One hundred extracted molar teeth with discoloured fissures but without any visible carious cavitation were selected from a large supply, using standardized criteria. The teeth were mounted, placed on a mechanical balance, and probed with a force of 500 g in every fissure, at as many places as possible. Every time the probe was found to stick, the spot was marked. After probing colour slides were made of the occlusal faces. Subsequently, the crowns were embedded in epoxy resin. 700 μm thick sections were cut in a facial-lingual direction with a diamond wheel. From the sections X-rays were taken which were scored as follows: a measuring grid was placed on the X-ray image of a section, and the caries score (0-4) for every millimetre was determined. By scoring every section of a tooth in this way, an overview was obtained of the location of all caries lesions in the occlusal surface. By comparing this overview with the colour slide of the tooth, the relationship between the sticky spots and the lesions was visualized. The results indicate that only 24% of the caries lesions were discovered by probing for stickiness (low sensitivity), but that the probe seldom stuck in a sound fissure (high specificity, > 99%). Probing proved to be unreliable for the diagnosis of fissure caries.

No abstract