Indian Journal of Dental Research

: 2019  |  Volume : 30  |  Issue : 5  |  Page : 783--787

Microleakage in posterior teeth with different materials and different types of cavities

Everton Ribeiro dos Santos, Adair Luis Busato 
 Department of Restorative Dentistry, Universidade Luterana do Brasil, Canoas, Brazil

Correspondence Address:
Prof. Everton Ribeiro dos Santos
Avenida Lavras, 622/402, Porto Alegre, Rio Grande do Sul


Background: The purpose of this study was to investigate the behavior of a resin-based 2,2-bis (p-[2′-hydroxy-3′-methacryloxypropoxy] phenylene) propane and a composite resin modified with oxirane, regarding the ability of marginal sealing, both with direct restorations and indirect restorations. Materials and Methods: To achieve this, mesio- and disto-occlusal cavities were made on the same tooth, totalling 100 cavities. These cavities were restored with two materials, Filtek P90 (3M ESPE) and Opallis (FGM). Then, they were divided into two groups: Fifty direct restorations, each sample restored with the two materials, following the manufacturer's protocol and fifty indirect restorations, and then cemented with a paste obtained by diluting the resin in its adhesive. The templates were thermally cycled, 30 s at 5°C and 30 s at 55°C for 500 cycles and then were immersed in methylene blue for 24 h. Afterward, they were trimmed using a plaster trimmer from occlusal toward a gingival direction. These samples were analyzed with stereomicroscope (×3) by two blinded observers. Then, a system which considers the average number of surrounding walls that have suffered infiltration on a scale of 0–3 was used. Results: Data were subjected to statistical analysis (Mann–Whitney and Kruskal–Wallis tests) with a significance level of 5% maximum. Conclusion: It was concluded that the indirect restorations showed the greatest number of restorations with score 0 and 1 (34) and lowest score 2 and 3 (16). This suggests that indirect restorations regardless of the material used behaved more effectively regarding the marginal sealing of restorations.

How to cite this article:
Santos ER, Busato AL. Microleakage in posterior teeth with different materials and different types of cavities.Indian J Dent Res 2019;30:783-787

How to cite this URL:
Santos ER, Busato AL. Microleakage in posterior teeth with different materials and different types of cavities. Indian J Dent Res [serial online] 2019 [cited 2020 Sep 25 ];30:783-787
Available from:

Full Text


Esthetic materials have always been an obsession in dentistry, but one stands out, the resin composite created by Bowen.[1] An important advance was made by Nakabayashi et al.,[2] obtaining adhesion to dentin. This material made a number of breakthroughs in terms of physical properties, but it still has not been able to eliminate a problem that remains present until this day: polymerization shrinkage. This obstacle can be one of the major causes of failures in restorations. Flaws on the interface of the tooth and restoration, due to inappropriate sealing are responsible for postoperative sensitivity, degradation of the hybrid layer, marginal staining, bacteria, and oral fluid infiltration, which can lead to secondary caries.

Several alternatives are present in literature to try to eliminate this effect such as: incremental insertion technique, polymerizing restorations outside the cavity (indirect) and altering the chemistry of the composite resins. Recently, the methacrylate base was substituted for oxirane, which, according to manufacturers has low shrinkage, reducing the possibility of failures in restorations.

The aim of this study was to verify the behavior of a resin-based 2,2-bis (p-[2′-hydroxy-3′-methacryloxypropoxy] phenylene) propane (Bis-GMA) and a resin modified with oxirane, regarding the ability of marginal sealing, both in direct restorations and indirect restorations.

 Materials and Methods

Thisin vitro study was approved by the ULBRA Ethics Committee under the CAAE number: 08170712.4.0000.534. Fifty extracted human third molars with surgical indication by orthodontics were used. Patients signed an informed consent form allowing the use of these teeth. The inclusion criteria were: third molars extracted healthy, without decay or restorations and age from 16 to 25 years.

The teeth were embedded in self-polymerizing acrylic resin Jet (Classic, São Paulo, Brazil) and were measured with an analog pachymeter (Eccofer, Curitiba, Brazil) in the buccolingual and mesiodistal directions.

We created two groups with 25 teeth each, one for direct restorations and the other for direct/indirect. In each tooth, two cavities were made: one MO cavity preparation and another OD, being called cavity I and cavity II randomly alternating side and restorative material-Opallis (FGM, Joinville, Brazil) and Filtek P90 (3M/ESPE, St. Paul, USA). By the end, there were fifty cavities for IR and fifty cavities for DR.

