Objective: This in vitro study was aimed to evaluate the performance of DIAGNOdent (Dd) by using the two tips of the device for the detection of carious lesions on approximal surfaces. Methods: The material comprised 42 approximal surfaces with white or brown spot lesions of 38 recently extracted human teeth. Each carious surface was marked on the images obtained by a digital camera to standardize the measured points for Dd method. Dd values were measured with both a cone-shaped tip (A) and a broad tip (B) by two examiners. Measurements were repeated 1 day apart on two occasions. Following these measurements, samples were sectioned on the spots measured by the laser device and Confocal Laser Scanning Microscope (CLSM) (gold standard) was used to determine lesion presence and extent. Proportion of agreement (Accuracy), Cohen's Kappa and Spearman's Rho Statistics were calculated between CLSM and Dd measurements. Results: Accuracy found for the first examiner for both tips was 50% but for the second examiner was 45.3% "for A" tip and 47.6% for "B" tip. The intraobserver agreements were 78.5% and 66.7% for "A" tip and 59.5% and 47.6% for "B" tip for the first and for the second examiner. For the first measurements, the interobserver reproducibility was 69.0% for "A" tip and 61.9% for "B" tip. At the level of enamel and dentin caries, the sensitivity values were 86% for "A" tip and 52% for "B" tip for the first observer, and for the second observer they were 71% and 48%, respectively. The specificity values for the first observer were 24% for "A" tip and 76% for "B" tip and for the second observer 57% for "A" tip and 71% for "B" tip. Conclusion: Accuracy of Dd tips was low in the presence of white and brown spot lesions on proximal surfaces.

Back to the Scientific Program Program
Back to the Joint Meeting of the Continental European, Israeli, and Scandinavian (NOF) Divisions of the IADR (August 25 -- 28, 2004)

Top Level Search