Indian Journal of Dental Research

: 2022  |  Volume : 33  |  Issue : 1  |  Page : 85--89

X-ray microtomography analysis of gaps and voids in the restoration of non-carious cervical lesions with different composite resins

Lanna Cristina Gonçalves da Costa Vieira, Adriano Rocha Campos, Plinio Mendes Senna, Cesar dos Reis Perez 
 Department of Prosthesis, School of Dentistry of State University of Rio de Janeiro, Rio de Janeiro, RJ, Brazil

Correspondence Address:
Prof. Cesar dos Reis Perez
Department of Prosthesis, School of Dentistry of State University of Rio de Janeiro, Boulevard 28 de Setembro, 157 . Vila Isabel, Rio de Janeiro - 20551-030, RJ


Context (Background): Resin composites are the most widely used material for restoring cervical defects. However, the high failure rate of these restorations is still a concern. Aims: The aim of this in vitro study was to evaluate, using microtomography (μCT), the interfacial gap and voids formation in Class V cavities in premolars restored with materials with lower polymerization shrinkage combined with different restorative techniques. Settings and Design: Cervical defects were created in 30 intact premolar and were randomly distributed to be restored by one of the following techniques (n = 6): Composite resin with two increments (CR), organic modified polymer (ORMOCER) with single (OR1) or two increments (OR2, or low viscosity bulk-fill composite resin with single (BF1) or two increments (BF2). Methods and Material: Each tooth was scanned before filling to determine the volume of interest (VOI) to be applied in the second μCT after restoration and to control the cavity volume among the groups. In the μCT after filling, the volume of interfacial gaps and voids was calculated for each group. Statistical Analysis: The groups were compared using one-way and Tukey HSD post hoc test (α = 0.05). Results: It was possible to identify higher gap formation in the OR1 group and higher void formation in CR group (P < 0.05). OR2 group showed better results than the group with one increment. BF2 showed the best filling capacity. Conclusions: It was possible to conclude that the material and the number of increments directly influenced the internal adaptation and voids formation of Class V restorations.

How to cite this article:
da Costa Vieira LC, Campos AR, Senna PM, Reis Perez Cd. X-ray microtomography analysis of gaps and voids in the restoration of non-carious cervical lesions with different composite resins.Indian J Dent Res 2022;33:85-89

How to cite this URL:
da Costa Vieira LC, Campos AR, Senna PM, Reis Perez Cd. X-ray microtomography analysis of gaps and voids in the restoration of non-carious cervical lesions with different composite resins. Indian J Dent Res [serial online] 2022 [cited 2022 Dec 5 ];33:85-89
Available from:

Full Text


The restoration of non-carious cervical lesions (NCCLs) is a common procedure with high failure rates associated with loss of retention, microleakage, and secondary caries.[1],[2] Therefore, it is encouraged the adoption of restorative materials and techniques that lead to fewer defects at the interface with the dental structure (gaps) and fewer defects within the material (voids) to improve the reliability of NCCLs restorations.[3],[4]

Composite resins have a relevant drawback inherent to the polymerization shrinkage, leading to gap formation.[5] The polymerization shrinkage can be influenced not only by factors related to the material properties such as the elastic modulus, total volume but also by operative variables such as the cavity geometry, the light-curing method, and restorative technique.[6],[7],[8]

New restorative materials were developed with better wettability, lower polymerization shrinkage and lower contraction, such as the organic modified polymers (ORMOCERs)[9],[10] and low-viscosity bulk-fill composite resins.[11],[12],[13]

Microtomography is a valuable tool to evaluate internal adaptation and reveal the behavior during and after the restorative procedure.[7] The aim of this in vitro study was to evaluate the interfacial gap and voids formation in Class V restorations using different materials and incremental techniques.

The research null hypotheses were as follows:

There would be no difference in the gap volume between the different materials.There would be no difference in the void´s volume between the different materials.


This in vitro study was approved by the ethics committee of Pedro Ernesto University Hospital – UERJ (process no. 10522918.2.0000.5259).

Sample size calculation

A priori power analysis was used to calculate the sample size. It was based on the results of a previous study that used microtomography evaluation.[7] Considering type 1 error of 0.05 and power of 0.8, the two-tailed t test determined a minimum of 6 samples in each group to achieve a medium effect size (G*Power; Universität Düsseldorf, Germany).

