This is an in-vitro study conducted on human molars after obtaining the Institutional Ethical committee clearance from Manipal College of Dental Sciences, Mangalore, India. The study was conducted in March and April 2018.

Sample selection and storage: Forty extracted human molars with no caries, wear defects, fracture line or cracks were included in the study. Surface debridement was done with hand-scalers, cleaning with a rubber cup applied with slurry of pumice, and was subsequently stored in distilled water at room temperature until use.

Sample preparation: Class V cavities, rectangular in shape (2 mm in occluso-gingival height, 4 mm in mesio-distal width and 2 mm in depth) were prepared with high-speed air-water cooled handpiece on the buccal side of extracted molars, using tungsten carbide straight fissure bur. Cavity preparations were located in the Cemento-Enamel Junction (CEJ), with the coronal margin in the enamel region, extending cervically to the cementum region. From the pool of forty teeth with class V cavities prepared, 10 teeth were randomly assigned to the four study groups.

Restorative procedure in each group: All the cavities were restored with composite resin Filtek Z350 XT (3M ESPE, St. Paul, MN, USA).

Preparation of the samples for microleakage: After restorative procedures were done, the samples were polished using (Sof-Lex (3M ESPE, St. Paul, MN, USA) and was kept in saline when not in use. The apices of the samples were sealed and two coats of nail varnish were used to cover the specimens, excluding the restoration surface and the surrounding 2 mm area. The specimens were inverted and placed in a solution of 2% Rhodamine-B (Lobachemie, India) dye for 24 hours. After removal of the specimens from the dye solution, longitudinal sections of the teeth were done bucco-lingually through the centre of the restorations using a low-speed diamond disc (Brasseler USA, Savannah, GA). The sections were then evaluated with a fluorescent microscope (Olympus CX41, Olympus Microscopy Europa) at 10x magnification to determine the extent of dye penetration at the cavity margins by two evaluators who were blinded to the study.

Dye scoring criteria: The level of dye penetration was evaluated according to the scoring system mentioned in [Table/Fig-1] [7]. In Score 0 there is no dye penetration; in Score 1, the dye penetration is present less than half of the cavity depth along the external wall; in Score 2 the dye is present along the external wall involving more than half of the cavity without extending on to the axial wall and in Score 3, the dye penetration is present along the full cavity depth and extending onto the axial wall.

[Table/Fig-1]:

Scoring criteria: (a) No dye penetration (Score 0); (b) Dye penetration along the external wall which is less than half the cavity depth (Score 1); (c) Dye penetration along the external wall which is more than half the cavity depth but not extending onto the axial wall (Score 2); (d) Dye penetration along the full external wall extending onto the axial wall (Score 3).

Microleakage scores based on the adhesive strategy used and the bonded substrate were recorded and the data obtained was subjected to statistical analysis using Chi-square test. The statistical analysis was carried out using SPSS Version 16.0 (SPSS Inc., Chicago IL). For the analysis, the level of significance was set at p<0.05.

The inter-examiner agreement of scores of enamel microleakage assessment was 23/34 which was 67.64% with a kappa value of 0.552, indicating moderate agreement (p=<0.001). For the microleakage score of the cementum wall, the agreement of scores was 17/27, which was 62.9% with a kappa value of 0.552, indicating moderate agreement. The microleakage scores for enamel and cementum have been given in [Table/Fig-2]. No significant difference was detected between the groups (p>0.05). Thus the null hypothesis was accepted. However, Group 3 (self-etch without dye) showed the least percentage of microleakage with none of the restorations showing grade 3 leakage score. Though not statistically significant, a slightly higher incidence of increased leakage was seen in the total etch group, especially in the cementum margin.

Microleakage is a big concern in the dental world as it is one of the major causes of apical periodontitis and an important cause of root canal treatment failure [8]. Various methodologies have been introduced to assess degree of microleakage. Dye penetration method is found to be dependable in microleakage assessment [9]. It is highly feasible and easily reproducible [10]. Guidelines were followed by emphasising the need to standardise tooth quality, type of cavity preparation, and method to evaluate microleakage at the margin to maintain transparency. Rhodamine B dye, which is a fluorescent dye, was used to assess degree of microleakage due to its fluorescent nature. Rhodamine B dye is also more sensitive than a conventional dye. On the other hand, Rhodamine B dye was used instead of methylene blue as it has smaller and more surface active molecules [11].

In the present study, microleakage was observed to be slightly more in cemental margin, compared to the enamel margin, though not statistically significant. It was in accordance with the study by Owens BM et al., [12]. This could be due to the presence of coarse collages fibres in cementum resulting in less and inhomogeneous penetration of resin monomers into the etched zone [6,13].

The present study showed increased microleakage in groups using the total etch system compared to the self-etch system. This result is in accordance with the study done by Gupta A et al., [14]. In total etch adhesives; the smear layer is completely dissolved. However, in self-etch adhesives, the acidic active components are responsible for dissolving the smear layer. The acidity is buffered by the mineral content of dentin, thus chemical interaction between some functional monomers and the remaining hydroxyapatite crystals along the collagen fibrils improve bond durability [15]. Self-etch adhesives are easy to use and theoretically able to etch and infiltrate at the same time, decreasing the inconsistency between demineralisation and infiltration [16].

In the present study, the use of caries detecting dye showed an increase in microleakage percentages compared to the absence of the dye. Although no statistical difference was shown through the results, the microleakage percentages in samples which used caries detecting dyes, were shown to be higher than the ones without dye. The higher incidence of dye leakage when compared to the study by Piva E et al., could be due to the difference in the composition of the caries disclosing solution used [7]. According to Demarco FF et al., the dye can be trapped in the dentin even after washing or acid etching [17]. This dye particle can affect the bonding of composite to the tooth structure. Thus, use of caries detecting dye should be done with caution, especially with adhesive restorative materials.

Being an in-vitro study, this study had its own limitations. Intraoral variables like temperature changes as well as effects of masticatory stresses and fatigue load were not taken into consideration. Further studies need to be conducted to evaluate the adhesion of composite resin, especially on caries affected dentin, after the use of various caries detecting dye to simulate the clinical scenario better.

Within the limitations of this study, it was possible to conclude higher leakage, though not statistically significant was seen in the total etch group than the self-etch adhesive group. In addition, the cavities exposed to caries detecting agent showed more leakage than those without its usage.