To evaluate interfacial three-dimensional adaptation and internal voids of different flowable materials before and after cyclic fatigue in a simulated deep-margin elevation scenario. Eighty (n = 80) extracted premolars were selected and two Class II cavities were prepared. The mesial one with cervical margin 1 mm above the cementum-enamel junction (CEJ) and the distal one with cervical margin 1 mm below the CEJ. After performing adhesive procedures, specimens were divided into four groups according to the employed materials for 2 mm horizontal deep-margin relocation: nanohybrid composite (Clearfil ES2, Kuraray); conventional viscosity flowable composite (Tetric Flow, Ivoclar); medium viscosity flowable composite (Majesty ES2 Low Flow, Kuraray); high viscosity flowable composite (Majesty ES2 Super Low Flow, Kuraray). All restorations were finalized by oblique layering with nanohybrid composite (Clearfil ES2, Kuraray). To reveal interfacial and internal gap progression, specimens were scanned with a micro-CT (SkyScan 1172), before and after 500,000 cycles of mechanical chewing simulation (50 N, 1 Hz). Data were imported into Mimics software after smoothing and region growing. Only the 2 mm margin relocation volumes were considered. Obtained masks were analyzed for noise removal and volume calculation. At baseline, interfacial gap progression and internal voids, expressed in mm3, were collected and statistically analyzed with two-way ANOVA (α 0.05) for the variables substrate and restorative materials followed by Tukey post-hoc test. An additional two-way ANOVA test, followed by Tukey post-hoc test, was performed to evaluate variation in interfacial gap progression after mechanical aging. At baseline, the ANOVA test showed a significant difference for the variable restorative materials (p = 0.01). More specifically, the Tukey post-hoc test revealed that the highly filled medium viscosity composite performed better than the conventional viscosity composite at baseline for the interfacial gap. The internal voids ANOVA test at baseline reported no significant differences for the variable tested. Analysis of variance for internal gap progression after thermocycling showed no differences for both substrate and restorative material employed. Highly filled medium viscosity composite performed significantly better than the conventional viscosity flowable composite for what concern baseline interfacial gaps. Artificial aging with a chewing simulator and thermocycling did not affect interfacial gap progression on enamel and dentin. The tested restorative materials performed equally after aging.