Journal of Investigative and Clinical Dentistry (2013), 4, 265–270

ORIGINAL ARTICLE
Conservative Dentistry

Two-year clinical study on postoperative pulpal

complications arising from the absence of a glass-ionomer
lining in deep occlusal resin-composite restorations
Danuchit Banomyong1 & Harold Messer2
1 Department of Operative Dentistry, Faculty of Dentistry, Mahidol University, Bangkok, Thailand
2 Melbourne Dental School, The University of Melbourne, Melbourne, Victoria, Australia

Keywords
clinical trial, deep occlusal cavities, glass-
ionomer cement, pulpal complication, resin
composite.
Correspondence
Dr D. Banomyong, Department of Operative
Dentistry, Faculty of Dentistry, Mahidol
University, 6 Yothee Road, Rajthevee,
Bangkok 10400, Thailand.
Tel: +66-2-203-6461
Fax: +66-2-203-6463
Email: dtdby@mahidol.ac.th
Received 14 November 2011; accepted
9 June 2012.
doi: 10.1111/j.2041-1626.2012.00160.x
Abstract
Aim: To observe the effects of glass-ionomer cement (GIC) lining on the risk
of pulpal complications in deep occlusal cavities with resin-based restorations.
Methods: Fifty-three patients, aged 18–30 years, who had one or two deep
occlusal carious lesions (  3 mm in depth) in molars, were recruited. Dental
caries were removed, and the prepared cavity was restored with resin composite
using one of two restorative procedures: (a) without GIC lining; and (b) with
(resin-modified) GIC lining. Restored teeth were evaluated for any pulpal com-
plications (subjective symptoms, objective signs or loss of tooth vitality) at
1 month (baseline), 1 year, and 2 years after restoration.
Results: After excluding shallow cavities, 31 restorations without GIC lining,
and 31 restorations with GIC lining, were placed and recalled at baseline with-
out any pulpal complications. At the 1- and 2-year recalls, six patients who
had restorations in group 1, and 13 in group 2, had dropped out. None of the
remaining teeth in the two groups exhibited pulpal complications at either
recall period, regardless of GIC lining placement.
Conclusions: The absence of GIC lining does not increase the risk of pulpal
complications in deep occlusal cavities restored with resin-based restorations in
either the short or long term.

barrier is not necessary for the resin-based restorations.8

Introduction
A liner/base material should generally be used when
restoring a deep cavity in order to protect the pulp.1 For
an amalgam restoration, placing a liner/base can reduce
the incidence of postoperative sensitivity, as the liner/base
might act as a thermal barrier beneath this un-bonded
metal restoration.2 Usually, the liner/base currently used
is glass-ionomer cement (GIC), due to its excellent bio-
compatibility and chemical adhesion to the tooth struc-
ture.3,4 Conventional GIC has a better biocompatibility
than resin-modified GIC,5 as resin monomers cause a
higher degree of cytotoxicity.6
In contrast, resin composite, a resin-based material, is
non-thermally conductive and has the ability to bond to
the tooth structure when used with a dental adhesive.7
Thus, it seems logical that the use of a liner as a thermal
However, it is still usually recommended that a liner/base
should be placed in a deep cavity in order to reduce
the risk of postoperative hypersensitivity,9 as it might
improve the seal of the restoration.10 Furthermore, the
pulpal inflammatory response to the restorative procedure
might be reduced by placing a glass-ionomer lining
cement, which prevents the diffusion of resin monomers
across dentinal tubules to the dental pulp.1 However, it
has been reported that the internal adaptation to normal
dentin at the cavity floor is negatively affected if a liner/
base is placed,11 or not improved when applied on caries-
affected dentin.12 Gap formations on the pulpal floor
beneath the liner might lead to the presence of clinical
postoperative sensitivity.11 In addition, placing a GIC
liner has no positive effect on the pulpal inflammatory
response if a self-etching adhesive is used.3

