In vitro and in vivo feasibility study of ultrasound for monitoring tooth surface loss

This work attempted to appraise the usefulness of ultrasound as a diagnostic tool for use in dentistry. A number of possibilities were modelled and the most promising, at this time, was monitoring tooth surface loss (TSL). TSL is a serious dental condition affecting patients worldwide. Current methods used to monitor TSL in the dental surgery are subjective and unreliable. Laboratory-based monitoring methods are time consuming and costly. Ultrasound is a non-invasive, non-destructive method that is mainly used in the medical field. Its use in dentistry started in the 1950s but is still limited to therapeutics (e.g. periodontics and endodontics) and head and neck imaging. There are several modes of ultrasound imaging. Of particular note are amplitude mode (A-mode) and brightness mode (B-mode) ultrasound. In A-mode, a single beam is sent to an object, and its reflected echo is captured. This mode does not produce an image but is rather displayed as a waveform in time and amplitude domains. B-mode ultrasound has two spatial axes and therefore is used as a cross-sectional imaging tool.
To date, there are no in vivo studies investigating the use of ultrasound to directly monitor TSL in the dental surgery. The aim of this thesis was to assess the feasibility and optimisation of ultrasound as a potential clinical dental tool in monitoring erosive TSL in vivo. This thesis investigated the coupling efficiency of various dental and other materials and their suitability as couplants. The results showed that Perspex was a suitable ultrasonic couplant for the purpose of enamel thickness measurements and the tightness of the coupling at an interface was of importance for efficient transmission of ultrasound energy into an object. However, a purpose-built apparatus was required for this as the ultrasound echoes were angle dependent. It further investigates the angle dependency of echoes arising from premolars compared to synthetic maxillary central incisors, as natural incisors were not available. The results demonstrated that the more planar incisors reflected ultrasound more readily and were less angle dependent than premolars (p-value < 0.001).
B-mode ultrasound imaging was then investigated to measure intact enamel in human teeth and validated with μ-computed tomography (μ-CT). Two systems were evaluated for this purpose; the first was an in-house ultrasound apparatus and the second was a commercial ultrasound scanner, with the data obtained validated with μ-CT. The results showed that the commercial ultrasound scanner was more accurate than the in-house scanner with Bland-Altman 95% limits of agreement of -0.48 to 0.47 mm and -1.21 to 0.87 mm respectively. However, the B-mode images produced were not of sufficient clarity and consequently the accuracy of the enamel thickness measurements was not suitable for monitoring progressive enamel loss. Therefore, the simpler A-mode ultrasound approach was investigated for enamel thickness measurements and validated with histological sections of the same teeth. A study of speed of sound (SOS) variations in enamel was also performed. It was found that A-mode ultrasound was able to measure enamel thickness in vitro with an accuracy of 10% compared to histology and the mean SOS in enamel was 6191 ±199 ms-1.
Finally, A-mode ultrasound was assessed in vivo (n = 30) to determine if it could monitor enamel thickness reliably and reproducibly on the labial surface of maxillary central incisors. The results showed that ultrasound was a highly reproducible and reliable technique for monitoring enamel thickness with 95% limits of agreement of -0.04 to 0.05 mm. The results demonstrated for the first time in vivo that A-mode ultrasound had sufficient precision (0.05 mm) to allow it to be used as a direct method for serial assessment of erosive TSL. The preferable site for making ultrasonic measurements was the cervical site (site 1) followed by the mid-buccal site (site 2). Therefore ultrasound is a promising and simple method to monitor early erosive changes in thickness of the enamel layer, especially in vulnerable patients with frequent acidic intake or in patients with gastro-oesophageal reflux disease (GORD).