Previous studies have shown that various factors such as ionic composition or pH of the extraction medium may significantly influence leaching of components from restorative materials. Therefore, it was the aim of this investigation to determine the release of fluoride from a resin-modified glass-ionomer cement (GIC) following storage in various extraction media, including an esterase buffer. Specimens of the resin-modified GIC, Fuji II LC, were stored for 144 h in deionized water, acidic buffer (pH 4.2), neutral buffer (pH 7.0), and neutral buffer supplemented with porcine liver esterase. Fluoride release into the various media was measured every 48 h over a 6-day period. In addition, activity of porcine esterase in neutral buffer (artificial saliva) was measured for up to 144 h. The data were statistically evaluated by three-way ANOVA using the Student-Newman-Keuls test (P < 0.05). It was found that esterase activity in neutral artificial saliva decreased during the first 24 h to approximately 40% of the baseline value and then remained constant for up to 6 days. Fluoride release into the various storage media varied significantly (P < 0.05). The highest amounts of fluoride were released into deionized water (30.9 ppm +/- 1.1) and acidic buffer (26.9 ppm +/- 0.7) after 48 h. In addition, significantly more fluoride leached into esterase-containing neutral artificial saliva (6.9 ppm +/- 0.2) than into neutral buffer without enzyme (6.3 ppm +/- 0.2) after 96 h. Our data indicate that fluoride release from the resin-modified GIC investigated may be increased under acidic conditions and by hydrolysis in saliva.