The objective of this study was to evaluate the effects of glass-ceram
ic inserts on reducing the marginal gaps caused by polymerization shri
nkage in composite restorations. A light microscope was used to measur
e the largest gap at margins around restorations made in glass cylinde
rs and tooth cavities with and without adhesion promoters. Where the c
ylinder was not silanated, the average gap was less in samples contain
ing an insert than in those without. Two preparations were made in the
dentin of 20 human molars. In each molar one cavity was restored with
a dentin bonding agent and composite and the other with a dentin bond
ing agent and an insert seated in the composite. The average maximum g
ap width of restorations containing inserts was statistically less tha
n for those with only composite (paired t-test, P < 0.0001). When cons
idering the volume of composite displaced by the insert, these results
indicate that the use of a glass-ceramic insert decreased the margina
l gaps resulting from polymerization shrinkage.