ª 2013 Wiley Publishing Asia Pty Ltd 265

Absence of lining and pulpal complication
Results from clinical trials are controversial. Postopera-
tive hypersensitivity and pulpal complications associated
with resin-based restorations might be unaffected,8,13
reduced,9 or even made worse14 by the placing of a liner/
base. Based on short-term results, the absence of a GIC
liner/base does not increase the risk of postoperative hyper-
sensitivity associated with occlusal resin-composite restora-
tions.8,13 It seems that the benefit of liner/base placement is
questionable. However, the long-term outcome should be
observed to confirm the short-term findings.
In several studies, the data include results from shallow
restorations, in which the chance of postoperative hyper-
sensitivity or pulpal complication is very low, and the use
of a liner/base is not commonly indicated. Thus, the pre-
vious results might not be totally applicable to the clinical
situation where a liner/base is placed only in a deeply-
restored cavity. Therefore, the purpose of this clinical
study was to observe the effect of GIC lining on the risk
of pulpal complication in deep occlusal cavities restored
with resin-based restorations after 2 years.
Materials and methods
This clinical study was conducted in the Postgraduate
Clinic of the Faculty of Dentistry, Mahidol University,
Bangkok, Thailand. The project was approved by the eth-
ics committee of Mahidol University, Thailand (ethics ID:
MU 2007-109).
Participant recruitment
Fifty-three patients, aged 18–30 years, who had at least
one deep occlusal carious lesion (more than 3 mm in
depth, but no pulpal exposure) in a first or second maxil-
lary/mandibular molar, were recruited. Each participant
was informed about the purpose of the study, and con-
sent was obtained. Participants were enrolled in the pres-
ent study if they met the inclusion criteria. Participants
with any medical problems or those with orofacial pain
were excluded. Investigated teeth were diagnosed as hav-
ing deep (primary) occlusal caries without other defects.
Each tooth had at least one opposing tooth, and the peri-
odontal tissues were healthy or mildly inflamed. Included
teeth did not exhibit any previous signs or symptoms of
pulpal and periapical diseases. However, preoperative
hypersensitivity might be present, but relieved immedi-
ately after stimulus removal, and no spontaneous pain
was observed. In addition, the teeth were excluded if
either the cavity depth after caries removal was less than
3 mm or a carious pulpal exposure was found.
The sample size was calculated based on the following
input conditions: power, 0.9; level of significance,
P = 0.05; and estimated standard deviation on a visual
266
D. Banomyong and H. Messer
analog scale, 14. The difference in clinical significance was
20 on the scale.8 The calculated minimum sample size
was 12 restorations per group. However, the sample size
at baseline was increased to compensate for the number
of participants who dropped out at recall periods.
Restorative procedures
Most restorations were made without rubber dam isola-
tion, but moisture contamination was carefully controlled
by placing cotton rolls, and using a saliva ejector in com-
bination with high-power suction. If requested, a local
anesthetic, 2% mepivacaine hydrochloride, with epineph-
rine 1:100 000 (Scandonest 2% special; Septodent, Saint-
Maur-des-Fosses Cedex, France) was administered. Dental
caries were removed using a conservative preparation, as
follows. To obtain entrance to the dentinal carious lesion,
the enamel caries was initially removed using a round or
fissure, high-speed diamond bur (Intensiv SA, Grancia,
Switzerland) under an air–water coolant. Next, the den-
tine caries was stained with a caries detector dye (Caries
Detector; Kuraray Medical, Okayama, Japan) for 10 s and
rinsed according to the manufacturer’s instructions. The
stained surface was removed using slow-speed, round
steel burs (Emil Lange, Engelskirchen, Germany) and
spoon excavators (Sci-Dent, Algonquin, IL, USA). The
staining procedure was repeated two or three times until
the dentine was minimally stained and relatively hard.
After complete caries removal, the greatest cavity depth
(in mm) was measured from the mesial or distal margin
using a periodontal probe (PCP-UNC 15; Hu-Friedy,
Chicago, IL, USA). In addition, the cavity length and
width were also measured in millimeters at the greatest
dimensions.
One of the two restorative procedures was randomly
allocated. Each participant was unaware of the restoration
placed; however, blinding the operator (DB) to which
intervention (lining or no lining) was allocated was not
possible. The restorative procedures were: (a) resin com-
posite (CO) without GIC lining. No GIC lining was
applied before bonding with an adhesive; and (b) GIC/
CO, the cavity was lined with a liquid-paste, resin-modi-
fied GIC liner (Fuji Lining LC Paste Pak; GC, Tokyo,
Japan) in 0.5–1 mm thickness over the entire dentin sur-
face, and then bonded with an adhesive. After excluding
the shallow cavities, 31 restorations were placed with no
GIC lining, while GIC lining was used for 31 restorations
in total. For bonding application, the cavity was bonded
with a two-step, total-etching adhesive (Single Bond 2;
