Influence of examiners' experience on the reproducibility of different gold standard techniques and histological criteria for validation of the diagnosis of occlusal caries lesions.

Abstract

The aims of this study were to evaluate the influence (1) of the examiner experience and (2) three histological classification criteria on the reproducibility of two gold standard techniques (non-dye or dye) for validation of the diagnosis of occlusal caries. This study comprised a sample of 210 digital images of 105 permanent teeth (105 rhodamine B dyed and 105 dye-free hemisections) and six examiners. Images were evaluated on a laptop computer and categorised according to three different histological classification criteria (proposed by Ekstrand, Lussi or Downer) and repeated in order to allow reproducibility calculation. For data analysis, the six participants were divided into two groups: G1: examiners with previous experience in histological evaluation and G2: examiners with no experience in histological evaluation. Statistical significance was set at 0.05. Results: The mean intra-examiner reproducibility values in G1 were higher than G2 in all variables. Intra-examiner reproducibility was lower for the dye-free technique in both groups. The mean values of inter-examiner reproducibility in G1 ranged from 0.60 to 0.68, and in G2 values ranged from 0.34 to 0.69. Conclusion: It can be concluded that: (1) the examiners' experience seems to influence the reproducibility of the two gold standard techniques studied and reproducibility tends to be lower when the dye-free technique is used, and (2) the histological classification criterion does not appear to influence the reproducibility for validation of the diagnosis of occlusal caries.