5
CONCLUSION
Under the demineralization challenge in this
in vitro study, glass-ionomer/composite
open-sandwich restorations exhibited a pro-
tective effect on adjacent dentin, observed
as an inhibition zone at the restoration inter-
face and a lower mineral loss in the vicinity of
0.25 mm from the cavity margin, while a
composite restoration without glass-ionomer
liner did not.
ACKNOWLEDGMENTS
This study was supported in part by a
Nontenured Faculty Grant, 3M Foundation
and the Minnesota Dental Research Center
for Biomaterials and Biomechanics. The
authors would like to thank Dr A. Versluis for
his advice.
REFERENCES
1. Fontana M, Gonzalez-Cabezas C. Secondary caries
and restoration replacement: An unresolved prob-
lem. Compend Contin Educ Dent 2000;21:1530.
2. Hicks J, Garcia-Godoy F, Donly K, Flaitz C. Fluoride-
releasing restorative materials and secondary
caries. Dent Clin North Am 2002;46:247276.
3. Mjr IA, Moorhead JE, Dahl JE. Reasons for replace-
ment of restorations in permanent teeth in general
dental practice. Int Dent J 2000;50:361366.
4. Bernardo M, Luis H, Martin MD, et al. Survival and
reasons for failure of amalgam versus composite
posterior restorations placed in a randomized clini-
cal trial. J Am Dent Assoc 2007;138:775783.
5. Tyas MJ, Anusavice KJ, Frencken JE, Mount GJ.
Minimal intervention dentistryA review. FDI
Commission Project 197. Int Dent J 2000;50:112.
6. Randall RC,Wilson NH. Glass-ionomer restoratives: A
systematic review of a secondary caries treatment
effect. J Dent Res 1999;78:628637.
7. Mjr IA, Dahl JE, Moorhead JE. Placement and
replacement of restorations in primary teeth. Acta
Odontol Scand 2002;60:2528.
2009 Quintessence Publishing Co, Inc.
All Rights Reserved
VOLUME 40 NUMBER 4 APRIL 2009
Q U I N T E S S E N C E I N T E R N AT I O N A L
Ta n t b i r o j n e t a l
8. Mjr IA. Glass-ionomer cement restorations and
secondary caries: A preliminary report.
Quintessence Int 1996;27:171174.
9. Donly KJ, Segura A, Kanellis M, Erickson RL. Clinical
performance and caries inhibition of resin-modi-
fied glass ionomer cement and amalgam restora-
tions. J Am Dent Assoc 1999;130:14591466.
10. Tyas MJ. Cariostatic effect of glass ionomer cement:
A five-year clinical study. Aust Dent J 1991;
36:236239.
11. McComb D, Erickson RL, Maxymiw WG, Wood RE. A
clinical comparison of glass ionomer, resin-modi-
fied glass ionomer and resin composite restora-
tions in the treatment of cervical caries in xeros-
tomic head and neck radiation patients. Oper Dent
2002;27:430437.
12. Haveman CW, Summitt JB, Burgess JO, Carlson K.
Three restorative materials and topical fluoride gel
used in xerostomic patients: A clinical comparison. J
Am Dent Assoc 2003;134:177184.
13. McLean JW, Powis DR, Prosser HJ,Wilson AD.The use
of glass-ionomer cements in bonding composite
resins to dentine. Br Dent J 1985;158:410414.
14. Mount GJ. Clinical requirements for a successful
sandwichDentine to glass ionomer cement to
composite resin. Aust Dent J 1989;34:259265.
15. Croll TP. The sandwich technique. J Esthet Restor
Dent 2004;16:210212.
16. Andersson-Wenckert IE, van Dijken JW, Hrstedt P.
Modified Class II open sandwich restorations:
Evaluation of interfacial adaptation and influence
of different restorative techniques. Eur J Oral Sci
2002;110:270275.
17. Vilkinis V, Hrsted-Bindslev P, Baelum V. Two-year
evaluation of class II resin-modified glass ionomer
cement/composite open sandwich and composite
restorations. Clin Oral Investig 2000;4:133139.
18. van Dijken JW, Kieri C, Carln M. Longevity of exten-
sive class II open-sandwich restorations with a
resin-modified glass-ionomer cement. J Dent Res
1999;78:13191325.
19. 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;172:103107.
20. Croll TP, Nicholson JW. Glass ionomer cements in
pediatric dentistry: Review of the literature. Pediatr
Dent 2002;24:423429.
21. Mitra SB. In vitro fluoride release from a light-cured
glass-ionomer liner/base. J Dent Res 1991;70:7578.
22. Forss H, Sepp L. Prevention of enamel demineral-
ization adjacent to glass ionomer filling materials.
Scand J Dent Res 1990;98:173178.
23. Tantbirojn D, Douglas WH, Versluis A. Inhibitive
effect of a resin-modified glass ionomer cement on
remote enamel artificial caries. Caries Res
1997;31:275280.
293
Q U I N T E S S E N C E I N T E R N AT I O N A L
Ta n t b i r o j n e t a l
24. Tantbirojn D, Feigal RJ, Ko CC, Versluis A.
Remineralized dentin lesions induced by glass
ionomer demonstrate increased resistance to sub-
sequent acid challenge. Quintessence Int 2006;
37:273281.
25. Angmar B, Carlstrom D, Glas JE. Studies on the ultra-
structure of dental enamel. IV. The mineralization of
normal human enamel. J Ultrastruct Res 1963;
8:1223.
26. Thewlis J. The structure of teeth as shown by x-ray
examination. Medical Research Council Special
Report Series 1940;238:582.
27. Hicks MJ, Flaitz CM, Silverstone LM. Secondary
caries formation in vitro around glass ionomer
restorations. Quintessence Int 1986;17:527532.
28. ten Cate JM, van Duinen RN. Hypermineralization of
dentinal lesions adjacent to glass-ionomer cement
restorations. J Dent Res 1995;74:12661271.
29. Donly KJ, Segura A, Wefel JS, Hogan MM. Evaluating
the effects of fluoride-releasing dental materials on
adjacent interproximal caries. J Am Dent Assoc
1999;130:817825.
30. Kielbassa AM, Schulte-Monting J, Garcia-Godoy F,
Meyer-Lueckel H. Initial in situ secondary caries for-
mation: Effect of various fluoride-containing
restorative materials. Oper Dent 2003;28:765772.
2009 Quintessence Publishing Co, Inc.
All Rights Reserved
294
31. 3M ESPE Dental Products.Vitrebond Plus Light Cure
Liner/Base Technical Product Profile. St Paul: 3M,
2005.
32. Mller U, Kielbassa AM, Schulte-Mnting J, Hellwig
E. Fluoride release from light-curing restorative
materials. Am J Dent 2000;13:301304.
33. Tantbirojn D. Surface modulation of dental hard tis-
sues [thesis]. Minneapolis: University of Minnesota,
1998.
34. Nikiforuk G. Understanding Dental Caries. Vol 1:
Etiology and Mechanisms, Basic and Clinical
Aspects. Basel: Karger, 1985:92.
35. Berg JH, Farrell JE, Brown LR. Class II glass
ionomer/silver cermet restorations and their effect
on interproximal growth of mutans streptococci.
Pediatr Dent 1990;12:2023.
36. Kreulen CM, de Soet JJ, Weerheijm KL, van
Amerongen WE. In vivo cariostatic effect of resin
modified glass ionomer cement and amalgam on
dentine. Caries Res 1997;31:384389.
VOLUME 40 NUMBER 4 APRIL 2009