Specimen preparation

Thirty intact human maxillary premolars with no crack, decay, fracture, abrasion, previous restorations, or structural deformities were selected from the Human Teeth's Bank of the School of Dentistry of the State University of Rio de Janeiro. They were freshly extracted for orthodontic purposes within a month before the study. The teeth were cleaned with an ultrasonic scaler 1 week before examination and cleaned with a pumice and rubber cup. Then, all the teeth were stored in distilled water at room temperature. Using a diamond bur (6835KR, Komet, Germany), cavities with approximately 3 mm occlusal-gingival height, 3 mm mesiodistal length, and 2 mm cavity depth were prepared at the buccal surface with air/water spray, without enamel bevel. A digital caliper was used to measure cavities' dimensions. The C-factor of the cavity simulated a clinical situation of medium stress.[14] The burs were changed after five preparations. All the preparations were performed with the gingival margin placed 0.5 mm below the cement-enamel junction (CEJ).

Then, the teeth were positioned and fixed into the artificial sockets present in the acrylic models. The gingival margins of the cavities were planned to be located 1.0 mm below the artificial gingiva level, aiming to simulate a clinical situation as close as possible.

The periodontal ligament was simulated using the method described by Soares et al.[15]

Restorative procedures

The resin blocks with the teeth were randomly assigned to the five groups (n = 6) through drawing. Only one operator performed all the restorative procedures. Another examiner evaluated the microtomography images and was blind to the examined group. The same adhesive strategy was used for all specimens. Single Bond 2 was applied according to the manufacturer's instructions: Acid etching on enamel and dentin for 15 s, rinsed with water for 10 s, and gently air-dried. Then, the adhesive was applied with an applicator for 10 s, followed by air-thinning for 5 s and light-cured for 10 s with a light-emitting diode light-curing unit (Radii Cal, SDI, Bayswater, Australia), presenting an irradiance of 1000 mW/cm2. The light intensity was controlled during the whole process using a radiometer (Demetron LED Radiometer, Kerr Corp.). Each tooth received an identifying number, but the test machine operator was blind to it, and the samples were divided into five groups:

Composite Resin Group (CR): Cavities restored with two oblique increments of Filtek™ Z350 XT (with gingival increment first), light-cured buccally for 20 s each.ORMOCER with single increment Group (OR1): Cavities restored with a single increment of Admira Fusion x-tra, and light-cured buccally for 20 s.Bulk-fill Composite Resin single increment Group (BF1): Cavities restored with a single increment of x-tra Base Bulk-fill, which was inserted after the placement of a mylar matrix stabilized with wood wedges and a photo-cured gingival barrier,[16] and light-cured buccally for 20 s.ORMOCER with two increments Group (OR2): Cavities restored with two oblique increments of Admira Fusion x-tra (with gingival increment first), light-cured buccally for 20 s each.Bulk-fill Composite Resin with two increments Group (BF2): Cavities restored with two increments of x-tra Base Bulk-fill. The first was inserted after the placement of a mylar matrix stabilized with wood wedges, a photo-cured gingival barrier,[16] and light-cured buccally for 20 s, followed by a second increment.

No additional finishing or polishing procedures were performed since the analysis was attained to the internal aspects of the restorations. The materials used are described in [Table 1].{Table 1}

X-ray microtomography (micro-CT) analysis

Each tooth was scanned twice using a μCT (Skyscan 1174; Bruker, Kontich, Belgium), before and after filling using a sample holder to allow perpendicular incidence of x-rays and similar tridimensional position on both scans. The operating condition of the μCT equipment was 50 kV voltage, 800 μA current, 0.5-mm aluminum filter, 10000 ms exposure time, 16 μm pixel size, and 1304 × 1024 pixels sensor resolution. Image acquisition was performed using rotation step of 0.7-degree, average of 3 frames. Scan duration was approximately 120 min. The x-ray projections were reconstructed (NRecon®, version; Bruker, Kontich, Belgium) using the following settings: Smoothing 2 (20%), automatic misalignment compensation, ring artifact reduction 4 (20%), and 40% beam-hardening correction. Extra care was taken to define the same dynamic range on both scanning of each tooth.

Each tooth image datasets (before and after filling) were loaded into DataViewer software program (version; Bruker, Kontich, Belgium) for tridimensional registration of both datasets. The volume of interest of the Class V cavity was defined and applied on the second dataset (VOIrestored) into CTan software program (version; Bruker, Kontich, Belgium) and the filling material was binarized. Next, the mathematical calculation of volume of interface gaps and voids was performed using the '3D analysis' tool in the software.

Statistical analysis

After checking data distribution with Shapiro–Wilks test, the groups were compared using one-way ANOVA and Tukey HSD post hoc test at a significance level of 0.05 (SPSS, v. 20; IBM).


