An in vitro study was performed to evaluate the effect of
two different proximal restoration techniques with different matrix systems on
the marginal seal and microhardness of Class II composite restorations. Results
indicated that the lowest, however, not significantly different, microleakage
was achieved in totally bonded deep Class II restorations prepared with margins
surrounded by enamel when using transparent matrices and reflective wedges in
combination with the centripetal buildup technique. Highest surface hardness of
composite resin was related to transparent matrices and reflecting wedges.
Marginal leakage appears to be an inherent shortcoming of
all dental restorations.1-5 Various techniques have been advocated
to enhance the marginal adaptation and reduce the microleakage of composite
restorations. Multilayer techniques, in contrary to bulk packing methods, have
decreased marginal gap formations.6,7 The size reduction of the
composite material, the diminution of polymerization shrinkage, and the
enlargement of the free surface area in relation to the volume are of great importance
in this context.8 The three-sided light-curing technique enables
initial progress in this direction; it is questionable if other techniques can
maintain better results.9 While promising results have been achieved
with centripetal buildups, comparative microleakage tests were not conducted.10
An important benefit of this procedure is that a thin proximal layer placed
towards the matrix band is cured before adjacent composite increments are
applied into the cavity. This can reduce the V/A ratio, where V is the cavity
volume and A is the area of the cavity walls. When the whole margin area is
first filled with an increment, fewer contraction gaps at the margins can be
expected using the centripetal technique versus the incremental technique. Even
if such a gap does develop, the next increment is likely to fill this gap.
In addition to methods, various materials (eg, light-curing
tips, matrix systems) have also been investigated.11,12 Different
experimental designs for microhardness and marginal analysis can be found in
current literature.13-15 To investigate marginal leakage of dental
restorations, isotope or dye penetration has been used.16,17
Following penetration, one mesiodistal section of the restoration was without a
description of the penetration patterns at the buccal or lingual sides of the
restoration.18,19 Although several investigations of microleakage
have been conducted, most of the
investigations involved Class II cavities in extracted teeth. These specimens
were not, however, mounted in contact with adjacent teeth to ensure their
movement during separation techniques.17,18
Microhardness analyses in specimens that were irradiated
from all sides to obtain polymerization with a high conversion rate in light-curing
composite are found in the literature. In most instances, cavities are
simulated with standardized metal blocks to obtain flat surfaces for microhardness
analysis.13,14,20 Under clinical conditions, however, flat
restoration surfaces can only be obtained in exceptional cases. The results,
therefore, may not have reflected the leakage and microhardness pattern of
Class II cavities, which are larger and more complex under optimized in vivo
simulating conditions. Some clinical studies show that a number of hindrances
remain in the attempt to develop an optimal method for placing composite
restorations that will remain intact in the oral cavity over extended function.4,21,22
Provided the preparation margins are surrounded by enamel, most multilayer
placement techniques can achieve adequate results in marginal adaptation of composite
materials.16,18,23 Deeper cavities with gingival floors that end in
dentin constitute a challenge for the adhesive mechanism.24,25
One purpose of this investigation was to create an
experimental design that allows (under simulated in vivo conditions) the
examination of different restorative techniques (incremental versus centripetal
technique) for the approximal box of Class II cavities. This study also
examined the effect of an opaque matrix system versus a transparent
matrix/wedge on the marginal seal of composite restorations. In addition, the
Vickers hardness of the approximal surfaces of composite restorations was
measured under simulated clinical conditions, but in a different experimental
Materials and Methods
One hundred sixty unrestored, extracted molar teeth free of
caries and fracture lines were randomly divided in two experimental designs: 96
specimen were used for the evaluation of the marginal seal, and 64 teeth were
employed for the evaluation of surface microhardness of composite restorations.
Once the specimen were cleaned with pumice and water, they were mounted in
silicon models in proximal contact with other specimens, which ensured the
movement of teeth during separation techniques that simulated clinical
conditions. Teeth were divided into two groups according to the following
specifications: in the first group, the gingival floor was surrounded by
enamel, in the second group by dentin. Each tooth was prepared with two Class
II cavities using parallel buccal and lingual walls on the occlusal aspect and
in the proximal boxes.
