Digital subtraction radiography after stannous fluoride treatment for occlusal caries diagnosis.

The material in this study consisted of 38 fully erupted, extracted third molars without clinical cavitation in the occlusal surface. A radiograph was made of each tooth before and after 5, 10, and 20 minutes of stannous fluoride treatment. The radiographs were digitized and subtraction performed between the images obtained after stannous fluoride treatment and the pretreatment image. Two observers assessed the stannous fluoride treated radiographic and the subtraction images on a monitor: 0 = no change, 1 = intensity increase (white area interpreted as a carious lesion) in dentinoenamel area. Caries was assessed on conventional film radiographs made before treatment: 0 = no caries in dentin, 1 = caries in dentin. The presence of caries in dentin was validated histologically. Sensitivity for intensity increase as a sign of caries was overall higher for the subtraction images based on 20-minute treatment than for the radiographic images (0.025 > p > 0.01) but not significantly higher than for the conventional radiographs. However, neither observer gave false-positive scorings in the subtraction images, whereas observer 1 had five false-positive scorings on the conventional films. Observer 2 had none. The subtraction method did not provide a higher sensitivity for dentinal occlusal caries than conventional film radiography, but the intensity increase could be trusted more than the traditional radiolucency as a sign of a dentinal lesion.