For the realization of the direct restorations cavities diamond burs #1093 (KG Sorensen, Alphaville, Brazil) were used at high speed with cooling, and for the indirect restorations cavities diamond burs #4130 (KG Sorensen, Alphaville, Brazil) were used at high speed with cooling. These cavities had an inclination of about 6°. To restore with the direct technique using material derived from methacrylate, the following protocol was used: acid etching with 37% phosphoric acid, acid gel (Villevie, Joinville, Brazil), first the enamel and then the dentin for 15 seconds. Then, the cavity was washed with water and dried with mild jets of air. Immediately afterwards, a layer of adhesive ambar (FGM, Joinville, Brazil) was applied and spread for 10 seconds. Then, a second layer of the same adhesive was applied and cured for 10 more seconds. The resin Opallis color Da2 was used in two increments with one increment reaching the buccal, lingual, axial, and pulp walls equivalent to half of the cavity walls, which was polymerized for 40 seconds. The second increment was added and reached the buccal, lingual, and proximal pulp walls completing the remainder of the cavity, which was also polymerized for 40 seconds. The restored teeth sat for 10 min and were then immersed in distilled water. The restorative technique for the cavities that had been restored with the silorane-based material followed the recommendation of the manufacturer, the self-etch primer was used (3M/ESPE St. Paul, USA), applied with disposable applicator cavibrush (FGM, Joinville, Brazil) for 15 seconds. The primer was dried with air jets at a distance for 10 seconds and light cured for 10 seconds. After this, the specific adhesive bond (3M/ESPE, St. Paul, USA), was applied with a disposable brush. The adhesive was applied over the whole surface and spread with air jets for 10 seconds, and then light cured for 10 seconds. The restorative material Filtek P90 was applied in two increments, with the first increment reaching the buccal, lingual, axial, and pulp walls equivalent to half of the cavity walls, which was polymerized for 40 seconds. The second increment was added and reached the buccal, lingual, and proximal pulp walls completing the remainder of the cavity, which was also polymerized for 40 seconds with the Radii-cal (SDI, São Paulo, Brazil) device that has been tested by a radiometer (Gnatus, Ribeirão Preto, Brazil) and has a light intensity exceeding 800 mW/cm2. The polymerization was carried out in one direction, from occlusal to gingival. The restored teeth sat for 10 min and were then immersed in distilled water.

For the indirect restorations, composite resins Opallis and Filtek P90 were used in a direct/indirect technique where restorations are modeled in the cavity, receive 10 seconds of curing, are removed from the cavity and light-cured for another 40 seconds before the cementation. The first 25 indirect restorations were made with composite resin Opallis, and the following 25 cavities were restored with composite resin Filtek P90. Initially, the cavities were isolated using KY (Johnson and Johnson, São José dos Campos, Brazil) then the cavity was completed with resin composite. Furthermore, a small handle, made of orthodontic wire (Morelli, Sorocaba, Brazil) was inserted, and this set was light cured for 10 seconds and removed from the cavity. Out of the cavity, it received another 40 seconds of curing and subsequent cementation. The cavity and restoration were washed with water/air spray to remove the isolation material. The protocol recommended by the manufacturer and previously mentioned was followed for the use of the adhesive systems. The cement was made by diluting the composite resin in its adhesive. For opallis, it was the adhesive ambar, and for Filtek P90, it was adhesive Bond, until proper cement consistency was achieved. The cemented restorations were light cured for 40 seconds from occlusal to gingival direction, sat for 10 min and were then immersed in distilled water.

The templates were thermally cycled for 500 cycles. A cycle being considered 30 seconds at 5°C and 30 seconds at 55°C for approximately 8 hours and 30 min. For the thermal cycling, Ética Model 521 (Ética Odontológica, São Paulo, Brazil) was used at the Federal University of Pelotas, Center of Control and Development of Biomaterials in Pelotas, Rio Grande do Sul. After the thermocycling was complete, the templates were immersed in distilled water for six days at room temperature. After this, they were immersed in methylene blue dye (Newprov, Pinhais, Brazil) for 24 hours according to the Ziehl–Neelsen method. The samples were then washed and rested for six days in distilled water to be cut in a plaster trimmer (VH, Araraquara, Brazil).

