Stereomicroscopy, film radiography, microradiography and naked-eye inspection of tooth sections as validation for occlusal caries diagnosis.

In establishing a trustworthy validation for occlusal caries lesions, various methods have been used after sectioning teeth. The aim of the present study was to compare the accuracy of four such methods: stereomicroscopy (SM), film radiography (FR), microradiography (MR) and naked-eye inspection (NEI) for the detection of caries in occlusal tooth surfaces. Further, the interobserver agreement between three observers' registrations of occlusal caries by these methods was evaluated. The material consisted of 18 unerupted third molars known to be sound with respect to caries, as they were embedded in bone prior to removal, and 20 erupted third molars with an unknown 'true state of disease'. The teeth were serially sectioned, coded and examined blindly for occlusal caries by three independent observers by the four validation methods under study. SM was demonstrated to be the only method by which all observers correctly identified all 18 unerupted teeth as sound, resulting in a specificity of 1.00. By the other three methods, 1 or more of the unerupted teeth were falsely classified as carious. In the erupted teeth, 73, 58, 52 and 32%, respectively, of the teeth were determined carious by SM, MR, NEI and FR. As it was possible by SM to detect caries in nearly three quarters of the erupted teeth and at the same time to identify all unerupted teeth as sound, this method seemed to be the most trustworthy of the validation methods under study.