3M ESPE, St Paul, MN, USA) or a two-step, self-etching
adhesive (Clearfil SE Bond; Kuraray Medical, Okayama,
Japan). All restorative materials were applied according to
the manufacturers’ instructions.
ª 2013 Wiley Publishing Asia Pty Ltd
 D. Banomyong and H. Messer
The bonded cavity was filled incrementally (<2 mm in
thickness) with a nano-filled resin composite (Filtek
Supreme XT; 3M ESPE, USA), and light cured for 40 s
using a LED light-curing unit (Bluephase; Ivoclar Viva-
dent AG, Schaan, Liechtenstein) in the high-power mode,
with a light intensity of 1100 mW/cm2. Next, any occlusal
interference was checked and corrected using high- and
slow-speed finishing diamond burs (Intensiv SA, Switzer-
land) under an air–water coolant. Finally, the surface of
the restoration was finished and polished with a series of
abrasive-impregnated silicone polishing points (Astropol;
Ivoclar Vivadent AG, Liechtenstein) under copious water
spray.
Evaluations at baseline and recalls
The participants were recalled, and the restorations were
examined by one evaluator (DB) at 1 month (baseline),
1 year, and 2 years after restoration. At these recalls, the
evaluator was blinded to the restorations that were exam-
ined. Each restoration was evaluated for any pulpal com-
plications or postoperative hypersensitivity following
these criteria: objective signs, subjective symptoms, and
no response to electrical pulp testing. For the objective
signs evaluation, teeth were carefully checked for any pos-
sible signs or symptoms of pulpal and periapical disease;
that is, pain to percussion, the presence of tooth mobility,
or gingival swelling (with or without sinus tract), and
then recorded as “presence” or “absence”. For the subjec-
tive symptoms evaluation, tooth pain or postoperative
hypersensitivity to any conditions (occlusal force, cold/
hot stimuli, or other) was identified by interviewing the
patient, and then recorded as “presence” or “absence”. If
present, the level of tooth pain or hypersensitivity was
recorded on a visual analog scale, from 0 (not at all) to
100 (worst), and the stimuli evoking sensitivity (if any)
were indicated. Finally, vitality test with an electric pulp
tester (Vitality Scanner; Sybron Endo, Orange, CA, USA)
was employed to confirm the vitality of each examined
tooth and recorded as a “response” or “no response”
result. Postoperative radiographic examination of the per-
iradicular area was not used due to the limitation of eth-
ics approval. As an alternative, the electric pulp test was
used to identify any tooth that would be diagnosed as
Table 1. Details of participants in the two experimental groups, and the number of restorations at baseline, and the 1- and 2-year recall periods
Mean  SD of
patients’ ages
Groups Patients (n) (years)
1. CO (no lining) 28 (22.5  4.9)
2. GIC/CO (lining) 25 (23.0  3.8)
CO, resin composite; GIC, glass-ionomer cement; SD, standard deviation.
ª 2013 Wiley Publishing Asia Pty Ltd
Absence of lining and pulpal complication
pulpal necrosis or asymptomatic apical periodontitis with-
out any signs or symptoms.
Results
The details of the participants and placed restorations are
shown in Tables 1 and 2. The means of the patients’ ages
and maximum cavity depths were similar between the
two experimental groups. At baseline, 31 resin-based
restorations were placed without GIC lining, and 31 res-
torations were placed with a GIC lining. At the 1- and 2-
year recalls, six patients who had restorations in group 1,
and 13 in group 2, were lost from the study, as the
patients did not attend the recalls. For the remaining res-
torations (25 restorations for the lining group, and 18
restorations for the non-lining group), none exhibited
any signs or symptoms of pulpal and periapical diseases
following the three evaluation criteria—subjective symp-
toms, objective signs, and no response to electrical pulp
testing—at the 1- and 2-year recall periods (Table 3).
Due to the high percentage of participants lost to follow
up at 2 years (approximately 30%), the intention-
to-treat analysis was performed (Table 4). Unlike the
results from the per-protocol analysis in Table 3, the inten-
tion-to-treat analysis was used to determine the success rate
calculated from all patients who were enrolled at baseline
(including patients who dropped out due to non-compli-
ance, those who did not follow the protocol, or those who
withdrew). In this analysis, all patients who dropped out
from the study were questioned via telephone by a dental
nurse, who questioned whether any signs or symptoms of
pulpal and periapical diseases were present or absent. None
of the patients reported such problems occurring with the
restored tooth at either the 1- or 2-year recall period.
Discussion
In the present study, the number of patients who dropped
out was relatively high. This might affect the clinical eval-
uations, as patients with any postoperative complications
might be more likely to withdraw from the clinical trial.
However, the patients in this study who did not attend
clinical follow up did not experience any signs or symp-
toms of pulpal and periapical diseases, as confirmed by
Sex Restorations
Restorations at 1 and
Males Females at baseline 2 years
13 15 31 25
9 16 31 18
267
Absence of lining and pulpal complication D. Banomyong and H. Messer

