Objectives:
To investigate how surface treatment affects fracture force, flexural strength, and dynamic loading cycles until failure of 3D-printed restorations.

Materials and methods:
Specimens (7 groups; n = 8 per group) were 3D-printed from an acrylate-based crown and bridge material. After cleaning and post-polymerization, specimens were treated with either silicon carbide paper (1000 grit; 1000/4000 grit) or blasting (Al2O3; 1 bar/125 µm; 2 bar/125 µm; 1 bar/250 µm) to simulate laboratory treatment. Surface roughness (Arithmetic mean Sa/maximum roughness height Sz; ISO 25178-2); fracture force (FF) and biaxial flexural strength (BFS; ISO 6872) were determined. The number of dynamic load cycles (LC) to failure was determined under cyclic loading in a BFS staircase approach. Statistics: ANOVA, Bonferroni-test, Kaplan-Meier survival, Pearson correlation; α = 0.05.

Results:
BFS ranged between 94.4 MPa and 199.9 MPa, FF between 260.6 N and 428.6 N and Sa/Sz between 0.0/1.0 μm and 1.8/18.4 μm. BFS, FF and Sa/Sz showed significant differences between the treatments (p < 0.001) and individual groups (p ≤ 0.013). Mean LC ranged between 204,364 and 267,637 cycles. ANOVA comparisons (p = 0.706) and Log Rank test (Chi2: 10,835; p = 0.094; Fig. 2) revealed no significant differences between the loading cycles. Surface treatment with either silicon carbide papers or blasting protocols had a significant influence on FF, BFS, Sa, and Sz, but not on LC.

Conclusions:
Surface treatment affected the fracture force and biaxial fracture strength of a 3D-printed crown. It showed no influence on the long-term dynamic behavior.

Clinical relevance:
Smooth surfaces improve the stability of a restoration fabricated from 3D-printing resins. Extensive surface roughness treatment before cementation can reduce the stability of a crown.