The cut was performed from occlusal to gingival, where it was possible to view buccal and lingual walls and the axial wall restoration. The depth of the cut was around 1.5 mm. Then, the cuts were evaluated by two blinded observers on a stereomicroscope (×3) (Optika, Ponteranica, Itália), using a scale of scores: 0: no surrounding walls infiltrated; 1: one surrounding wall showed infiltration, 2: two surrounding walls showed infiltration and 3: three walls showed infiltration. The nonparametric Mann–Whitney test and the nonparametric Kruskal–Wallis tests were used. For the tests mentioned above the maximum level of significance was set at 5% (P ≤ 0.05) and the software used for statistical analysis was SPSS version 13.0 (Spss Inc., Chicago, USA).


The nonparametric Mann–Whitney test was used to compare the amount of walls with infiltration between the direct/indirect and direct techniques, and between Opallis and Filtek P90. The nonparametric Kruskal–Wallis test was used to compare the amounts of infiltrated walls between four different combinations of techniques/materials: Opallis/Indirect Opallis/Direct, and P90/Indirect P90/Direct.

For the analysis of walls infiltrated considering restoration technique of cavity I, it was found that the amount of infiltrated walls for direct restorations type was significantly superior to indirect restorations type (P = 0.001). For the analysis of walls infiltrated regarding restorative material in cavity I, it was found that the number of walls infiltrated with Filtek P90 was significantly higher than Opallis restorative material (P = 0.000).

As for cavity II, regarding the number of walls infiltrated and restoration technique, it was found that the amount of infiltrated walls for the direct restorations presents no significant difference from the IR (P = 0.649). When a comparison is made between walls infiltrated and restorative material, it appears that the amount of infiltrated walls for Filtek P90 was significantly higher than Opallis restorative material (P = 0.000).

When comparing infiltrated sides between the four technique/equipment combinations using the nonparametric test Kruskal–Wallis, it was observed that cavity I in all groups differ where a greater number of walls is infiltrated. The combination P90/Direct was followed by P90/Indirect group and Opallis/Direct group. For cavity II, using the nonparametric Kruskal-Wallis test, it was verified that there is a significant difference between the number of walls infiltrated among the four groups: Opallis/Indirect, Opallis/Direct, and P90/Indirect P90/Direct. A higher number of walls was infiltrated with the P90/Direct combination, followed by the P90/Indirect group. The Opallis/Direct and Opallis/Indirect groups have the lowest amount of infiltrated and walls and do not differ (P = 0.000), with a lower number of walls infiltrated group Opallis/Indirect (P = 0.000).


The sequence for composite resin restorations using the direct technique is rigid, and any failure in the steps can lead to failures in the restorative procedure.[3]

It has been said that with the use of the indirect technique, polymerization shrinkage occurs on the model, other than in the cavity, eliminating tension on the tooth and free monomers.[4],[5],[6] Another alternative is the direct/indirect restorations.[7] This technique has the advantage of reducing the polymerization shrinkage because the polymerization itself occurs outside the cavity leaving only the contraction of the resin cement that will bond this restoration to tooth structure.

This study agrees with studies showing that indirect restorations have less infiltration when compared to direct restorations[4],[5],[6] and disagrees with the research that shows infiltration similar between direct restorations and indirect restorations.[8] In this particular case, direct/indirect restorations were performed because they cost less, the technique is simple and also because the procedure is done in a single appointment.

Although several laboratory studies show that silorane-based material has lower polymerization shrinkage and consequent lower infiltration, comparing direct restorations[9],[10],[11] this research does not corroborate the results regarding leakage because it was found that both direct restorations and indirect restorations with this material showed higher microleakage than the methacrylate-based material by the stress generated by its adhesive[12],[13] associated with a high modulus of elasticity[14] leading to marginal leakage.

Regarding the present research, the aim was to evaluate the performance of direct restorations and indirect restorations. Results show that regardless of the cavity, there was more leakage when the material used was the composite resin Filtek P90, and one can also see that the leakage was greater in direct restorations. This data are surprising, since the technical information of the material alleged low polymerization shrinkage and consequently low thermal dimensional change, which would cause low levels of infiltration to occur, differing from what actually happened. One of the probable reasons regarding the direct restorations is the fact that there was no previous etching, which is not ideal for enamel bonding. Infiltration begins by the enamel, and most likely the failure occurred due to the lack of pretreatment of the enamel. For indirect restorations, composite resin Filtek P90 has no adhesive cement, and in the procedure, the actual resin diluted with the adhesive was used as a cementing agent. This may also have caused the flaws. On the other hand, the composite Opallis showed regular behavior, varying little between direct restorations and indirect restorations.

