Diagnostic performance of a new red light LED device for approximal caries detection

Abstract

The aim of this study was to test a newly developed LED-based fluorescence device for approximal caries detection in vitro. We assembled 120 extracted molars without frank cavitations or fillings pairwise in order to create contact areas. The teeth were independently assessed by two examiners using visual caries detection (International Caries Detection and Assessment System, ICDAS), bitewing radiography (BW), laser fluorescence (LFpen), and LED fluorescence (Midwest Caries I.D., MW). The measurements were repeated at least 1 week later. The diagnostic performance was calculated with Bayesian analyses. Post-test probabilities were calculated in order to judge the diagnostic performance of combined methods. Reliability analyses were performed using kappa statistics for nominal data and intraclass correlation (ICC) for absolute data. Histology served as the gold standard. Sensitivities/specificities at the enamel threshold were 0.33/0.84 for ICDAS, 0.23/0.86 for BW, 0.47/0.78 for LFpen, and 0.32/0.87 for MW. Sensitivities/specificities at the dentine threshold were 0.04/0.89 for ICDAS, 0.27/0.94 for BW, 0.39/0.84 for LFpen, and 0.07/0.96 for MW. Reliability data were fair to moderate for MW and good for BW and LFpen. The combination of ICDAS and radiography yielded the best diagnostic performance (post-test probability of 0.73 at the dentine threshold). The newly developed LED device is not able to be recommended for approximal caries detection. There might be too much signal loss during signal transduction from the occlusal aspect to the proximal lesion site and the reverse.

Effect of Different Contact Materials on Approximal Caries Detection by Laser Fluorescence and Light-Emitting Diode Devices

“…For approximal caries, the specificity and sensitivity were 0.87 and 0.32, respectively at the enamel threshold and 0.89 and 0.39, respectively at the dentine threshold. 6 In the present study, at the T1 threshold, the specificity of the LED device tended to increase in all conditions, but this increase was significant only for composite restorations. At the T2 thresholds, the specificity of the LED device significantly increased only in the teeth that had contact with zirconia.…”

“…2,5 Approximal lesions are usually detected with a visual tactile inspection combined with radiography. 6 Visual inspection is noninvasive and highly specific, but has low sensitivity and reproducibility. 2,[7][8][9] Another method is separating the teeth temporarily with elastomeric separators, which allows for direct inspection of proximal sites.…”

Alternative methods to visual and radiographic examinations for approximal caries detection

“…The presence of a carious lesion will lead to changes in these properties. Another fiber optic cable serves as a photodetector that transmits the captured light to a microprocessor, which compares the signals to defined parameters (73). A caries detection method based on this principle was introduced as Midwest Caries I.D.…”

“…One in vitro study (73) evaluated the performance of LED fluorescence, LF pen, radiography, and visual examination for approximal enamel and dentin caries detection. Visual examination and radiography performed better in terms of Sn and Sp for both enamel and dentin lesions.…”

Diagnostic agreement between radiologist, dentist and dental students for radiographic detection of approximal caries.

“…These values fall within expected ranges considering the high diagnostic variability reported for the clinical evaluations of caries. Recent studies show interobserver agreement using unweighted Kappa values that range from 0.31 to 0.61 in the clinical diagnosis using the criteria ICDAS 17 , from 0.63 to 0.81 using different types of radiographs 18 , and from 0.32 to 0.87 using laser fluorescence in approximal lesions 19 .…”