Marginal Integrity: Is the Clinical Performance of Bonded Restorations Predictable In Vitro?

Purpose: In vitro testing of dental materials is daily routine for the preclinical investigation of restoratives. Although clinical trials remain the ultimate instrument, preclinical screenings are still important. However, it is still not fully understood whether clinical performance is predictable in the lab. The aim of this paper is to combine known facts and recent results to answer the question concerning in vitro predictability of clinical marginal integrity and related outcome. Materials and Methods: (a) The literature in the field from 1990-2005 was analyzed regarding marginal integrity in vitro and in vivo, especially in frequently cited papers. (b) Five different adhesives, a 4-step etch-and-rinse adhesive (Syntac), a 3-step etch-and-rinse adhesive (OptiBond FL), a 2-step etch-and-rinse adhesive (Single Bond), a 2-step self-etching adhesive (Clearfil SE Bond), and a 1-step self-etching adhesive (Xeno III) were used for bonding of a resin composite (Tetric Ceram) in Class I cavities (n = 8 in vitro and n = 8 in vivo). In vitro, the restorations were thermomechanically loaded (TML; 100,000 with 50 N and 2500 cycles of 5°C/55°C) according to a previously published protocol. Replicas of restorations were analyzed initially and after TML (in vitro) and two years of clinical service (respectively). Results: (a) Marginal integrity is reliably predictable in laboratory in vitro studies by simulating clinical circumstances. However, marginal analyses of direct restorations in vitro still suffer from not being able to determine a lower boderline, ie, actually worse in vitro results may still result in acceptable restorations in vivo. (b) The in vitro-in vivo comparison revealed significantly better marginal adaptation in (enamel) margins when etch-and-rinse adhesives were used for bonding. After 2 years of clinical service, restorations bonded with self-etching adhesives did not clinically fail but exhibited significantly more marginal gaps. Conclusion: Clinical behavior of restoration margins is predictable. However, marginal adaptation is only one among several important aspects in restorative dentistry, ie, clinical outcome is not predictable from marginal integrity alone. Palabras clave: marginal integrity, gap, etch-and-rinse, self-etching, adhesives, resin composites.