Objective: Bitewing radiographs are the main tool for approximal caries detection. However, interpretation of findings is affected by numerous factors, such as beam angulation, exposure conditions, morphologic variations, etc. The objective of the present study was to compare the ability to detect virtually identical, artificial lesions in different tooth surfaces.
Methods: The in vitro model included preparation and radiologic evaluation of approximal artificial lesions in premolars and molars in depths ranging from 0.00 mm to 1.50 mm.
Results: Significant differences were found in the prospect of a lesion detection between the different lesion depths (p < 0.0000) and between the different surfaces examined (p < 0.05). The interaction between lesion depth and tooth surface was significant at the 0.0001 level (ANOVA with repeated measures).
Conclusions: The results indicate that initial approximal lesions of uniform size are more readily detected in premolars than in molars and that the chance of an incipient lesion to remain undetected is twice as high in molars than in premolars.