Table 2. Details of restored cavities, teeth, and adhesives in the two experimental groups at the 1- and 2-year recall periods

Tooth positions
Cavity sizes (mean  SD in mm) (molars) Adhesive systems used

Total Self-
Groups Depth Length Width Upper Lower etching etching

1. CO (no lining, n = 25) 3.1  0.3 3.9  1.2 2.8  1.1 13 12 14 11

2. GIC/CO (lining, n = 18) 3.2  0.5 4.2  1.1 3.1  1.3 9 9 7 11

CO, resin composite; GIC, glass-ionomer cement; SD, standard deviation.

Table 3. Subjective symptoms, objective signs, and no response to electrical pulp testing at baseline and the 1- and 2-year recall periods

Evaluated restorations showing any pulpal complications (n)

Subjective symptoms Objective signs No response to vitality test

Groups Baseline 1 year 2 years Baseline 1 year 2 years Baseline 1 year 2 years

1. CO (no lining, n = 25) 0 0 0 0 0 0 0 0 0

2. GIC/CO (lining, n = 18) 0 0 0 0 0 0 0 0 0

CO, resin composite; GIC, glass-ionomer cement; no response to vitality test, patient did not sense any stimuli with an electric pulp tester; objec-
tive signs, presence of pain to percussion, tooth mobility, or gingival swelling (with or without sinus tract); SD, standard deviation; subjective
symptoms, presence of tooth pain or hypersensitivity to any conditions (occlusal force, cold/hot stimuli, or other).

Table 4. Intention-to-treat analysis, including the telephone interview

results from patients who dropped out of the study
Patient(s) reporting
any signs or symptoms
of pulpal & periapical
diseases at 1- and 2-year
recall periods
Groups Present Absent
1. CO (no lining) 0 28
2. GIC/CO (lining) 0 25
All patients (28 patients with 31 restorations in the non-lining group,
and 25 patients with 31 restorations in the lining group) were inter-
viewed in person or by telephone at the 1- and 2-year recall periods,
as to whether any signs or symptoms of pulpal and periapical diseases
were present or absent.
CO, resin composite; GIC, glass-ionomer cement.
the telephone interview in every case. In fact, the reasons
obtained from the patients for withdrawal from this clini-
cal trial were not related to the restored tooth. Thus, the
results of this study are still valuable and are not strongly
affected by the high drop-out rate. The results from the
intention-to-treat analysis, which indicated that none of
the patients reported postoperative complications, did not
differ from those of the per-protocol analysis.
Cavity depth is a key factor that plays an important
role in determining postoperative complications; the dee-
per the cavity, the higher the chance of postoperative sen-
sitivity and pulpal complications.15 The size and
complexity of the cavity also influence the risk of postop-
erative sensitivity. Occluso-proximal restorations have a
higher incidence of postoperative sensitivity than occlusal
restorations.16 In this clinical study, the absence of a
resin-modified GIC liner in restoring a deep occlusal cav-
ity did not increase the risk of postoperative sensitivity or
pulpal complications. This result corresponds with those
from the other clinical studies,14,17 in which a lining/base
was reported to not reduce the incidence of postoperative
sensitivity.
If a resin-based adhesive is used correctly, dentin can
be sealed without significant gap formation,11 and there-
fore, does not become sensitive. Moreover, sclerotic den-
tin at the pulpal floor, which is a result of the natural
defense response to dental caries, makes dentin imperme-
able and non-sensitive.12 In addition, resin composite
should be placed incrementally to prevent void entrap-
ment18 and to ensure complete light polymerization.
When the restorative procedure is carefully done, postop-
erative sensitivity or pulpal complication is less likely to
occur.8,13
In contrast, one clinical study showed the advantage
of using a resin-modified GIC liner/base beneath a
resin-based restoration to reduce short-tem postopera-
tive sensitivity.9 In that study, a total-etch adhesive was
used, which involves a technique-sensitive application
procedure.19 Thus, it is possible that the higher chance
of postoperative sensitivity in the group without the
GIC lining might be due to minor errors in adhesive
procedure, which might lead to an imperfect seal.

