Strain Distribution at the Composite-Tooth Interface: An In Vitro Study Using Digital Image Correlation and Linear...

To compare polymerization-induced interfacial strain among two high-viscosity bulk-fill resin composites and a conventional reference composite using digital image correlation (DIC), and to evaluate regional differences in strain.

Thirty extracted human molars were prepared and randomly allocated to three restorative materials (n = 10 per group): Filtek One Bulk Fill, Tetric PowerFill, and Tetric EvoCeram as the conventional reference composite. Polymerization-induced interfacial strain was assessed using DIC. Eight strain values were exported for each of three predefined cavity regions (mesial, central, and distal) per tooth, yielding 720 values from 30 independent specimens. Data were analyzed using linear mixed-effects models with material and cavity region as fixed effects and random intercepts for tooth and tooth-by-region. Pairwise comparisons were performed using estimated marginal means with Tukey adjustment.

Mean strain was lowest for Filtek One Bulk Fill (0.68% ± 0.13%), followed by Tetric PowerFill (0.74% ± 0.17%) and Tetric EvoCeram (1.12% ± 0.23%). The final additive model yielded significant material (F[2, 715] = 419.34, p < 0.001) and region effects (F[2, 715] = 8.97, p < 0.001). Tukey-adjusted comparisons indicated differences between all materials, including Filtek One Bulk Fill versus Tetric PowerFill (adjusted p = 0.0054). The interaction did not improve model fit. Regional means were identical across teeth within each material group, and both random-intercept variances were zero, resulting in a singular fit and limiting inferential precision.

Under the present in vitro conditions, the analyzed dataset showed lower interfacial strain for both tested bulk-fill composites than for Tetric EvoCeram, with a small difference between the bulk-fill materials. Interpretation is constrained by the repeated regional means and singular model fit, and clinical relevance remains undetermined. The implemented DIC workflow provides a basis for further methodological refinement.