Purpose: This randomized controlled trial aimed to evaluate the 14-year clinical performance of a HEMA-free 1-step self-etch adhesive (1SEa) compared with that of a 3-step etch-and-rinse adhesive (3E&Ra).
Materials and Methods: 267 non-carious cervical lesions in 52 patients were restored with the microhybrid composite Gradia Direct (GC), bonded in random order either with the HEMA-free 1SEa G-Bond (GB; GC) or the 3E&Ra Optibond FL (OFL; Kerr), which is considered the gold-standard E&Ra (control). The restorations were followed over 14 years for retention, marginal adaptation and discoloration, and caries occurrence. Statistical analysis involved a logistic regression model with generalized estimating equations (2-way GEE model).
Results: The patient recall rate at 14 years was 63%. In total, 79 restorations (39 GB, 40 OFL) failed because of retention loss (GB: 19.4%, OFL: 19.6%), severe marginal defects, discoloration and/or caries (GB: 21.7%; OFL: 22.5%). The overall clinical success rate was 58.9% and 57.9% for GB and OFL, respectively. The number of restorations with an unacceptable marginal defect (GB: 14.5%; OFL: 19.2%) and deep marginal discoloration (GB: 18.2%; OFL: 13.2%) increased during the last 5 years. No significant difference in overall clinical performance was recorded between the two adhesives (p > 0.05). Changes in the medical health of some patients and recurrence of abrasion/erosion/abfraction increased the failure rate and retention rate.
Conclusion: After 14 years, restorations bonded with the HEMA-free 1SEa performed as well as those bonded with the 3E&Ra gold standard. Unacceptable marginal deterioration was the main reason for failure, followed by loss of retention.