Diagnostic Accuracy and Reproducibility of Conventional and Novel Caries Detection Methods as Determined by Histology and Micro-CT; A Study In Vitro

Abstract

Background: Dental caries is considered as one of the most common human diseases. In its early stages, it can be treated preventively where the lesion can be arrested or even remineralised. As such, early caries detection and monitoring is required to determine the severity and activity of the lesion over time and to evaluate the effectiveness of the preventive treatment modalities. Traditionally, caries detection is carried out primarily with a visual examination and bitewing radiographs (BW). Other caries detection devices have been developed to provide objective measurements, for example the DIAGNOdent and more recently the DIAGNOdent Pen, and a newer device, the CarieScan PRO, which relies on Electrical Impedance Spectroscopy.

Over the last three decades, there has been a revolution in the medical CT technology in parallel with the invention of compact machines such as cone beam CT which can be used in the dental practice by producing three dimensional images of the teeth and related structures. Micro-CT is considered a novel (3D) digital technology that shows great promise for detecting subtle changes within structures, and has the potential to be an alternative to routine histological sectioning in laboratory studies for validation of caries detection methods without the need for associated destruction of teeth.

Aims and Objectives:
1. To determine the reproducibility of conventional visual (ICDAS) and BW radiographs and novel caries detection methods (Fiberoptic Transillumination (FOTI), Cone BeamCT (CBCT), DIAGNOdent Pen (DD) and CarieScan PRO (CS) for the detection of occlusal and proximal caries in permanent teeth.
2. To evaluate the relationship between histological sections and corresponding MicroCT tomographic slices, for the assessment of the presence and extent of dental caries on
both the occlusal and proximal surfaces in permanent premolar and molar teeth.
3. To evaluate the diagnostic accuracy of the caries detection methods as determined by histological and Micro-CT validation.
4. To devise a quantitative method for volumetric analysis of occlusal and proximal lesions from Micro-CT images and to develop a novel method to determine the percentage
mineral loss in the lesion.

Materials and Methods: A preliminary inspection was carried out to select 200 teeth with a total of 244 occlusal and 400 proximal investigation sites. Eleven examiners were recruited to take part in this study, who were all qualified and experienced dentists. All examinations were carried out with teeth set up in anatomical arches except for CarieScan PRO where teeth were examined individually. Subsequently, 140 teeth underwent MicroCT scanning and then all teeth were serially sectioned in a mesio-distal direction. For qualitative analysis of lesions, a precise and reliable locating technique was developed to accurately link occlusal investigation sites to histological section and corresponding Micro-CT image. For quantitative analysis of lesions, a new method was devised for caries lesion area measurement and volumetric analysis in three dimensions using MicroCT. In addition, a novel method was developed for the measurement of mineral content within the enamel of carious lesions.

Results:
1. The inter- and intra-examiner reproducibility of the DIAGNOdent Pen and ICDAS were superior to the other examination methods for occlusal and proximal caries detection respectively and showed strong to substantial agreement. However, the CarieScan PRO and the DIAGNOdent Pen were inferior to other examination methods for occlusal and
proximal caries detection respectively showing poor agreement between and within examiners.2. Using two currently accepted histological classification systems (Downer and Ekstrand, Ricketts, Kidd (ERK)), there was a strong relationship and agreement between Micro-CT tomographic images and corresponding histological sections.3. In relation to diagnostic accuracy, the ICDAS and CBCT were superior compared with other methods for occlusal and proximal caries detection.
4. The diagnostic accuracy of DIAGNOdent Pen and digital BW radiographs were poor at both the D1 and D3 diagnostic thresholds for occlusal and proximal caries detection.
5. There were highly significant differences in the area, volume and percentage of mineral loss between enamel and dentine lesions on the occlusal surfaces according to Downer
classification system. However, the differences were less significant on the proximal surfaces.

Conclusion: ICDAS has the greatest potential of the detection methods tested for clinicians to use in clinical practice. The use of the DIAGNOdent Pen and CarieScan PRO as an adjunct over and above ICDAS is questioned. CBCT also performed well in relation to diagnostic accuracy, and whilst it should not be used solely for the purpose of caries detection, it is important to report on dental caries if requested for any other clinical reason. For research purposes, Micro-CT would be a valuable method for validation of
new caries detection methods based on quantitative measurements rather than qualitative depth analysis of caries lesions.

Date of Award	2016
Original language	English
Supervisor	David Ricketts (Supervisor) & Andrew Hall (Supervisor)