The obtained data were assessed based on the recorded volume (mm3). Cavity volumes were similar between groups (p > 0.05). [Table 2] shows the volume analysis.{Table 2}

Two aspects were analyzed: Internal gap volume between restorative composites and cavity walls, and the volume of internal voids present in the materials. Micro-CT-based gap and void formation volumes with standard deviations are shown in [Table 3].{Table 3}

The representative two-dimensional (2D) μCT images of all tested groups are shown in [Figure 1]. Considering void formation volume, CR presented the highest values (P < 0.05). OR1 showed a significantly higher gap formation volume than the other groups, and BF2 the lower, although not significant to BF1.{Figure 1}


The present study results demonstrate that there were statistically significant differences in the volume of micro gaps and the voids among the experimental groups. Consequently, the first null hypothesis of the study—that there is no significant difference in micro gap volume between the tested materials and the dental substrate—was rejected. The second null hypothesis was also rejected since there was significant difference among the materials considering the internal void's volume.

The restoration of non-carious cervical lesions presents several peculiarities that justify its high failure rate: The difficulties in removing the causal factor (and the consequent persistence of elements that induce the recurrence of lesions), the intrinsic access limitations according to the region, the difficulties in isolation and the numerous variables related to adhesion.[1],[2]

In this work, the adhesion variables were not approached, since the teeth were extracted without lesions, which were created artificially with bur. So, dental adhesion was uniformed through the same adhesive strategy. Also, the teeth were positioned in artificial arches simulating the clinical situation. Thus, every effort was performed to assess factors related to the restorative materials, directly influencing the restoration's performance.

It should be noted that the frequency of occurrence and the volume of gaps. It is impossible not to be surprised by the rate of incomplete internal sealing of the restorations, observable only through x-ray microtomography. This technology reveals how far from the ideal is the complete sealing of restorations.[7],[13] New technologies are being developed that will allow evaluating the phenomenon in real time, allowing a complete understanding of it.[17] Previous studies have reported that the gaps' volume depended more on the initial gap due to poor wetting than on polymerization shrinkage.[18] Although the initial gap formed by incomplete wetting during material insertion was not evaluated in this work, a better wetting of the walls in the BF groups could be expected, considering its flowable consistency. Also, O1 showed the highest gap volume. This result could be expected since the material was inserted in a single portion. Taking into account that this material presents a worse wetting ability compared with flowable resins. When the same material was inserted in two increments (O2), a similar gap formation was observed compared to TCR, which was also inserted in two increments.

Although there was no statistical difference between these two groups, the lower gap formation in O2 could be justified by the alleged lower volume contraction of the ORMOCER. The bulk fill material present in BF1 and BF2 presented the smaller gap volume, with BF2 showing significantly better than the other tested groups. This result could be justified because the flowable bulk fill material would present less polymerization contraction and better wetting.[9],[10]

TCR showed the highest index formation concerning the internal void's formation, probably related to the two increments insertion. Despite this, O2 presented significantly fewer voids, although it was also inserted in two increments and had a similar consistency.

Some works consider that the incremental placement with oblique layers, which is recommended to decrease polymerization shrinkage, tends to create more internal voids between resin layers.[3],[19] Anterior works showed that voids could also be justified by the presence of pre-existing bubbles in the inserted material.[7],[13]

Although there is a wide variation in the shape and the depth of NCCLs clinically, the present work used preparations with large diameters and depths, with a configuration which simulates a situation of medium challenge concerning the polymerization's contraction.[9] Additionally, the fixation in an artificial hemi-arch allowed a simulation of what occurs clinically, adding some peculiarities that would be absent if the restorative technique was performed over an individualized tooth. The access limitations imposed by the simulation directly influenced the material's insertion seeking to mimic what happens clinically.

Adopting an alternative technique for stabilizing a mylar matrix[16] in BF1 allowed the flowable resin's insertion, filling the cavity. In BF2, the technique was also used but only for the first increment. Clinically, this alternative could be more relevant since the intrinsic limitations in access to the cavity are more pronounced due to the anatomical and technical peculiarities of the cervical area. An injectable material could provide good wetting and complete filling if it could be contained.

Bulk fill and ORMOCER resin composites, in flowable or conventional presentations, showed lower polymerization shrinkage when compared with conventional resin composites in many in vitro studies.[3],[4],[7] This property, associated with the increased depth of cure, is interesting for restoring NCCLs, with more minor polymerization contraction issues and the possibility of more straightforward and faster restoration placement.[3],[7] However, it should be noted that the performance was better when two increments were used and that internal defects, although minimized, are still present.


The bulk-fill flowable resin showed better results considering the gaps and the void's volume when applied in one or two increments. The better wetting presented by the flowable material associated with the alleged lower contraction was considered the reason for these results.

The ORMOCER applicated in a single increment showed the higher gap volume indicating that a bigger material volume and, mainly, a poor wetting of the internal walls directly affected this aspect, overcoming the beneficial effect of its alleged lower polymerization shrinkage. When it was used in two increments, the gap volume declined dramatically.