The preparations were cut with a diamond bur in a high-speed
handpiece with water coolant. The finishing procedure was performed in a
similar fashion. After six preparations, a new bur was used. To ensure as much
uniformity among the preparations as possible, a periodontal probe was utilized
as a guide: the depth of the gingival floors of the preparations was estimated
from 4.5 mm (Group 1) to 7.5 mm (Group 2); the pulpal floor extended 3 mm into
dentin; all proximal extensions were estimated at 6 mm for both groups. No
bevels were prepared. After the preparations were completed, the teeth were
randomly assigned to eight groups of 24 teeth each. The teeth in all groups
were restored with a hybrid composite resin using the total-etch technique as
recommended by the manufacturer. The combination of composite insertion
technique and the matrix system varied with each group. Each layer was
subjected to a 40-second exposure to the curing unit. Prior to each use of this
curing unit, a curing radiometer was employed to measure light output in the
400 nm to 500 nm wavelength range. The measured light intensity varied from 800
mW / cm2 to 1000 mW / cm2.
In the groups EIM, EIT, DIM, and DIT, the composite resin
was placed with an incremental technique: the first layer of composite resin
was placed on the gingival floor, the second and third layers were placed
diagonally, and the last increment was used to complete the filling in the
occlusal portion of the cavity. The incremental technique in
cavities with a depth of 4.5 mm was performed in the following sequence: 1.5 mm
+ 2 mm + 1 mm. For cavities with a depth of 7.5 mm, the following sequence was
used: 1.5 mm + 2 mm + 2 mm + 2 mm.
In the groups ECM, ECT, DCM, and DCT, the hybrid resin was
placed in a centripetal technique: a first layer of resin (1 mm thick) was placed
towards the matrix band, and the subsequent increments (2 mm thick) were
applied horizontally towards the occlusal area of the cavity (Figure 1). Since the same number of
increments was used for buildup (depending on the depth of the cavities), this
investigation evaluated the layering technique used.
Opaque Matrix System
In the groups EIM, ECM, DIM, and DCM, a 0.05-mm Tofflemire
metal matrix band and a No. 15 in a retainer with wooden wedges were applied to
the specimen. Each increment of composite for these specimens was cured only
from the occlusal side with visible light for 40 seconds. Following removal of
the matrix system, the restorations were cured for 40 seconds from the buccal
and occlusal aspects for both techniques.
In the groups EIT, ECT, DIT, and DCT, a precontoured 0.05-mm
transparent matrix band in a retainer was applied with reflective wedges. Each
increment of composite was cured with visible light for 40 seconds. With the
incremental technique, the first layer was cured indirectly through the light
wedge; the second and third layers were polymerized from the buccal and oral
direction in order to ensure that the shrinkage vectors were directed toward
the cavity margins. The last increment was polymerized from the occlusal
aspect. With the centripetal technique, the first layer was polymerized from
occlusal direction, the second layer through the light wedge, the third and
fourth layers from the buccal and occlusal direction as previously described,
and last layer from the occlusal aspect. Following the removal of the matrix
system, no postcuring was performed for the definitive restoration.
All restorations were finished immediately after placement
with fluted carbide burs, soft polishing disks, and silicon polishing under
water coolant. Plastic finishing strips were used for the finishing of the
interproximal surface. Postcuring was not performed.
Evaluation of Marginal
The specimens were removed from their mounting and
thermocycled for 5000 cycles (5¡C to 55¡C) with a dwell time of one minute at
each temperature. After thermocycling, two coats of colored fingernail varnish
were applied to all specimens, excluding the restoration margins of 1.5 mm.
Microleakage was assessed with 2% methylene blue diffusion
for 24 hours at 37¡C. The teeth were then embedded in a self-curing,
transparent epoxy resin, longitudinally sectioned with a diamond saw in a
buccolingual direction at the approximal box of the restoration, and dye
penetration was evaluated by light microscopy at (32 magnitude. The gingival
margins of the embedded specimens in transparent resin were marked from the
mesiodistal view for the first section. From here, two sections of 500 µm were obtained with a mean loss of 280
µm per section. Microleakage was
calculated in percent of the total length of the gingival margins of the
Evaluation of Surface
Sixteen teeth of each group were embedded in resin and
sectioned on the approximal sides to receive a parallel plane area. This flat
surface was necessary to facilitate Vickers hardness tests. Afterwards,
different cavities with already described sizes were prepared into the
specimens. A strip of the matrix band used (transparent or metal) was glued to
the resin blocks without forming marginal gaps between the prepared cavities
and the matrix band. Restorations were applied, as previously described, and
all curing procedures were performed from the occlusal surface of the cavity.
After curing, the matrix band was removed, no postcuring technique was used,
and the Vickers hardness of the planed approximal composite surfaces was
measured with microhardness meter at a load of 0.3 kg for 30 seconds. The
Vickers hardness was measured on the approximal composite surface 24 hours
following resin placement. Six measurements were made at predetermined,
regularly distributed sites in the cervical area of the approximal surface. The
values were evaluated by light microscopy at (200 magnitude. Hardness was
calculated by dividing the applied load by the surface area of the indentation.
