Korrelation des 'Calcivis® Caries Activity Imaging System' mit visuellen Kariesdetektionssystemen zur Erfassung von aktiven kariösen Läsionen – eine In-vitro-Studie

The Aim: A proper diagnosis of caries includes the decision about the possible activity of the lesion apart from its detection, modification and extent. For assessing the activity of caries first of all visual-tactile methods are applied, with the help of which the probability or rather the tendency of the progression of a carious lesion can be estimated. The study on hand aims at checking the qualification of a new type of a digital process to categorize the activity of a lesion by means of the Calcivis Caries Activity Imaging System. Material and Method: 46 extracted permanent human teeth were available for this study (35 molars, 11 pre-molars). The extent and activity of a lesion at an investigation site on each tooth were classified in accordance with the criteria of visual caries detection methods (NYVAD et al. and ICDAS). All investigation sites were recorded with a prototype of the Calcivis System according to the manufacturer’s specifications. Thus resulting digital images were reviewed with regard to a possible activity (Bioluminescence, blue areas at the investigation sites). Afterwards the teeth were hemisectioned through the investigation site, randomized, and split up into five groups. Different methods for inactivation lesions were used on the tooth halves (fluoride, ICON caries infiltrant, Curodont PROTECT, Curodont REPAIR, control group). After that further images with the Calcivis System were produced and again assessed concerning the activity/inactivity. The gold standard used in the evaluation of activity or inactivity was the color transition of the tooth sections by means of methyl red solution. The kappa statistic and the rank correlation coefficient according to Spearman (rs) were used for the statistical analysis of the data. In order to categorize the diagnostic accuracy of the process, Receiver- Operation-Characteristic-Curves (ROC curves) were constructed and the area under the curves (AUC) were compared with each other. The level of significance was α = 0.05. Results: By the two visual criteria (ICDAS, Nyvad) 41 investigation sites as active lesions and 5 sites as inactive caries were categorized. The kappa value of the agreement between Nyvad and ICDAS criteria with regard to the activity of the lesion was at 0.78. The kappa values for the agreement of the visual findings to Calcivis were: Nyvad/Calcivis = 0.78 and ICDAS/Calcivis = 1.0. The correlation of the visual methods and the bioluminescence method was significantly positive (p < 0.001, two-sided), where rs (Nyvad/Calcivis) = 0.776 and rs (ICDAS/Calcivis) = 1.0. The diagnostic accuracy (AUC) of Calcivis was at 0.89 (NYVAD criteria as reference value) and at 1.0 (ICDAS criteria as reference value). The difference was not statistically significant (p > 0.05). The correlation of the Calcivis measurements to histology for the treated tooth halves was rs = 0.47 (p = 0.002), and for the untreated tooth halves rs = -0.01 (p > 0.05). Discussion and Conclusion: The use of a Calcivis prototype provids good results with regard to activity assessment of a carious lesion in the field of the coronal caries in-vitro. Referring to the assessment of caries activity the process is proven to be appropriate even if the first detection of caries should performed visually, as the camera of the Calcivis system seems to be less suitable. The classification of the caries activity is especially advantageous with initial lesions so that teeth need not urgently be treated invasively, remineralizing or rather minimal-invasive methods can be applied instead. The study on hand may to a large extent, but not yet finally clarify if and in how far the success of such processes also by means of the bioluminescence method like Calcivis is feasible; further studies will have to verify this.