June 2006
Volume 2, Issue 5

In Vivo Versus in Vitro Microtensile Bond Strength of Axial Versus Gingival Cavity Preparation Walls in Class II Resin-Based Composite Restorations

Howard E. Strassler, DMD

Purk JH, Dusevich V, Glaros A, et al. J Am Dent Assoc. 2004;135(2):185-193.


BACKGROUND: Gingival margins in Class II composite restorations are a site of frequent failure. The purpose of the authors’ study was to compare the microtensile dentin bond strength of gingival and axial restored cavity preparation walls of Class II composite restorations under in vivo and in vitro conditions. METHODS: After obtaining informed consent, the authors placed Class II resin-based composite restorations in 14 premolar teeth from five patients, under in vivo or in vitro conditions. The teeth were sectioned to obtain rectangular specimens from axial and gingival walls with a surface area of approximately 0.5 square millimeters. The authors tested 85 microtensile adhesive samples from the 14 teeth on a testing instrument (Universal Instron, Model 125, Instron, Canton, MA) until failure. RESULTS: The mean (+/ standard deviation) mircotensile dentin bond strengths in mega-pascals were as follows: in vivo axial, 36.5 (14.9); in vivo gingival, 17.6 (11.6); in vitro axial, 49.5 (13.9); in vitro gingival, 34.0 (13.1). A two-way analysis of variance found a statistically significant difference between in vitro and in vivo conditions and between the axial and gingival walls (P < or = .001). Eighty-eight percent of the fractured samples involved the adhesive layer as observed under scanning electron microscopy up to x2,500. Seventeen of the gingival samples and two of the axial samples debonded during the preparation phase and could not be tested. CONCLUSION: The dentinal microtensile strength of adhesive/resin-based composite bonded to the gingival wall was significantly weaker than the bond to the axial wall, and in vivo conditions produced significantly weaker bond strengths than did in vitro conditions. CLINICAL IMPLICATIONS: The dentinal adhesive bond of resin-based composite to gingival walls is significantly weaker and thus more subject to failure than the bond to axial walls. In vitro bond strength studies may overestimate the bond strength of adhesives in in vivo applications. between a self-etch adhesive and total-etch adhesive at 2 weeks. The use of a flowable composite did not decrease postoperative sensitivity.


There is no doubt that the trend in restoring posterior teeth is moving from the use of amalgam to composite resin. Amalgam has proved to be a very forgiving restorative material. Resin-based restoratives require multiple steps in their placement, including adhesion and light-curing. The protocol of this study was well done, and the products used are those that have provided success in past clinical studies and have a proven track record. In fact, the light-curing times used in the study, 40 seconds from the buccal, lingual, and occlusal directions are probably greater than most clinicians use. I suspect that shorter curing times would have provided worse data for the gingival wall. This study should provoke some thoughts on how to maximize the longevity of Class II resin-based restorations. The data presented is unsettling and should be viewed with some important considerations, such as the fact that adhesive resin-based restorations will most likely have better survivability when the margins end on enamel. When the gingival margins end on dentin, these restorations must be evaluated thoroughly at every recall for any evidence of degradation of the gingival margin.

About the Author
Howard E. Strassler, DMD
Professor and Director of Operative Dentistry
Department of Endodontics, Prosthodontics and Operative Dentistry
University of Maryland Dental School, Baltimore, Maryland

© 2015 AEGIS Communications | Privacy Policy