To evaluate the role of compomers in secondary caries formation in vitro.
36 human teeth with caries-free enamel (coronal) and root surfaces underwent a fluoride-free prophylaxis. Class V cavities were prepared in buccal and lingual coronal and root surfaces. The teeth were divided into two groups with 3 treatment subdivisions [12 teeth per restorative material (6 with coronal restorations and 6 with root surface restorations)]). The preparations were restored with: (1) filled composite (Filtek Z250 with Single Bond), (2) fluoride-releasing compomer [Hytac with Prompt L-Pop (fluoride-releasing water-based adhesive)]; or (3) fluoride-releasing compomer [Dyract AP with Prime & Bond NT (fluoride-releasing acetone-based adhesive)]. 24 cavities (6 teeth with 2 cavities/tooth) were available with each treatment. The teeth were thermocycled (500 cycles at 5-50 degrees C) in synthetic saliva prior to artificial caries formation. After lesion formation, 5 longitudinal sections were taken from each specimen for polarized light evaluation. Mean surface lesion depths and wall lesion frequencies were determined.
The Filtek Z250 group had significantly (P < 0.05, ANOVA, DMR) greater mean body of the lesion depths in coronal (205 microm) and root surfaces (221 microm) than either the Hytac (119 microm coronal and 161 microm root surfaces) or the Dyract AP (127 microm coronal and 149 microm root surfaces) groups. Likewise, wall lesion frequency was significantly higher (P < 0.05, ANOVA, DMR) for the Filtek Z250 group (41% coronal and 37% root surfaces) compared with either the Hytac (31% coronal and 25% root surfaces) or Dyract AP (26% coronal and 21% root surfaces) groups. No differences in outer body of the lesion depths or wall lesion frequencies were found between Hytac and Dyract AP for coronal and root surface restorations (P > 0.05, ANOVA, DMR).