The clinical performance of chairside caries risk assessment kits.

OBJECTIVE: Caries risk assessment is an important component of clinical practice. The goal of this project was to evaluate the diagnostic performance of two commercially available kits (the Saliva-Check mutans and the Plaque-Check pH kit; GC-America) in distinguishing caries-active from caries-free individuals. The effect of following instructions not to eat, drink, or perform oral hygiene, prior to testing, was also investigated on the effectiveness of these kits. METHOD AND MATERIALS: The subjects included 26 caries-affected children and 20 caries-free adults. Plaque and saliva samples were collected for analysis by the two kits following the manufacturer's instructions. For 63 additional subjects, instructions not to eat, drink, or practice oral hygiene were omitted prior to sample collection. The sensitivity and specificity in distinguishing caries-free from caries-active subjects was calculated for both kits. RESULTS: The sensitivity (88%) and specificity (90%) of the Saliva-Check mutans kits were satisfactory in the subjects refraining from eating etc. In contrast, both the sensitivity and the specificity of this kit were reduced when subjects were tested without use instructions. When tested in subjects that followed instructions, the pH kit's sensitivity and specificity were 72% and 55% respectively. This modest diagnostic performance was reduced when pH values were measured in subjects not following use instructions. CONCLUSION: These kits, particularly the Saliva-Check mutans kits, can contribute to patient education by documenting microbial caries risk factors. Though difficult to implement in practice, subjects should refrain from eating or other activities that disturb the oral environment prior to sample collection.

In vitro study of the diagnostic performance of the Spectra Caries Detection Aid.

OBJECTIVE: The Spectra Caries Detection Aid is a fluorescent camera (FC) caries detector device that has been demonstrated to detect occlusal caries extending into dentin with good sensitivity and specificity. This in vitro study examined the diagnostic performance of this device with the goal of defining the numerical reading that best indicates that a tooth has dentin caries. METHODS: Ninety extracted human third molars, that either appeared clinically intact or had early occlusal caries, were used. Teeth were photographed then analyzed using the Spectra Caries detector. Following Spectra readings, the teeth were sectioned perpendicular to the long axis of the tooth. Cuts were made in such a way as to expose the dentino-enamel junction below the pits and fissures. The sectioned teeth were scored as being caries-free, having enamel caries, or dentin caries. Using each tooth's Spectra reading and histological diagnosis, the sensitivity and specificity for various potential dentin caries cut-off values were calculated. Also, the receiver operator curve (ROC) was plotted and the area under the curve calculated. RESULTS: At the manufacture's recommended cut-off for dentin caries diagnosis of 2, the sensitivity is 0.68 and the specificity is 0.78. At a lower cut-off value in the study reported here of 1.8, it was observed that the sensitivity is higher at 0.87, but that the specificity is lower at 0.7. The overall diagnostic performance of the Spectra is good with an ROC area under the curve of 0.82. CONCLUSION: Although lower cut-off values may have higher sensitivity, the use of the manufacture's established cut-off of 2 will result in fewer instances where dentin caries will be diagnosed incorrectly. The use of higher thresholds will discourage restorative treatment of early lesions that are best managed by sealants or other preventive approaches.