Marginal and internal adaptation of different flowable composite restorations in class V cavities after thermomechanical cyclic loading: in vitro study

This in vitro study aimed to evaluate and compare the marginal and internal adaptation of five different flowable composite restorations in Class V cavities after thermomechanical cyclic loading. One hundred freshly extracted human premolars were selected. Standardized Class V cavities were prepared on the facial surface, with gingival margin located 1 mm coronal to cementoenamel junction. Teeth were randomly assigned into five groups (n = 20) according to the restorative material: Group I: Short fiber-reinforced flowable( NanovaPro Flow); Group II: Self-adhesive flowable(Vertise Flow); Group III: ORMOCER-based bulk-fill flowable(Admira fusion X-base); Group IV: Resin-based bulk-fill flowable(Venus Bulk Flow); Group V: Injectable flowable(Gaenial Universal Injectable). Each group was subdivided (n = 10/ subgroup) for evaluation either at baseline (Non-Thermomechanical cyclic loading, N-TMC) or after thermomechanical cyclic loading (TMC).The TMC protocol involved 5000 thermal cycles (5°±1℃ to 55°±1℃) and 100,000 cycles of mechanical loading (100 N, 4 Hz). Marginal and internal adaptation were evaluated using a scanning electron microscope (SEM) at 200x magnification, with gaps > 1 μm considered defective. Data were statistically analyzed using the Monte Carlo test to compare the studied groups, and the McNemar test was used to compare N-TMC and TMC results, with a statistically significant level set at (P ≤ 0.05). At baseline (N-TMC), all groups showed 100% perfect marginal adaptation (< 1 μm gap), with no significant differences between them (p = 1.0). After TMC, a significant deterioration in marginal adaptation occurred in all groups (p ≤ 0.007 for all within-group comparisons), with no significant differences between the groups (p = 0.08). For internal adaptation, there was a statistically significant difference in N-TMC results when comparing groups with p < 0.001, and no statistically significant difference between the groups after TMC p = 1.0. Finally, there was a statistically significant difference when comparing N-TMC with TMC in all groups with p < 0.001 except group V, where there was no statistically significant difference with P = 1.0. Within the limitations of this in vitro study, while all materials demonstrated good initial marginal and internal sealing, thermomechanical cycling significantly impaired the adaptation of both marginal and internal interfaces. The short fiber-reinforced flowable composite exhibited the most favorable performance, showing the least degradation after aging. This finding suggests that this material offers a promising solution for increasing the durability of Class V restorations.