Evaluation of Digital Imaging Fiber Optic Trans-Illumination (DIFOTI) and FOTI

Objectives: To determine and compare the ability of DIFOTI and FOTI to detect primary caries. Methods: The study was initiated with 119 children 8-12 years of age. Deciduous molars were examined at 6-month intervals throughout a 2-year study period. Except for sites with cavitations, restorations and sealants, the following were examined using DIFOTI and FOTI at predetermined sites: three or four sites from each buccal and lingual surface, and the mesial, central and distal fossae of occlusal surfaces. Presence or absence of caries was recorded. One hundred-eighty exfoliated teeth were collected, sectioned through predetermined sites, and examined for the presence or absence of caries using polarized light microscopy (PLM). A total of 1008 sites were observed. Results: PLM found 791 sound and 217 carious sites (126 on smooth surfaces, and 91 on occlusal surfaces). With respect to all surfaces for DIFOTI, sensitivity was 27.4% [95% confidence interval (CI) 18% to 37%] and specificity was 93.8% (CI: 92% to 96%) and these values for FOTI were 27.4% (CI: 18 to 37%) and 95.9% (CI: 94 to 98%), respectively. The sensitivity and specificity on smooth surfaces were 12.1% (CI: 2 to 23%) and 98.3% (CI: 97 to 100%) for DIFOTI, 17.5% (CI: 5 to 30%) and 97.1% (CI: 95 to 100%) for FOTI, respectively; and these values on occlusal surfaces were 41.4% (CI: 29 to 54%) and 84.8% (CI: 80 to 90%) for DIFOTI and 36.6% (CI: 24 to 49%) and 94.0% (CI: 91 to 97%) for FOTI, respectively. DIFOTI and FOTI showed moderate agreement (Kappa: 0.5). Conclusion: Sensitivity of DIFOTI and FOTI techniques was low and specificity was high. The caries detection ability on occlusal surfaces for both techniques appears to be better than on smooth surfaces. This study funded by NIH/NIDCR PO1 DE13540.