Discs of 300-400 渭m thickness including enamel with cross-cut or long-cut prisms and mid-coronal dentine with cross-cut, long-cut or oblique-cut tubules were obtained from 28 non-carious freshly extracted human third molars sectioned at different directions. Each disc was placed over a metal plate and B-scan images were obtained by a swept-source OCT at 1330 nm centre wavelength (Dental SS-OCT Prototype 1, Panasonic Healthcare Co., Ltd, Japan). Refractive index (n) calculated from optical path length and real thickness, and OCT signal slope (OCTSS) in relation to structural orientation were statistically compared among groups (伪 = 0.05). The discs were sectioned and observed under SEM to confirm prism or tubule orientations.
No significant differences were found in n and OCTSS among cross-cut and long-cut enamel groups (p > 0.05). However, in dentine, there was a significant difference in both variables among different regions. In cross-cut tubule orientation, the average n and OCTSS values were significantly lower compared to other two groups (long-cut and oblique-cut) (p < 0.05). There was a significant difference in both variables between enamel and dentine (p < 0.05).
Unlike enamel, refractive index and OCT signal patterns in dentine vary according to structural orientation, with dentine tubules playing the role. Attenuation of OCT signal intensity was small in enamel. The findings may contribute to a better understanding of the interactions of light with the dental tissue.
Precise records of refractive indices and OCT signal patterns may be important for clinical diagnosis of caries and measurement of structural depth for operative purposes using this technology. Effects of dentine structural orientation on refractive index and scattering pattern must be considered when observing human teeth by OCT cross-sectional imaging.