A systematic review and meta-analysis was conducted to compare the retention rates of 3-step etch-and-rinse (3ER) adhesives with 1-step self-etch (1SE) adhesives in noncarious cervical lesions (NCCLs). The secondary outcomes were marginal integrity and marginal discoloration. Only randomized clinical trials (RCTs) that compared 1SE with 3ER in NCCLs were included. Controlled vocabulary and keywords were combined in the search strategy for PubMed/Medline, LILACS, BBO, Web of Science, Cochrane Library, grey literature, and IADR abstracts (1990-2018). The Cochrane Risk of Bias tool (RoB) was applied to eligible studies. Meta-analyses were conducted for retention rate and secondary outcomes at different follow-up times, using the random effects model. Heterogeneity was assessed with the Cochran Q test and I2 statistics. The GRADE approach was used to assess the quality of the evidence. After the removal of duplicates, title and abstract screening, 18 studies remained. Of these, 15 studies were used for meta-analysis. Fourteen out of these 15 were judged at "unclear" risk and 1 at "low" risk of bias. No significant differences between groups were observed in the different follow-up periods for retention rates 12 to 24 months (p = 0.66), 24 to 36 months (p = 0.21) and 60 months (p = 0.96). A significant difference in marginal integrity was found at 12 to 24 months (p = 0.04) and in marginal discoloration at 12 to 24 months (p = 0.003). There is no evidence that 3-step ER adhesives have better retention rates than 1-step SE adhesives in NCCLs.
Read full abstract