Résumé
Objective: To study the effect of hydroxyapatite (HA) based agents on the bonding properties of universal adhesive with different application modes, and to provide evidence for the use of adhesives after desensitization treatment. Methods: Sixty impacted third molars were extracted and selected (acquired from Department of Oral and Maxillofacial Surgery, College of Stomatology, Xi'an Jiaotong University). Four third molars were used to prepare 1 mm thick dentin disks and treated with 1% citric acid to simulate sensitive tooth models. The dentin surfaces were observed by scanning electron microscope (SEM) after treating with no desensitization (control group), desensitized by HA based toothpaste Biorepair and Dontodent Sensitive respectively (desensitizing toothpaste A group and B group), or HA paste treatment (desensitizing paste group ) (n=2 per group). The remaining teeth were selected to expose the mid-coronal dentin and establish dentin sensitivity models. Then, the specimens were divided into 4 former groups and received corresponding treatment. Each group was randomly divided into 2 subgroups, and intermediately strong universal adhesive (G-Premio Bond) was applied on the desensitized dentin by either etch-and-rinse mode or self-etch mode. Resin-dentin slice specimens (n=4 per subgroup), microtensile specimens (n=20 per subgroup) and slice specimens (n=6 per subgroup) were prepared. The microstructure and nanoleakage of the adhesive interfaces were observed by scanning electron microscopy (SEM). The microtensile strength (bond strength) and fracture mode were tested and recorded. The water permeability of the adhesive interface was observed by laser scanning confocal microscopy (LSCM). Results: SEM showed that desensitizing toothpaste and desensitizing paste could partially or entirely occlude most of the dentin tubules. For the etch-and-rinse mode, the bond strength of specimens treated with toothpaste A [(40.98±4.60) MPa], toothpaste B [(40.89±4.64) MPa] and HA paste [(41.48±3.65) MPa] was significantly higher than that of the control group [(38.58±4.28) MPa] (F=3.89,P<0.05). There was no statistically significant difference in bond strength among the 4 subgroups for self-etch modes (F=0.48,P>0.05). After desensitization, the bond strength of the control group and desensitizing groups in the self-etch mode was significantly higher than that in the etch-and-rinse mode (P<0.05). The overall fracture modes were mixed failure and interfacial failure in the control group and desensitizing groups. SEM showed speckled silver-stained particles deposited along the bottom of the hybrid layer on the bond interface of etch-and-rinse mode, and there were few silver-stained particles deposited on the bond interface of self-etch mode. LSCM showed continuous linear penetration in the hybrid layer of etch-and-rinse mode subgroups and discontinuous linear penetration in the hybrid layer of self-etch mode subgroups. Conclusions: HA based desensitizers have no adverse effect on the bond strength of intermediately strong universal adhesive and show good bonding performance accompanied with the self-etch mode.