268 ª 2013 Wiley Publishing Asia Pty Ltd

 D. Banomyong and H. Messer Absence of lining and pulpal complication

Adhesion to deep dentin of a total-etch adhesive is also
negatively affected by wetness from dentinal fluid, while
that of GIC is not.20 Thus, the adaptation of GIC to
deep dentin is likely to be more constant than that of
the adhesive, and as a result, a lower incidence of
short-term postoperative sensitivity might be obtained
in such situations.
In the present study, we recruited patients aged 18–
30 years, because age is also an important factor that
should be considered when a liner/base is placed. In
elderly patients, large amounts of impermeable sclerotic
dentin are commonly observed.21 Moreover, secondary
dentin, as well as tertiary dentin, gradually deposit at the
pulpal side over time. This leads to recession of the pul-
pal horns, and the thickness of remaining dentin after
caries removal is usually enough to protect the pulp and
reduce the incidence of pulpal exposure. Thus, the chance
of postoperative pulpal complications occurring among
elderly patients might be much lower than in a younger
group, and placing a liner/base might not be necessary in
elderly patients.
Although it seems that placing a GIC liner/base has no
benefit in reducing postoperative pulpal complications or
sensitivity,8,13 other advantages might still justify its use.
Carious dentin might be partially left, unintentionally or
intentionally, and placing a GIC liner/base might promote
remineralization of this remaining carious dentin;22 how-
ever, the effect of GIC on the remineralization of carious
dentin is still controversial.23 Furthermore, GIC is a fluo-
ride-releasing material and has an antibacterial effect, so
it is useful when used in combination with resin compos-
ite (as a liner/base or the open-sandwich technique) in
high caries-risk patients.24
In conclusion, the absence of GIC lining did not
increase the risk of postoperative hypersensitivity or pul-
pal complications in deep occlusal cavities restored with
resin-based restorations at the 2-year evaluation. Never-
theless, this medium-term result should be further investi-
gated and confirmed at a longer time period. In addition,
the results obtained from a well-controlled clinical trial
might not be totally applicable for use in general practice,
where technical errors in restorative procedures might
occur. In the present study, the restorative procedure was
done carefully so that none of the patients had postopera-
tive pulpal complications. In fact, errors in the restorative
procedure (especially adhesive application) of resin-
composite restoration can lead to complication after res-
torations. Furthermore, the results obtained from the
occlusal restorations in the present study might not
totally apply to the occluso-promixal restorations, in
which greater postoperative complications are expected.
Finally, a GIC liner/base can still be used without any
negative clinical effect.8,13 Thus, in situations in which the
bond to dentin with a dental adhesive is not reliable, a
less technique-sensitive GIC liner/base might be selectively
placed to achieve a more predictable restoration.

