Bonding to Dentin Contaminated with Ceramic-Repair Primers/Etchants.

PURPOSE: To evaluate bonding to dentin contaminated with primers/etchants used for adjacent ceramic repair. MATERIALS AND METHODS: Mid-coronal dentin of sound human third molars was exposed and allocated to 10 experimental groups. The universal adhesive (UA) Single Bond Universal ('SBU,' 3M Oral Care), applied either in etch-and-rinse (E&R) or self-etch (SE) bonding mode, and the considered gold-standard SE adhesive Clearfil SE Bond X ('CSE,' Kuraray Noritake) were bonded to dentin contaminated with either Monobond Etch & Prime ('MEP,' Ivoclar) or IPS Ceramic Etching Gel ('HF,' Ivoclar) following 10 scenarios: phosphoric acid (PA)+SBUE&R (uncontaminated E&R UA control), HF+PA+SBUE&R, MEP+PA+SBUE&R, PA+MEP+SBUE&R, SBUSE (uncontaminated SE UA control), HF+SBUSE, MEP+SBUSE, CSESE (uncontaminated SE control), HF+CSESE, MEP+CSESE. Upon adhesive and composite application, the specimens were stored in artificial saliva at 37°C. After 1 week, all specimens were sectioned into resin-bonded dentin sticks, which were randomly distributed over two groups to be subjected to a microtensile bond-strength test immediately at 1 week or upon aging by storage in artificial saliva for 6 months. Statistics involved linear mixed-effects modeling with Bonferroni correction (P 0.05). RESULTS: E&R bonding to dentin contaminated with MEP or HF did not significantly differ from bonding to non-contaminated dentin (controls). However, SE bonding to MEP- and HF-contaminated dentin was significantly less effective than to non-contaminated dentin (controls). Aging for 6 months did not reduce E&R bonding as compared to the 1-week data, while SE bonding was significantly less effective upon 6-month aging. E&R bonding was affected more when dentin was contaminated with MEP before phosphoric acid (PA) etching than when dentin was contaminated with MEP after PA etching. CONCLUSIONS: Dentin contamination with MEP and HF impacted self-etch (SE) bonding but not etch&rinse (E&R) bonding.