Method: Two premolars with NCCLs mostly in dentin were randomly assigned, pumice cleaned only and restored by two clinicians in 30 adult subjects using either a self-etch 1-step or etch-and-rinse 3-step adhesive and nanofilled composite (Bond Force/Estilite Sigma Quick-BF/ESQ, Tokuyama; Optibond FL/Premise™-OPFL/P). Cotton and cord was used for isolation in all but four restorations. Esthetic (surface luster and stain, marginal stain, color match, anatomic form), functional (retention, marginal adaptation, patient’s view) and biologic properties (sensitivity, caries, tooth integrity, periodontal response, adjacent mucosa) were collected by two blinded and calibrated clinical evaluators at baseline, 6-, 12-, 24- and 36-months using Hickel et al., 2007 and 2010 criteria. McNemar’s test and Bowker's test of symmetry were used to compare restorations and Kaplan-Meier survival analysis was conducted to evaluate the survival distributions of length of overall clinical success between the two restorations (statistically significant if p<0.05, marginally significant if 0.05
Result: Retention over 36-months: OPFL/P one partially lost and BF/ESQ four lost restorations with all failed restorations from one clinician. 23 patients were available for recall at 36-months. Within groups as compared to baseline: BF/ESQ had significantly worse marginal stain (p=0.0186), retention (p=0.0455), marginal adaptation (p=0.0018); OPFL/P had improved periodontal response (p=0.0455). Between groups BF/ESQ had significantly better color match at baseline (p=0.0107) and significantly worse marginal adaptation (p=0.0252) at 36-months. A marginally significant difference between restoration survival (p=0.0568 log-rank test and p=0.0751 Wilcoxon test) was found with estimated 36-month probabilities of survival for BF/ESQ and OPFL/P of 0.75 and 0.92, respectively.
Conclusion: OPFL/P had significantly better clinical performance in NCCLs through 36-months as compared to BF/ESQ.