Enamel specimens in this in-vitro-study were prepared from bovine incisors, which have been used and discussed in multiple studies investigating erosive/abrasive wear of dental hard tissues and were shown to be a suitable substitute for human enamel [24, 27, 28]. Tooth collection was carried out in accordance with relevant guidelines and regulations. Additional approval by the associated Swiss Ethics Committee was not required.
Surface wear was measured using profilometric surface analysis, which is proven to be a reliable and accurate method in order to quantify erosive/abrasive enamel wear and has been used and discussed in numerous other studies investigating this issue [14, 29]. The reflux simulating cycling model and all applied parameters aimed to simulate realistic clinical conditions. Thus, a HCl perfusion was used to imitate regurgitation of gastric fluid reaching the oral cavity. The quantity (11 times a day during 12 hours) and duration (30 s) of the acid attacks, as well as flow rate of 2 ml/min and pH-value 3.0 also aimed to simulate realistic conditions.
Between the erosive attacks, all specimens were perfused with artificial saliva. While immediately after each attack, the flow rate was enhanced to simulate increased salivary flow during acid exposure in-vivo  and to wash away the acid and stop the erosive process, the regular flow rate was 0.5 ml/min which corresponds with normal unstimulated salivation .
Following recommendations made for patients suffering from erosion, daily toothbrushing was performed before erosion and one hour after the last erosive attack . The quantity
(2 times/day), as well as the amount (n = 15), frequency (1 stroke/second) and application force (2 N) of the performed brushing strokes were based on recommendations by Wiegand and Attin . The required overall duration of the applied cycling model (20 days) to see a sound effect was verified in a recent study with similar set up .
However, it has to be considered that the occurrence of gastroesophageal reflux in-vivo is not limited to a defined number and time of a day and may also happen while sleeping at night . As no reflux was simulated during 12 h (overnight) in this study, this might be a limitation of the present cycling model. Furthermore, gastric fluid is mainly, but not only composed of HCl and may contain various enzymes such as the proteolytic enzyme pepsin . Other than in a previous study , the acid was not enriched with pepsin in this study, as the amount of organic matrix in enamel is much lower compared to dentine and no influence of pepsin admixture on the erosive/abrasive tooth wear was observed even for dentin . Other modifying factors such as pellicle formation, bacteria and fluoride in saliva and plaque fluid were also not regarded in this study. Additionally, it has to be considered that the storage of specimens in artificial saliva between the brushing periods does not adequately imitate in-vivo-mineralising processes .
In general, saliva might enhance the abrasive wear resistance and support remineralisation through calcium und phosphate precipitation and thus lead to a stabilisation of eroded enamel . However, saliva induced remineralisation must be regarded as a slow process with mineral gain mainly taking place in the surface layer of the lesion . This might be the reason, why only a minor remineralising effect of previously eroded enamel  and still increased susceptibility to abrasion of previously eroded enamel after a remineralisation period of one hour  is described in literature. Another study reports partial re-hardening of softened enamel surface within two hours of salivary exposure but no significant further remineralisation after 12 hours . Thus, it is questionable if the specimen storage in artificial saliva between the erosive attacks and during 12 hours (overnight) in this study might have led to a notable remineralisation.
After 20 days of daily application, all investigated amine/sodium and sodium fluoride gels (12.500 ppm) were able to significantly reduce gastroesophageal reflux induced loss of enamel. This finding is in agreement with the results of other studies investigating the potential of different fluoride gels to protect from DE [44, 45]. The amine/sodium fluoride gel Elmex Gelée (G2) showed the overall best protection, though the amine/sodium fluoride gel in G3 and sodium fluoride gel in G4 revealed almost equally good results. Still, enamel wear in G5 (sodium fluoride gel, Sensodyne ProSchmelz Fluoride Gelée) was significantly higher compared to the amine/sodium fluoride gel in G2 and the sodium fluoride gel in G4. The reasons for these differences remain unclear. Nevertheless, it might rather not be attributed to the compound of fluoride as the same kind of fluoride (sodium fluoride) was applied in G4. Anyway, the potential to protect from gastroesophageal reflux induced DE in G5 still was significantly higher compared to the untreated control group and is in conformance with other studies describing a protective effect of sodium fluoride [46, 47]. The concentration of fluoride and pH-value may also have an influence. It is evident, that a low pH and high fluoride concentration support the uptake of fluoride into dental hard tissues and the formation of a CaF2-like layer . All tested fluoride gels in this study had the same high fluoride concentration of 12.500 ppm which has been proven to be effective in reducing erosive toothwear . Regarding the differing pH-values of the gels, it has to be taken into consideration that enamel specimens in this study were eroded before the fluoride gels were applied. Erosion causes an enlargement of the enamel surface which enables higher fluoride uptake , thus making the pH of the applied fluorides less important.
Overall, the effects of fluoride on the oral cavity and the entire human organism are well known and investigated . Therefore, the use of a 12.500 ppm fluoride gel once a day can be recommended in order to reduce GERD-induced dental erosion.