Grey discolouration of the enamel around a filling and marginal fracture are often reasons to replace restorations due to suspicion of secondary caries. The aim of this study was to establish the validity of grey discolouration and marginal fracture for the diagnosis of caries at the enamel-dentine junction (EDJ) next to an amalgam filling. The occlusal surfaces of 161 extracted molars with occlusal amalgam fillings were photographed on colour slides, from which grey discolourations and marginal fracture were recorded. The size of the fracture was scored using a modified Mahler scale with six categories: score 1 = no marginal fracture ( < 30 μm), score 6 = fracture width > 200 μm. Secondary caries at the EDJ was scored on radiographs of 700 μm sections of the molars. Both radiolucencies and radiopacities in dentine were scored as caries. The sensitivity of grey discolouration for detection of secondary caries was 50% and the specificity 91%. The positive and negative predictive values were 71 and 80%, respectively. Marginal fracture was evaluated at five different levels of severity. Although there was caries present more frequently at the EDJ under severely fractured margins, marginal fracture was found of little diagnostic value at all levels. Only marginal fracture at threshold level V provided a positive predictive value (50%), which was significantly higher than the caries prevalence (31%). It is concluded that width of marginal fracture, as defined by the Mahler scale, has hardly any value for the diagnosis of secondary caries. However, grey discolouration may be a useful diagnostic aid.

This investigation was carried out to establish the validity of bite-wing radiographs for the diagnosis of secondary caries in teeth with occlusal amalgam restorations. One hundred and fifty-nine extracted molars with occlusal amalgam fillings and characteristics that might be indicative of the presence of secondary caries were selected. The characteristics were a blueish-gray discolouration of enamel, a brownish discolouration of the amalgam-enamel margin, marginal breakdown of the restoration, and/or deep fissures at the margin of the restoration. The teeth were radiographed in a bucco-lingual direction to obtain images comparable to bite-wings. Subsequently, the crowns were sectioned (700 μm) and glued on plastic sheets. The sections were recorded radiographically and then examined by independent examiners. Radiolucencies as well as radiopacities adjacent to the restoration were considered to be caries. The validation of the bite-wing radiographic diagnosis was achieved by comparison with the radiographs of the sections. A sensitivity of 64% and a specificity of 98% were calculated. Large lesions were always visible on the bite-wing radiograph (100%), the majority of medium-sized lesions (89%) and 40% of small lesions. Small radiopacities were detected more often than small radiolucencies.

Radiopacities, caused by tin or zinc deposits in partly demineralized dental tissue, are frequently seen under amalgam restorations. The aim of this study was to determine to what extent these radiopaque areas could be identified by Caries Detector (1% acid red in propylene glycol) which is claimed to stain the irreversibly demineralized dentine. Twenty-eight extracted teeth showing radiopacities under amalgam fillings were selected. The restorations were removed, and Caries Detector was applied. Caries was excavated until the dentine did no longer stain with the Caries Detector. Standardized radiographs were taken at different stages. In all teeth the radiopaque areas stained with the Caries Detector. Visual inspection of the radiographs, taken after excavation, revealed that the radiopacities had disappeared completely in 6 teeth; in 5 teeth a very small part of the radiopaque area remained; in 17 teeth the cavity floor appeared as a thin white line on X-ray. Overall, line scan analysis confirmed the data obtained by visual observation. The residual radiopacities and radiopaque lines were a very small fraction of the initial radiopacities. Therefore, it is concluded that the radiopaque zone under amalgam fillings represents almost entirely an area of irreversibly demineralized dentine as indicated by the Caries Detector.

In the present in vitro study, the validity was determined of bite-wing radiographs for the diagnosis of secondary caries in approximal parts of teeth with class II amalgam restorations in vitro. The study was performed on 47 molars and 49 premolars providing 68 and 69 filled approximal surfaces, respectively. The teeth were radiographed in a buccolingual direction to obtain images comparable to bite-wings. Radiolucencies and radio-opacities were scored as caries. Subsequently the teeth were sectioned (700 μm) in transverse or sagittal direction and then the sections were radiographed. Validation was performed against these radiographs of sections, on which also radiolucencies and radio-opacities were scored. Statistical analysis of the results revealed significant differences between the validity parameters for molars and premolars, being for sensitivity 73 and 80%, respectively, for specificity 90 and 95%, for positive predictive value 95 and 92%, and for negative predictive value 58 and 86%. For molars, the results did not differ significantly from the results of our previous study on caries diagnosis in teeth with occlusal fillings. The present experiment showed that bite-wing radiographs are useful for the diagnosis of secondary caries under approximal parts of class II amalgam restorations, just as for secondary caries diagnosis adjacent to class I restorations.

Grey discolouration around occlusal amalgam restorations has been found to be useful for the diagnosis of secondary caries. The aim of the present in vitro study was to establish the validity of grey discolouration of buccal and lingual enamel next to the approximal amalgam of class II restorations for the diagnosis of secondary caries. One hundred buccal and lingual areas next to the approximal parts of a class II filling in molars and premolars were included in the study. Fifty-one areas showed grey discolouration. Validation was performed against radiographs of transverse sections. Both radiolucencies and radiopacities were scored as being caries. The results were: sensitivity 92%, specificity 55%, positive predictive value 22% and negative predictive value 98%. It was concluded that grey discolouration of the approximal walls next to class II amalgam fillings is not useful for the detection of secondary caries, but the absence of grey discolouration gives a good prediction for the absence of secondary caries. This is in contrast to grey discolouration around occlusal amalgam restorations, which was considered to be useful for the diagnosis of secondary caries.