Adhesive strategies in cervical lesions: systematic review and a network meta-analysis of randomized controlled trials

Abstract

A network meta-analysis (NMA) was performed to assess which adhesive strategy is most clinically effective in treating non-carious cervical lesions (NCCLs). Studies were identified by a systematic search of electronic databases including MEDLINE via PubMed, Brazilian Library in Dentistry (BBO), Cochrane Library, EMBASE, Latin American and Caribbean Health Sciences Literature database (LILACS), Scopus, and Web of Science without restrictions on publication year or language. The grey literature was also consulted. Only randomized clinical trials that compared different adhesive strategies in NCCLs in adult patients were included. The risk of bias was evaluated by using the Cochrane Collaboration tool. A random-effects Bayesian mixed treatment comparison model was used to compare adhesive strategies (3ER, 2ER, 2SE, and 1SE) at different follow-up times. The surface under cumulative ranking curve (SUCRA) was estimated for each strategy. Heterogeneity was assessed by using the Cochran Q test and I2 statistics. The quality of evidence was evaluated using the GRADE approach. A total of 5058 studies were identified, 66 of which met the eligibility criteria and of these 5 were judged "low" risk of bias and 57 were meta-analyzed. We did not observe significant differences in the NMA analysis for any two pairs of adhesives, except for the shortest follow-up for 2ER vs 3ER. The material 2SE ranked highest, although it differed only slightly from the other bonding strategies. No bonding strategy is better than the others. Adhesive efficacy cannot be characterized by its bonding strategy.