The material composition, its viscosity, and the number of increments directly influenced the internal adaptation of the restorations.

Financial Support and Sponsorship

This study was financed in part by the Coordenação de Aperfeiçoamento Pessoal de Nível Superior – Brasil (CAPES) – Finance Code 001.

Conflicts of interest

There are no conflicts of interest.


1Perez CDR, Gonzalez MR, Prado NAS, Ferreira de Miranda MS, de Andrade Macêdo M, Fernandes BMP. Restoration of noncarious cervical lesions: When, why, and how. Int J Dent 2012;2012:687058.
2Santos MJMC, Ari N, Steele S, Costella J, Banting D. Retention of tooth-colored restorations in non-carious cervical lesions--A systematic review. Clin Oral Investig 2014;18:1369-81.
3Correia AMO, Tribst JPM, Matos FS, Platt JA, Caneppele TMF, Borges ALS, et al. Polymerization shrinkage stresses in different restorative techniques for non-carious cervical lesions. J Dent 2018;76:68-74.
4Ichim IP, Schmidlin PR, Li Q, Kieser JA, Swain MV. Restoration of non-carious cervical lesions. Part II. Restorative material selection to minimize fracture. Dent Mat 2007;23:1562-9.
5Braga RR, Ballester RY, Ferracane JL. Factors involved in the development of polymerization shrinkage stress in resin-composites: A systematic review. Dent Mater 2005;21:962-70.
6Borges ALS, Borges AB, Xavier TA, Bottino MC, Platt JA. Impact of quantity of resin, C-factor, and geometry on resin composite polymerization shrinkage stress in class V restorations. Oper Dent 2014;39:144-51.
7Almeida Júnior LJS, Penha KJS, Souza AF, Lula ECO, Magalhães FC, Lima DM, et al. Is there correlation between polymerization shrinkage, gap formation, and void in bulk fill composites? A μCT study. Braz Or Res 2017;31:e1-10.
8dos Santos GO, da Silva AH, Guimarães JG, Sampaio AALEM, Silva EM. Analysis of gap formation at tooth-composite resin interface: Effect of C-factor and light-curing protocol. J Appl Oral Sci 2007;15:270-4.
9Alqudaihi FS, Cook NB, Diefenderfer KE, Bottino MC, Platt JA. Comparison of internal adaptation of bulk-fill and increment-fill resin composite materials. Oper Dent 2019;44:E32-44.
10Tauböck TT, Jäger F, Attin T. Polymerization shrinkage and shrinkage force kinetics of high- and low-viscosity dimethacrylate- and ormocer-based bulk-fill resin composites. Odontology 2019;107:103-10.
11Rizzante FAP, Duque J, Duarte MAH, Mondelli RFL, Mendonça G, Ishikiriama SK. Polymerization shrinkage, microhardness and depth of cure of bulk fill resin composites. Dent Mater J 2019;38:403-10.
12Marovic D, Tauböck TT, Attin T, Panduric V, Tarle Z. Monomer conversion and shrinkage force kinetics of low-viscosity bulk-fill resin composites. Acta Odontol Scand 2014;73:474-80.
13Sampaio CS, Fernández AJ, Atria PJ, Cáceres E, Pardo Díaz CF, Anderson Z, et al. Volumetric polymerization shrinkage and its comparison to internal adaptation in bulk fill and conventional composites: A μCT and OCT in vitro analysis. Dent Mat 2019;35:1568-75.
14Feilzer AJ, de Gee AJ, Davidson CL. Setting stress in composite resin in relation to configuration of the restoration. J Dent Res 1987;66:1636-9.
15Soares CJ, Pizi ECG, Fonseca RB, Martins LRM. Influence of root embedment material and periodontal ligament simulation on fracture resistance tests. Braz Oral Res 2005;19:11-6.
16Perez CR. Alternative technique for class V resin composite restorations with minimum finishing/polishing procedures. Oper Dent 2010;35:375-9.
17Buurlage JW, Marone F, Pelt DM, Palenstijn WJ, Stampanoni M, Batenburg KJ, et al. Real-time reconstruction and visualisation towards dynamic feedback control during time-resolved tomography experiments at TOMCAT. Sci Rep 2019;9:18379.
18Ferracane JL, Mitchem JC. Relationship between composite contraction stress and leakage in class V cavities. Am J Dent 2003;16:239-43.
19Kim RJY, Kim YJ, Choi NS, Lee IB. Polymerization shrinkage, modulus, and shrinkage stress related to tooth-restoration interfacial debonding in bulk-fill composites. J Dent 2015;43:430-9.