The mean values and standard deviations were calculated for
each group. Data were analyzed with nonparametric statistics; the
Kruskal-Wallis Multiple-Comparison Z-Value Test - including the Bonferroni
Correction - were used with alpha = 0.05
for statistical analysis of the results.
(Continued from page 1 )
Evaluation of Marginal
The results of the dye penetration of the various groups of
composite restorations were recorded (Table 1). Marginal microleakage or its
absence was observed in each group evaluated in this study. Each restoration
demonstrated marginal microleakage on the gingival wall. Highest values were
found in the groups DCM (92.42%) and DIT (89.31%), which were significantly
different from the groups EIT (59.4%), ECM (54.26%), and ECT (52.37%). Since
buccolingual sections were obtained in order to encompass the entirety of the
cervical shoulder in the buccolingual direction, the method of sectioning may
have influenced the results obtained. An occasional high standard deviation was
observed as a result of the either perfect seal or high values of penetration
that were obtained in this evaluation.
Evaluation of Surface
The results of the microhardness measurements were listed
(Table 2). For all groups investigated, there was a strong correlation between
increased microhardness and the use of transparent matrices. When transparent
matrices were used, no statistically significant correlation was found between
the use of the incremental technique versus centripetal technique or of the
location of the cavity preparation. For the group DCM (opaque matrix system in
combination with the centripetal technique in deep cavities), a significant
decrease in microhardness values was noted compared with all seven possible
As a composite resin material undergoes polymerization
shrinkage, the force of this shrinkage may exceed the bond strength of the
material to tooth structure.26 Shrinkage-free resins that would
permit the placement of perfectly adapted and sealed restorations are not yet
available. Factors allowing the optimization of the marginal adaptation of
composite resin restorations can be found in the composite resin material,
cavity preparation, and placement technique. A multistep insertion technique,
in combination with transparent matrices and reflective wedges, has been
designed to enhance the marginal quality of Class II composite restorations.8
In addition, the marginal adaptation can be significantly improved by the use of
a buildup base material that reduces the size of the composite restoration,
thus increasing the free surface-to-volume ratio.27,28 Based on
these facts, and in combination with various parameters previously described, a
comparison was made between the centripetal and incremental techniques. It was
found that the centripetal technique showed better marginal adaptation in
cavities prepared in enamel than did the common incremental technique. Several
authors have indicated that one of the most important principles for
incremental placement is to reduce the V/A ratio by applying the first
increment to only one cavity wall.29 Other reports have indicated
that the application of composite in oblique layers resulted in fewer
contraction gaps at the margins.30,31 There has been disagreement
concerning the relative merits of apical oblique versus coronal oblique
incremental patterns, although differences between these are of less
significance than the need for incremental rather than single-bulk technique.32
In the present study, it was found that neither the
technique nor the matrix band material had statistical significant influence in
the marginal microleakage. Only the preparation depth influenced the results.
Eakle and Ito reported that significantly less leakage occurred under gingival
margins when the proximal box ended on enamel than when it terminated on
cementum.19 The same behavior was found in this study, where the
highest microleakage value in the enamel-surrounded groups 75.83% penetration
(EIM), as opposed to the highest value of 92.42% (DCM) in deeper cavities.
