A pilot study of a modified radiographic technique for detecting early proximal cavities.

Diagnosing the necessity of cavity preparation in demineralized proximal areas has been considered as a challenge in restorative treatment planning. The purpose of this study was to investigate the clinical performance of a modified radiographic technique for the detection of proximal cavities. The sample consisted of 44 proximal surfaces in 38 dental students. The patients had radiolucent proximal lesions restricted to the inner half of enamel or the outer third of dentine in bitewing radiographs, and there was doubt regarding the presence or absence of cavities in the approximal areas. The suspected surfaces were then examined by secondary bitewing radiographs which were taken after pressing radiopaque material into the proximal areas. Finally, orthodontic elastic separators were placed in the contact areas to provide enough space for direct visual and tactile examination, thus detecting any proximal cavity (reference standard). The sensitivity, specificity and accuracy of the modified bitewing radiography were calculated against the reference standard. Overall, 7 surfaces presented caries cavities according to the reference standard. All proximal radiolucencies observed in the inner half of enamel and 46% of those extended to the outer third of dentine were not cavitated when evaluated by direct visual and tactile examination. The sensitivity, specificity and accuracy of bitewing radiography with opaque material for detecting proximal cavities (n = 7) were 86, 100 and 98%, respectively. The tested radiographic technique displayed good validity in this pilot study for detecting proximal cavities in posterior teeth and should be further investigated.