Zur Sensitivität und Spezifität der digitalen Fluoreszenzkamera VistaProof für die Detektion der okklusalen Karies

Aim: The accurate and timely diagnosis of dental caries is an everyday challenge for dental practitioners. Due to caries decline, there is an increasing need for more subtle diagnostic methods able to e.g. detect initial lesions. In addition to visual inspection, several methods including radiography, laser or light fluorescence based techniques and electrical impedance measurements are available. The aim of this study was to assess the sensitivity and specificity of the fluorescence based camera VistaProof in the detection of occlusal caries. Histological evaluation after serial sectioning was used as the gold standard. Material & Methods: 53 freshly extracted human teeth (47 molars, 6 pre-molars) were available for this study. Prior to extraction, informed consent had been obtained from the patients about the use of the teeth for research purposes. After the teeth had been cleaned, the occlusal surfaces were photographed and 1-3 investigation sites per surface were marked in the digital images (n=99). All investigation sites were visually inspected according to the ICDAS-II-criteria by two examiners. Afterwards, both examiners independently examined the teeth using the fluorescence based camera system VistaProof. For this purpose, digital images were taken and analyzed by the appropriate software. The teeth were then serially sectioned in a vestibulo-oral direction for the histological examination, resulting in 9-12 slices per tooth with a thickness of 300µm (± 30 µm) each. Lesion depth was assessed by both examiners using microscopy. Statistics: The statistical evaluation was carried out with SPSS, version 14.0. Correlation between the fluorescence measurements and the histological results was determined by means of Spearman’s rank correlation coefficient. Sensitivity and specificity were assessed at D1- (enamel and dentin caries) and D3-level (dentin caries), and receiver operating characteristic (ROC) curves were generated. Results: Spearman’s rank correlation coefficient was moderate for both examiners (rs = 0.47 and 0.55) and there was significant positive correlation (p < 0.01). The cutoff-limits were defined according to the highest sum of sensitivity and specificity at each threshold, resulting in the 1.2/1.3 cutoff for D1-level and the 1.3/1.4 cutoff for D3-level. At the D1 diagnostic threshold, sensitivity was 0.71 (Examiner A) and 0.77 (Examiner B) while specificity was 0.76 (A) and 0.68 (B). At the D3 diagnostic threshold, sensitivity values were 0.87 (A) and 0.91 (B). Specificity was 0.67 (A) and 0.56 (B). The areas under the receiver operating characteristic curves were 0.77 (A) and 0.75 (B) at D1-level, while they were 0.81 (A) and 0.77 (B) at D3-level. Conclusion: Between the fluorescence based and the histological examinations, moderate correlation could be observed. Concerning sensitivity and specificity, the diagnostic performance can be improved by defining the cutoff-limits according to the highest sum of sensitivity and specificity at D1- and D3-level. The fluorescence based camera system VistaProof appears to be a valuable tool for the objective caries detection and for monitoring purposes as well as motivating patients and communicating with them. The system is equally applicable by experienced examiners and in student education. In order to further investigate and, if necessary, redefine the cutoff-limits provided by the manufacturer, in-vivo studies should be conducted. For follow-up studies it would be useful to involve more examiners and investigation sites to acquire more data.