The composite Opallis in indirect restorations showed no restoration infiltration level 3, and 84% of the restorations showed no infiltration. Only 15.4% had infiltration on two walls. In this case, when referring to two walls it means surrounding walls, never the axial. In direct restorations with Opallis, there was a severe change in the pattern of infiltration, and 62.5% of the restorations showed infiltration of one wall (regardless of whether the buccal or lingual) and 25% of the restorations showed no leakage. Since it is the same material, it is presumed that these levels of leakage actually are due to the polymerization shrinkage even if the restoration was executed in two increments.

Regarding the composite resin Filtek P90, no cavity or restoration had zero infiltration. In the indirect restorations, regular levels were found: 41.7% infiltration on one wall; 33.3% infiltration in two walls, and 25.0% infiltration on three walls. This result is coherent since polymerization shrinkage did not occur on the tooth structure, when subjected to cycling it behaved with reasonable thermal dimensional changes. On the other hand, in direct restorations, there was no restoration without infiltration as well, but also no restoration infiltration level 1 was found. In this situation, 41.25% of the restorations had infiltration on two walls, and 58.8% had infiltration on three walls. The reason of the substitution of Bis-GMA was the problem with the presence of water responsible for the polymerization shrinkage, which does not seem to have produced the expected effect with the introduction of silorane, because with the present method of marginal leakage, the material showed a very debatable pattern, when compared to the traditional Bis-GMA system. It is possible to find variations of marginal leakage throughout the cavity, either in the mesial or distal, as well as infiltration on one side only, and in 90% of these cases, it involved the buccal wall, perhaps because it has less contact with the saliva and the minerals that produce the aprismatic layer.

In circumstances in which the study was conducted, we conclude that the type of cavity is less important than the restorative material and that the restorative technique can influence the overall performance of the restorations.

After numerous studies[15],[16] it was concluded this technique was not viable with deep cavities that had less than 1 mm of dentin separating the cavity floor from pulp. With this concern in mind[2] a new approach was proposed, where the etching was eliminated because the presence of smear was important. The suggestion was to use the acidity of the primer and the adhesive and modify the smear layer, making it more permeable to the penetration of the adhesive, without removing it. Thus was created the integration technique. This adhesive protects from sensitivity, but the adhesive strength is lower, and the enamel is not modified as much as in the technique that uses the etching. It is necessary to evaluate pros and cons. Greater protection but lower bonding strength, which may also lead to higher levels of marginal leakage with this material, as described by Marins de Carvalho et al.[15]

It is important to point out that marginal leakage in clean mouths has little meaning if we consider the possibility of recurrence of caries, however, the postoperative sensitivity can be a discomfort for the patient. The execution of the bevel is a controversy still. Baratieri[17] argues that it has little meaning in terms of resistance, as does Coelho-de-Souza.[18]

In this study, it was opted to not bevel, because it was agreed that this increases the size of the cavity, and in this case, our main focus was the assessment of the restorative material.

The recent study of Apolônio et al.[19] agrees with the present paper. It also concluded that Silorane-based resin is not more effective in terms of marginal sealing then the methacrylate-based resins, however, restorative Silorane-based resin has an important clinical advantage because it can be placed in horizontal layers, uniting the buccal wall to the lingual wall, due to its low polymerization shrinkage.

Moreover in the end, there is an important question: what is the true impact of microleakage in restorations if our patient has a clean and healthy mouth? The risk of caries is almost none because there are no conditions for the disease to appear. On the other hand, to solve postoperative sensitivity ideally, there should not be leakage that might occur due to the lack of marginal sealing because with the ingestion of sugars and cold foods the patient will inevitably feel what can be confused with pain. The stress of this patient with this phenomenon may cause a tumultuous relationship, although sensitivity cases, immediately when the cause is removed.


