Comparative evaluation of cervical microleakage in class II composite restorations using different liner and placement techniques: An in vitro study Authors: Shikha Ahirwal, Vathsalya Shetty, Jisha Pillai, Gunjan Gautam, Kushali Goswami, Krutika Chudasama, Sanpreet Singh Sachdev Published: 2025 ABSTRACT Background:Marginal integrity at sub-cementoenamel junction (CEJ) margins remains a clinical challenge in posterior composite restorations. Polymerization shrinkage and interfacial stress contribute to microleakage and restoration failure. The present in vitro study compared cervical microleakage among four restorative approaches in Class II cavities extending below the CEJ.Materials and Methods:Forty extracted human maxillary first premolars were prepared with standardized Class II box-only cavities (gingival margin 2 mm below CEJ) on both sides, thereby yielding a total of n = 80 restorations. Specimens were divided into four groups (n = 20) according to the restorative technique: Resin-modified glass ionomer cement (RMGIC) liner + nanohybrid composite (Group 1); flowable liner + nanohybrid composite (Group 2); incremental composite technique (Group 3); and bulk-fill composite (Group 4). Following thermocycling and fluorescein dye immersion, sections were examined under a stereomicroscope (×40), and leakage was scored on a four-point scale. Data were analyzed using Kruskal–Wallis and Dunn’s post hoc tests (P < 0.05).Results:All groups showed some degree of leakage. Mean dye penetration scores were: Group 1 = 3.75 ± 0.55, Group 2 = 3.65 ± 0.67, Group 3 = 1.90 ± 1.77, and Group 4 = 2.00 ± 1.72 (P < 0.05). Incremental and bulk-fill techniques demonstrated significantly less cervical microleakage than RMGIC-and flowable-lined restorations. No significant difference existed between Groups 3 and 4.Conclusion:Incremental and bulk-fill placement techniques achieved superior marginal sealing compared with restorations using RMGIC or flowable liners. Liners did not improve sub-CEJ sealing and may increase interfacial stress. Simplified adhesive strategies using low-shrinkage composites can enhance long-term marginal integrity. INTRODUCTION The pursuit of an ideal posterior restoration has evolved significantly from the era of silver amalgam toward the use of tooth-colored, adhesive resin composites. Early in the 1960s, Bowen introduced the first successful resin-based composite, revolutionizing restorative dentistry with materials that combined esthetics and adhesion to tooth structure.[1] Despite these advancements, posterior composite restorations continued to face clinical challenges such as polymerization shrinkage, marginal leakage, postoperative sensitivity, and secondary caries.[2] The challenges were even more prominent when the gingival margins extended below the cementoenamel junction (CEJ) owing to the uncontrollable infragingival environment.[3] Marginal integrity remains a critical determinant of restorative success, especially in Class II cavities where dentin or cementum forms the cervical margin.[45] Numerous studies have shown that polymerization shrinkage generates interfacial stress, compromising the adhesive bond and leading to microleakage and recurrent caries. Researchers have explored diverse strategies to mitigate this issue, such as using low-shrinkage composites, snowplow and preheating techniques, and bonded-base approaches using resin-modified glass ionomer cement (RMGIC) or flowable liners.[678910] While these techniques have shown variable success in improving marginal sealing, results remain inconsistent owing to differences in material chemistry, adhesive systems, and experimental conditions. Recent innovations in bulk-fill composites, including flowable and sculptable variants such as Smart dentin replacement (SDR), Tetric EvoCeram Bulk Fill, and Filtek Bulk Fill, claim to simplify placement while reducing polymerization stress through modified resin matrices and enhanced translucency.[11] However, questions persist regarding their performance at sub-CEJ margins and whether liners like RMGIC or flowable resin actually enhance cervical sealing or merely complicate restorative protocols. In this context, determining the optimal restorative strategy for Class II cavities with gingival margins in cementum is warranted. The present study, therefore, aimed to comparatively assess microleakage among four restorative approaches, including RMGIC liner, flowable liner, incremental composite, and bulk-fill composite techniques. The goal of the study was to provide clinical insight into the most effective method for minimizing cervical leakage and enhancing restoration longevity. MATERIALS AND METHODS The present in vitro experimental study was conducted in the department of conservative dentistry and endodontics, in collaboration with the department of clinical pharmacology, from January 2025 to March 2025. Forty freshly extracted, intact human maxillary first premolars indicated for orthodontic extraction were selected. Teeth with caries, restorations, fractures, cracks, or developmental defects were excluded. All samples were disinfected using 0.5% sodium hypochlorite and stored in 0.9% saline at room temperature until use, following infection-control protocols recommended by OSHA and CDC.