Statement of problem: intact, prepared, and restored human maxillary premolars.
Material and methods: Fifty intact, noncarious human maxillary premolars were divided into 5 groups of 10 and were mounted with their roots imbedded in autopolymerized acrylic. In the first group, the teeth were intact with no preparation. In the other 4 groups, Class II MOD preparations were made with a water-cooled high-speed hand piece. In 1 group, the cavity preparations were restored with bonded CAD/CAM ceramic inlays. In 2 groups, the preparations were restored with bonded CAD/CAM composite inlays (acid etched or air particle abraded). In the final group, the teeth were prepared but unrestored. Specimens were tested individually in a universal testing machine, in which a 4.82-mm-diameter steel sphere plunger was mounted in the crosshead moving at 0.5 mm/min. The plunger contacted the facial and lingual triangular ridges beyond the margins of the restorations. Peak load to fracture (N) was measured for each specimen. Means were calculated and analyzed with analysis of variance (P</=.05).
Results: MOD preparations weakened the teeth by approximately 59%. Restoring the teeth with ceramic or composite inlays did not significantly strengthen the teeth under this testing system. Of the restored teeth, those restored with indirect composite inlays cemented following manufacturer's recommendations had the highest fracture resistance.
Conclusion: Within the limitations of this study, under static compression load testing, bonded inlay restorations did not strengthen maxillary premolars with large MOD preparations.