Permeability of demineralized dentin to HEMA.

OBJECTIVES: The objectives of this study were to develop a macromodel of the demineralized layer of dentin that is formed after acid-etching, and to determine if the amount of HEMA uptake was responsive to air drying. The hypothesis that was tested was that HEMA uptake by demineralized dentin depends upon the degree of expansion of the collagen fibril network. METHODS: Dentin cubes (2 x 2 x 2 mm) were prepared in coronal dentin from extracted unerupted human third molars. They were incubated in 100% HEMA for up to 1000 min and then removed and blotted free of excess adherent HEMA. The HEMA taken up by each cube was extracted in 2 ml of water for 1 hr with shaking. This was repeated and the extracts combined. HEMA was quantitated spectrophometrically. The dentin cubes were then demineralized in 0.5 M EDTA for 10 days and the HEMA uptake remeasured at 1, 10, 100 and 1000 min. Then the cubes were air-dried and the HEMA uptake remeasured. After re-expanding the specimens in water, the cubes were stiffened in ascending concentrations of acetone of 100%, then allowed to air dry again in a more expanded state, and HEMA uptake was remeasured. Finally, after rehydration, the dentin cubes were "acid-etched" for 1 or 10 min with 37 wt% phosphoric acid and HEMA uptake remeasured. RESULTS: Before demineralization, the dentin took up little HEMA (ca. 4.8 x 10(-7) moles min-3) compared to after demineralization when the uptake for 10, 100 and 1000 min was 27.4, 43.8 and 51.4 x 10(-3) moles mm-3, respectively. Acid etching for 1 or 10 min had no effect on HEMA uptake. Air drying produced a 72% volumetric shrinkage but a 97% reduction in uptake. When the demineralized matrix was stiffened in acetone prior to air-drying, the volumetric shrinkage was only 27% and the HEMA uptake only fell 16% compared to the wet, fully expanded condition. SIGNIFICANCE: The results support the hypothesis that the uptake of HEMA by demineralized dentin depends on the degree of expansion of the dentin matrix.