[1213] Experimental setup A customized jig was fabricated using an addition silicone impression material (DMG Silagum Putty STD) within a plastic mold to simulate intraoral conditions and provide structural support. The extracted teeth were mounted in groups of three, replicating natural proximal contact with a maxillary canine on the mesial side and a maxillary second premolar distally. Standardized class II box-only cavities were created on both mesial and distal surfaces using a diamond-point bur (SF11, Mani Inc., ISO 111/012; 0.8 mm diameter, 8 mm length) operated in an air–water-cooled high-speed turbine handpiece (NSK Pana, Japan). To maintain consistent cutting efficiency, a fresh bur was used after every eight preparations. In total, 80 Class II box-only cavities were prepared on 40 teeth, each measuring 4 mm bucco-lingually, 8 mm vertically, and 1.6 mm in width. The gingival floor of each cavity was positioned 2 mm apical to the CEJ, ensuring placement in dentin/cementum. The desired cavity dimensions were standardized by marking the diamond point prior to preparation and verifying measurements with a Williams periodontal probe (API AshooSons, India). The axial depth corresponded to twice the diameter of the instrument to maintain uniformity across samples. Each prepared surface was conditioned with 37% phosphoric acid for 15 s, rinsed thoroughly, and gently air-dried. A fifth-generation bonding agent (Adper Single Bond 2, 3M ESPE) was applied in two coats and light-polymerized for 20 s. Transparent matrix bands and light-transmitting wedges were positioned to achieve optimal contour adaptation [Figure 1]. (a) Matrix band and wedge placed after preparation of proximal cavities, (b) Acid etching, (c) Application of bonding agent, (d) Light curing of the bonding agent Grouping and restorative procedures The 40 teeth were randomly divided into four groups (n = 10 per group) according to the restorative protocol illustrated in Figure 2: Placement of the first increment of (a) Resin-modified glass ionomer cement liner, (b) Flowable resin composite, (c) Packable composite by incremental technique, and (d) Bulk fill composite resins Group 1: A 2 mm layer of RMGIC (GCFujiCEM) was placed on the gingival floor and overlaid with nanohybrid composite resin (Tetric N-Ceram, Ivoclar Vivadent) using a horizontal incremental technique in three 2 mm layers from the cervical to the occlusal surfaceGroup 2: A 2 mm base of flowable composite (Tetric N-Flow, Ivoclar Vivadent) was applied to the gingival floor, followed by nanohybrid composite resin (Tetric N-Ceram, Ivoclar Vivadent) using the same incremental horizontal technique as Group 1Group 3: Cavities were restored entirely with nanohybrid composite resin (Tetric N-Ceram, Ivoclar Vivadent) in three horizontal increments (each 2 mm thick) from cervical to occlusal directionGroup 4: Cavities were restored with flowable bulk-fill nanohybrid composite (SDR bulk-fill, Dentsply) in a single 4 mm increment, followed by a 2 mm occlusal capping layer of composite resin (Esthet-X HD, Dentsply). Group 1: A 2 mm layer of RMGIC (GCFujiCEM) was placed on the gingival floor and overlaid with nanohybrid composite resin (Tetric N-Ceram, Ivoclar Vivadent) using a horizontal incremental technique in three 2 mm layers from the cervical to the occlusal surface Group 1: A 2 mm layer of RMGIC (GCFujiCEM) was placed on the gingival floor and overlaid with nanohybrid composite resin (Tetric N-Ceram, Ivoclar Vivadent) using a horizontal incremental technique in three 2 mm layers from the cervical to the occlusal surface Group 2: A 2 mm base of flowable composite (Tetric N-Flow, Ivoclar Vivadent) was applied to the gingival floor, followed by nanohybrid composite resin (Tetric N-Ceram, Ivoclar Vivadent) using the same incremental horizontal technique as Group 1 Group 2: A 2 mm base of flowable composite (Tetric N-Flow, Ivoclar Vivadent) was applied to the gingival floor, followed by nanohybrid composite resin (Tetric N-Ceram, Ivoclar Vivadent) using the same incremental horizontal technique as Group 1 Group 3: Cavities were restored entirely with nanohybrid composite resin (Tetric N-Ceram, Ivoclar Vivadent) in three horizontal increments (each 2 mm thick) from cervical to occlusal direction Group 3: Cavities were restored entirely with nanohybrid composite resin (Tetric N-Ceram, Ivoclar Vivadent) in three horizontal increments (each 2 mm thick) from cervical to occlusal direction Group 4: Cavities were restored with flowable bulk-fill nanohybrid composite (SDR bulk-fill, Dentsply) in a single 4 mm increment, followed by a 2 mm occlusal capping layer of composite resin (Esthet-X HD, Dentsply). Group 4: Cavities were restored with flowable bulk-fill nanohybrid composite (SDR bulk-fill, Dentsply) in a single 4 mm increment, followed by a 2 mm occlusal