Numerous researchers noted that neither of the one-bulk incremental placement
techniques was able to produce consistently leak-free margins, even on etched
Through the use of the centripetal technique, the V/A ratio
could be reduced. This differed from the incremental technique, where the
complete apical area of the cavity was filled with the first layer of composite
resin material. In the incremental technique, this first layer had less contact
with the lateral walls than did the resin in the first layer of the centripetal
buildup technique. Alternatively, the first layer of the centripetal technique
had no contact to the pulpoaxial walls and thus had less tendency to contract
toward this wall and away from the cervical floor during polymerization. In the
proximal box, the polymerization shrinkage tended to pull this first horizontal
increment away from the cervical margin. The second layer of the incremental
technique, which was a diagonal layer, was not able to cover the first portion
in the cervical area, which did occur with the second layer of the centripetal
In deep cavities, the dentin adhesive material was unable to
inhibit dye penetration. Lui et al determined that the worst marginal
adaptation of restorations was found in the cervical area and attributed to the
effect of polymerization shrinkage, inadequate adaptation of noncondensable
resin, difficulty of placement at the proximal box, and shrinkage toward the
light source.23 In the present study, this behavior was only
registered in groups EIM, ECM, and DCM. It is, however, surprising that all
cavities in the Lui study were filled with different techniques and in
different depths yet always with metal matrices and wooden wedges. Wooden
wedges, when properly used, enhance the adaptation to the cavity walls and
provide firm contact areas and anatomical proximal contours.33
Scherer et al registered that restorations made with transparent matrices and
reflective wedges exhibited less microleakage than those delivered with opaque
matrices and wooden wedges.25 Furthermore, Lutz et al proved that
the concept of directing the polymerization shrinkage vectors toward the
margins of a cavity by using light-reflective wedges with reflective cores was
In another study,12 it was shown that the worst
marginal qualities were obtained in cavities filled with composite in a
one-step technique and with opaque wedges and matrices (62.2% excellent
margins) in contrast to composite fillings with transparent matrices and
laterally reflecting wedges (79.4% excellent margins). The values were not
significantly different. In that study, however, the cavities were placed in a
one-bulk technique and opaque matrices were only cured from the occlusal
direction. No postcuring was performed following the replacement of the opaque
matrices and wooden wedges. In the present study, only the restorations
delivered with metal matrix bands were postcured. For this purpose, postcuring
was performed for 40 seconds from the buccal and occlusal aspects. The
additional use of reflective wedges in the proximal area during the postcuring
period was not investigated in this study, as the authors believed that a
postcuring situation with reflective wedges in the proximal area would not
accurately reflect the differences in the aforementioned techniques.
A statistically significant increase in microhardness was
obtained in all groups treated with transparent matrices and reflective wedges.
This could be attributed to the higher degree of conversion and crosslinking of
the resin.13,34 There is also a correlation between the curing light
intensity and the depth of cure.
von Beetzen et al investigated the microhardness of
composite restorations and determined that the Herculite composite material,
also used in this study, was characterized by Vickers values ranging from 44 HV
to 59.8 HV, depending on the polymerization technique.13 In the
experimental set-up of their investigation, standardized Class II cavities in
brass blocks were filled with composite in 2.5-mm increments, which were then
cured for 60 seconds from the occlusal aspect of the cavity. Plastic matrix
bands were used and, after the increments had been cured from the occlusal
aspects, no postcuring was done - as in the present study. Although higher
values (59.8 HV) were obtained when using a transparent cone for the
polymerization,13 comparable values were also achieved for the
restorations using both metal matrices (42.65 HV for group EIM) and transparent
matrices (64.48 HV for group EIT) in the current investigation. This is due to
the 1.5-mm increments in this study, the preparation of extracted teeth rather
than metal blocks, and the use of an additional curing unit. Questions were raised,
however, regarding the examination of surface hardness conducting different
polymerization procedures, as in evaluation of the marginal seal. In the
experimental design of microhardness analysis, the different matrix bands were
glued to the flat surface of the specimen. Although securing and utilization of
different wedges was not conducted in this investigation, it would be
interesting, to do this in further investigations. A postcuring polymerization
technique, which could have improved the results, was also rejected.
The authors concluded that none of the insertion techniques
and matrix bands used in this study were able to prevent extensive microleakage
at the cervical margins of Class II composite restorations. The marginal
integrity of composite resins placed in cavities ending in enamel and restored
with the centripetal technique and transparent matrices was exceptional (though
not significantly different) to those filled in the incremental technique in
combination with either reflective or wooden wedges.
Cavities prepared with their marginal aspects in dentin
showed no significant differences in their microleakage behavior either in
dependence of the matrix band material nor of the placement technique. In deep
cavities, best (although not statistically significant) results were obtained
when the centripetal technique was used in combination of transparent matrices.
The preparation depth significantly influenced the results, with less leakage
observed in margins located within the enamel. The results also determined that
the highest mean hardness values for composite resin restorations were achieved
using transparent matrices.
*Assistant Professor, Department of Operative Dentistry, Johann Wolfgang
Frankfurt am Main, Germany.
- Bergenholtz G, Cox CF, Loesche WJ, Syed SA.
Bacterial leakage around dental restorations. Its effect on the pulp. J Oral
- Brannstrom M. Communication between the oral
cavity and the dental pulp associated with restorative treatment. Oper Dent
- Cheung GS. Reducing marginal microleakage of
posterior composite resin restorations: A review of clinical techniques. J Prosthet
- Eriksen HM, Pears G. In vitro caries related to marginal
leakage around composite resin restorations. J Oral Rehabil 1978;5(1):15-20.
