Surface-specific electrical occlusal caries diagnosis: reproducibility, correlation with histological lesion depth, and tooth type dependence.

Electrical conductance measurements are being used experimentally for occlusal caries detection. Recently, it was suggested to cover the fissure system with a conducting medium, resulting in a surface-specific measurement. It was the aim of this study to determine in vitro the reproducibility of this modified technique for occlusal caries in posterior teeth, to determine for a large study sample the correlation between the electrical measurements and histological lesion depth, and to evaluate the difference between results for premolars and molars. For the reproducibility determination, surface-specific electrical resistance measurements were made using a sample of 68 posterior teeth. Eight operators performed measurements on all teeth, and repeated measurements on 24 teeth. The validity study included the previous sample and the collected samples from two more studies, resulting in a total sample of 325 posterior teeth. One operator had performed electrical resistance measurements on all teeth in the sample. Reproducibility was good: mean Pearson's correlation coefficient 0.89 (+/-0.05) for interexaminer correlation, and 0.86 (+/-0.12) for intra-examiner correlation, using log (resistance) as the result parameter. The correlation coefficient between log(resistance) and histological lesion depth for the large sample was -0.78 for all teeth, -0.64 for premolars, and -0.73 for molars. The regression line for molars was located below the regression line for premolars: at a hypothetical histology score of 2.5 (a dentine caries threshold) the estimated resistance threshold would be 507 kOmega for premolars, and 233 kOmega for molars. Converted to Electronic Caries Monitor (ECM) readings, the difference is about 1.4 on the ECM scale. It was concluded that the in vitro reproducibility of the described surface-specific method for electrical resistance measurement is very good, even for inexperienced operators. The correlation between measurements and histological lesion depth is moderate to good. The method is sensitive to electrode area differences, which will result in different clinical cut-offs for caries diagnosis in premolar and molar teeth.