SUMMARY: Biofilm (bacterial plaque) accumulation on the surface of restorative materials favors the occurrence of secondary caries and periodontal inflammation. Surface characteristics of restorations can be modified by finishing and/or polishing procedures and may affect bacterial adhesion. The aim of this systematic review was to characterize how finishing and polishing methods affect the surface properties of different restorative materials with regard to bacterial adhesion and biofilm formation. Searches were carried out in MEDLINE-PubMed, EMBASE, Cochrane-CENTRAL, and LILACS databases. From 2882 potential articles found in the initial searches, only 18 met the eligible criteria and were included in this review (12 with in vitro design, four with in situ design, and two clinical trials). However, they presented high heterogeneity regarding materials considered and methodology for evaluating the desired outcome. Risk bias analysis showed that only two studies presented low risk (whereas 11 showed high and five showed medium risk). Thus, only descriptive analyses considering study design, materials, intervention (finishing/polishing), surface characteristics (roughness and surface free energy), and protocol for biofilm formation (bacterial adhesion) could be performed. Some conclusions could be drawn: the impact of roughness on bacterial adhesion seems to be related not to a roughness threshold (as previously believed) but rather to a range, the range of surface roughness among different polishing methods is wide and material dependent, finishing invariably creates a rougher surface and should always be followed by a polishing method, each dental material requires its own treatment modality to obtain and maintain as smooth a surface as possible, and in vitro designs do not seem to be powerful tools to draw relevant conclusions, so in vivo and in situ designs become strongly recommended.

To evaluate the substrate masking ability of different ceramic systems used for complete crowns and assess the influence of their association with an opaque resin-based luting agent (OLA). Eleven ceramic groups were tested (n=10). Bilayer groups were: ZrPc - zirconia+porcelain; CAD-onHT - zirconia+high translucent lithium disilitace; CAD-onLT - zirconia+low translucent lithium disilicate; LDPc - high opaque lithium disilicate+porcelain. Monolithic groups were: TZ - high translucent zirconia; TLS - translucent, zirconia-reinforced lithium silicate; HTLS - high translucent, zirconia-reinforced lithium silicate; LTLD - low translucent lithium disilicate; HTLD - high translucent lithium disilitace; LGC - leucite-reinforced glass ceramic; FC - feldsphatic ceramic. The substrates adopted were resin composites shaded A1 (control), A3.5 and C4; coppery and silvery metals. Color differences (ΔE00) and translucency parameter (TP00) were assessed with the CIEDE2000 formula. ΔE00 was obtained assessing the ceramic influence only and in association with OLA. Zirconia groups presented lower ΔE00 and TP00. CAD-on structures associated to OLA led to the lower ΔE00 when tested over metals. ΔE00 for LDPc were similar to zirconia when associated to OLA. Monolithic groups presented significant higher ΔE00 and TP00 compared to bilayers, with ΔE00 below the acceptability threshold only when associated to OLA over substrate shade A3.5. Ceramic type and the white-opaque resin-based luting agent presented a significant effect on the substrate masking ability. All the discolored substrates tested are adequately masked with veneered zirconia or with LDPc (preferably associated to OLA). The CAD-on technique associated with OLA improved the masking ability over metallic substrates.