- Qvist V. Correlation between marginal adaptation
of composite restorations and bacterial growth in cavities. Scand J Dent
- Navarro Majo JL. Marginal adaptation of
posterior resins: Effect of hygroscopic expansion and layer placement technique
on Class II and V fillings. Stoma (Lisb) 1991;2:47-48.
WP, Strang R, Ahmad I. Effect of composite resin placement and use of an
unfilled resin on the microleakage of two dentin bonding agents. Am J Dent
- Lutz F, Krejci I, Oldenburg TR. Elimination of
polymerization stresses at the margins of posterior composite resin
restorations: A new restorative technique. Quint Int 1986;17(12):777-784.
- Krejci I, Sparr D, Lutz F. Three-layer light
hardening procedure with traditional composites for Black Class II
restorations. Quintessenz 1987;7:1217-1229.
- Bichacho N. The centripetal build-up for
composite resin posterior restorations. Pract Periodont Aesthet
E, Denehy G, Boyer DB. Effect of matrix systems and polymerization
techniques on microleakage of Class II resin composite restorations. Am J Dent
- Lutz F, Krejci I, Barbakow F. The importance of
proximal curing in posterior composite resin restorations. Quint Int 1992;23(9):605-607.
- von Beetzen M, Li J, Nicander I, Sundstršm F.
Microhardness and porosity of Class 2 light-cured composite restorations cured
with a transparent cone attached to the light-curing wand. Oper Dent
- Simonsen RJ, Kanca J. Surface hardness of
posterior composite resins using supplemental polymerization after simulated
occlusal adjustment. Quint Int 1986;17(10):631-633.
- Unterbrink GL, Muessner R. Influence of light
intensity on two restorative systems. J Dent 1995;23(3):183-189.
- Hinoura K, Setcos JC, Philips RW. Cavity design
and placement techniques for Class 2 composites. Oper Dent 1988;13(1):12-19
DR, Menezes JM. The effect of retention grooves on
posterior composite resin restorations: An in vitro microleakage study. Oper
- Donly KJ, Wild TW, Jensen ME. Posterior
composite Class II restorations: In vitro comparison of preparation designs and
restoration techniques. Dent Mater 1990;6(2):88-93.
- Eakle WS, Ito RK. Effect of insertion technique
on microleakage in mesio-occlusodistal composite resin restorations. Quint Int
- Neo JC, Denehy GE, Boyer DB. Effects of
polymerization techniques on uniformity of cure of large-diameter, photo-initiated
composite resin restorations. J Am Dent Assoc 1986;113(6):905-909.
AB, Chosack A, Eidelman E.
Assessment of marginal leakage around Class II composite restorations in
retrieved primary molars. Pediatr Dent 1990;12:24-27.
- Opdam NJ, Roeters JJ, Kuijs R, Burgersdijk RC.
Necessity of bevels for box only Class II composite restorations. J Prosthet Dent 1998;80(3):274-279.
JL, Masutani S, Setcos JC, et al. Margin quality and microleakage of
Class II composite resin restorations. J Am Dent Assoc 1987;114(1):49-54.
- 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. Quint Int 1995;26(10):717-727.
W, Caliskan F, Kaim JM, et al. Microleakage comparison in opaque and
transparent matrix systems. Gen Dentistry 1989;37(6):482-484.
- Lambrechts P, Braem M, Vanherle G. Buoncore
memorial lecture: Evaluation of clinical performance for posterior composite
resins and dentin adhesives. Oper Dent 1987;12:53-78.
- Aboushala A, Kugel G, Hurley E. Class II
composite resin restorations using glass-ionomer liners: Microleakage studies.
J Clin Pediatr Dent 1996;21(1):67-70.
- Lutz F, Krejci I, Luescher B, Oldenburg TR.
Improved proximal margin adaptation of Class II composite resin restorations by
use of light-reflecting wedges. Quint Int 1986;17(10):659-664.
- Davidson CL. Resisting the curing contraction
with adhesive composites. J Prosthet Dent 1986;55(4):446-447.
- Hansen EK. Effect of cavity depth and
application technique on marginal adaptation of resins in dentin cavities. J Dent
- Hansen EK, Asmussen E. Comparative study of
dentin adhesives. Scand J Dent Res 1985;93(3):280-287.
- Bryant RW. Direct posterior composite
restorations: A review. 2. Clinical technique. Aust Dent J 1992;37(3):161-171.
- Eli I, Weiss E, Kozlovsky A, Levi N. Wedges in
restorative dentistry: Principles and applications. J Oral Rehabil 1991;18(3):257-264.
- Kanca J. Visible light activated composite resin
for posterior use: A comparison of surface hardness and uniformity of cure.
Update. Quint Int 1985;16(10):687-690.