capping layer of composite resin (Esthet-X HD, Dentsply). All restorations used shade A2 composite. Each increment was light-cured for 20 s from the buccal and lingual embrasures and an additional 20 s from the occlusal aspect using an LED curing unit (Koden LED light-curing device, 450 mW/cm2 intensity). The curing intensity was verified with a radiometer (Litex) before and during use. After polymerization, the matrix bands and wedges were carefully removed. Restorations were stored for 24 h, then finished with fine diamond burs and polished with composite discs (Super-Snap Mini-Kit, Shofu). The details of each material used in the present study are elaborated in Annexure 1. Microleakage assessment Specimens underwent thermocycling for 250 cycles between 5°C and 60°C with a 15-s dwell time in each bath. After thermocycling, the samples were detached from the silicone jig and coated with two layers of nail varnish (Maybelline), leaving the gingival margins of the restorations exposed. The root apices were sealed with modeling wax (MAARC, ISO 9001:2008). The specimens were then immersed in fluorescein dye (Samarth Life Sciences, India) and incubated at 37°C for 24 h. After immersion, they were rinsed under running water for 10 min to eliminate any superficial dye residue. Microscopic evaluation and scoring A total of 80 restorations (20 per group) were examined under a stereomicroscope (Leica L2) at ×40 magnification to evaluate dye penetration. Microleakage was graded on a five-point scale as follows: 0 = no dye penetration; 1 = penetration limited to enamel; 2 = penetration extending up to two-thirds of the cervical wall; 3 = penetration beyond two-thirds but not reaching the pulpal wall; and 4 = penetration involving the pulpal wall. Statistical analysis Data were analyzed using SPSS software (version 25.0) (v25.0, IBM Corp., Armonk, NY). Descriptive statistics (mean, standard deviation) were calculated for all groups. The Kruskal–Wallis test was used to determine intergroup differences, followed by Dunn’s post hoc pairwise comparisons. P < 0.05 was considered statistically significant. RESULTS All specimens exhibited some degree of marginal leakage along the gingival wall, varying according to the restorative technique used. The frequency distribution of dye penetration scores [Table 1] showed that Groups 1 (RMGIC liner) and 2 (flowable liner) had the highest percentage of severe leakage (Score 4), at 80% and 75%, respectively. Conversely, Group 3 (incremental composite technique) and Group 4 (bulk-fill composite) exhibited lower frequencies of high-grade leakage, with 35% and 30% of samples, respectively, showing Score 4 penetration. None of the specimens in Groups 1 and 2 showed complete absence of leakage (Score 0), whereas 35% of the restorations in Groups 3 and 4 exhibited no detectable dye penetration. Table 1Frequency distribution of dye penetration scores among study groupsScoreGroup 1 (RMGIC liner), n (%)Group 2 (flowable liner), n (%)Group 3 (incremental technique), n (%)Group 4 (bulk-fill technique), n (%)0007 (35)7 (35)1003 (15)1 (5)21 (5)2 (10)2 (10)3 (15)33 (15)3 (15)1 (5)3 (15)416 (80)15 (75)7 (35)6 (30)Total (n)20202020RMGIC: Resin-modified glass ionomer cement Frequency distribution of dye penetration scores among study groups RMGIC: Resin-modified glass ionomer cement The mean dye penetration values were 3.75 ± 0.55 for Group 1, 3.65 ± 0.67 for Group 2, 1.90 ± 1.77 for Group 3, and 2.00 ± 1.72 for Group 4 [Table 2]. The Kruskal–Wallis test revealed a statistically significant difference among the four groups (χ2 = 22.793; P = 0.000), indicating that the restorative technique significantly influenced microleakage. Subsequent pairwise comparisons using Dunn’s post hoc test demonstrated statistically significant differences between the mean depth of penetration of Group 1 and Groups 3 and 4 (P < 0.05 each), and between Group 2 and Groups 3 and 4 (P < 0.05 each). In contrast, no significant difference was observed between Groups 1 and 2 or between Groups 3 and 4 [Table 3]. Table 2Descriptive statistics of dye penetration among the four restorative techniquesGroup n MeanSDSEMinimumMaximumGroup 1 (RMGIC liner)203.750.550.1224Group 2 (flowable liner)203.650.670.1524Group 3 (incremental technique)201.901.770.4004Group 4 (bulk-fill technique)202.001.720.3804Total802.821.560.1704SD: Standard deviation, SE: Standard error, RMGIC: Resin-modified glass ionomer cement Descriptive statistics of dye penetration among the four restorative techniques SD: Standard deviation, SE: Standard error, RMGIC: Resin-modified glass ionomer cement Table 3Dunn’s post hoc pairwise comparison among groupsComparison P SignificanceGroup 1 versus Group 21.00NSGroup 1 versus Group 30.0025SignificantGroup 1 versus Group 40.0025SignificantGroup 2 versus Group 30.0045SignificantGroup 2 versus Group 40.0045SignificantGroup 3 versus Group 41.00NSP<0.05. NS: Not significant Dunn’s post hoc pairwise comparison among groups P<0.05. NS: Not significant Visual examination under stereomicroscopy corroborated