RESUMO
The present study investigated whether variability in fluoride release was reduced by use of a capsulated system as compared with a hand-mixed system based on the same qualitative and quantitative glass-ionomer formulation. Five operators independently prepared five disks of each type of glass-ionomer restorative filling material. The amount of fluoride released in water by each specimen after 1, 2, 3, 7, and 14 days was determined. The variance and the mean value of the fluoride release at a given time were independent of the operator, regardless of the glass-ionomer system considered. However, the variance and the mean value of the fluoride release at a given time were considerably greater for the capsulated system than for the hand-mixed system. A regression analysis further showed that the cumulative amount of fluoride released, [F], as a function of the time t conformed to [F] = a[1 - exp (-bt)] + ct0.5 for each glass-ionomer specimen, suggesting the simultaneous occurrence of two processes. Whereas the parameter b was the same for the hand-mixed and capsulated system and was independent of both a and c, the latter parameters were positively correlated. Moreover, the values for a and c were significantly greater for the capsulated system. On the basis of these results, it is concluded that the mixing process drastically influences the short-term as well as the long-term fluoride release.