Randomized Clinical Trial of Four Adhesion Strategies in Non-Carious Cervical Lesion Restorations: Five-Year Follow-Up.

Several adhesion strategies have been proposed for restoring non-carious cervical lesions (NCCLs). Long-term clinical trials are essential to guide clinicians toward the most effective approach. PURPOSE: To evaluate the 5-year clinical performance of NCCLs restored using different adhesion strategies. MATERIALS AND METHODS: Four strategies were assessed: an adhesive system (Scotchbond Universal; SBU/Filtek Z350XT) applied without (SBU) and with selective enamel conditioning (E-SBU), and a resin-modified glass-ionomer cement (Vitremer; RMGIC) applied without (RMGIC) and with ethylene-diaminetetraacetic acid pretreatment (E-RMGIC). Two hundred restorations were evaluated using United States Public Health Service criteria after 5 years. Kappa test, Wilcoxon Signed-Rank test, Kaplan-Meier analysis, equality tests of two proportions, and multiple logistic regression were applied (α = 0.05). RESULTS: One hundred and nineteen restorations were re-evaluated after 5 years. No significant differences were observed in restoration survival among groups. E-SBU exhibited more Bravo scores for marginal integrity compared to the ionomer groups and for marginal discoloration compared to RMGIC. Ionomer groups had higher Bravo scores for surface texture. RMGIC showed greater wear than E-SBU, being statistically similar to SBU. No significant differences were found for color and secondary caries over time. Multiple logistic regression analysis revealed that restoration retention was significantly influenced by the type of tooth (premolars), the degree of dentin sclerosis, and the degree of marginal tissue recession. CONCLUSIONS: All four adhesion strategies demonstrated similar 5-year survival. Marginal defects were more frequent in the selective enamel etching group than in ionomeric restorations. Surface luster was reduced in ionomer restorations. The retention of NCCL restorations may be affected by tooth type, degree of dentin sclerosis, and marginal tissue recession. The four adhesion strategies had similar survival after 5 years; however, composite resin restorations applied after selective enamel etching promote more initial marginal defects than ionomer-based restorations following the manufacturer's instructions. Clinical factors such as lesion location, dentin sclerosis, and gingival recession can influence the retention of NCCL restorations.