References
1 Costa CAD, Giro EMA, do Nascimen-
to ABL, Teixeira HM, Hebling J.
Short-term evaluation of the pulpo-
dentin complex response to a resin-
modified glass-ionomer cement and a
bonding agent applied in deep cavi-
ties. Dent Mater 2003; 19: 739–46.
2 Gordan VV, Mjor IA, Moorhead JE.
Amalgam restorations: postoperative
sensitivity as a function of liner treat-
ment and cavity depth. Oper Dent
1999; 24: 377–83.
3 Costa CAD, Teixeira HM, do Nasci-
mento ABL, Hebling J. Biocompatibil-
ity of resin-based dental materials
applied as liners in deep cavities pre-
pared in human teeth. J Biomed Mater
Res B Appl Biomater 2007; 81B: 175–84.
4 Coutinho E, Yoshida Y, Inoue S et al.
Gel phase formation at resin-modified
glass-ionomer/tooth interfaces. J Dent
Res 2007; 86: 656–61.
5 Sidhu SK, Schmalz G. The biocom-
patibility of glass-ionomer cement
materials. A status report for the
American Journal of Dentistry. Am J
Dent 2001; 14: 387–96.
6 Lan WH, Lan WC, Wang TM et al.
Cytotoxicity of conventional and
modified glass ionomer cements. Oper
Dent 2003; 28: 251–9.
7 Tyas MJ, Burrow MF. Adhesive
restorative materials: a review. Aust
Dent J 2004; 49: 112–21.
8 Burrow MF, Banomyong D, Harnirat-
tisai C, Messer HH. Effect of glass-io-
nomer cement lining on postoperative
sensitivity in occlusal cavities restored
with resin composite—a randomized
clinical trial. Oper Dent 2009; 34: 648
–55.
9 Akpata ES, Sadiq W. Post opera-
tive sensitivity in glass-ionomer
versus adhesive resin-lined poster-
ior composites. Am J Dent 2001;
14: 34–8.
10 Ratih DN, Palamara JEA, Messer HH.
Minimizing dentinal fluid flow associ-
ated with gap formation. J Dent Res
2006; 85: 1027–31.
11 Banomyong D, Palamara JEA, Mes-
ser HH, Burrow MF. Sealing ability
of occlusal resin composite restora-
tion using four restorative proce-
dures. Eur J Oral Sci 2008; 116: 571
–8.
12 Banomyong D, Palamara JEA, Messer
HH, Burrow MF. Fluid flow after
resin-composite restoration in
extracted carious teeth. Eur J Oral Sci
2009; 117: 334–42.
13 Banomyong D, Harnirattisai C, Bur-
row MF. Posterior resin composite
restorations with or without resin-
modified, glass-ionomer cement lin-
ing: a 1-year randomized, clinical
trial. J Investig Clin Dent 2011; 2:
63–9.
14 Unemori M, Matsuya Y, Akashi A,
Goto Y, Akamine A. Self-etching

ª 2013 Wiley Publishing Asia Pty Ltd 269

Absence of lining and pulpal complication
adhesives and postoperative sensitiv-
ity. Am J Dent 2004; 17: 191–5.
15 Auschill TM, Koch CA, Wolkewitz M,
Hellwig E, Arweiler NB. Occurrence
and Causing stimuli of postoperative
sensitivity in composite restorations.
Oper Dent 2009; 34: 3–10.
16 Briso ALF, Mestrener SR, Delicio G
et al. Clinical assessment of postoper-
ative sensitivity in posterior compos-
ite restorations. Oper Dent 2007; 32:
421–6.
17 Unemori M, Matsuya Y, Akashi A,
Goto Y, Akamine A. Composite resin
restoration and postoperative sensitiv-
ity: clinical follow-up in an undergrad-
uate program. J Dent 2001; 29: 7–13.
18 Miguez PA, Pereira PNR, Foxton
RM, Walter R, Nunes MF, Swift EJ.
270
Effects of flowable resin on bond
strength and gap formation in class
I restorations. Dent Mater 2004; 20:
839–45.
19 Ferrari M, Tay FR. Technique sensi-
tivity in bonding to vital, acid-
etched dentin. Oper Dent 2003; 28:
3–8.
20 Banomyong D, Palamara JEA, Bur-
row MF, Messer HH. Effect of den-
tin conditioning on dentin
permeability and micro-shear bond
strength. Eur J Oral Sci 2007; 115:
502–9.
21 Tagami J, Hosoda H, Burrow MF,
Nakajima M. Effect of aging and
caries on dentin permeability. Proc
Finn Dent Soc 1992; 88(Suppl 1):
149–54.
D. Banomyong and H. Messer
22 Duque C, Negrini TD, Sacono NT,
Spolidorio DMP, Costa CAD, Hebling
J. Clinical and microbiological perfor-
mance of resin-modified glass-iono-
mer liners after incomplete dentine
caries removal. Clin Oral Invest 2009;
13: 465–71.
23 Kim YK, Yiu CKY, Kim JR et al. Fail-
ure of a Glass ionomer to remineral-
ize apatite-depleted dentin. J Dent Res
2010; 89: 230–5.
24 van Dijken JWV, Kieri C, Carlen M.
Longevity of extensive class II open-
sandwich restorations with a resin-
modified glass-ionomer cement.
J Dent Res 1999; 78: 1319–25.
ª 2013 Wiley Publishing Asia Pty Ltd
Copyright of Journal of Investigative & Clinical Dentistry is the property of Wiley-Blackwell
and its content may not be copied or emailed to multiple sites or posted to a listserv without
the copyright holder's express written permission. However, users may print, download, or
email articles for individual use.