Sensitivity and Specificity of DIAGNOdent in Occlusal Caries Diagnosis

Objective: The purpose of this study was to evaluate the in-vitro performance of visual and tactile inspection (VT), VT and radiographic examination (R) and VT plus R and laser fluorescence (D) for the detection of occlusal caries. Methods:  50 non-cavitated molar teeth with deep pits and fissures were selected. The teeth were evaluated by three independent examiners visually and with an explorer. The teeth were imaged with a bitewing radiograph (D speed film) and again evaluated with the radiographic information. Following measurements with laser fluorescence (DIAGNOdent™ KaVo, Biberach, Germany), the teeth were evaluated a third time with all data available to the evaluators. The teeth were sectioned buccolingually for a ‘gold standard' histologic evaluation with microscopy to determine the presence or absence of caries that would require surgical intervention for tooth restoration.  Binary evaluation to restore or not restore was made and compared to the gold standard. Results: Sensitivity (SE) and Specificity (SP) values (%) for each group were:

The mean kappa values for inter-examiner agreement were: VT- 0.32 (0.13-0.52), VT+R- 0.32 (0.13-0.52) and VT+R+D- 0.37 (0.34-0.39). A paired t-test revealed no significant difference between kappa agreements (p>0.05).

Conclusion: Under the conditions of this experiment, DIAGNOdent did not improve the sensitivity of caries detection or significantly improve inter-examiner agreement. Supported by the Health Future Foundation.