In 12 healthy volunteers the pharmacokinetic parameters of two new sustained-release formulations of disodiummonofluorophosphate (MFP) (B and C) were compared with those of a nonsustained-release reference preparation (A). This randomized study had a single-dose, triple-cross over design and consisted of 3 trial days separated by a 1-week washout period. Serial blood samples were obtained over a period of 24 hours and 24-hour urine was collected. Serum and urine fluoride concentrations were determined using an ion-sensitive electrode (Orion Research). The results of this study showed a significant reduction of the area under the serum concentration versus time curve (AUC) for the sustained-release formulations (AUC B: 1487 +/- 354 ng/ml x hour, AUC C: 1369 +/- 384 ng/ml x hour) compared with the reference preparation (AUC A: 2374 +/- 652 ng/ml x hour) (B/A: 63%, C/A: 58%) (P < 0.001). Furthermore, the peak serum concentrations of fluoride (Cmax) for B and C (CmaxB: 166 +/- 42 ng/ml, CmaxC: 110 +/- 48 ng/ml) were significantly lower than for A (CmaxA: 380 +/- 77 ng/ml) (P < 0.001). The 24-hour urine fluoride recovery rates were 5.6 +/- 0.7 mg fluoride for A, 3.6 +/- 0.8 mg for B, and 3.2 +/- 1.1 mg for C and corresponded well to the relative fluoride bioavailability, as concluded from the serum fluoride concentration. In conclusion, the sustained-release preparations of MFP led to a decrease of fluoride bioavailability and avoided high peak serum concentrations.