In this paper the in vitro remineralization of human dentine with 0, 0.5, 2 and 10 ppm F in the remineralization solution is presented and analyzed by microradiography. Furthermore, the remineralization data of human and bovine dentine are quantitatively compared. The main results of this paper are: (1) the efficacy of human (and bovine) remineralization is about proportional to the square root of the F level in the rem-ineralizing solution, and (2) the amount of mineral effectively deposited in the dentine is most likely controlled by the diffusion of fluoride into the tissue. This study shows furthermore that although numerical differences exist in the remineralization of human and bovine dentine, the general remineralization behaviour is quite similar. The presence of F^–– in the parts per million range is essential for the remineralization efficacy in vitro and also for the ‘overremineralization’ of the outer dentine surface.

In this paper the acid resistance of hyper-remineralized dentine was quantified by means of fluoride and mineral measurements. Hyper-remineralization was achieved by demineralization of dentine in an acidic gel system (pH 5) for 3 weeks followed by remineralization in a solution containing 1.5 mM Ca, 0.9 mM phosphate and 10 ppm F at pH 7 and 37°C for 8 days. The samples were subsequently again demineralized in the gel system mentioned for 1,2 and 3 weeks. Analysis for fluoride was done by means of the microdrill biopsy technique and to obtain information on the fluoride distribution by secondary ion mass spectrometry (SIMS); mineral was assessed by microradiography. The results showed that in hyper-remineralized dentine the original fluoride content was approximately 30,000 ppm F. This value was still in the same order after the acid challenge of 3 weeks at pH 5. These values were substantially higher than the baseline values in sound or in demineralized dentine being about 900 ppm F. The main microradiographic result was that there was no significant mineral change in the hyper-remineralized dentine due to the acid challenges. There was, however, a tendency for mineral redistribution deeper into the dentine lesion leading to lamination phenomena. The SIMS experiments on some of the samples showed a fluoride distribution in agreement with the microdrill fluoride data. Furthermore a fluoride redistribution took place in deeper parts of the lesion due to the acid challenges. The combined fluoride concentration and microradiographical data indicate that fluoride-enriched and highly mineralized hyper-remineralized dentine is more acid resistant than sound or demineralized dentine.

From intact roots of human cuspids 83 dentine specimens were cut. The specimens with known location and thickness were embedded in a holder of polymethyl methacrylate and microradiographic images were made. From the longitudinal microradiography (LMR) measurements the average mineral density (kg/m³) of the sound human dentine specimens was calculated. All specimens were subsequently demineralized using a constant composition method in a solution containing 3 mM CaCl₂·2H₂O, 3 mM KH₂PO₄, 50 mM CH₃COOH and 0.2 ppm fluoride as NaF at pH = 5 for 24 h. After demineralization the LMR measurements were repeated to calculate the amount of mineral lost. The data show that there is no correlation between: (1) the location of a dentine specimen in the root and the mineral density of sound human dentine, (2) the location of a dentine specimen in the root and the degree of demineralization after 24 h and (3) the mineral density of sound human dentine and the demineralization degree. This information is useful for future in vitro and in vivo studies on human roots.