Given the methodology used and the results obtained and submitted to statistical analysis, we can conclude that: despite the lower polymerization shrinkage of silorane composite resin Filtek P90 it showed higher values of microleakage compared to composite resin based Bis-GMA, indirect restorations showed lower values of microleakage when compared to direct restorations, and only the Filtek P90 provided infiltration level 3.


The authors would like to thank Universidade Federal de Pelotas, Pelotas, Brazil.

Financial support and sponsorship


Conflicts of interest

There are no conflicts of interest.


1Bowen RL. Dental Filling Material Composing Vinyl Silane Treated Fused Silica and a Binder Consisting of a Reaction Product of Bisphenol and Glycidyl Acrylate. United States Patent US 3066112; 27 November, 1962.
2Nakabayashi N, Kojima K, Masuhara E. The promotion of adhesion by the infiltration of monomers into tooth substrates. J Biomed Mater Res 1982;16:265-73.
3de Andrade OS, de Goes MF, Montes MA. Marginal adaptation and microtensile bond strength of composite indirect restorations bonded to dentin treated with adhesive and low-viscosity composite. Dent Mater 2007;23:279-87.
4Cassin AM, Pearson GJ. Microleakage studies comparing a one-visit indirect composite inlay system and a direct composite restorative technique. J Oral Rehabil 1992;19:265-70.
5Duquia Rde C, Osinaga PW, Demarco FF, de V Habekost L, Conceição EN. Cervical microleakage in MOD restorations:In vitro comparison of indirect and direct composite. Oper Dent 2006;31:682-7.
6Fruits TJ, Knapp JA, Khajotia SS. Microleakage in the proximal walls of direct and indirect posterior resin slot restorations. Oper Dent 2006;31:719-27.
7Fahl Júnior N. The direct/indirect composite resin veneers: A case report. Pract Periodontics Aesthet Dent 1996;8:627-38.
8Soares CJ, Celiberto L, Dechichi P, Fonseca RB, Martins LR. Marginal integrity and microleakage of direct and indirect composite inlays: SEM and stereomicroscopic evaluation. Braz Oral Res 2005;19:295-301.
9Yamazaki PC, Bedran-Russo AK, Pereira PN, Wsift EJ Jr. Microleakage evaluation of a new low-shrinkage composite restorative material. Oper Dent 2006;31:670-6.
10Ilie N, Jelen E, Clementino-Luedemann T, Hickel R. Low-shrinkage composite for dental application. Dent Mater J 2007;26:149-55.
11Al-Boni R, Raja OM. Microleakage evaluation of silorane based composite versus methacrylate based composite. J Conserv Dent 2010;13:152-5.
12Boaro LC, Gonçalves F, Guimarães TC, Ferracane JL, Versluis A, Braga RR, et al. Polymerization stress, shrinkage and elastic modulus of current low-shrinkage restorative composites. Dent Mater 2010;26:1144-50.
13Lopes MB, Valarini N, Moura SK, Guiraldo RD, Gonini Júnior A. Photoelastic analysis of stress generated by a silorane-based restoration system. Braz Oral Res 2011;25:302-6.
14Monteiro GQ, Montes MA. Evaluation of linear polymerization shrinkage, flexural strength and modulus of elasticity of dental composites. Mater Res 2010;13:51-5.
15Marins de Carvalho R, Carrilho MR, Pereira LC, Garcia FC, Marquezini L Jr., Silva SM, et al. Sistemas adesivos: Fundamentos para aplicação clínica. Bio Odontol 2004;2:6-86.
16Gallo JR 3rd, Henderson M, Burgess JO. Shear bond strength to moist and dry dentin of four dentin bonding systems. Am J Dent 2000;13:267-70.
17Baratieri LN, Canabarro S, Lopes GC, Ritter AV. Effect of resin viscosity and enamel beveling on the clinical performance of class V composite restorations: Three-year results. Oper Dent 2003;28:482-7.
18Coelho-de-Souza FH, Klein CA Jr., Papaleo F, Deibler J, Barbosa NA, Macedo RP. Bevel and restorative material effects on fracture strength of lower premolars with cervical restorations. Stomatos 2008;14:26-35.
19Apolônio FM, Ramalho MS, Souza LC, Lima FC, Rodrigues LK, Saboia VP. Evaluation of marginal leakage on restorations of methacrilate and silorane-based composites. Rev Fac Odontol Univ Pernambuco Fed 2011;16:312-6.