these results, as specimens in Groups 1 and 2 exhibited deep dye penetration extending to the pulpal wall, whereas Groups 3 and 4 displayed dye limited to the cervical wall or enamel margins [Figure 3]. Stereomicroscopic image of fluorescein dye penetration in (a) Resin-modified glass ionomer cement liner, (b) Flowable composite liner, (c) Incremental technique, and (d) Bulk-fill technique DISCUSSION The results of the present study indicated that restorations placed with RMGIC or flowable liners demonstrated higher microleakage than those placed using incremental or bulk-fill methods. Among all groups, the incremental technique (Group 3) recorded the least mean microleakage, closely followed by the bulk-fill technique (Group 4). The findings suggest that the use of liners beneath composite restorations at sub-CEJ margins may not enhance marginal sealing and, in fact, may exacerbate polymerization stress and interfacial gap formation. These findings are consistent with the growing body of evidence emphasizing that marginal adaptation at sub-CEJ levels is predominantly governed by polymerization shrinkage stress and adhesive performance rather than the mere presence of a liner layer.[14] The greater microleakage observed in RMGIC and flowable liner groups can be attributed to multiple interfacial boundaries that introduce additional zones of potential stress accumulation.[15] The hydrophilic nature and limited cohesive strength of RMGIC make it susceptible to interfacial failure under polymerization contraction, particularly when bonded to resin composites that undergo volumetric shrinkage.[16] Similar findings were reported by Loguercio et al. and Tabatabaei et al., who demonstrated that open-sandwich restorations permitted greater dye ingress along the gingival interface due to micro-fracture of the RMGIC layer during curing stress transmission.[1718] Flowable liners, although intended to act as stress-absorbing intermediates, exhibit higher polymerization shrinkage and a lower elastic modulus owing to reduced filler loading, which can negate any theoretical stress-relief benefits.[15] The incremental and bulk-fill techniques in the present study produced superior marginal seals. Incremental layering reduces the configuration factor (C-factor) by limiting the bonded surface area per increment, thereby distributing contraction stress more evenly. Comparable results have been reported by Jacob and Goud and Ranjini et al., who observed that multiple incremental placements significantly reduced cervical leakage in deep proximal boxes.[67] Similarly, bulk-fill composites such as SDR contain modified urethane dimethacrylate monomers with stress-relieving molecular structures that lower polymerization contraction without compromising depth of cure. Feiz et al. and Al-Nahedh corroborated this observation, noting that bulk-fill systems demonstrated lower microleakage than conventional nanohybrids owing to their enhanced flow, increased translucency, and polymerization kinetics optimized for deep curing.[819] From a clinical standpoint, these outcomes highlight that restorative techniques minimizing interfacial complexity and polymerization stress, such as incremental or bulk-fill placement, are more effective in maintaining the marginal integrity of Class II restorations placed below the CEJ.[20] The results also suggest that the use of liners may not always be advantageous in posterior composites, as they can inadvertently introduce mechanical and adhesive incompatibilities between materials. Clinicians should, therefore, prioritize adhesive strategies and resin systems with low shrinkage potential rather than relying on intermediary layers for stress modulation.[21] The limitations of the present study include its in vitro design, which does not fully simulate intraoral conditions such as masticatory load, occlusal wear, and long-term hydrothermal aging. Only one brand and formulation of each material type were tested, which may restrict the generalizability of the results to all commercial systems. Future studies should conduct longitudinal in vivo evaluations comparing universal adhesives, newer bioactive liners, and novel low-shrinkage resin matrices to validate the clinical translation of these findings. CONCLUSION Within the limitations of this in vitro study, it can be concluded that Class II composite restorations placed with incremental or bulk-fill techniques exhibited significantly less cervical microleakage than those restored using RMGIC or flowable liners. The incorporation of an intermediate liner did not improve marginal sealing at sub-CEJ margins and may increase interfacial stress due to a mismatch in polymerization shrinkage. The findings emphasize that minimizing interfacial complexity and employing low-shrinkage, stress-relieving composite systems are key to achieving durable marginal integrity. Further long-term in vivo investigations are warranted to substantiate these outcomes under functional oral conditions. Conflicts of interest There are no conflicts of interest.